HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Alonso, R Articles by Bouza, E Search for Related Content PubMed PubMed Citation Articles by Alonso, R Articles by Bouza, E Agricola Articles by Alonso, R Articles by Bouza, E

Toxigenic status of Clostridium difficile in a large Spanish teaching hospital

Department of Clinical Microbiology and Infectious Diseases, Hospital General Universitario ‘‘Gregorio Marañón’', C/Doctor Esquerdo, 46, 28007 Madrid, Spain

Correspondence R. Alonso ralonso.hgugm{at}salud.madrid.org

Received July 2, 2004
Accepted September 27, 2004

The aim of this study was to evaluate the toxigenic status of circulating strains of Clostridium difficile in a large teaching hospital. Overall 220 isolates were studied of which 199 (90.5 %) produced both large clostridial toxins detected by conventional methods. Ten more strains (4.5 %) had toxin A and B genes detectable by PCR. Eleven (5.0 %) variant strains (A^–B⁺) were detected among the isolates studied and 10 strains (4.5 %) had the binary toxin genes (cdtA and cdtB).

Abbreviation: LCTs, large clostridial toxins.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Clostridium difficile is the main aetiologic agent of antibiotic-associated diarrhoea (CDAD) (Bartlett, 1992; Kelly et al., 1994) and is responsible for both sporadic cases and epidemic outbreaks (Johnson et al., 1990). Most cases are nosocomially acquired, but community-acquired CDAD is being increasingly recognized (Hirschhorn et al., 1994; Kyne et al., 1998).

C. difficile can produce two large clostridial toxins (LCTs), toxin A (TcdA), which is mainly enterotoxic, and toxin B (TcdB), which is highly cytotoxic. Both toxins disrupt the cytoskeleton by acting on regulatory proteins involved in actin polymerization (Lyerly et al., 1988). Traditionally, it was considered that toxigenic strains produced both LCTs whereas non-toxigenic strains did not produce either. Some years ago, the isolation of toxigenic strains that produced toxin B only was reported (A^–B⁺) (Lyerly et al., 1992) and it was soon demonstrated that this phenomenon was not as rare among clinical isolates as previously thought (Kato et al., 1998; Depitre et al., 1993; Rupnik et al., 2003). CDAD is usually caused by strains producing both LCTs although A^–B⁺ isolates may also cause the disease (Alfa et al., 2000).

The presence of an additional toxin in C. difficile has recently been detected. This actin-specific ADP-ribosyltransferase toxin has been designated binary toxin (CDT) due to its two independent proteins, CDTa, the catalytic component, and CDTb, the binding component. C. difficile binary toxin is related genetically, immunologically and functionally to the group of clostridial binary toxins, which includes the well known iota toxin of Clostridium perfringens (Popoff et al., 1988). It is not clear how the production of the binary toxin by strains of C. difficile can determine its virulence. It has been suggested that, although strains that produce CDT only have a relatively low virulence, the toxin could act synergistically in strains that produce both LCTs (Stubbs et al., 2000; Geric et al., 2003). Data regarding the prevalence of CDT in C. difficile are scarce, but figures range from 4 to 12.0 % (Perelle et al., 1997; Gulke et al., 2001).

The aim of this study was to evaluate the toxigenic status of circulating strains of C. difficile in a large Spanish hospital with a high prevalence of CDAD.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Strains and identification.

Two hundred and twenty isolates of C. difficile were collected prospectively over a 6-month period (January–June 2001) from 1154 diarrhoeic stool samples submitted for C. difficile investigation (hospitalized patients treated with antibiotics). Samples were cultured on CCFA (cycloserine–cefoxitin–fructose agar) plates which were incubated under anaerobic conditions at 37 °C for 48 h. C. difficile isolates were presumptively identified by their colony morphology, yellow colour, ground-glass texture, characteristic horse-dung smell and Gram-stain appearance. Additional biochemical tests were also used (ATB 32A; bioMérieux). Only one isolate was collected from each positive sample.

Toxin detection.

The presence of C. difficile toxin B was determined by demonstrating a specific cytopathic effect on MRC-5 cells, as described previously (Chang et al., 1979; Bowman & Riley, 1988; Bartlett, 1994), either directly from fecal samples or, if negative, from pure cultures of the micro-organism.

An enzyme immunoassay (CdTOX A OIA; BioStar) was used to detect the presence of toxin A in fecal samples. The test was repeated from pure cultures when a negative result was observed in the direct clinical specimen.

Molecular methods.

LCTs and cdt genes were detected by PCR assays. DNA was extracted from pure C. difficile cultures using a Chelex resin-based commercial system (InstaGene Matrix; Bio-Rad) following the manufacturer's recommendations. The tcdA gene was detected using a previously published PCR assay (Kato et al., 1991). Briefly, oligonucleotides 5'-CCC AAT AGA AGA TTC AAT ATT AAG CTT-3' and 5'-GGA AGA AAA GAA CTT CTG GCT CAC TCA GGT-3' were used to prime PCR reactions. Amplifications were performed by 35 cycles at 95 °C for 15 s, 50 °C for 20 s and 72 °C for 40 s. Positive samples produced an amplification product of 251 bp. The tcdB gene was detected using the method of Wolfhagen et al. (1994). Amplification primers were 5'-TAA TAG AAA ACA GTT AGA AA-3' and 5'-TCC AAT CCA AAC AAA ATG TA-3'. Amplification was carried out with 40 cycles at 94 °C for 1 min, 50 °C for 1 min and 72 °C for 1 min. In positive samples, a fragment of 301 bp was amplified. Binary toxin was also detected by PCR (Stubbs et al., 2000). Two reactions were needed to provide evidence of both components of the toxin. cdtA was detected with primers 5'-TGA ACC TGG AAA AGG TGA TG-3' and 5'-AGG ATT ATT TAC TGG ACC ATT TG-3', whereas cdtB was detected using primers 5'-CTT AAT GCA AGT AAA TAC TGA G-3' and 5'-AAC GGA TCT CTT GCT TCA GTC-3'. In both cases, reactions were subjected to 30 cycles of denaturation at 94 °C for 45 s, annealing at 52 °C for 1 min and elongation at 72 °C for 20 s. Amplification products were detected as bands of 375 and 510 bp for cdtA and cdtB, respectively. All PCR reactions were carried out in a 9700 Applied Biosystems thermocycler. In all cases, amplification products were detected in agarose gels stained with ethidium bromide under UV illumination.

Control strains C. difficile ATCC 9689 (TcdA⁺ TcdB⁺), C. clostridioforme 3268 (TcdA^– TcdB^–), C. difficile 48489 (CDT⁺) and C. difficile 48752 (CDT^–) were included.

Bacterial typing.

The genetic diversity of strains sharing the same toxin patterns was analysed by PCR ribotyping, with amplification of the 16S and 23S intergenic regions (Bidet et al., 2000). The primer sequences were 5'-GTG CGG CTG GAT CAC CTC CT-3' (16S primer) and 5'-CCC TGC ACC CTT AAT AAC TTG ACC-3' (23S primer). Amplification conditions consisted of 35 cycles of 1 min for denaturation at 94 °C, 1 min for primer annealing at 57 °C and 1 min for extension at 72 °C. Ribotyping patterns were separated by electrophoresis on MS-8 agarose (3.0 %) at 100 V for 3 h and were analysed with UV light after ethidium bromide staining.

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
The results of LCT detection in the 220 isolates studied are summarized in Table 1. A total of 199 isolates (90.5 %) produced toxin A and in 21 (9.5 %) the toxin was not detected. The PCR method confirmed the presence of the tcdA gene in all 199 isolates and was positive also in 10 of the 21 enzyme immunoassay-negative isolates (95.0 %).

The cytotoxicity assay identified 210 isolates that produced toxin B (95.5 %), and 10 (4.5 %) that did not. All 220 isolates were PCR-positive for the tcdB gene (100 %). Amplification products of tcdA and tcdB genes can be seen in Fig. 1(a) and Fig. 1(b), respectively.

View larger version (86K): [in this window] [in a new window]   Fig. 1. Amplification products of tcdA (a) and tcdB (b) genes. Lanes: 1, 14 and 26, molecular mass marker; 2, reference strain ATCC 9689T; 12, negative control (C. clostridioforme 3268); 3–11, 13, 15–25, clinical isolates.

C. difficile binary toxin was detected in 10 isolates (4.5 %). In all cases, both components of the CDT were detected (Fig. 2). All 10 CDT-positive isolates were also toxigenic for toxin A and toxin B. Fig. 2 shows examples of positive and negative strains for both components of CDT.

View larger version (89K): [in this window] [in a new window]   Fig. 2. PCR products of CDTa (left) and CDTb (right). Lanes: 1, molecular mass marker; 2 and 5, control strain (C. difficile 48489); 3 and 6, negative control (C. difficile 48752); 4 and 7, positive clinical isolate.

Strains harbouring both LCTs only showed high genetic homogeneity and most of them (170 isolates, 78.0 %) belonged to a single ribotype (R1) (data not shown). The 11 variant strains (TcdA^– TcdB⁺) were grouped into five ribotypes (data not shown) and were not associated with any specific unit in the hospital (Table 2). The 10 isolates which were positive for binary toxin showed a high genetic homogeneity belonging to two ribotypes (seven to R8 and three to R19; data not shown), despite the fact that they spread throughout different wards and were not clustered chronologically either (Table 2).

Considering only phenotypic criteria for toxigenicity, 10 of our 220 isolates (4.5 %) would have been considered non-toxigenic, but they harboured toxin genes. This apparent disagreement seems to question our present ‘gold standard’ (cytotoxicity assay) for diagnosis. Another consideration is the clinical significance of detecting genes that could be unexpressed or even incomplete. It is worth noting that all our strains were obtained from patients with diarrhoea, which may explain why we did not find any non-toxigenic strains.

Binary toxin genes were detected in 4.5 % of our strains. Other authors estimate proportions ranging from 4.0 to 12.0 % (Goncalves et al., 2004; Stubbs et al., 2000). All our strains had both components of the binary toxin, which was not always the case in the reports of other authors (Perelle et al., 1997). In our cases, binary toxin was present only in strains with LCTs, but it has also been reported in strains without LCTs (Geric et al., 2003).

Data regarding prevalence of different toxins and genes in C. difficile strains should always be interpreted on the basis of the clonal distribution in each institution. The genetic diversity of strains producing both LCTs varies in different series, depending on the sporadic or epidemic character of each one (Kato et al., 2001; Spigaglia et al., 2001). Several authors report a high homogeneity of variant strains (Alfa et al., 2000; Pituch et al., 2001) and one study reported that strains producing binary toxin are usually epidemiologically unrelated and not derived from a common ancestor (Goncalves et al., 2004), although in this report the isolates were obtained from 17 different institutions.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
This paper has been supported in part by ‘Red Española de Investigación en Patología Infecciosa’ (REIPI – CO3/14) and by ‘Fondo de Investigaciones Sanitarias’ (FIS 02/1049). We are grateful to Dr Maja Rupnik for kindly providing the reference strains for the binary toxin detection experiments (C. difficile 48489 and 48752). We also thank Thomas O'Boyle for his help with the correction of the English version of this manuscript.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES