OUP user menu

Evidence that Tn5565, which includes the enterotoxin gene in Clostridium perfringens, can have a circular form which may be a transposition intermediate

Sigrid Brynestad, Per Einar Granum
DOI: http://dx.doi.org/10.1111/j.1574-6968.1999.tb13385.x 281-286 First published online: 1 January 1999


The Clostridium perfringens enterotoxin gene is on a transposon-like element, Tn5565, integrated in the chromosome in human food poisoning strains. The flanking IS elements, IS1470 A and B, are related to IS30. The IS element found in the transposon, IS1469, is related to IS200 and has been found upstream of cpe in all Type A strains. PCR and sequencing studies from cell extracts and plasmid isolations of C. perfringens indicate that Tn5565 can form a circular form with the tandem repeat (IS1470)2, similar to the transposition intermediates described for a number of IS elements.

  • Clostridium perfringens
  • Enterotoxin
  • Transposition
  • IS1470
  • IS1469

1 Introduction

Bacterial insertion sequences (ISs) have been found in all major bacterial species, and play a significant part in the plasticity of the bacterial genome and in the acquisition and dissemination of genes involved in virulence and pathogenicity, antibiotic resistance, as well as catabolic pathways (reviewed in [1]). The level of transposition is generally tightly regulated at a low level, but there is evidence that certain elements can undergo transient bursts of transposition activity [2] or transposase production [3].

Clostridium perfringens is a Gram-positive, anaerobic, sporulating, ubiquitous bacterium which is important both in human and veterinary medicine. Diarrhoea is caused by the ingestion of strains containing the enterotoxin gene (cpe), which can produce large amounts of CPE under sporulation [4]. The enterotoxin gene is found on a 6.3-kb transposon-like element, Tn5565, in human food poisoning strains which includes flanking copies of IS1470 (A and B) (IS30 family), IS1469 (IS200-like) and cpe [5]. The transposon is integrated into the chromosome while the enterotoxin gene is located on plasmids in other isolates [68] and then cpe is often associated with IS1151 [9], and in Type A strains IS1469 is upstream from cpe [7]. IS1470 has not been found on plasmids, it is located in 0–10 copies on the chromosome in tested Type A strains [10]. IS1470 has a single ORF which codes for a putative transposase of 346 aa with homology to IS30 transposase, has 26-bp inverted repeats and appears to duplicate 14 bp upon insertion [5].

There have been a number of models proposed for the movement of IS elements. It is becoming clear that the formation of an (IS)2 intermediate can stimulate transposition and act as a regulatory mechanism for transposase expression. The dimer formed is very unstable. IS dimers have been shown to be transposition intermediates in IS30 [11], IS21 [12] and IS2 [13]. IS1 [14], IS2 [13, 15], IS3 [16], IS150 and IS911 [3] have all been shown to form circles, which are proposed to transposition intermediates. The generation of (IS)2 or a circular form of the IS element leads to a burst of transposition activity, postulated to be due to the formation of a strong promoter which is created by the abutted right and left IRs in front of transposase gene [3]. Studies done with IS30 showed that the N-terminal domain of the transposase interacts with the terminal IRs [17], and the carboxy-terminal domain is involved in the formation and resolution of the (IS30)2 dimer [18]. The conserved C-terminal active site, DD(35)E, is found in many types of transposases and integrases [19], and rather strong homology in the C-terminal has been found in the IS4 family of transposases [20] and between IS30-related elements [3].

Most of the work done on (IS)2 dimers has been performed in artificial systems. We wanted to see if we could detect naturally occurring dimerisation of IS1470 and circle formation of Tn5565 in cultures of C. perfringens to help explain the movement of the enterotoxin gene.

2 Materials and methods

2.1 Bacteria strains and growth conditions

NCTC 8239, 790-94 and 44071-C05 are enterotoxin positive C. perfringens described in [5]. The first two have cpe on Tn5565. 44071-C05 carries cpe on a large plasmid, and has one copy of IS1470 in the genome between the same two genes as Tn5565. This strain was used as a control for the plasmid isolation protocols and for PCR. Strains were grown anaerobically in BHI medium (Difco) or DS sporulation medium as described in [5]. Escherichia coli strains, other than those included in kits, were Epicurian Coli SURE cells from Stratagene.

2.2 Isolation of DNA

Plasmids from E. coli were isolated using QIA prep or QIAGEN plasmid Midi kits (QIAGEN, Germany), or as described by [21]. ‘Plasmid’ DNA from C. perfringens was isolated using QIA prep or S.N.A.P. mini-prep kit (Invitrogen) but the cells were incubated with 5 µl of 50 mg ml−1 lysozyme for 1 h first. Total DNA from Clostridia was isolated using ADVAMAX beads and the protocol supplied by the manufacturers for bacterial genomic DNA purification (Advanced Genetics Technologies, Gathersburg, MD.)

2.3 Polymerase chain reaction

PCR was carried out in an MJ Research Minicycler. The annealing temperature used was from 40 to 55°C, depending on the primers used. The standard programme was: 92°C for 1 min, annealing temperature for 1 min and 72°C for 1 min for 30 cycles. The low annealing temperatures are due to the high AT content of the C. perfringens genome. DYNAZyme DNA polymerase and buffer and 10 mM dNTP from Finnzymes OY were used. Primers used (see also Fig. 1): 1470mr, TGAAAACCGTGAAGAATTTGG; 1470AR, GGAAAGACTATTCACATATAG; 1469r, TGCTAGAGTTAAAAGTGTCC; 1470BF, GCTACCTAGATAGATTCAAGAATCTTC; 1470ef, GCTTAGTAGTACCACGGCGTATTTCG; cpeend, CAGTCCTTAGGTGATGGAGT (forward end of gene, not shown on Fig. 1).

Figure 1

Top of the figure: the chromosomal configuration of Tn5565 in NCTC8239 is shown with arrowheads indicating sets of PCR primers used in determination of circular intermediates. Inverted repeats of IS1470, and the stem-loop structures flanking IS1469 are indicated. Black arrows: 1470mr (left) and 1470ef (right) which are internal to IS1470. The 1000-bp product indicated the formation of a (IS1470)2 dimer, illustrated in A. White arrows: 1470AR (left) and 1470BF (right) gave a product of 1300 bp. This product was sequenced and shown to be IS1470A, indicating the excision (and likely expression) of IS1470B. The configuration of this circle is given in B. Gray arrows: 1469r (left) and 1470BF (right) gave a product of 700 bp, which indicated the presence of a circular form where both copies of IS1470 had been excised, seen in C.

2.4 Cloning of PCR products and PCR screening of clone

PCR products were cloned using the pMOSBlue T-vector kit and pMOSBlue blunt kit (Amersham) following the manufacturer's instructions. Clones from transformations were screened using PCR. Colonies were picked and transferred to new LB-amp agar plates. Small amounts of cells were suspended in 100 µl water, and microwaved for 3 min. Two to 5 µl from this tube was used in a PCR reaction.

2.5 DNA sequencing

Sequencing was performed on a Perkin Elmer ABI Prism 377 automatic sequencer. ABI PRISM Dye Primer or Dye Terminator Cycle Sequencing Ready Reaction Kit were used following the manufacturer's instructions.

3 Results and discussion

We wanted to investigate the possibility that Tn5565 could form (IS1470)2 dimers with a resulting circular intermediate carrying cpe and IS1469 in a putative transposition intermediate. From the results of IS30 studies [11, 17, 18, 22, 23], we assumed that the level of transposition in natural systems would be low, and the dimer very unstable in the presence of its own transposase, so the only feasible mean of detection of circular structures from a natural system was by a PCR approach. PCR primers were selected so that no product is produced by linear copies of Tn5565, but will give products if the transposon is found in a circular form. Total DNA was isolated from overnight cultures of strains 790-94 and NCTC8239 in BHI and sporulating cultures in DS. Initial trials showed that sporulating cells and stationary vegetative phase cells gave essentially the same results, and only BHI cultures were used in further experiments. As there are other copies of IS1470 in the genome, so the PCR product could conceivably stem from a (IS1470)2 dimer that is not connected with cpe. After the initial trials with total DNA plasmid preparations were used to minimise the chance of a PCR product resulting from a non-circular form. Plasmid DNA was isolated from NCTC8239, 790-94, and as a control for successful plasmid isolation from 44071-C05 to confirm that the PCR products formed most likely stem from covalently closed circular molecules. Plasmid preparations were run on an agarose gel, and only the expected plasmids from 44071-C05 were visible, while there was no visible DNA in the other samples, though there is the possibility of small amounts of linear DNA in the samples.

The internal IS1470 primers selected only produce a product if (IS1470)2 has been formed. This product was visible from fresh (same day) DNA preparations, but was not produced from frozen or samples held at 4°C of the same DNA a few days later, indicating that the structure is unstable. Sequencing showed that the four bases found in the junction between the inverted repeats were identical to the two bases on each side, alternatively three bases from 1470A IRR end and one from IS1470B IRRL end, of the transposon in the chromosome. The sequences were such that no conclusions can be drawn as to the mechanism of initiation of transposition, and further studies will have to be done to determine if a single-stranded or a double-stranded cut is used. Primers from DNA immediately outside of the IR of IS1470A and B (1470AR and 1470BF) and primers from IS1469 and cpe were used to check that Tn5565 was found in a circular form and that IS1469 and cpe were still associated with IS1470. The PCR products from these outside primers showed that there is a circular form, but that one copy or both copies of IS1470 had been excised from most of the circles. These products were also found in DNA preparations which had been frozen, indicating that these circles are more stable than the dimer configuration. The PCR product from 1470AR/1470BF from two different clones was sequenced and were found to contain IS1470A. The PCR product from using primers 1469mr and 1470BF should span both IS1470 elements, but the 600-bp product indicated that IS1469 and cpe can exist as a circular intermediate without IS1470. For a summary of PCR results see Fig. 1.

In dimer formation studies of IS30, over 80% of the isolated plasmids contained a single copy of the IS element, suggesting that a major segregation product contained a remnant IS30 element [11]. Although caution has to be used when interpreting PCR results, this remnant IS element configuration also seems to be a major product formed from Tn5565.

It has been proposed that the formation of a (IS)2 dimer could also lead to the formation of a strong promoter [3]. From the IR sequences, it would appear that the formation of (IS1470)2 could indeed form a putative promoter that is much stronger than otherwise available (Fig. 2). The question here is if this consensus E. coli promoter is equally as strong a promoter in the Clostridium host cell. That IS1470B is excised in those circles which have one IS1470 suggests that this IRR–IRL junction promoter is active, as the promoter should stimulate transposase production from IS1470B. As many transposases act preferably in cis, [1] the observed results agree with IS1470B transcription.

Figure 2

The top part of the figure shows how two different promoter could function in Tn5565 where IS1470 ORF is translated on the lower strand in the 5′ to 3′ direction. The two alternative promoter regions indicated by underlining and in bold. The shaded bases are found on each side of Tn5565 (two on each end) on the genome. The bottom part of the figure shows potential promoters in the abutted ends of IS1470 and the related elements IS30 and IS1070. The underlined X are the bases included (spacer) from the surrounding DNA upon excision. IS30 and IS1070 sequences are from [3].

The formation of a circular form of Tn5565 has important implications in the possible movement of cpe. In human food poisoning strains, the transposon is found in the same chromosomal location in all strains tested. The configuration of the transposon indicates that Tn5565 has integrated as a unit. The other possibility is that the circle with one IS1470 element integrates into a site with a single copy of IS1470, assuming that IS1470 has the preference, as IS30 does, for the ends of the element as a ‘hot spot’ for integration [24]. It may be difficult to distinguish products of integration of the entire transposon from a integration of other configurations. It is worth noting, however, that there are many chromosomal copies of IS1470 in some C. perfringens strains, but the Tn5565 integration site is conserved. Localisation of the transposon could involve other factors in addition to base pair sequence recognition in the choice of integration site.

Strain 44071-C05 has IS1470C (a little different in sequence for A and B) in the same chromosomal location as Tn5565 in NCTC8239, while cpe and IS1469 are found on a large episome. The circular form of IS1469 and cpe, where both copies of IS1470 are excised, is interesting in that IS1469 codes for what should be an active transposase. Integration of this circular form could account for the structure seen in animal isolates of enterotoxin positive Type A strains where cpe is found on an episome [7, 9].

Evidence of the presence of a circular form of Tn5565, and circular forms of segregation products of the transposon in this PCR study fits well into the pattern seen from IS30 and IS3 family studies. Most of these studies have been performed with ‘artificial’ transposons, and in Gram-negative bacteria. The presence of circular forms with (IS)2 dimers in a natural system, with a virulence factor, from a Gram-positive bacteria should strengthen the theory that (IS)2/circle formation is a commonly found structure which can play a role in transposition. Our results are a good start in the examination of cpe movement. Further investigations using IS cassettes are underway so that the movement of these IS elements can be followed more closely than is feasible in the natural system in C. perfringens.


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