Clostridium difficile - Semantic Scholar

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 2006, p. 7311–7323 0099-2240/06/$08.00⫹0 doi:10.1128/AEM.01179-06 Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Vol. 72, No. 11

The Mosaic Nature of Intergenic 16S-23S rRNA Spacer Regions Suggests rRNA Operon Copy Number Variation in Clostridium difficile Strains䌤 Nourkhoda Sadeghifard,1† Volker Gu ¨rtler,2* Michael Beer,1 and Robert J. Seviour1 Biotechnology Research Centre, La Trobe University, Bendigo, Victoria 3552, Australia,1 and Microbiology Department, Austin Hospital, Studley Road, Heidelberg, Melbourne, Victoria 3084, Australia2 Received 22 May 2006/Accepted 7 September 2006

Clostridium difficile is a major spore-forming environmental pathogen that causes serious health problems in patients undergoing antibiotic therapy. Consequently, reliable and sensitive methods for typing individual strains are required for epidemiological and environmental studies. Ribotyping is generally considered the best method, but it fails to account for sequence diversity which might exist in intergenic 16S-23S rRNA spacer regions (ISRs) within and among strains of this organism. Therefore, this study was undertaken to compare the sequence of each individual ISR in five strains of C. difficile to explore the extent of this diversity and see whether such information might provide the basis for more sensitive and discriminatory strain typing methods. After targeted PCR amplification, cloning, and sequencing, the diversity of the ISRs was used as a measure of rRNA operon copy number. In C. difficile strains 630, ATCC 43593, A, and B, 11, 11, 7, and 8 ISR length variants, respectively, were found (containing different combinations of sequence groups [i to xiii]), suggesting 11, 11, 7, and 8 rrn copies in the respective strains. Many ISRs of the same length differed markedly in their sequences, and some of these were restricted in occurrence to a single strain. Most of these ISRs did not contain any tRNA genes, and only single copies of the tRNAAla gene were found in those that did. The presence of ISR sequence groups (i to xiii) varied between strains, with some found in one, two, three, four, or all five strains. We conclude that the intergenic 16S-23S rRNA spacer regions showed a high degree of diversity, not only among the rrn operons in different strains and different rrn copies in a single strain but also among ISRs of the same length. It appears that C. difficile ISRs vary more at the inter- and intragenic levels than those of other species as determined by empirical comparison of sequences. The precise characterization of these sequences has demonstrated a high level of mosaic sequence block rearrangements that are present or absent in multiple strain-variable rrn copies within and between five different strains of C. difficile. and the cdtA and cdtB genes, encoding TcdA and TcdB or CDT, respectively, in stool samples, have been described (6, 22, 38, 52). For C. difficile strain identification, essential for epidemiological studies, culturing and molecular characterization are necessary. Many different typing methods have been applied to strains of this organism, including repetitive sequence-based PCR using repetitive extragenic palindromic primers, multilocus sequence typing, and pulsed-field gel electrophoresis (37, 47, 60), but there seems to be general, if not universal, agreement that PCR ribotyping, where the variable intergenic 16S23S rRNA spacer region (ISR) is amplified (26, 29, 31), is the most attractive method for strain identification and has been used most frequently for this purpose (7, 10, 21, 42, 62, 65). Bacterial typing of both pure cultures and environmental samples by fragment analysis of the ISR (PCR-ribotyping) has become a popular method because the ISR is more variable in both its length and sequence than the 16S rRNA gene (1). However, PCR-ribotyping will not reveal the true extent of this ISR diversity, since identical-length ISRs of different sequences may produce identical ISR fragment patterns after electrophoresis on agarose gels (25). Sequence analysis of the ISR has revealed the mosaic nature of the ISR in Acinetobacter baylyi (12), Haemophilus parainfluenzae (23, 54), and Photobacterium damselae (48) typified by the presence or absence of sequence blocks up to ⬃100 nucleotides in length between

Clostridium difficile is a major spore-forming environmental pathogen that is found in soil and is nosocomially acquired during outbreaks of diarrhea in hospital patients who are immunocompromised or for whom antibiotics are prescribed (43, 53, 68). Suggestions are that as many as 300,000 patients suffer from C. difficile-associated disease each year in the United States (68). Pathogenesis is associated with several toxins produced by toxigenic strains of C. difficile (22, 52, 55). Genes encoding TcdA, an enterotoxin, and TcdB, a cytotoxin, are both located within the pathogenicity locus and are both diverse, while the binary toxin CDT is located elsewhere on the chromosome. Most strains synthesize only TcdA and less frequently TcdB, but some of these also synthesize CDT, whose role in the disease symptoms is still unclear (52, 55). In the diagnosis and typing of C. difficile, culture techniques are seldom used, which may result in underestimating the detection of this organism (19). Instead, several immunological (2, 8, 65, 66) and PCR-based methods, including those directed at identifying and in some cases quantifying the tcdA and tcdB genes

* Corresponding author. Mailing address: Microbiology Department, Austin Hospital, Studley Road, Heidelberg, Melbourne, Victoria 3084, Australia. Phone: 61 3 9496 3136. Fax: 61 3 9457 2590. E-mail: [email protected]. † Present address: Microbiology Department, Ilam University of Medical Sciences, Ilam, Iran. 䌤 Published ahead of print on 15 September 2006. 7311

7312

SADEGHIFARD ET AL.

multiple ISR copies of a genome. Furthermore, the wholegenome sequence analysis of C. difficile 630 has also revealed a mosaic genome characterized by a larger number of mobile genetic elements than in other whole bacterial genomes (59). Eleven copies of the rrn operon are known to occur in C. difficile strain 630 from whole-genome sequencing and annotation (59). In this study the sequence data of these 11 rrn operons from strain 630 were analyzed to confirm previous data which showed that considerable variations exist in both the sizes and sequences of ISRs found in different C. difficile strains, including 630 (26). However, little is known about rrn copy numbers and intragenomic ISR sequence variability in other strains of this organism. Therefore, the aim of this study was to determine the extent of ISR sequence diversity in three other strains of C. difficile compared to that in C. difficile 630 and to see if such diversity might provide some basis for moreprecise determination of rrn copy number and identification of individual C. difficile strains.

