Toxin A-negative, toxin B-positive Clostridium difficile

Transcription

1 International Journal of Infectious Diseases (2007) 11, PERSPECTIVE Toxin A-negative, toxin B-positive Clostridium Denise Drudy a, *,Séamus Fanning a, Lorraine Kyne b a Centre for Food Safety, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland b Department of Medicine for the Older Person, Mater Misericordiae University Hospital, Dublin 7, Ireland Received 30 November 2005; received in revised form 31 March 2006; accepted 17 April 2006 Corresponding Editor: Timothy Barkham, Singapore KEYWORDS A B + Clostridium ; Infectious diarrhea; Bacterial genotyping Summary Clostridium is a major cause of infectious diarrhea in hospitalized patients. Many pathogenic strains of Clostridium produce two toxins TcdA and TcdB, both of which are pro-inflammatory and enterotoxic in human intestine. Clinically relevant toxin A-negative, toxin B-positive (A B + ) strains of Clostridium that cause diarrhea and colitis in humans have been isolated with increasing frequency worldwide. This perspective describes these important toxin variant strains and highlights the need to use Clostridium diagnostic methods that can detect both TcdA and TcdB. # 2006 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. Introduction Clostridium is a common nosocomial pathogen and a major cause of infectious diarrhea in hospitalized patients. 1,2 Colonization with C. is associated with a wide spectrum of clinical presentations ranging from asymptomatic carriage to fulminant pseudomembranous colitis. 3 In recent years the incidence of C. and the number of cases associated with severe outcomes have increased worldwide. 4 8 This may be related to several factors, including changing demographics of patients admitted to hospital, changing infection control policies, or the emergence of more virulent strains of C.. 7,9,10 Two structurally similar toxins denoted as TcdA (308 kda) and TcdB (270 kda) are the main virulence determinants * Corresponding author. Tel.: ; fax: address: denise.drudy@ucd.ie (D. Drudy). associated with Clostridium -associated disease (CDAD) and most pathogenic strains of C. produce both toxins (A + B + ). Both of these large toxins are glucosyltransferases that catalyze the monoglucosylation of threonine 35/37 of small GTP-binding proteins Rho, Rac, and Cdc42 within target cells, and thus modulate several physiological cellular events resulting in cell death. Structurally these toxins have three domains: the enzymatic activity is located in the N terminus of the protein, the middle domain is the putative translocation region, and the C terminus is involved in receptor binding. The N termini of TcdA and TcdB demonstrate 74% homology that accounts for their similar substrate specificity. The carboxy-terminal domain of both toxins carry a number of short homologous regions termed combined repetitive oligopeptides (CROPs). The role of these toxins in the pathogenesis of CDAD has been well described. 11 Both toxin A and B are proinflammatory, cytotoxic, and enterotoxic in the human colon. 12,13 These toxins are encoded by two genes tcda and tcdb, /$32.00 # 2006 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. doi: /j.ijid

2 6 D. Drudy et al. Figure 1 Schematic representation of the Clostridium pathogenicity locus of Clostridium and the alterations found in the four A B + strain types described to date. Filled bars represent the repetitive sequences of both toxin genes. All four types differ in the length of the deletion (indicated by the triangle and open bar) in tcda and the size of the deletion where known is indicated. The presence of the 3 0 end of tcda in strain XVII has not been confirmed and is therefore indicated with an asterisk. mapping to a 19.6-kb pathogenicity locus (PaLoc) containing additional regulatory genes. 14 C. isolates with varying genetic modifications within the PaLoc have been investigated and 28 different toxinotypes have been described. 15,16 These include variant C. that have deletions, insertions, or polymorphic restriction sites in one or more of the genes on the PaLoc but which can still produce functional toxin proteins TcdA and TcdB and toxin-variant strains (A B + ) which fail to produce detectable toxin A 15,17 (Figure 1). Toxin variant strains that produce either toxin A or toxin B are the most commonly isolated variant C. with modifications in their PaLoc. Until recently, it was thought that all toxigenic strains associated with disease produced both TcdA and TcdB. Early studies in animal models indicated that TcdA and TcdB acted synergistically but that TcdA was required to produce the initial damage in the colon. 