• Hemolytic uremic syndrome;
  • restriction fragment length polymorphisms;
  • Shiga toxin-producing Escherichia coli


  1. Top of page

Using culture-independent technology, PCR-RFLP were used to identify and type STEC in the stool of a patient with HUS. Fecal PCR-RFLP patterns were identical to those of the STEC O157:H7 isolated from the patient.

List of Abbreviations: 

ethylenediaminetetraacetic acid

Escherichia coli

E. coli

HGT agarose

high gelling temperature agarose


hemolytic uremic syndrome


polymerase chain reaction restriction fragment length polymorphisms


pulsed-field gel electrophoresis




sodium dodecylsulfate


sorbitol MacConkey


Shiga toxin


Shiga toxin-producing Escherichia coli





STEC O157:H7 is an important human pathogen (1), which causes diarrhea, bloody diarrhea, and HUS. It is important to diagnose infected patients as rapidly as possible, because timely identification can accelerate disease control efforts by public health authorities (2) and prompt appropriate medical intervention such as volume expansion (3). However, despite the importance of making this diagnosis, the identification of most patients still relies on standard culture technology, such as plating of the stool on SMAC agar. In this study, we directly and rapidly identified and typed STEC in a human specimen without using culture, by employing a recently described PCR-RFLP technique (4–6) based on the genetic diversity of the regulatory region of Stx-phage.

Seventeen stool specimens were collected from patients with diarrhea and/or HUS. DNA was extracted from the stool specimens using guanidinium hydrochloride. Each stool specimen (350 mg) was dissolved in 350 μl of TE buffer and mixed vigorously with 350 μl of 4 M guanidium hydrochloride solution. The same volume of TE saturated phenol was added and mixed well. Subsequently the same volume of chloroform/isoamyl alcohol (24:1) was added and mixed well. The mixture was centrifuged and the supernatant was collected. These steps were repeated twice. DNA was precipitated with sodium acetate and isopropyl alcohol. After centrifugation, the pellet was dried, dissolved with proteinase K solution (100 μg/ml in 0.1 M Tris-HCl [pH 7.4], containing 50 mM NaCl, 10 mM EDTA and 0.2 % SDS) and incubated at 37°C for 2 hr. Then DNA was treated by phenol/chloroform and precipitated with ethanol. After drying, the pellet was dissolved in RNase A solution (100 μg/ml) and incubated at 37°C for 30 min. DNA was again treated by phenol/chloroform and precipitated with ethanol. After drying, the DNA was dissolved in TE buffer and the quantity and quality of the purified DNA were determined by spectrophotometer with A260/A280 and 0.7 % agarose gel electrophoresis. A concurrent attempt was also made to isolate E. coli O157 by plating the stool specimen and/or enrichment tryptic soy broth culture of the stool specimen on SMAC agar and/or CT-SMAC and incubating at 37°C overnight. Sorbitol-negative colonies were examined by enteric bacteria identification test strip API 20E (bioMerieux, Lyon, France), and their serotype determined with O157 and H7 antisera. The presence of stx1 and stx2 genes was confirmed by multiplex PCR (7). Both DNA extracted from the stool specimen and E. coli O157:H7 isolated were amplified by PCR using TaKaRa LA-Taq (Takara-Bio, Shiga, Japan) and primers targeting the regulatory region of Stx-phage as described previously (4). The PCR product was analyzed by 0.4 % agarose gel electrophoresis using HGT agarose and/or by field inversion gel electrophoresis with a 1.0 % pulsed-field certified agarose gel in 0.5 × TBE electrophoresis buffer for 26 hr, followed by ethidium bromide staining. The PCR product was subjected to enzymatic digestion with either BglI or EcoRV to generate the PCR-RFLP. DNA from stool specimens, as well as cultures of nine commensal E. coli isolates (three each from one healthy adult and two healthy children) processed by the same methods, were used as negative controls for PCR.

Among the 17 samples examined, E. coli O157:H7 was isolated from two stool specimens. The specific PCR amplification was obtained from only one of these two, a specimen from a 14-year-old girl who had been diagnosed as having early HUS. She presented to an ambulatory office with abdominal pain, vomiting and mild diarrhea. On the second day of illness, she had a fever, and intravenous lincomycin hydrochloride and oral cefaclor monohydrate were administered. On the third day of illness, the patient had a hematocrit of 31.9 %, a platelet count of 89 000/mm3, and a serum creatinine of 0.9 mg/dL, thereby establishing a diagnosis of early HUS. STEC infection was suspected, stool was immediately collected, and DNA was directly extracted from the stool specimen as above. The patient remained hospitalized for two weeks and received intravenous fluid, oral fosfomycin and a lactobacillus. Abdominal ultrasonography showed thickening of the ileocecal mucosa as previously reported (8).

DNA (50 ng) extracted from the stool was amplified by PCR using primers targeting the regulatory region of Stx-phage (4), resulting in two bands of about 8.2 and 12.7 kb (Fig. 1a). These bands were then digested by restriction enzymes (4). Six bands ranging in size from 0.52 to 9.4 kb, and nine bands ranging in size from 0.65 to 4.1 kb, were produced by digestion with BglI and EcoRV, respectively (Fig. 1b, c). E. coli O157:H7 was isolated and identified. The E. coli O157:H7 isolate was positive for both loci by multiplex PCR (7). Comparison of the PCR-RFLP results of a pure culture of the resulting E. coli O157:H7 demonstrated the same PCR products and RFLP patterns as those obtained with genomic DNA purified from the stool specimens (Fig. 1). No amplicons were produced using DNA purified from stool specimens from controls or from cultures of the nine commensal E. coli isolates (data not shown).


Figure 1. (a) Field inversion gel electrophoresis of LA-PCR products of purified genomic DNA of stool from a HUS patient and of E. coli O157:H7 isolated from enrichment culture. Lanes: 1, 2.5 kb DNA ladder; 2, stool specimen; 3, STEC O157 isolate. Field inversion gel electrophoresis of BglI digest (b) and EcoRV digest (c) of LA-PCR products of purified genomic DNA of stool specimen obtained from HUS patient and of STEC O157 isolated from enrichment culture. Lanes: 1, λ-HindIII digest; 2, 1-kb DNA ladder; 3, stool specimen; 4, STEC O157 isolate.

Download figure to PowerPoint

Molecular typing by PCR-RFLP has at least two distinct advantages. First, we have demonstrated that isolation of bacteria is not essential for initial identification of the pathogens. It has been reported that isolation of E. coli O157:H7 from diarrheal patients and food samples is sometimes difficult. For instance, Tarr et al. (9) reported that if stool samples of patients with HUS were cultured within six days of onset of diarrhea for EHEC O157:H7 the recovery rate was nearly 100 %, but this rate declined to 33.3 % if the stool was cultured after a week of illness. However, we do not believe that non-culture methods should replace culture technology, because it is important to confirm assay positive signals, and to recover the infecting strain for typing purposes. In the largest outbreak of STEC O157:H7 reported to date, which occurred in Japan in 1996 (10), radish sprouts were suspected to be responsible for the outbreak, but this pathogen could not be isolated from the sprouts or their seeds. However, PCR on an enrichment culture sample with seed of radish sprouts as template DNA, followed by Southern blot hybridization, demonstrated specific amplification of stx1, stx2 and O157-specific genes after, but not before, cultivation (Shimizu T. and Takeda Y., unpublished).

Second, the time required for molecular typing can be shortened. For example, molecular typing by PCR-RFLP without isolation of bacteria can be completed within 16 hours of obtaining a stool specimen. However, it takes several days to complete molecular typing by PCR-RFLP on bacterial isolates obtained from stools. If PFGE analysis is employed, an additional day may be required. Furthermore, PFGE has several disadvantages such as genetic change caused by clonal turnover (11, 12) and by repeated subculture prolonged storage (13), and a limitation in the number of isolates that can be analyzed at one time. We recently reported that PCR-RFLP could solve confusion caused by PFGE caused by clonal turnover (5) and by repeated subculture prolonged storage (6).

PCR-RFLP could be a very powerful technique for concurrent identification and molecular typing of STEC strains, because identification and typing data can be available the next day. However, timing of collection of stool specimen is crucial to successful performance of PCR-RFLP without isolation of bacteria. In this study, there was a sample from which we were able to isolate E. coli O157:H7 but could obtain no specific amplification. Very few colonies of E. coli O157:H7 were obtained by CT-SMAC from this sample, indicating that the number of E. coli O157:H7 present in the stool may have been very low. At least 10 pg of purified genomic DNA of E. coli O157:H7 is probably the minimum requirement for PCR-RFLP to get a PCR product which can be used for further restriction analysis. There may have been less than 10 pg of genomic DNA of E. coli O157:H7 in the sample cited above. Therefore, it is important to collect stool specimens as early as possible, at latest within six days of onset of diarrhea. Moreover the sensitivity and specificity of this methodology need to be determined by testing more infected and control human specimens.

In summary, we describe a method to detect E. coli O157:H7 directly in freshly shed stool using PCR-RFLP. This methodology is feasible because of its simplicity, speed, and ease of use and could be refined to provide point-of-presentation diagnosis and typing, which would have public health and clinical value. Moreover, this approach might avert the confusion engendered by clonal turnover within the gastrointestinal tracts of infected humans. Future studies should be directed to determining its sensitivity and specificity in groups of infected subjects and controls.


  1. Top of page

We are grateful to Phillip I. Tarr, Washington University School of Medicine, St Louis, for helpful discussions and critical reading of the manuscript. This work was supported in part by A Grant-in-Aid for Scientific Research (C) 19590455 from the Japan Society for the Promotion of Science and the Program of Founding Research Centers for Emerging and Reemerging Infectious Diseases, MEXT Japan.


  1. Top of page
  • 1
    Tarr P.I., Gordon C.A., Chandler, W.L. (2005) Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 365: 10731086.
  • 2
    Bell B.P., Goldoft M., Griffin P.M., Davis M.A., Gordon D.C., Tarr P.I., Bartleson C.A., Lewis J.H., Barrett T.J., Wells J.G., et al. 1994. A multistate outbreak of Escherichia coli O157:H7-associated bloody diarrhea and hemolytic uremic syndrome from hamburgers. The Washington experience. JAMA 272: 13491353.
  • 3
    Ake J.A., Jelacic S., Ciol M.A., Watkins S.L., Murray K.F., Christie D.L., Klein E.J., Tarr P.I. (2005) Relative nephroprotection during Escherichia coli O157:H7 infections: association with intravenous volume expansion. Pediatrics 115: 673680.
  • 4
    Shima K., Terajima J., Sato T., Nishimura K., Tamura K., Watanabe H., Takeda Y., Yamasaki S. (2004) Development of a PCR-restriction fragment length polymorphism assay for the epidemiological analysis of Shiga toxin-producing Escherichia coli. J Clin Microbiol 42: 52055213.
  • 5
    Shima K., Yoshii N., Akiba M., Nishimura K., Nakazawa M., Yamasaki S. (2006) Comparison of PCR-RFLP assay and PFGE for determining the clonality of enterohemorrhagic Escherichia coli strains. FEMS Microbiol Lett 257: 124131.
  • 6
    Shima K., Wu Y., Sugimoto N., Asakura M., Nishimura K., Yamasaki S. (2006) Comparison of PCR-restriction fragment length polymorphism (PCR-RFLP) assay to pulsed-field gel electrophoresis: effect of repeated subculture and prolonged storage on RFLP patterns of Shiga toxin-producing Escherichia coli O157:H7. J Clin Microbiol 44: 39633968.
  • 7
    Khan A., Yamasaki S., Ramamurthy T., Pal A., Datta S., Chowdhury N.R., Das S.C., Sikdar A., Tsukamoto T., Bhattacharya S.K., Takeda Y., Nair G.B. (2002) Prevalence and genetic profiling of virulence determinants of non-O157 Shiga toxin-producing Escherichia coli isolated from cattle, beef and humans Calcutta, India. Emerg Infect Dis 8: 5462.
  • 8
    Tarr P.I., Weinberger E., Hatch E.I. Jr., Christie D.L. (1992) Bacterial ileocecitis caused by Escherichia coli O157:H7. J Pediatr Gastroenterol Nutr 14: 261263.
  • 9
    Tarr P.I., Neill M.A., Clausen C.R., Watkins S.L., Christie D.L., Hickman R.O. (1990) Escherichia coli O157:H7 and the hemolytic uremic syndrome: importance of early cultures in establishing the etiology. J Infect Dis 162: 553556.
  • 10
    Michino H., Araki K., Minami S., Takaya S., Sakai N., Miyazaki M., Ono A., Yanagawa H. (1999) Massive outbreak of Escherichia coli O157:H7 infection in schoolchildren in Sakai City, Japan, associated with consumption of white radish sprouts. Am J Epidemiol 150: 787796.
  • 11
    Akiba M., Sameshima T., Nakazawa M. (2000) Clonal turnover of enterohemorrhagic Escherichia coli O157 in experimentally infected cattle. FEMS Microbiol Lett 184: 7983.
  • 12
    Karch H., Rüssmann H., Schmidt H., Schwarzkopf A., Heesemann J. (1995) Long-term shedding and clonal turnover of enterohemorrhagic Escherichia coli O157 in diarrheal diseases. J Clin Microbiol 33: 16021605.
  • 13
    Iguchi A., Osawa R., Kawano J., Shimizu A., Terajima J., Watanabe H. (2002) Effects of repeated subculturing and prolonged storage at room temperature of Enterohemorrhagic Escherichia coli O157:H7 on pulsed-field gel electrophoresis profiles. J Clin Microbiol 40: 30793081.