APPL. ENVIRON. MICROBIOL. 20 ␮l. After an initial ramp of 96°C, samples were submitted to 25 cycles of 96°C for 10 s, 50°C for 55 s, and 60°C for 4 min. Sequence data analysis. The vector sequence was trimmed from all sequences using the program SeqMan 4.05 (DNASTAR, Inc.), and with the same program, the sequences were assembled and checked for any conflicts between the overlapping sequences generated with the forward and reverse primers. The Australian National Genomic Information Service was used to perform BLASTn (3), to do text searches for ISR DNA sequences and perform sequence alignments of ISR DNA sequences (between regions 4 and 5) using Clustal W (63), and to obtain restriction enzyme fragment sizes of the ISR (between regions 4 and 5) for TaqI using tacg (which analyzes a DNA sequence for restriction enzyme sites) (41). The ISR sequences were then annotated, separated, stored, and sorted using MacClade (39), FileMakerPro 8, Sequin 6.2 (http://www.ncbi.nlm.nih.gov /Sequin/index.html), and CLC Free Workbench 2.5.2 (www.clcbio.com). The sorted information was analyzed further with Microsoft Excel 2003. Maps of the ISRs were imported from CLC Free Workbench 2.5.2 into Illustrator 10. The C. difficile 630 genome sequence was generated by the Sanger Institute Pathogen Sequencing Unit in XBase (14, 59). Nucleotide sequence accession numbers. The sequences reported in this paper have been assigned GenBank accession numbers DQ487223 and DQ487224, DQ487228 to DQ487230, DQ487232 and DQ487233, DQ487236 to DQ487243, DQ487245 to DQ487249, DQ487251 to DQ487260, DQ487262, DQ487264 to DQ487268, and DQ487270 to DQ487276.

MATERIALS AND METHODS Organism and growth conditions. Clostridium difficile strain ATCC 43593 and two clinical strains (A and B) isolated from the Austin Hospital, Melbourne, Australia, and identified by culture, latex agglutination (Oxoid), and cytotoxin cell culture assay (9, 26) were selected for this study. The strains were grown on brain heart infusion agar medium supplemented with 10% horse blood in an anaerobic atmosphere at 37°C for 48 h. DNA extraction. Chromosomal DNA was extracted from colonies of C. difficile by using the UltraClean soil DNA kit (Mo Bio) according to the manufacturer’s instructions. The DNA obtained was resuspended in Tris-EDTA buffer and electrophoresed on a 1% agarose gel to determine its integrity before being stored at ⫺20°C until required. Amplification of the 16S-23S rRNA intergenic spacer region. PCR amplification of the ISR was carried out using two universal primers complementary to conserved regions in the 16S and 23S rRNA genes, as recommended previously (31). The forward and reverse primer sequences are located at nucleotide positions 1477 to 1493 on the 16S rrnA gene of C. difficile strain 630 (region 4, 5⬘-GGC TGG ATC ACC TCC TT-3⬘; region 5, 5⬘-TAG TGC CAA GGC ATC CGC CCT-3⬘) complementary to positions 21 to 41 on the 23S rRNA gene, respectively (31). DNA templates were amplified in a total reaction volume of 50 ␮l containing 2.5 U of AmpliTaq Gold thermostable polymerase (Roche), 50 pmol of each primer, 200 ␮M of each deoxynucleotide, 1.5 mM MgCl2, 10 mM Tris-HCl (pH 8.3), and 50 mM KCl. Amplification was carried out in a GeneAmp 2400 thermal cycler (Applied Biosystems) with denaturation at 94°C for 10 min, followed by 30 cycles according to the following program: 94°C for 1 min, 54°C for 1 min, 72°C for 2 min, and a final extension of 10 min at 72°C to complete partial polymerizations. The resulting amplification products were purified with a rapid PCR purification kit (Marligen; Biosciences) according to the manufacturer’s instructions and the products analyzed on a 2% agarose gel, stained with ethidium bromide, and viewed on a UV transilluminator. Cloning, sequencing, and RFLP of the 16S-23S rRNA ISR. The purified PCR amplification products were cloned into the pGEM-T Easy Vector System II (Promega) in accordance with the manufacturer’s instructions. Recombinant colonies were picked off and grown in 3 ml of LB medium containing ampicillin (100 ␮g/ml) for 14 to 16 h at 37°C. Plasmid DNA was extracted from the clones using the QIAprep Spin Miniprep kit (QIAGEN), again following the manufacturer’s instructions. A total of 126 clones were screened by digestion with the restriction endonuclease EcoRI or by PCR amplification using the M13f and M13r primers and subsequent digestion with TaqI. Digestion products from the PCR products were run on 1% agarose gels to check that clones contained an insert and to reveal any possible variations in sizes of these inserts. In the case of TacI digestion, products were separated on 10% polyacrylamide in TBE (40 mM Tris, 20 mM borate and 1 mM EDTA) at 200 V for 1 h. The resulting restriction fragment length polymerases (RFLPs) were analyzed by the database management program GelCompar II upgrade 3.5 (Applied Maths, Belgium). Sequencing of selected clones was carried out using an ABI DNA sequencer model 377a (Applied Biosystems) with Big-Dye terminator kits (Applied Biosystems). Each sequencing reaction contained 6 ␮l of template DNA, 2 ␮l of primer (1.6 ␮M), and 6 ␮l of Big-Dye reaction mix in a total reaction volume of

RESULTS 16S-23S rRNA gene ISR sequence characterization for strains of C. difficile. When searched for highly conserved rrn sequences (i.e., regions 4 and 5) (31), the published whole genome of C. difficile strain 630 was found to contain 11 rrn operons with ISRs each of variable length and sequence (Table 1). After the ISR regions of C. difficile ATCC 43593 and clinical strains A and B were PCR amplified and cloned, a total of 126 clones were screened. Of these ISR clones generating different EcoRI digestion patterns (consisting of 19 from ATCC 43593, 16 from clinical strain A, and 12 from clinical strain B), 47 were sequenced. The data showed that these sequences were complete, except for those from four of the clones, where an incomplete insert had been ligated. Complete sequence analysis of the full set of 55 ISR sequences shown in Table 1 (derived from (i) 43 remaining ISR clones [17 from C. difficile strain ATCC 43593, 16 from clinical strain A, and 10 from clinical strain B], (ii) 11 ISRs from the whole genome of C. difficile strain 630, and (iii) a single ISR from strain ATCC 43597 [7]) revealed 26 different rrn operon sizes or sequence arrangements (Table 1). Their ISR sizes (from region 4 to region 5), including those derived from whole-genome sequence data from strain 630 (Tables 1 and 2), varied from 238 to 566 bp, with 11, 7, 8, and 11 different ISR sizes recognizable in C. difficile ATCC strain 43593, strain A, strain B, and strain 630, respectively (Table 2). These different-size ISRs were sequence mosaics (see next sections); therefore, they could correspond only to alleles with different genomic locations, the total for each strain corresponding to an estimate of the number of rrn operons per genome. C. difficile ISR sequence alignments. In order to determine consensus homology between the 55 ISR sequences from five C. difficile strains, 200 nucleotides of homologous flanking regions in the 16S and 23S rRNA genes were included in the alignment. In order to aid alignment, detection, and further analyses of the short blocks of sequence similarity that were variably present in only certain clones, the aligned sequences were arbitrarily divided into 13 different groups (i to xiii), as summarized in Fig. 1 and 2. Alignment groups i to xiii ranged

VOL. 72, 2006

MOSAIC 16S-23S rRNA SPACER REGIONS IN C. DIFFICILE

7313

TABLE 1. Properties of 43 ISR clones from isolates A, B, and ATCC 43593, 11 ISR sequences from the whole genome of isolate 630, and 1 ISR sequence from ATCC 43597 showing the distribution of variants in each of the 13 ISR groups (i to xiii) shown in Fig. 1 and 2a to da

a Symbols: ⴱ, not in 630; ¶, ISR nucleotide sequence length in bp from the start of region 4 to the end of region 5, which includes the end of the 16S rRNA gene (30 bp) and the beginning of the 23S rRNA gene (36 bp); ‡, this sequence (from strain ATCC 43597) was obtained directly from GenBank (7); it was not included in the comparative analysis of rrn operons between strains and is included only to demonstrate that other sequence variants may be found when other strains are investigated; †, the allele designation for isolate 630 were named rrnA to -K and correlated by size and sequence to the other strains (A, B, and ATCC 43593); §, the C. difficile 630 genome sequence was generated by the Sanger Institute Pathogen Sequencing Unit in XBase (14, 59).

in size from 14 to 77 bp (Table 1). Variations in each of the alignment groups and their respective sequence variants are detailed in Table 1 and Fig. 1 and 2. For quick reference to different sequence groups, Fig. 2 is color coded so that highly

homologous blocks are blue and blocks with low sequence homology are red. Some groups of the alignment were similar in more than 50% of sequences (increasing shades of blue for increasing homology in groups i, ii, vii, x, xi, xii, and xiii). In

7314

SADEGHIFARD ET AL.

APPL. ENVIRON. MICROBIOL.

TABLE 2. Estimate of the number of rrn operons per genome from numbers of allele variants detected in C. difficile isolates A, B, 630, and ATCC 43593a

a

For definitions of symbols, see footnote a of Table 1.

stark contrast, other groups of the aligned sequences from the five strains were similar in less than 50% of sequences (increasing shades of red for decreasing homology in groups ii, iii, iv, v, vi, and ix). ISR sequence similarities within and between strains of C. difficile. Sequence analysis revealed that some ISRs of the same length in the five different strains shared identical sequences (Table 1 and Fig. 2). For example, the sequence of ISR clones AT-10, A-2, A-8, B-10, and B-19 were identical (rrnB), and clones A-1, AT-17, and B-21 were identical (rrnK1). Furthermore, the sequences of some ISRs of the same length in a single strain but with different sense were also identical (e.g., clones A-7 and A-16 [rrnG1], AT-5 and AT-18 [rrnJ], and A-2 and A-8 [rrnB]). Finally, in strain 630, operons rrnJ and rrnB had identical ISR sequence groups (i-1, ii-2, iii-1, vi-7, vii-1, viii-3, x-1, xi-1, xii-1, and xiii-1), and operons rrnC and rrnA differed in only one sequence group (i-1, ii-1, iii-1, iv-1, v-1, vi-1, vii-1, viii-1 ix-1[rrnC] or ix-2 [rrnA], x-1, xi-1, xii-1, and xiii-1). Therefore, it is possible that some alleles that were found to be identical from a single strain could be located within different rrn operons (Table 2). ISR sequence differences within and between strains of C. difficile. On the other hand, many ISRs of the same or very similar lengths had markedly different sequences within or between these strains (Table 1) and were thus designated as residing at different chromosomal locations. For example, the

respective ISR sequence groups for rrnE and rrnE1 were xi-2 and xi-1; those for rrnF and rrnF1 were xi-1 and xi-2; those for rrnI1, rrnI2, rrnI3, and rrnI4 were iii-2, vi-3, and xi-2; iii-4, vi-4, and xi-1; iii-4, vi-4, and xi-2; and iii-5, vi-5, and xi-1; those for rrnK, rrnK1, and rrnK2 were iii-6, viii-2, and xi-4); iii-1, viii-1, x-1, and xi-1; and iii-1, viii-3, x-1, and xi-1; those for rrnG and rrnG1 were xi-1 and xi-2; and those for rrnN, rrnO, and rrnP were i-2 and xi-5; i-1 and xi-6; and i-1 and xi-5. Mosaics of ISR sequence groups suggests operon number differences between strains. The different sequence types within each sequence group (i to xiii) vary significantly in sequence and length (Table 1; Fig. 1 and 2). The differences are so great that the combinations of the sequence groups among the 55 ISR sequences suggest a unique chromosomal origin corresponding to 26 different rrn operons in four C. difficile strains (Table 2). An estimation of the number of operons in each of these strains has been made in Table 2 with the qualification that it is not certain that all operons have been isolated even though a large number of clones were screened, increasing the chance of complete detection. The number of operons could be an underestimate, because in some cases two or more sequences were detected for unique operons (e.g., three copies of rrnG1 in strain A and four copies of rrnJ in strain ATCC 43593). Therefore, the ISR mosaic and operon copy number differences between C. difficile strains needs further investigation by parallel whole-genome sequencing (59),

VOL. 72, 2006


7315

FIG. 1. Complete map and GenBank annotation of 16S-23S ISR group variants from four C. difficile isolates. The ISR has been divided into 13 groups (i to xiii) with the variants for each group (1 to n, listed in Table 1) shown together by length and sequence block. See Table 1 for variant numbering nomenclature. The map on the top shows the primer sites corresponding to regions 4 and 5 (31), annotations for groups i to xiii (green blocks), rRNA (left red block corresponds to 16S, and right left block corresponds to 23S), tRNAAla (blue block), and the TaqI sites in groups iv and xiii (see Fig. 3). The blue graph at the bottom of the figure shows the sequence conservation within each of the annotated groups.

oligonucleotide microarray analysis (11, 16, 18, 24, 59), or multiplex ligation-dependent probe amplification (32) from multiple C. difficile strains. Presence of tRNA genes in the ISRs of C. difficile. Most of the ISRs in the three strains examined here lacked tRNA genes. Those that contained them possessed just a single copy of the tRNAAla gene, as seen with ATCC strain 43593 with two ISRs (clones AT-6 and AT-13), strain A with three ISRs (clones A-3, A-5, and A-15), and strain B with two ISRs (clones B-11 and B-15). In contrast, in C. difficile strain 630, its whole-genome sequence data revealed that five rrn operons, rrnA, rrnC, rrnD, rrnE, and rrnF, each contained a single copy

of the tRNAAla gene. These sequences were identical in 75 bp (groups iii-1 and iv-1 [positions 76 to 150]), and one had a single nucleotide difference (A-15 [position 96]). When a BLASTn search was performed with this sequence, many other Clostridium species (56–58), including C. tetani (11), were found to contain within their ISRs a sequence of high homology (90 to 98%). Differentiating ISRs by RFLP typing. The feasibility of distinguishing between C. difficile ISRs of identical size but differing in their sequences based on their restriction fragment length polymorphisms was examined (Fig. 3). Differences in the lengths and sequences of some of the ISRs seen in strains

SADEGHIFARD ET AL.

FIG. 2. Sequence alignments of variants for ISR groups i to iv (a), v to viii (b), ix to xi (c), or xii and xiii (d), shown together in Fig. 1. The sequence names are the same as in Fig. 1 and Table 1, and the variant numbering (Table 1) is shown at the right side of each group. The sequence alignment is shaded in increasing levels of blue, corresponding to increasing homology above 50% conservation, and the decreasing shade of red corresponds to decreasing homology, less than 50% conservation. The numbering (shown at the top of the alignment) starts at the beginning of region 4 (Fig. 1), but only the ISR is included (from start to finish). In the sequence alignments, a dot represents nucleotide identity to the top consensus sequence and a dash represents a gap or absence of the respective nucleotide in the consensus sequence. The blue graph at the bottom of the figure shows the sequence conservation within each of the annotated groups.

7316 APPL. ENVIRON. MICROBIOL.


FIG. 2—Continued.

VOL. 72, 2006

7317

SADEGHIFARD ET AL.


FIG. 2—Continued.

7318


7319

FIG. 2—Continued.

VOL. 72, 2006

7320

SADEGHIFARD ET AL.


FIG. 3. The variable TaqI site within group variants xiii-1, iv--, and iv-1 for the differentiation of rrnE1 sequence variants. From left to right: phylogram; electrophoresis composite compiled in GelCompar II of TaqI ISR restriction fragments; strain, clone, group, allele, size (bp) of regions 4 and 5, and TaqI restriction fragment sizes in bp; maps of TaqI restriction fragments found in each of the six ISRs shown in the electrophoresis composite. Mutations in region xiii at position 492 responsible for the presence or absence of the TaqI site are shown as follows: ⴱ, deletion at 492 (TaqI site absent); †, G at 492 (TaqI site present); ¶, A at 492 (TaqI site absent).

A and B were also often very small. For example, rrnE1 from clones A-3, A-15, and B-11 (strains A and B, respectively) had ISRs of 502, 502, and 503 nucleotides in length (Fig. 3). Close examination of the ISR sequences revealed that the size difference between clones from strains A and B was due to a nucleotide deletion at position 492 of group xiii in clones A-15 and A-3. This single-nucleotide difference enabled strains A and B to be distinguished by the ISR from allele rrnE1 with Taq1 restriction enzyme digestion, recognizing the cut site T⬘CGA, which was present in clone B-11 but was absent in A-3 and A-15 (Fig. 3). For comparison, digestion patterns from several other clones that could also be distinguished by size and absence of the TaqI site (clones A-5, A-6, and B-12) are included in Fig. 3. Oligonucleotide design based on linkage of conserved pairs of group variants to contiguous ISRs. To distinguish between strains, the data shown in Table 2 identify the following five strain-dependent sequence group types that can be used for the design of mosaic oligonucleotides: (i) sequences found in only one strain (e.g., iii-6 and xi-4 in strain 630); (ii) sequences found in two strains (e.g., xi-6 in strains A and B); (iii) sequences found in three strains (e.g., the whole of rrnD is found in strains A, B, and 630); (iv) sequences found in four strains (e.g., the whole of rrnB is found in strains A, B, ATCC 43593, and 630); (v) sequence block xii, which is found in all five strains (A, B, ATCC 43593, ATCC 43597, and 630). The mosaic nature of the ISR (Table 2) demonstrates 26 different rrn alleles in 5 different strains. When the 13 sequence blocks in these 26 unique alleles were examined, different combinations were together on different alleles (Table 2). The first class of sequence groups (Table 2) contained two or more variants that were linked only with two or more variants from one or more other groups (the only example of this class of sequence groups was the linkage of variants viii-3, vi-7, and ii-2 in all four strains on alleles rrnB, rrnJ, rrnJ1, and rrnK2). The second class of sequence groups (Table 2) contained variants linked with a number of different variants in another sequence group. This was most evident in the conserved or moderately conserved sequence groups (i, ii, vi, vii, xii, and xiii). The sequence group variant vi-7 is one of many examples that was

linked to different sequence group variants (vii-1 or vii-2; viii-1, viii-2, or viii-3; x-2 or x-1; xi-1, xi-2, xi-4, xi-5, or xi-6; and xiii-1 or xiii-2). The variability revealed by this analysis of each of the sequence groups shows clearly the mosaic nature of the ISRs in C. difficile, which has been described for many other bacterial species (see Discussion for references). Apart from describing the widespread molecular biological phenomenon of mosaic 16S-23S ISR sequences of the rrn operon, the practical implications for these sequence group variants are in designing the following types of oligonucleotide for use in PCR and microarrays. (i) The conserved pairs of sequence variants show the most potential for revealing sequences that might be used as oligonucleotides for the detection of all C. difficile strains, because many of these sequences are highly conserved in C. difficile strains. (ii) Compared with highly variable alignment groups (e.g., iii, vi, and xi), there are fewer conserved sequence group pair combinations because fewer variants occur within these alignment groups, thereby reducing the number of possible combinations. (iii) The number of combinations between the highly variable alignment group variants will be much greater than the number of conserved sequence group pair combinations shown in Table 2. Combinations of these highly variable sequences can therefore be used for the detection of specific strains, groups of strains, specific operons, and groups of operons. DISCUSSION The ISR sequence data for the five C. difficile strains examined here show that considerable variation exists between them in (i) their rrn copy numbers; (ii) the absolute presence or absence of sequence block types; and (iii) the association of multiple sequence block types with specific rrn operons. The amount of variation seen in the five C. difficile strains from this study appears to be considerably greater than in all other bacterial species so far examined. Bacterial rRNA (rrn) operons are generally organized in the order of 16S rrnA-ISR-23S rrnA-ISR-5S rrnA, and their copy numbers vary considerably between 1 and 15 (31, 45, 64, 67, 69). In addition, the 16S-23S ISRs are known to vary in both

VOL. 72, 2006


their length and sequence, and consequently this region has been used with some success to differentiate between closely related species and different strains within a single species (31). This study aimed to see to what extent ISR sequences differed among strains of C. difficile and between the different operons in a single strain. Three strains (ATCC 43593 and two clinical strains [A and B]) were analyzed and the data compared with those for C. difficile strain 630, whose whole-genome sequence contains 11 rrnA operons. Their sequences have been deposited in GenBank. If each generated ISR clone of a different size, different sequence, or different sense is assumed to represent a single rrn operon, it can be deduced that strains A and B and ATCC strain 43593 contained 7, 8, and 11 operons, respectively. The rrn copy number is not always conserved in different strains of the same species. For example, Streptococcus thermophilus contains six rrn operons, but some strains have been reported to have only five rrn copies (49). Similarly, rrn copy numbers of 7, 8, and 9 occur within different Vibrio cholerae strains (17, 36). Recombination events between rrn operons may be the reason for rrn operon copy number variations among strains of the same species (1, 25, 28, 36). It is still not clear what benefit there might be to an organism in having multiple rrn copies. Klappenbach et al. (34, 35) thought it might facilitate an organism’s ability to adapt better to environmental change, and some evidence has been presented suggesting that from soil organisms, those with higher rrn copy numbers dominated in soil communities exposed to complex carbon sources like pesticides. These strains also grew faster in complex media, while slow growers generally had lower rrn copy numbers, suggesting that rrn copy number is growth rate related (34). Genes coding for tRNAAla are present only in the ISRs of operons rrnA, rrnC, rrnD, rrnE, and rrnF in C. difficile strain 630, and all are coded by the positive-sense DNA strand. They were detected also in ISR clones AT-6, AT-13, A-3, A-5, A-15, B-11, and B-15. The known number of tRNA genes occurring in bacterial rrn operons varies from zero to five (4, 5, 12, 17, 27, 44, 46, 48, 50, 51). However, most of the ISRs in these C. difficile strains did not contain any tRNA genes. In some bacteria, ISRs encoding different tRNAs have been reported, as seen for Vibrio cholerae, Vibrio mimicus, and Photobacterium damselae, where tRNAGlu, tRNALys, tRNAVal, and tRNAAla were revealed (17, 48). The sequence data analyzed from the 55 ISRs within these 5 C. difficile strains revealed they had a mosaic-like structure, as was also reported for the ISRs in Salmonella enterica, Haemophilus parainfluenzae, Staphylococcus aureus, Vibrio cholerae, Vibrio mimicus, Vibrio parahaemolyticus, Myxobacterium spp., and Acinetobacter baylyi (4, 5, 12, 17, 25, 27, 44, 46, 48, 50, 51). The highly mosaic nature of the C. difficile ISR is entirely consistent with the recent finding that the complete genome of C. difficile is also mosaic in nature, with 11% of the genome comprised of mobile genetic elements that in many cases contain repetitive sequences (59). Twenty-six different types of ISRs existed among these five strains of C. difficile. When the 16S rRNA and 23S rRNA flanking regions were excluded from the analyses, the resulting data showed that all ISRs shared sequences of 44 bp at the 5⬘ end and of 136 bp at the 3⬘ end (ii-1 was specific for rrnA, -C, -D, -E, and -F, and ii-2 was specific for the remainder of the

7321

alleles). In V. parahaemolyticus, all six different ISRs share a 180-bp sequence at the 3⬘ end (40), whereas in P. damselae, a 21-bp sequence was shared at their 3⬘ end (48). The mosaic nature of the ISR seen in all these organisms is analyzed in a unique way in this study, because (i) all the variants are analyzed together, (ii) mosaic ISRs are used to deduce operon variants, and (iii) operon variants are used to deduce rrn copy number within strains. These sequence data suggest that intraspecific diversity in the ISRs in C. difficile is much greater than that reported for any other bacterial species (4, 5, 12, 15, 17, 20, 27, 30, 40, 50, 51). Many authors have compared ISRs between different species and strains of a single bacterial species (15, 17, 33, 44, 46). In some cases the ISR diversity between species of a genus is lower than the diversity seen between the strains of C. difficile examined here. Sequence differences between ISRs of the same size most likely arise from nucleotide substitution events (4, 5, 25, 30). Also, some ISRs were of similar length (e.g., rrnE and rrnF of S. aureus) but had sequence blocks with different nucleotide sequences, probably because of recombination events (25, 27, 28). Some of the ISRs were identical in length, and any differences in their sequences were due to single nucleotide deletions, insertions, or substitutions (4, 5, 25, 30). Although PCR amplification of ISRs in C. difficile has been used for typing and differentiation of C. difficile strains, many strains give identical PCR band patterns on agarose gels (7, 13, 21, 26, 42, 55, 61). This study suggests that these apparent similarities may mask considerable levels of heterogeneity not revealed by these fingerprinting methods. Thus, both strains A and B gave similar PCR band patterns on agarose gels, but sequencing these revealed markedly different ISRs in the two strains. For example, ISR clones A-3, A-15, A-11, and A-12 were found only in strain A, while ISR clones B-11, B-12, and B-14 just existed in strain B. It was possible to differentiate between these ISR clones and hence strains which contain them. Examination of restriction endonuclease maps of the ISRs with Biomanager software indicated that the TaqI restriction enzyme could differentiate between B-11 of strain B and A-3 and A-15 of strain A on the basis of different size restriction fragments (Fig. 3). It should also be added that several other restriction enzymes could be used for the differentiation of other ISR clones of strain A from ISR clones of strain B. Our work has shown that a high degree of diversity existed among ISR sequences from five different strains of C. difficile with respect to rrn copy number and sequence variation. Other studies have shown that significant ISR length variations exist between a large number of other C. difficile strains (13, 26, 61). The sequences of the ISRs in C. difficile are similarly diverse, and it is likely that more diversity will be uncovered when more strains are analyzed at the sequence level. Strain-specific oligonucleotide microarrays have been described for ISR sequences of streptococci (16), for industrial wastewater treatment plant microorganisms in the study of microbial diversity within complex environmental samples (18), and for the detection of different species of the genus Kitasatospora (24). The further characterization of ISR sequence mosaics in C. difficile by oligonucleotide-mosaic arrays is likely to uncover an even higher level of mosaic sequence block rearrangements as more strains of C. difficile are analyzed.

7322

SADEGHIFARD ET AL. ACKNOWLEDGMENTS

This work was funded by the Biotechnology Research Centre, La Trobe University. We are grateful to Christopher Maslunka for his assistance in locating the rrn operons in the whole-genome sequence of C. difficile strain 630. REFERENCES 1. Acinas, S. G., L. A. Marcelino, V. Klepac-Ceraj, and M. F. Polz. 2004. Divergence and redundancy of 16S rRNA sequences in genomes with multiple rrn operons. J. Bacteriol. 186:2629–2635. 2. Ackermann, G., S. Thomalla, F. Ackermann, R. Schaumann, A. C. Rodloff, and B. R. Ruf. 2005. Prevalence and characteristics of bacteria and host factors in an outbreak situation of antibiotic-associated diarrhoea. J. Med. Microbiol. 54:149–153. 3. Altschul, S. F., M. S. Boguski, W. Gish, and J. C. Wootton. 1994. Issues in searching molecular sequence databases. Nat. Genet. 6:119–129. 4. Anton, A. I., A. J. Martinez-Murcia, and F. Rodriguez-Valera. 1999. Intraspecific diversity of the 23S rRNA gene and the spacer region downstream in Escherichia coli. J. Bacteriol. 181:2703–2709. 5. Anton, A. I., A. J. Martinez-Murcia, and F. Rodriguez-Valera. 1998. Sequence diversity in the 16S-23S intergenic spacer region (ISR) of the rRNA operons in representatives of the Escherichia coli ECOR collection. J. Mol. Evol. 47:62–72. 6. Belanger, S. D., M. Boissinot, N. Clairoux, F. J. Picard, and M. G. Bergeron. 2003. Rapid detection of Clostridium difficile in feces by real-time PCR. J. Clin. Microbiol. 41:730–734. 7. Bidet, P., F. Barbut, V. Lalande, B. Burghoffer, and J. C. Petit. 1999. Development of a new PCR-ribotyping method for Clostridium difficile based on ribosomal RNA gene sequencing. FEMS Microbiol. Lett. 175:261–266. 8. Blake, J. E., F. Mitsikosta, and M. A. Metcalfe. 2004. Immunological detection and cytotoxic properties of toxins from toxin A-positive, toxin B-positive Clostridium difficile variants. J. Med. Microbiol. 53:197–205. 9. Bowman, R. A., S. A. Arrow, and T. V. Riley. 1986. Latex particle agglutination for detecting and identifying Clostridium difficile. J. Clin. Pathol. 39:212– 214. 10. Brazier, J. S. 2001. Typing of Clostridium difficile. Clin. Microbiol. Infect. 7:428–431. 11. Bruggemann, H., S. Baumer, W. F. Fricke, A. Wiezer, H. Liesegang, I. Decker, C. Herzberg, R. Martinez-Arias, R. Merkl, A. Henne, and G. Gottschalk. 2003. The genome sequence of Clostridium tetani, the causative agent of tetanus disease. Proc. Natl. Acad. Sci. USA 100:1316–1321. 12. Carr, E. L., V. Gurtler, and R. J. Seviour. 2004. Variation of 16S-23S rRNA intergenic spacer regions (ISRs) in Acinetobacter baylyi (strain B2) isolated from activated sludge. Syst. Appl. Microbiol. 27:478–491. 13. Cartwright, C. P., F. Stock, S. E. Beekmann, E. C. Williams, and V. J. Gill. 1995. PCR amplification of rRNA intergenic spacer regions as a method for epidemiologic typing of Clostridium difficile. J. Clin. Microbiol. 33:184–187. 14. Chaudhuri, R. R., and M. J. Pallen. 2006. xBASE, a collection of online databases for bacterial comparative genomics. Nucleic Acids Res. 34:D335– D337. 15. Chen, C. C., L. J. Teng, and T. C. Chang. 2004. Identification of clinically relevant viridans group streptococci by sequence analysis of the 16S-23S ribosomal DNA spacer region. J. Clin. Microbiol. 42:2651–2657. 16. Chen, C. C., L. J. Teng, S. Kaiung, and T. C. Chang. 2005. Identification of clinically relevant viridans streptococci by an oligonucleotide array. J. Clin. Microbiol. 43:1515–1521. 17. Chun, J., A. Huq, and R. R. Colwell. 1999. Analysis of 16S-23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus. Appl. Environ. Microbiol. 65:2202–2208. 18. Cook, K. L., A. C. Layton, H. M. Dionisi, J. T. Fleming, and G. S. Sayler. 2004. Evaluation of a plasmid-based 16S-23S rDNA intergenic spacer region array for analysis of microbial diversity in industrial wastewater. J. Microbiol. Methods 57:79–93. 19. Delmee, M., J. Van Broeck, A. Simon, M. Janssens, and V. Avesani. 2005. Laboratory diagnosis of Clostridium difficile-associated diarrhoea: a plea for culture. J. Med. Microbiol. 54:187–191. 20. Doignon-Bourcier, F., A. Willems, R. Coopman, G. Laguerre, M. Gillis, and P. de Lajudie. 2000. Genotypic characterization of Bradyrhizobium strains nodulating small Senegalese legumes by 16S-23S rRNA intergenic gene spacers and amplified fragment length polymorphism fingerprint analyses. Appl. Environ. Microbiol. 66:3987–3997. 21. Fawley, W. N., P. Parnell, P. Verity, J. Freeman, and M. H. Wilcox. 2005. Molecular epidemiology of endemic Clostridium difficile infection and the significance of subtypes of the United Kingdom epidemic strain (PCR ribotype 1). J. Clin. Microbiol. 43:2685–2696. 22. Geric, B., M. Rupnik, D. N. Gerding, M. Grabnar, and S. Johnson. 2004. Distribution of Clostridium difficile variant toxinotypes and strains with binary toxin genes among clinical isolates in an American hospital. J. Med. Microbiol. 53:887–894.

APPL. ENVIRON. MICROBIOL. 23. Giannino, V., G. Rappazzo, A. Scuto, O. Di Marco, A. Privitera, M. Santagati, and S. Stefani. 2001. rrn operons in Haemophilus parainfluenzae and mosaicism of conserved and species-specific sequences in the 16S-23S rDNA long spacer. Res. Microbiol. 152:461–468. 24. Gunther, S., I. Groth, S. Grabley, and T. Munder. 2006. Design and evaluation of an oligonucleotide-microarray for the detection of different species of the genus Kitasatospora. J. Microbiol. Methods 65:226–236. 25. Gu ¨rtler, V. 1999. The role of recombination and mutation in 16S-23S rDNA spacer rearrangements. Gene 238:241–252. 26. Gu ¨rtler, V. 1993. Typing of Clostridium difficile strains by PCR-amplification of variable length 16S-23S rDNA spacer regions. J. Gen. Microbiol. 139: 3089–3097. 27. Gu ¨rtler, V., and H. D. Barrie. 1995. Typing of Staphylococcus aureus strains by PCR-amplification of variable-length 16S-23S rDNA spacer regions: characterization of spacer sequences. Microbiology 141:1255–1265. 28. Gu ¨rtler, V., and B. C. Mayall. 2001. Genetic transfer and genome evolution in MRSA. Microbiology 147:3195–3197. 29. Gu ¨rtler, V., and B. C. Mayall. 1994. Genotyping of Clostridium difficile isolates. J. Clin. Microbiol. 32:3095. 30. Gu ¨rtler, V., Y. Rao, S. R. Pearson, S. M. Bates, and B. C. Mayall. 1999. DNA sequence heterogeneity in the three copies of the long 16S-23S rDNA spacer of Enterococcus faecalis isolates. Microbiology 145:1785–1796. 31. Gu ¨rtler, V., and V. A. Stanisich. 1996. New approaches to typing and identification of bacteria using the 16S-23S rDNA spacer region. Microbiology 142:3–16. 32. Hogervorst, F. B., P. M. Nederlof, J. J. Gille, C. J. McElgunn, M. Grippeling, R. Pruntel, R. Regnerus, T. van Welsem, R. van Spaendonk, F. H. Menko, I. Kluijt, C. Dommering, S. Verhoef, J. P. Schouten, L. J. van’t Veer, and G. Pals. 2003. Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method. Cancer Res. 63:1449–1453. 33. Houpikian, P., and D. Raoult. 2001. 16S/23S rRNA intergenic spacer regions for phylogenetic analysis, identification, and subtyping of Bartonella species. J. Clin. Microbiol. 39:2768–2778. 34. Klappenbach, J. A., J. M. Dunbar, and T. M. Schmidt. 2000. rRNA operon copy number reflects ecological strategies of bacteria. Appl. Environ. Microbiol. 66:1328–1333. 35. Klappenbach, J. A., P. R. Saxman, J. R. Cole, and T. M. Schmidt. 2001. rrndb: the ribosomal RNA operon copy number database. Nucleic Acids Res. 29:181–184. 36. Lan, R., and P. R. Reeves. 1998. Recombination between rRNA operons created most of the ribotype variation observed in the seventh pandemic clone of Vibrio cholerae. Microbiology 144:1213–1221. 37. Lemee, L., A. Dhalluin, M. Pestel-Caron, J. F. Lemeland, and J. L. Pons. 2004. Multilocus sequence typing analysis of human and animal Clostridium difficile isolates of various toxigenic types. J. Clin. Microbiol. 42:2609–2617. 38. Lemee, L., A. Dhalluin, S. Testelin, M. A. Mattrat, K. Maillard, J. F. Lemeland, and J. L. Pons. 2004. Multiplex PCR targeting tpi (triose phosphate isomerase), tcdA (toxin A), and tcdB (toxin B) genes for toxigenic culture of Clostridium difficile. J. Clin. Microbiol. 42:5710–5714. 39. Maddison, D. R., and W. P. Maddison. 2005. MacClade 4: analysis of phylogeny and character evolution, version 4.08. 40. Maeda, T., N. Takada, M. Furushita, and T. Shiba. 2000. Structural variation in the 16S-23S rRNA intergenic spacers of Vibrio parahaemolyticus. FEMS Microbiol. Lett. 192:73–77. 41. Mangalam, H. 2002. tacg—a grep for DNA. BMC Bioinformatics 3:8. [Online.] http://www.biomedcentral.com/1471-2105/3/8. 42. Martirosian, G., S. Kuipers, H. Verbrugh, A. van Belkum, and F. MeiselMikolajczyk. 1995. PCR ribotyping and arbitrarily primed PCR for typing strains of Clostridium difficile from a Polish maternity hospital. J. Clin. Microbiol. 33:2016–2021. 43. McFarland, L. V. 2005. Alternative treatments for Clostridium difficile disease: what really works? J. Med. Microbiol. 54:101–111. 44. Mora, D., G. Ricci, S. Guglielmetti, D. Daffonchio, and M. G. Fortina. 2003. 16S-23S rRNA intergenic spacer region sequence variation in Streptococcus thermophilus and related dairy streptococci and development of a multiplex ITS-SSCP analysis for their identification. Microbiology 149:807–813. 45. Mylvaganam, S., and P. P. Dennis. 1992. Sequence heterogeneity between the two genes encoding 16S rRNA from the halophilic archaebacterium Haloarcula marismortui. Genetics 130:399–410. 46. Nguimbi, E., Y. Z. Li, B. L. Gao, Z. F. Li, B. Wang, Z. H. Wu, B. X. Yan, Y. B. Qu, and P. J. Gao. 2003. 16S-23S ribosomal DNA intergenic spacer regions in cellulolytic myxobacteria and differentiation of closely related strains. Syst. Appl. Microbiol. 26:262–268. 47. Northey, G., M. Gal, A. Rahmati, and J. S. Brazier. 2005. Subtyping of Clostridium difficile PCR ribotype 001 by REP-PCR and PFGE. J. Med. Microbiol. 54:543–547. 48. Osorio, C. R., M. D. Collins, J. L. Romalde, and A. E. Toranzo. 2005. Variation in 16S-23S rRNA intergenic spacer regions in Photobacterium damselae: a mosaic-like structure. Appl. Environ. Microbiol. 71:636–645. 49. Pebay, M., Y. Roussel, J. Simonet, and B. Decaris. 1992. High-frequency deletion involving closely spaced rRNA gene sets in Streptococcus thermophilus. FEMS Microbiol. Lett. 98:51–55.

VOL. 72, 2006


50. Perez Luz, S., F. Rodriguez-Valera, R. Lan, and P. R. Reeves. 1998. Variation of the ribosomal operon 16S-23S gene spacer region in representatives of Salmonella enterica subspecies. J. Bacteriol. 180:2144–2151. 51. Perez-Luz, S., J. Fernandez, F. Rodriguez-Valera, L. Pascual, C. Moreno, A. Amo, D. Apraiz, and V. Catalan. 2002. Sequence diversity of the internal transcribed spacer (ITS) region of the rRNA operons among different serogroups of Legionella pneumophila isolates. Syst. Appl. Microbiol. 25:212–219. 52. Pituch, H., M. Rupnik, P. Obuch-Woszczatynski, A. Grubesic, F. MeiselMikolajczyk, and M. Luczak. 2005. Detection of binary-toxin genes (cdtA and cdtB) among Clostridium difficile strains isolated from patients with C. difficile-associated diarrhoea (CDAD) in Poland. J. Med. Microbiol. 54:143– 147. 53. Poxton, I. R. 2005. Clostridium difficile. J. Med. Microbiol. 54:97–100. 54. Privitera, A., G. Rappazzo, P. Sangari, V. Giannino, L. Licciardello, and S. Stefani. 1998. Cloning and sequencing of a 16S/23S ribosomal spacer from Haemophilus parainfluenzae reveals an invariant, mosaic-like organisation of sequence blocks. FEMS Microbiol. Lett. 164:289–294. 55. Rupnik, M., J. S. Brazier, B. I. Duerden, M. Grabnar, and S. L. Stubbs. 2001. Comparison of toxinotyping and PCR ribotyping of Clostridium difficile strains and description of novel toxinotypes. Microbiology 147:439–447. 56. Sasaki, Y., K. Yamamoto, K. Amimoto, A. Kojima, Y. Ogikubo, M. Norimatsu, H. Ogata, and Y. Tamura. 2001. Amplification of the 16S-23S rDNA spacer region for rapid detection of Clostridium chauvoei and Clostridium septicum. Res. Vet. Sci. 71:227–229. 57. Sasaki, Y., K. Yamamoto, A. Kojima, M. Norimatsu, and Y. Tamura. 2000. Rapid identification and differentiation of pathogenic clostridia in gas gangrene by polymerase chain reaction based on the 16S-23S rDNA spacer region. Res. Vet. Sci. 69:289–294. (Erratum, 70:93–94, 2001.) 58. Sasaki, Y., K. Yamamoto, A. Kojima, Y. Tetsuka, M. Norimatsu, and Y. Tamura. 2000. Rapid and direct detection of Clostridium chauvoei by PCR of the 16S-23S rDNA spacer region and partial 23S rDNA sequences. J. Vet. Med. Sci. 62:1275–1281. 59. Sebaihia, M., B. W. Wren, P. Mullany, N. F. Fairweather, N. Minton, R. Stabler, N. R. Thomson, A. P. Roberts, A. M. Cerdeno-Tarraga, H. Wang, M. T. Holden, A. Wright, C. Churcher, M. A. Quail, S. Baker, N. Bason, K. Brooks, T. Chillingworth, A. Cronin, P. Davis, L. Dowd, A. Fraser, T. Feltwell, Z. Hance, S. Holroyd, K. Jagels, S. Moule, K. Mungall, C. Price, E. Rabbinowitsch, S. Sharp, M. Simmonds, K. Stevens, L. Unwin, S. Whithead, B. Dupuy, G. Dougan, B. Barrell, and J. Parkhill. 2006. The multidrug-

60. 61.

62.

63.

64.

65.

66.

67. 68. 69.

7323

resistant human pathogen Clostridium difficile has a highly mobile, mosaic genome. Nat. Genet. 38:779–786. Spigaglia, P., and P. Mastrantonio. 2004. Comparative analysis of Clostridium difficile clinical isolates belonging to different genetic lineages and time periods. J. Med. Microbiol. 53:1129–1136. Stubbs, S. L., J. S. Brazier, G. L. O’Neill, and B. I. Duerden. 1999. PCR targeted to the 16S-23S rRNA gene intergenic spacer region of Clostridium difficile and construction of a library consisting of 116 different PCR ribotypes. J. Clin. Microbiol. 37:461–463. Svenungsson, B., L. G. Burman, K. Jalakas-Pornull, A. Lagergren, J. Struwe, and T. Akerlund. 2003. Epidemiology and molecular characterization of Clostridium difficile strains from patients with diarrhea: low disease incidence and evidence of limited cross-infection in a Swedish teaching hospital. J. Clin. Microbiol. 41:4031–4037. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680. Turova, T. P., B. B. Kuznetsov, E. V. Novikova, A. B. Poltaraus, and T. N. Nazina. 2001. Heterogeneity of nucleotide sequences of 16S ribosomal RNA genes from the Desulfotomaculum kuznetsovii type strain. Mikrobiologiia 70:788–795. (In Russian.) van den Berg, R. J., H. A. Ameen, T. Furusawa, E. C. Claas, E. R. van der Vorm, and E. J. Kuijper. 2005. Coexistence of multiple PCR-ribotype strains of Clostridium difficile in faecal samples limits epidemiological studies. J. Med. Microbiol. 54:173–179. van den Berg, R. J., L. S. Bruijnesteijn van Coppenraet, H. J. Gerritsen, H. P. Endtz, E. R. van der Vorm, and E. J. Kuijper. 2005. Prospective multicenter evaluation of a new immunoassay and real-time PCR for rapid diagnosis of Clostridium difficile-associated diarrhea in hospitalized patients. J. Clin. Microbiol. 43:5338–5340. Wang, Y., Z. Zhang, and N. Ramanan. 1997. The actinomycete Thermobispora bispora contains two distinct types of transcriptionally active 16S rRNA genes. J. Bacteriol. 179:3270–3276. Wilkins, T. D., and D. M. Lyerly. 2003. Clostridium difficile testing: after 20 years, still challenging. J. Clin. Microbiol. 41:531–534. Yap, W. H., Z. Zhang, and Y. Wang. 1999. Distinct types of rRNA operons exist in the genome of the actinomycete Thermomonospora chromogena and evidence for horizontal transfer of an entire rRNA operon. J. Bacteriol. 181:5201–5209.