18 Consequently much of the research on the pathogenesis of CDAD has focused on the inflammation modulated by toxin A. However many recent reports demonstrate the clinical importance of A-negative toxin B positive (A B + ) isolates Outbreaks of C. diarrhea, sporadic cases of infection and cases of pseudomembranous colitis (PMC) caused by A B + C. have been previously documented. 20,23 25 A B + Clostridium To date four different A B + Clostridium strains have been described and characterized using Rupnik s toxinotyping (PCR-RFLP) scheme. 15 In the early 1990s two A B + strain types were identified that produced TcdB but no detectable TcdA. Strain 8864, the first toxin variant strain to be characterized, caused hemorrhage, fluid accumulation, and tissue damage in ligated ileal rabbit loops. 26,27 Molecular analysis identified a large deletion of 5.9-kb in the PaLoc mapping to the 3 0 end of tcda and tcdc and which prevented production of TcdA at a transcriptional level. In addition there is a 1.1-kb insertion between tcda and tcde. 28 Toxinotyping designates this strain type as toxinotype X 15 (Figure 1). Early studies indicated that this strain type produced a modified form of TcdB 29 and subsequent analyses determined that TcdB-8864 induced a different cytopathic effect (CPE) compared to TcdB (TcdB-8864 glucosylates the Rho proteins Rac1, Rap 1/2, and Ral) 28 (Figure 2). Only one organism of strain type 8864 has ever been isolated (from an asymptomatic adult) and although the clinical significance of this strain in humans is questionable its virulence in animal models confirms its pathogenic potential. The second category of A B + strain types identified were serogroup F strains (type strain 1470) that were considered non-pathogenic due to their frequent isolation from asymptomatic infants and their lack of pathogenicity in animal models. 30,31 Like strain 8864 these are also truncated in the 3 0 region of the repetitive domain of tcda. They contain a 1.8-kb deletion in tcda, together with a mutation that introduces a stop codon corresponding to amino acid position 47, leading to the truncation of TcdA (Figure 1). Toxinotyping classified these organisms as type VIII. Analysis of TcdB-1470 indicated that this toxin also has polymorphisms in its tcdb gene resulting in altered glucosylation and differential cytopathic effects. 33 Sequence analysis of TcdB from strains 8864 and 1470 showed a high degree of sequence conservation in the enzymatic domain with substitutions at only 8 of 560 amino acid positions. 28 These novel cytotoxins are a hybrid between TcdB from reference strain VPI and Clostridium sordellii lethal toxin (TclS). 33 While the receptor binding domain shares 100% homology with TcdB the enzymatic domain is only 79% homologous (Figure 2). These hybrid toxins suggest that genetic exchange

3 Toxin A B + Clostridium 7 Figure 2 Receptor binding domain of variant Clostridium toxin B and toxin B from VPI demonstrate 99% homology. The enzymatic domain of variant Clostridium toxin B is only 79% homologous to toxin B from VPI Variant toxin B glucosylates different Rho proteins than toxin B from VPI This is further demonstrated by the differential cytopathic effect on fibroblast cells. and recombination may have occurred between large clostridial toxin (LCT) producing strains of clostridia. Most recently two additional A B + strains, toxinotypes XVI and XVII, have been described. 17 Interestingly these newer toxinotypes contain the genes that code the Clostridium binary toxin. The molecular mechanism responsible for the lack of toxin A production in these newer toxinotypes has not yet been described although these strains also have deletions in the CROP region of tcda 17 (Figure 1). Although four A B + strain types have been described, toxinotype VIII appears to be the most clinically significant toxin variant strain type. These have been isolated from four documented outbreaks, sporadic cases of diarrhea, and cases of pseudomembranous colitis. In contrast, only one strain of toxinotype XVI has been associated with clinical disease and the other two toxinotypes X and XVII have been isolated from asymptomatic patients. Outbreaks of A B + Clostridium To date there have been four documented outbreaks associated with A B + Clostridium. The first outbreak reported in 1998 by Alfa et al. occurred in a Canadian hospital over a three month period. 19,23 There were 16 cases of nosocomial C. diarrhea on four hospital wards; eight of the cases spent time on one hospital ward. This outbreak was associated with a 19% recurrence rate of CDAD and two deaths. Following the outbreak, the hospital infection control policy was changed and the laboratory testing for C. was changed from a toxin A ELISA, which failed to detect these A B + strains, to the cell culture cytotoxicity assay. A second A B + outbreak occurred in The Netherlands between 1997 and In this outbreak 24 patients with A B + C. were identified. The recurrence rate in this outbreak was 13% while there were seven episodes of severe diarrhea and one death. Clonality of the 16 available outbreak isolates was determined using 16 23S rrna ribotyping. This outbreak subsided following a change in the surgical prophylactic antibiotic policy such that clindamycin was withdrawn, along with the introduction of strategic infection control measures. In 2001, 10 patients with identical A B + C. strains were identified in a cancer care hospital in Japan. 34 Our group recently described an A B + C. outbreak in one Dublin hospital that affected 73 patients (Drudy et al., submitted for publication). Molecular typing of 90 recovered isolates determined that 95% of the isolates were A B + (PCR ribotype 017, toxinotype VIII). Disease spectrum and transmissibility of A B + Clostridium A B + Clostridium causes the same spectrum of disease that is associated with A + B + strains ranging from asymptomatic colonization through to the more severe fulminant colitis. The rate of C. recurrence observed in our study (36%) and other outbreaks due to toxin-variant strains, is similar to recurrence rates reported by us, and others, for toxin A + B + strains. However, outbreaks described by us and other authors have documented increased disease severity associated with these A B + isolates. 35 The mechanism(s) responsible for this observed increase in disease severity are not known. We have previously demonstrated that the immune response to toxin A was important in determining clinical presentation and course of CDAD. 35,36 Recent in vitro studies have indicated that like TcdA, TcdB is also a potent enterotoxin capable of inducing 10-fold more toxicity in human colonic explants than toxin A. 13,37 Warny

4 8 D. Drudy et al. and colleagues have demonstrated that emerging epidemic A + B + clones produce 23 times more TcdB and 16 times more TcdA compared to toxinotype 0 strains. 10 Although the kinetics of toxin B production in A B + isolates have not yet been measured these strains do not produce toxin A and therefore the pathology observed in cases of severe colitis are likely to result from host inflammatory events modulated by TcdB. These findings add to an increasing number of investigations that highlight the importance of toxin B in the pathogenesis of CDAD. The mechanisms of nosocomial acquisition, persistence, and transmission of Clostridium are complex and not fully understood for all strain types. Other investigators have determined that greater sporulation and spore germination capacity along with increased antimicrobial resistance may contribute to the persistence of A + B + clonal isolates in the nosocomial setting. 10,38 40 The mechanisms by which A B + strains persist have not been widely studied. However, the increased resistance to macrolide, lincosamide and streptogramin (MLS) antibiotics and fluoroquinolones seen in our outbreak isolates is likely to contribute to persistence in both the hospital environment and the human gut. Ongoing surveillance at our hospital indicates the persistence of this strain even though the incidence rate remains low. Surveillance following the outbreak in The Netherlands demonstrated that their A B + isolates did not persist. These A B + strains are no longer found in patients at that institution (E. Kuiper, personal communication). Antibiotic resistance in A B + Clostridium An early investigation by Delmée and Avesani showed that serogroup F strains (A B + ) were usually susceptible to clindamycin and erythromycin. 41 Since that time the emergence of resistance to MLS antibiotics has been documented. Sixteen A B + isolates from the outbreak in The Netherlands were investigated for antibiotic resistance and all were resistant to clindamycin, erythromycin, and tetracycline and contained the ermb gene that encoded MLS resistance. 20 The isolates from the Canadian and Japanese outbreaks were not tested for antibiotic susceptibility. The A B + isolates from our hospital were also resistant to MLS antibiotics and contained the ermb gene. In addition our A B + strains were resistant to a number of newer classes of fluoroquinolones including ciprofloxacin, ofloxacin, levofloxacin, moxifloxacin, and gatifloxacin. To our knowledge this is the first description of fluoroquinolone resistance in this A B + strain type. Pituch et al. described the emergence of a clindamycin resistant clone of A B + C. isolates in Polish hospitals. 42 As C. is not routinely cultured in many hospital laboratories the true incidence of antimicrobial resistance among C. is not known. A current prospective study undertaken by the European C. study group is investigating antimicrobial resistance in C. and will generate data with regard to antibiotic resistance in A B + and other C. strain types. Prevalence of A B + Clostridium These toxin variant strains have been isolated from several countries 22,43,44 with varying prevalence rates. 34,45 47 For example, rates of 0.2% have been reported from a multicenter study in the USA where 2 of 102 Clostridium isolates from six clinical settings were A B In the UK prevalence rates of 3% have been described representing 43 isolates from 9 of 35 hospitals that submitted strains for typing to the Anaerobic Reference Laboratory in Cardiff. 48 In France, rates of 3% from 25 different hospitals in Paris were reported. 45 In contrast, A B + prevalence rates as high as 39% have been described in one Japanese study. 46 Furthermore, a recent study in Israel documented A B + C. rates of 56%. 49 A recent study in Argentina indicated that A B + strains have completely replaced A + B + strains over a four-year time period wherein A B + isolates increased from 12.5% in 2000 to 97.9% in Laboratory diagnosis of A B + Clostridium Most laboratory diagnostic tests for Clostridium are based on either faecal culture or direct toxin detection in faecal samples. The cell culture cytotoxicity assay is considered the gold standard for toxin detection and can detect picogram quantities of toxin after 48 hours. In addition cell culture can identify these A B + strains due to the different CPE induced by these strains as a result of their different glucosyltransferase substrate specificities. Enzyme immunoassays (EIAs) which can detect either TcdA and/or TcdB are the most commonly used diagnostic approach due to their ease of use and quick turn-around time although these methods are less sensitive compared to the cytotoxicity assay. Toxin A-specific EIAs use antibodies targeted against the CROP region of tcda, which is deleted in A B + strain types, which therefore fail to react in these assays. A recent survey carried out by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) study group on C. determined that among European laboratories performing C. toxin detection, 58% used ELISAs that can only detect toxin A and therefore toxin A B + strains would fail to be detected in these clinical settings. 50 Failure to use standardized C. diagnostic methods that include both toxin A and B could lead to significant under-reporting of C.. C. culture lacks specificity due to the presence of non-toxigenic isolates and takes 48 hours to complete. However, culture is the only way to type strains and to test for antimicrobial susceptibility and therefore the only way in which the emergence and spread of more virulent or resistant strains can be identified and monitored. Molecular analysis of A B + Clostridium Several different molecular typing methods have been applied to A B + variants including restriction endonuclease analysis (REA), S rrna ribotyping, 52 amplified fragment length polymorphism (AFLP), 53 multi-locus sequence typing (MLST), 54 and toxinotyping (PCR-RFLP). 15 A recent study by Lemee et al. applied MLST to 72 C. isolates from various hosts, geographic locations, PCR ribotypes, and toxigenic types. The study isolates included eight A B + isolates from the USA, Japan, and France cultured from human adults and children. All eight isolates belonged to the same sequence type and clustered in a very homogenous MLST

5 Toxin A B + Clostridium 9 phylogenetic lineage despite their origins from unrelated patients. Furthermore, a null allele for one of the seven housekeeping genes investigated further supported the low genetic diversity of these A B + types. 54 van den Berg et al. used AFLP and 16 23S rrna ribotyping to analyze 39 A B + strains from seven different countries. AFLP recognized two genotypes among the 39 isolates while two different ribotyping methods could differentiate two and three PCR ribotypes, respectively. All three methods exhibited similar discriminatory power when typing A B + strain types. 53 One international study has compared the relatedness of 23 A B + strains using three typing methods: serogrouping, 16 23S rrna PCR ribotyping, and REA. 43 This study documented a high degree of similarity between A B + strains from the USA and Europe. Twenty-one of the 23 isolates had a 1.8- kb deletion in the 3 0 end of tcda corresponding to toxinotype VIII and 20 of the 21 examined were PCR ribotype 017. REA was the most discriminatory typing method distinguishing 11 REA types among the 23 isolates. This method could further subtype the 20 PCR 017/serogroup F strains into six different REA types. PCR ribotyping and serogrouping identified four and three distinct types respectively. The remaining two A B + strains investigated in this international study were strain 8864, which was unique using all three typing methods. In addition a PCR ribotype 110/serogroup X/REA DA1 strain had the same toxin genotype as VPI but did not produce any toxin A in vitro. Concluding remarks The incidence of A B + Clostridium strains appears to be increasing worldwide. These strain types now represent a substantial number of C. isolates. These isolates may be under-reported due to the widespread use of diagnostics that detect TcdA only. Although further studies are required to determine whether the pathogenesis of toxin variant CDAD (A B + ) differs from that seen with toxin A + B + strains, clinicians should be aware of the risk of severe disease associated with these variant strain types. TcdB is likely to play a more prominent role in the pathogenesis of C. than previously considered, and future work should focus on the expression and regulation of this toxin as well as modulation of the host immune response. However the lack of suitable genetic tools to manipulate C. and lack of isogenic mutants may hinder progress. New therapeutic approaches for CDAD such as toxin binders, passive immunotherapy or active immunization through vaccination will now need to target both TcdA and TcdB. Acknowledgements The authors acknowledge the financial support provided by the National Institute on Aging, The Health Research Board, Ireland, The Mater Foundation and The Newman Scholarship Programme, University College Dublin. Conflict of interest: No conflict of interest to declare. References 1. Bartlett JG. Clinical practice. Antibiotic-associated diarrhea. N Engl J Med 2002;346: Kelly CP. Immune response to Clostridium infection. Eur J Gastroenterol Hepatol 1996;8: Kyne L, Kelly CP. Recurrent Clostridium diarrhoea. Gut 2001;49: Archibald LK, Banerjee SN, Jarvis WR. Secular trends in hospitalacquired Clostridium disease in the United States, J Infect Dis 2004;189: Eggertson L. Clostridium : by the numbers. CMAJ 2004;171: Morris AM, Jobe BA, Stoney M, Sheppard BC, Deveney CW, Deveney KE. Clostridium colitis: an increasingly aggressive iatrogenic disease? Arch Surg 2002;137: Pepin J, Valiquette L, Alary ME, Villemure P, Pelletier A, Forget K, et al. Clostridium -associated diarrhea in a region of Quebec from 1991 to 2003: a changing pattern of disease severity. CMAJ 2004;171: Dallal RM, Harbrecht BG, Boujoukas AJ, Sirio CA, Farkas LM, Lee KK, et al. Fulminant Clostridium : an underappreciated and increasing cause of death and complications. Ann Surg 2002; 235: Pepin J, Valiquette L, Cossette B. Mortality attributable to nosocomial Clostridium -associated disease during an epidemic caused by a hypervirulent strain in Quebec. CMAJ 2005;173: Warny M, Pepin J, Fang A, Killgore G, Thompson A, Brazier J, et al. Toxin production by an emerging strain of Clostridium associated with outbreaks of severe disease in North America and Europe. Lancet 2005;366: Pothoulakis C. Pathogenesis of Clostridium -associated diarrhoea. Eur J Gastroenterol Hepatol 1996;8: Pothoulakis C, Lamont JT. Microbes and microbial toxins: paradigms for microbial-mucosal interactions II. The integrated response of the intestine to Clostridium toxins. Am J Physiol Gastrointest Liver Physiol 2001;280:G Savidge TC, Pan WH, Newman P, O Brien M, Anton PM, Pothoulakis C. Clostridium toxin B is an inflammatory enterotoxin in human intestine. Gastroenterology 2003;125: Cohen SH, Tang YJ, Silva Jr J. Analysis of the pathogenicity locus in Clostridium strains. J Infect Dis 2000;181: Rupnik M, Braun V, Soehn F, Janc M, Hofstetter M, Laufenberg- Feldmann R, et al. Characterization of polymorphisms in the toxin A and B genes of Clostridium. FEMS Microbiol Lett 1997;148: Rupnik M, Avesani V, Janc M, von Eichel-Streiber C, Delmee M. A novel toxinotyping scheme and correlation of toxinotypes with serogroups of Clostridium isolates. J Clin Microbiol 1998;36: Rupnik M, Kato N, Grabnar M, Kato H. New types of toxin A- negative, toxin B-positive strains among Clostridium isolates from Asia. J Clin Microbiol 2003;41: Lyerly DM, Saum KE, MacDonald DK, Wilkins TD. Effects of Clostridium toxins given intragastrically to animals. Infect Immun 1985;47: Alfa MJ, Kabani A, Lyerly D, Moncrief S, Neville LM, Al-Barrak A, et al. Characterization of a toxin A-negative, toxin B-positive strain of Clostridium responsible for a nosocomial outbreak of Clostridium -associated diarrhea. J Clin Microbiol 2000;38: Kuijper EJ, de Weerdt J, Kato H, Kato N, van Dam AP, van der Vorm ER, et al. Nosocomial outbreak of Clostridium -associated diarrhoea due to a clindamycin-resistant enterotoxin A-negative strain. Eur J Clin Microbiol Infect Dis 2001;20: Sambol SP, Merrigan MM, Lyerly D, Gerding DN, Johnson S. Toxin gene analysis of a variant strain of Clostridium that causes human clinical disease. Infect Immun 2000;68: van den Berg RJ, Legaria MC, de Breij ER, van der Vorm ER, Brazier JS, Kuijper EJ. Introduction of TcdA-negative, TcdBpositive Clostridium in a general hospital in Argentina.

6 10 D. Drudy et al. 15th European Congress of Clinical Microbiology and Infectious Diseases (ECCMID), Copenhagen, Denmark, April al-barrak A, Embil J, Dyck B, Olekson K, Nicoll D, Alfa M, et al. An outbreak of toxin A negative, toxin B positive Clostridium -associated diarrhea in a Canadian tertiary-care hospital. Can Commun Dis Rep 1999;25: Johnson S, Kent SA, O Leary KJ, Merrigan MM, Sambol SP, Peterson LR, et al. Fatal pseudomembranous colitis associated with a variant Clostridium strain not detected by toxin A immunoassay. Ann Intern Med 2001;135: Limaye AP, Turgeon DK, Cookson BT, Fritsche TR. Pseudomembranous colitis caused by a toxin A( ) B(+) strain of Clostridium. J Clin Microbiol 2000;38: Borriello SP, Wren BW, Hyde S, Seddon SV, Sibbons P, Krishna MM, et al. Molecular, immunological, and biological characterization of a toxin A-negative, toxin B-positive strain of Clostridium. Infect Immun 1992;60: Torres JF. Purification and characterisation of toxin B from a strain of Clostridium that does not produce toxin A. J Med Microbiol 1991;35: Soehn F, Wagenknecht-Wiesner A, Leukel P, Kohl M, Weidmann M, von Eichel-Streiber C, et al. Genetic rearrangements in the pathogenicity locus of Clostridium strain 8864 implications for transcription, expression and enzymatic activity of toxins A and B. Mol Gen Genet 1998;258: Lyerly DM, Barroso LA, Wilkins TD, Depitre C, Corthier G. Characterization of a toxin A-negative, toxin B-positive strain of Clostridium. Infect Immun 1992;60: Delmée M, Avesani V. Virulence of ten serogroups of Clostridium in hamsters. J Med Microbiol 1990;33: Depitre C, Delmee M, Avesani V, L Haridon R, Roels A, Popoff M, et al. Serogroup F strains of Clostridium produce toxin B but not toxin A. J Med Microbiol 1993;38: von Eichel-Streiber C, Zec-Pirnat I, Grabnar M, Rupnik M. A nonsense mutation abrogates production of a functional enterotoxin A in Clostridium toxinotype VIII strains of serogroups F and X. FEMS Microbiol Lett 1999;178: Chaves-Olarte E, Low P, Freer E, Norlin T, Weidmann M, von Eichel-Streiber C, et al. A novel cytotoxin from Clostridium serogroup F is a functional hybrid between two other large clostridial cytotoxins. J Biol Chem 1999;274: Sato H, Kato H, Koiwai K, Sakai C. A nosocomial outbreak of diarrhea caused by toxin A-negative, toxin B-positive Clostridium in a cancer center hospital. Kansenshogaku Zasshi 2004;78: Kyne L, Warny M, Qamar A, Kelly CP. Association between antibody response to toxin A and protection against recurrent Clostridium diarrhoea. Lancet 2001;357: Kyne L, Warny M, Qamar A, Kelly CP. Asymptomatic carriage of Clostridium and serum levels of IgG antibody against toxin A. N Engl J Med 2000;342: Riegler M, Sedivy R, Pothoulakis C, Hamilton G, Zacherl J, Bischof G, et al. Clostridium toxin B is more potent than toxin A in damaging human colonic epithelium in vitro. J Clin Invest 1995;95: Wilcox MH, Fawley WN. Hospital disinfectants and spore formation by Clostridium. Lancet 2000;356: Gerding DN. Clindamycin, cephalosporins, fluoroquinolones, and Clostridium -associated diarrhea: this is an antimicrobial resistance problem. Clin Infect Dis 2004;38: Gaynes R, Rimland D, Killum E, Lowery HK, Johnson 2nd TM, Killgore G, et al. Outbreak of Clostridium infection in a long-term care facility: association with gatifloxacin use. Clin Infect Dis 2004;38: Delmée M, Avesani V. Correlation between serogroup and susceptibility to chloramphenicol, clindamycin, erythromycin, rifampicin and tetracycline among 308 isolates of Clostridium. J Antimicrob Chemother 1988;22: Pituch H, Van Belkum A, Van Den Braak N, Obuch-Woszczatynski P, Verbrugh H, Meisel-Mikolajczyk F, et al. Recent emergence of an epidemic clindamycin-resistant clone of Clostridium among Polish patients with Clostridium -associated diarrhea. J Clin Microbiol 2003;41: Johnson S, Sambol SP, Brazier JS, Delmée M, Avesani V, Merrigan MM, et al. International typing study of toxin A-negative, toxin B- positive Clostridium variants. J Clin Microbiol 2003; 41: Pituch H, van den Braak N, van Leeuwen W, van Belkum A, Martirosian G, Obuch-Woszczatynski P, et al. Clonal dissemination of a toxin-a-negative/toxin-b-positive Clostridium strain from patients with antibiotic-associated diarrhea in Poland. Clin Microbiol Infect 2001;7: Barbut F, Lalande V, Burghoffer B, Thien HV, Grimprel E, Petit JC. Prevalence and genetic characterization of toxin A variant strains of Clostridium among adults and children with diarrhea in France. J Clin Microbiol 2002; 40: Komatsu M, Kato H, Aihara M, Shimakawa K, Iwasaki M, Nagasaka Y, et al. High frequency of antibiotic-associated diarrhea due to toxin A-negative, toxin B-positive Clostridium in a hospital in Japan and risk factors for infection. Eur J Clin Microbiol Infect Dis 2003;22: Lyerly DM, Neville LM, Evans DT, Fill J, Allen S, Greene W, et al. Multicenter evaluation of the Clostridium TOX A/B TEST. J Clin Microbiol 1998;36: Brazier JS, Stubbs SL, Duerden BI. Prevalence of toxin A negative/b positive Clostridium strains. J Hosp Infect 1999; 42: Samra Z, Talmor S, Bahar J. High prevalence of toxin A-negative toxin B-positive Clostridium in hospitalized patients with gastrointestinal disease. Diagn Microbiol Infect Dis 2002;43: Barbut F, Delmée M, Brazier JS, Petit JC, Poxton IR, Rupnik M, et al. A European survey of diagnostic methods and testing protocols for Clostridium. Clin Microbiol Infect 2003; 9: Clabots CR, Johnson S, Bettin KM, Mathie PA, Mulligan ME, Schaberg DR, et al. Development of a rapid and efficient restriction endonuclease analysis typing system for Clostridium and correlation with other typing systems. J Clin Microbiol 1993;31: Stubbs SL, Brazier JS, O Neill GL, Duerden BI. PCR targeted to the 16S 23S rrna gene intergenic spacer region of Clostridium and construction of a library consisting of 116 different PCR ribotypes. J Clin Microbiol 1999;37: van den Berg RJ, Claas EC, Oyib DH, Klaassen CH, Dijkshoorn L, Brazier JS, et al. Characterization of toxin A-negative, toxin B- positive Clostridium isolates from outbreaks in different countries by amplified fragment length polymorphism and PCR ribotyping. J Clin Microbiol 2004;42: Lemee L, Dhalluin A, Pestel-Caron M, Lemeland JF, Pons JL. Multilocus sequence typing analysis of human and animal Clostridium isolates of various toxigenic types. J Clin Microbiol 2004;42: