Molecular and phenotypic characterization of diarrhoeagenic Escherichia coli from Nicaraguan children

Abstract

Diarrhoeal diseases continue to be a foremost public health problem worldwide, with over 1.5 million deaths occurring each year, mostly in children in developing countries. In Nicaragua, the mortality due to diarrhoea in children less than 5 years of age is 11 per 10,000 inhabitants. Additionally, this group of children account for more than 120,000 thousands consultations due to diarrhoea per year, most of them occurring during the winter season. However, in Nicaragua, information about the prevalence of many of the enteric pathogens such as diarrheoagenic Escherichia coli (DEC), is still lacking. Thus, the studies presented in this thesis focused on the identification, molecular and phenotypic characteristics of diarrhoeagenic E. coli pathotypes as enteric pathogens associated with diarrhoea in children from León, Nicaragua.

The results from studies I to III are: Five DEC types have been identified in 526 faecal samples from children less than 5 years of age, using PCR based methods. Additionally, the majority of the E. coli positives for a DEC type occurred alongside with the winter season in Nicaragua. The ETEC pathotype was identified with significantly higher prevalence in children with 78/381 (20.5%) compared to children without 12/145 (8.3%) diarrhoea. The distribution of the identified ETEC was as follow: eltB 72/90 (80%), followed by eltB-estA 14/90 (15.6%) and finally estA 4/90 (4.4%) positives. Using the biochemical fingerprinting method PhP-RE, ETEC positive samples seem to belong to defined biochemical phenotypes (BPTs). In addition, facts that a limited number of BPTs grouped estA positives, and estA ETEC samples were only isolated from children with diarrhoea suggest the spread of a marked pathogenic clonal group. On the other hand, atypical EPEC (only eaeA+), was identified with similar frequencies in children with 61/381 (16%) and children without 30/145 (20.7%) diarrhoea.

Additionally, clonal group analysis of atypical EPEC isolates showed no differences between children with and without diarrhoea. EIEC was identified at low, but similar frequency in both children with 3/381 (0.8%) and without 2/145 (1.4%) diarrhoea. Conversely, EIEC positive samples seem to be a clonal group, yet the number of samples was rather small. EAEC was the most prevalent pathotype of the identified E. coli. However, no differences were observed between the children with 106/382 (27.8%) and without 48/145 (33.1%) diarrhoea in terms of prevalence and phenotypic fingerprinting characteristics. Although, differences were appreciated between the herein tested EAEC isolates from children with 73/116 (62.9%) and without 28/68 (41.2%) diarrhoea, that showed aggregative adherence to Caco-2 cells. Additionally, a great variety of other putative virulence markers combination was detected, confirming the heterogeneity of this pathotype.

EHEC was only identified in samples from children with 8/381 (2.1%) diarrhoea. The identified EHEC were distributed as follow: vt2 6/8 (75%), followed by vt2-eaeA 1/8 (12.5%) and vt1 1/8 (12.5%) positives, at similar frequencies. In addition, the low diversity (Di=0.829) obtained from PhPRE analysis of EHEC positive samples suggest the presence of pathogenic clonal groups. In addition, the majority of the group of children that required IRT 42/68 (61.8%) harboured EAEC, ETEC and EPEC either alone or in combination. Also, the diversity value (Di=0.937) of the isolates from these children suggest that these strains may represent virulent clones capable of causing a more severe case of diarrhoea.

In summary, the ETEC pathotype play an important role in diarrhoea in children less than 5 years of age in León, Nicaragua. Nonetheless, EAEC, EPEC and EHEC pathotypes are to some extent important pathogens associated with diarrhoea in those children. In addition to all the above, the structure of the Oantigen polysaccharides (PS) from the EAEC strain 94/D4 and the international type strain from Escherichia coli O82 were determined in study IV. The O-antigen is composed of tetrasaccharide repeating units with the following structure: [--->4)-alpha-D-Glcp6-P-2-D-GroA-(1--->4)-beta-D-Galp-(1--->4)-beta-D-Glcp-(1--->3)-beta-D-GlcpNAc-(1--->]18