OUP user menu

Prevalence of secreted autotransporter toxin gene among diffusely adhering Escherichia coli isolated from stools of children

Carla R. Taddei, Ana Carolina R. Moreno, Antônio Fernandes Filho, Liana P.G. Montemor, Marina B. Martinez
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00688-8 249-253 First published online: 1 October 2003


In this report, we analyzed the prevalence of the sat gene in 336 Escherichia coli samples collected from stools of children with and without diarrhea in Brazil and in 100 uropathogenic E. coli strains. The results show a high correlation between diffusely adhering E. coli (DAEC) and the presence of sat (44%) in intestinal isolates. DAEC strain FBC114 expresses a 107-kDa protein, which showed 98% homology with Sat.

  • Sat
  • Diffusely adhering Escherichia coli
  • Autotransporter toxin

1 Introduction

Diffusely adhering Escherichia coli (DAEC) is a newly proposed category of diarrheogenic E. coli [1]. Several recent studies have implicated DAEC strains as agents of diarrhea, while in other studies DAEC strains were not recovered more frequently from diarrheal patients than from asymptomatic controls [25]. An association of DAEC with diarrhea was found in older children. It has been shown that the relative risk of diarrhea associated with DAEC increased with age, from 1 year to 4–5 years [5]. In France, DAEC strains were found to be highly prevalent in diarrheal cases among hospitalized patients who had no other identified enteropathogen, suggesting that DAEC strains may be an important diarrheal pathogen in developed countries [3].

Data regarding cause–effect relationships between DAEC strains and diarrhea remain, at present, controversial, and the enteric pathogenicity of these strains is still in question. It has been thought that DAEC strains constitute a heterogeneous group, and many different DAEC clones probably coexist, all of them exhibiting the diffusely adhering phenotype in tissue culture assays, while only a sub-population of these microorganisms are pathogenic [6]. This theory may explain why DAEC strains have been isolated from non-diarrheic stools in epidemiologic matched control studies [3,7].

Several structures involved in virulence have been screened for in DAEC strains. So far, four adhesins related to the diffuse-adherent phenotype have been described, namely: AIDA-I [8], F1845 [9], a 57-kDa protein [10] and CF16k [11]. However, there were insufficient data to confirm the role of these adhesins in diarrhea.

Some DAEC strains were found to contain a homologue of the locus of enterocyte effacement pathogenicity island and to exhibit pathogenic properties characteristic of enteropathogenic E. coli (EPEC) strains, such as the secretion of EspC proteins [12].

The importance of Afa/Dr DAEC strains in urinary tract infections (UTI) has already been described [13]. Recent studies have shown that some DAEC strains harbor virulence factors found in uropathogenic E. coli strains (UPEC) [1417].

Sequences homologous to Sat (secreted autotransporter toxin), recently described as a new virulence determinant of UPEC, have been described for some categories of E. coli, such as: EPEC, enteroaggregative E. coli (EAEC) and enterohemorrhagic E. coli (EHEC) [18].

Autotransporter toxins are proteins that have all the requirements for secretion across the outer membrane within a single molecule, and secretion is energy-independent [19]. Sat appears to belong to the family of serine protease autotransporters of Enterobacteriaceae (SPATE) [20]. This family, which has specific and distinct functions, has been identified only in pathogenic bacteria [19,20], and includes a variety of virulence toxins such as Pet, Pic, EspC, SigA, SepA, Tsh, and EspP [19].

In order to study the presence of virulence determinants in DAEC strains, we analyzed the prevalence of the sat gene in 436 E. coli samples collected from the stools of children with and without diarrhea, and from the urine of UTI patients, and evaluated its correlation to DAEC.

2 Materials and methods

2.1 Bacterial strains

The E. coli strains used in this study were distributed into two groups: the first was comprised of DAEC strains and non-DAEC strains, both isolated from stools of children with and without diarrhea from sporadic cases in Brazil [3,21,22], and the second group was represented by UPEC. The DAEC group was composed of 145 strains. Of the 191 non-DAEC strains, 24 were EPEC, 24 were EAEC, 28 were enterotoxigenic E. coli (ETEC), 25 were enteroinvasive E. coli (EIEC), and 90 were non-adherent E. coli strains. All the E. coli strains were characterized in hybridization assays with virulence gene probes for diarrheogenic E. coli (inv, eae, bfp, EAF, LT, ST, Stx1, Stx2, AA, daaC, pic, pet, she, east1) [1,19]. To investigate the prevalence of sat in urine samples, 100 UPEC strains were selected (group two) from a collection in our laboratory. All of them were isolated from urine cultures which yielded more than 105 CFU ml−1. DAEC strain FBC114, which was isolated from a sporadic case of diarrhea in Brazil [22], was used as a prototype strain to study the expression of Sat.

2.2 Protein electrophoresis

The protein fractions above 50 kDa obtained from DAEC FBC114 culture supernatants using Ultrafree filters (cutoff 50 kDa, Millipore, Bedford, MA, USA) were analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) using the Laemmli [23] method under reducing conditions. Proteins separated by SDS–PAGE were stained with Coomassie blue (Sigma, St. Louis, MO, USA), or transferred to Immobilon membranes (Millipore), for amino acid sequencing. The proteins were subjected to N-terminal amino acid sequencing using an Applied Biosystems protein sequencer model 491 (Applied Biosystems, Foster City, CA, USA) at the Centro de Quı́mica de Proteı́nas, Faculdade de Medicina de Ribeirão Preto, Brazil.

2.3 Search for a sat homologous DNA sequence

All the E. coli strains described above were assayed by the polymerase chain reaction (PCR) amplification technique. Primers Sat1 (5′-GCAGCAAATATTGATATATCA-3′) and Sat2 (5′-GTTGTTGACCTCAGCAAGGAA-3′) were designed based on the sequence of the sat gene in E. coli [18]. Sat1 is located on the sense strand, 145 bp of the Sat passenger domain, and Sat2 is located on the antisense strand (775 bp) [18] (Fig. 1). These primers were designed to amplify a 630-bp fragment. Primers corresponding to the entire sat gene, Sat1 and Sat3 (5′-GTTGTTGACCTCAGCAAGGAA-3′) were designed based on the gene described by Guyer et al. [18]. Sat3, 3061 bp, is located on the antisense strand [18] (Fig. 1). All primers were synthesized by Invitrogen (Carlsbad, CA, USA). Amplification was done for 30 cycles on a DNA thermal cycler (Perkin-Elmer Gene Amp 2400, Perkin-Elmer, Foster City, CA, USA) and the PCR conditions included denaturation for 1 min at 94°C, annealing for 1 min at 57°C, and extension for 2 min at 72°C. All PCR products were purchased from Invitrogen. The homology of the nucleotide sequence of the 630-bp fragment with other SPATE genes was searched using the Standard BLAST program (http://www.ncbi.nlm.nih.gov/BLAST, July 2003), and there was no homology with pet, pic, espC, espP, tsh, sigA, sepA [19], epeA (AF422145), eatA (AY163491), eaaA and eaaC [24].

Figure 1

Structure of the sat gene and location of primer binding sites. 1: promoter; 2: signal sequence; 3: passenger domain containing N-terminal and sec-dependent regions; 4: autotransporter domain. P1: Sat1 primer; P2: Sat2 primer; P3: Sat3 primer.

2.4 HEp-2 cell adherence assays

All E. coli isolates were characterized by the pattern of adherence to HEp-2 cells in the presence of 1%d-mannose, according to the method described by Cravioto et al. [25]. After the HEp-2 cells had grown to 60% confluence, E. coli samples were added and the culture plate was incubated for 3 h. Weakly adherent isolates, or those whose adhesion followed no pre-defined pattern, were retested using the same method with an additional incubation for 3 h (6 h assay). The following E. coli strains were included as adherence controls: E2348/69 for LA [5], C1845 for DA [9] and 042 for AA [5].

2.5 Statistical analysis

Data were analyzed using the χ2 test (α=5%). A P value of less than or equal to 0.05 was considered as being statistically significant [26].

3 Results and discussion

The DAEC FBC114 culture supernatant was analyzed by SDS–PAGE and showed a protein of molecular mass 107 kDa. When this protein was subjected to N-terminal amino acid analysis, the first 13 amino acids showed 98% identity with Sat (Fig. 2). A fragment of 2.9 kb corresponding to the passenger domain of sat [18] was successfully amplified from the DNA of the DAEC FBC114 strain when Sat1/Sat3 primers were used. When the 2.9-kb fragment was sequenced, it showed 98% homology with sat, confirming the presence of this gene in the DAEC FBC114 sample (data not shown).

Figure 2

Comparison of amino acid sequences of the 107-kDa protein of FBC114 E. coli and other autotransporter toxins. Pet (EAEC) from [30]; EspC (EPEC) from [31].

To determine the prevalence of sat in the E. coli strains selected for this study PCR amplification using primers Sat1 and Sat2 was done. Among the 145 DAEC strains isolated from feces, the 630-bp fragment was amplified from 64 (44%), but only 14 of 191 (7.3%) non-DAEC strains produced the amplified sat fragment (P<0.001) (Table 1). Similar to the findings of Guyer et al. [18], the highest frequency of sat among the non-DAEC strains was in the 24 EAEC strains studied, of which eight (33.3%) presented the 630-bp fragment (Table 2). It is possible that EAEC might be a reservoir of the sat gene. However, further investigation as to the expression of this gene must be carried out to evaluate the significance of its high prevalence. None of the 25 EIEC strains studied produced the 630-bp PCR fragment. Among the 24 EPEC strains and 28 ETEC strains, the numbers of positive results were three (12.5%) and one (3.5%), respectively (Table 2). The frequency of sat was 27% in the UPEC strains, which were assayed in HEp-2 cells, and 21 (78%) showed a diffuse adherence pattern (Table 2).

View this table:
Table 1

Presence of sat among DAEC and non-DAEC E. coli isolates

Bacterial strainNumber of isolatesNumber positive for sat (%)
DAEC strains
Intestinal isolates14564 (44)
Urine isolates2121 (100)
Non-DAEC strains
Intestinal isolates19114 (7.3)
Urine isolates796 (7.6)
View this table:
Table 2

Distribution of sat among diarrheogenic E. coli non-DAEC strains

Non-DAEC strainNumber of isolatesNumber positive for sat (%)
EPEC243 (12.5)
ETEC281 (3.5)
EAEC248 (33.3)
Non-adherent902 (2.2)

Guyer et al. [18] reported the presence of sat in some diarrheogenic E. coli categories (EPEC and EHEC) and in Shigella strains, but not in such a high percentage as found among DAEC strains in this study. A further report has also shown sat in Shigella spp. [27]. However, none of these studies reported expression of Sat by Shigella spp. Interestingly, the 630-bp nucleotide sequence amplified with primers Sat1 and Sat2 showed 98% identity with a 560-bp nucleotide sequence of the Shigella flexneri complete chromosome [28], when analyzed in the Standard BLAST program. The probe used in those studies include the sat region that we amplified with the Sat1 and Sat2 primers. Therefore, those probes might have detected a chromosomal region present in S. flexneri 2a, which was sequenced by Jin et al. [28]. Until now, the function of the gene containing this sequence has not been described.

Sat was first described in UPEC strains, and the authors found a homologous sequence of sat in 38 out of 67 (55%) pyelonephritis strains but only six out of 27 (22%) fecal isolates [18]. However, our results indicated that the homologous sequences of sat are present significantly more often in DAEC strains, independent of the site of infection. In addition, we have observed that the DAEC prototype strain C1845 [9] also yielded the sat PCR product (data not shown).

The data presented here show a high correlation of DAEC and Sat. In general, UTI are caused by the entrance of pathogenic bacteria, present in the colon, into the urinary tract [29]. DAEC strains may represent the reservoir for UPEC, since several virulence factors of DAEC, such as adhesins and hemolysins, are found in UPEC isolates [14].

The mode of action of Sat is the induction of vacuolation and glomerular damage, demonstrated recently in a CBA mouse model of ascending UTI, indicating that Sat is a vacuolating cytotoxin [20]. If Sat is, indeed, a virulence factor expressed by some DAEC clones, the injuries seen in the urinary tract could be found in intestinal tissue, inducing diarrhea. To elucidate this question, we are working to determine the role of Sat in diarrheal disease and its effects on intestinal tissue.


This work was supported by grants from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP No. 00/11870-6 to C.R.T.). We thank Dr. Tania T.G. Amaral for providing some DAEC strains.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
View Abstract