Plasmid 58 (2007) 182–189

www.elsevier.com/locate/yplas

Disruption of a toxin gene by introduction of a foreign geneinto the chromosome of Clostridium perfringens

using targetron-induced mutagenesis

Yue Chen a,*, Lori Caruso a, Bruce McClane b, Derek Fisher b, Phalguni Gupta a

a Department of Infectious Diseases and Microbiology, Graduate School of Public Health, University of Pittsburgh, 430 Parran Hall,

GSPH, 130 DeSoto Street, Pittsburgh, PA 15261, USAb Department of Molecular Genetics and Biochemistry, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA

Received 22 February 2007, revised 18 April 2007Available online 5 June 2007

Communicated by Ananda Chakrabarty

Abstract

Clostridium perfringens (C. perfringens) has been developed as a potential oral delivery vehicle to deliver antigens ortherapeutic compounds to Gut-Associated Lymphoid Tissue (GALT). However, this recombinant C. perfringens carriesa plasmid-encoded expression system, which raises several safety concerns regarding possible horizontal plasmid transferand spread of plasmid-associated antibiotic resistant genes. Furthermore, this bacterium produces the extracellular thetatoxin, which poses a potential safety issue for general administration. Using a Clostridium-specific-targetron donor plas-mid, we inserted the Simian Immunodefiency Virus (SIV) p27 gene into the theta toxin gene (pfoA) on the C. perfringens

chromosome, which simultaneously inactivated the theta gene and introduced SIV p27 gene onto the bacterial chromo-some. Such mutant C. perfringens without an input plasmid or antibiotic resistant gene stably produced a large amountof SIV p27 protein during sporulation and did not produce theta toxin. Upon oral feeding of the mutant bacteria to mice,intact p27 protein was detected in the lower GI tract. The re-engineered C. perfringens provides a biosafe efficient oral vehi-cle to deliver antigen to the gastrointestinal tract.� 2007 Elsevier Inc. All rights reserved.

Keywords: Oral delivery vehicle; Clostridium perfringens; Gene inactivation; Mobile genetic elements

1. Introduction

Gut-Associated Lymphoid Tissues (GALT),which are the main target for many vaccines andtherapeutic interventions, are mainly located in thelower portion of the small intestine. The use of live

0147-619X/$ - see front matter � 2007 Elsevier Inc. All rights reserved

doi:10.1016/j.plasmid.2007.04.002

* Corresponding author. Fax: +1 412 624 4953.E-mail address: cheny@pitt.edu (Y. Chen).

bacteria to deliver antigens or therapeutic com-pounds to GALT is an attractive method due tointrinsic properties of many bacteria, low produc-tion costs and easy administration (Detmer andGlenting, 2006; Fischetti et al., 1996). The materialsthat are usually delivered to the gastrointestinal(GI) tract are immunogens or therapeutic com-pounds expressed from plasmids carrying antibioticresistance genes for selective pressure. Recently,Clostridium perfringens (C. perfringens) type A, a

.

Y. Chen et al. / Plasmid 58 (2007) 182–189 183

gram-positive, endospore forming bacterium foundin the normal intestinal flora of human and otheranimals, has been developed as an oral deliveryvehicle (Chen et al., 2004). Due to the unique char-acteristics of the C. perfringens enterotoxin (cpe)promoter and the biology of the organism, the vec-tor can express and deliver a large amount of for-eign protein in intact form to the terminal ileumafter oral administration (Chen et al., 2004). TheC. perfringens type A transformant used for thedelivery vehicle is negative for enterotoxin and b-2toxin and considered safe for oral administration(Fisher et al., 2005; Smedley et al., 2004). However,to ensure the safety of the C. perfringens based vehi-cle, the alpha toxin gene (plc) in C. perfringens chro-mosome has been inactivated using mobile group IIintron based targetron technology(Chen et al.,2005). In these recombinant clostridia, SimianImmunodefiency Virus (SIV) p27 antigens areexpressed from intracellular plasmids carrying anti-biotic resistance genes for selective pressure. How-ever, some safety concerns have been recentlyraised about this approach, because (1) there arepossibilities that input plasmid may transfer hori-zontally in the environment, integrate into the hostgenome and potentially cause oncogenesis(Detmerand Glenting, 2006) and (2) plasmid-associated anti-biotic resistant genes can be transferred to humansand could hamper the use of therapeutic antibioticsin clinic. In this study, we apply the targetron meth-odology to simultaneously inactivate the chromo-somal theta toxin gene (pfoA) and insert SIV p27gene into the C. perfringens chromosome, producinga recombinant C. perfringens mutant which does notproduce alpha or theta toxins, but stably expressesSIV p27 from bacterial chromosome without anyantibiotic pressure.

2. Materials and methods

2.1. Bacterial strain and growth conditions

To construct a pfoA/plc double knockout mutant pro-ducing p27, we used our previously constructed plc

knockout mutant of C. perfringens ATCC3624 (Chenet al., 2005), which is naturally enterotoxin- and b-2toxin-negative. Vegetative cultures of all C. perfringens

isolates were grown for 9 h at 37 �C in fluid thioglycolatebroth (FTG) (Difco). A modified Duncan–Strong med-ium (MDS) (Kokai-Kun et al., 1994) was used to obtainsporulating cultures of C. perfringens strains after 8 h cul-ture at 37 �C. Media were supplemented with 15 lg ofchloramphenicol (CM) per milliliter when necessary.

2.2. Theta toxin gene targetron donor plasmid construction

plc targetron donor plasmid pJIR750ai (Chen et al.,2005) was used in this study to construct a pfoA targetronfor inactivation of pfoA gene in bacterial chromosome.The Sigma Targetron Design web site (http://www.sigma-genosys.com/targetron/checksequence/) pre-dicts 10 targetron insertion sites in the pfoA gene openreading frame (ORF). The insertion site in the antisensestrand at position 32/33 from the initial ATG in pfoA

ORF was chosen for Targetron modification accordingto the manufacturer’s instructions. Briefly, a first PCRwas performed on pACD3 (Sigma–Aldrich) with theprimers IBS1/2 (AAAAAAGCTTATAATTATCCTTAATTGCCATACTTGTGCGCCCAGATAGGGTG) andLtrBAsEBS2 (CGAAATTAGAAACTTGCGTTCAGTAAAC) to generate a 257-bp amplicon. A second PCRwas then performed on pACD3 using the primers EBS1/delta (CAGATTGTACAAATGTGGTGATAA CAGATAAGTCATACTTGCTAACTTACCTTTCTT TGT) andEBS2 (TGAACGCAAGTTTCTAATTTCGATTGCAATTCGATAGAGGAAAGTGTCT) to amplify a117-bp amplicon. The 257-bp and 117-bp amplicons werethen used as templates for a second round of PCR withIBS1/IBS2 and EBS1/delta primers to generate a 353-bppfoA targetron. The amplified 353-bp fragment was thendigested with HindIII and BsrGI and ligated into thepJIR750ai vector digested with the same two restrictionenzymes to construct the theta toxin gene targetron(pJIR750hi).

2.3. Construction of pfoA gene targetron containing the

SIV p27 gene cassette

To introduce the SIV p27 gene cassette into pfoA tar-getron, the cassette including the cpe promoter, p27ORF and downstream sequence was PCR amplified fromplasmid cp27 (Chen et al., 2004) with the primers of F(ACGCGTACATTTCAACTTGATCTCTTTAACGTATATCTCTTTT) and R (ACGCGTTATATGGAAGGAGAAATTAAAGCTATAATACTTACTTTACAG)which contain MluI sites at their 5 0 ends. Then, the ampli-fied gene cassette (1340 bp) and pJIR750hi were digestedwith MluI and ligated by T4 ligase to construct plasmidpJIR750hi/p27.

2.4. C. perfringens transformation and detection of

targetron insertion

pJIR750hi or pJIR750hi/p27 were electroporated intoour plc inactivated C. perfringens ATCC 3624 as describedpreviously. CM-resistant colonies (growth on brain heartinfusion [BHI] agar plates containing 15 lg CM/ml)were screened for chromosomal insertion of the pfoA

targetron or the pfoA targetron containing the p27ORF by PCR using the primers theta F (5-AG

184 Y. Chen et al. / Plasmid 58 (2007) 182–189

GGGAAAATTAAAAAAAGGGGGATTTATA) andtheta MR (5-TTTTACCATTAAGATTGTAGGTCTATTTTCCACA), which hybridize to flanking sequencesof the insertion site, and primers theta EBS2 (5-TGAACGCAAGTTTCTAATTTCGATTGCAATTCGATAGAGGAAAGTGTCT) and p27 M (5-TGAACTAGTTGTTCCTGCAATATCTGATCC), which hybrid-ize to pfoA targetron and p27 sequences, respectively.Transformants with a chromosomal targetron insertionconfirmed by PCR were then plated on TSA II 5% sheepblood agar plates (Becton Dickinson, Sparks, MD) andincubated in anaerobic condition over night. Colonieswithout surrounding hemolytic zones were selected forfurther analysis.

2.5. Curing the plasmid from transformed mutants

plc/pfoA double mutants carrying p27 gene were seri-ally sub-cultured daily for 5 days in FTG medium lackingCM. The culture was then plated onto BHI agar plates.Colonies cured from input plasmid were selected by neg-ative colony PCR results with input plasmid-specific prim-ers 4931F (5-TCAATAAGGTCAAAAATCTTAAAGGCAAAGAAAA) and pJIR750R (5-CAGATGCGTAAGGAGAAAATACCGC).

2.6. DNA Sequence and southern blotting

Products of colony PCR using primers of Theta F andTheta MR were gel-purified and sequenced using an auto-mated sequencer. C. perfringens genomic DNAs werepurified using a MasterPure Gram Positive DNA Purifica-tion kit (Epicentre Biotechnologies, Madison, Wisconsin).For Southern hybridization, DNAs were digested withEcoRI and run a 0.8% agarose gel, blotted to a nylonmembrane and probed with a digoxigenin-labeled probespecific to the intron sequence. Probe generation and sig-nal detection were performed according to manufacture’sinstructions by using the DIG High Prime DNA Labelingand Detection Starter Kit II (Roche Diagnostics, Mann-heim, Germany).

2.7. Western blotting

Sporulating and vegetative C. perfringens cells wereharvested, washed, and lysed by lysing buffer. The celllysates were analyzed for the presence of p27 protein byWestern immunoblots as described previously (Chenet al., 2004).

2.8. Animals, oral feeding, and detection of SIV p27

expressed by C. perfringens

BALB/c female mice were purchased from the CharlesRiver Laboratories, Inc. (Wilmington, MA), housed in apathogen-free facility, and fed rodent chow and water

ad libitum. The mice were used between 6 and 10 weeksof age. Mice were orally fed with the sporulating C. per-

fringens recombinants. Ninety minutes after administra-tion, the entire small intestine was divided into six equalsegments and the contents of each segment were extractedvigorously with PBS containing protease inhibitors andthen pelleted down by centrifugation. The pellet fromeach segment was subjected to a Western blot analysis.

3. Results and discussion

C. perfringens is ubiquitous in the environmentand is found in the normal intestinal flora of humanand animals. C. perfringens is also a human and vet-erinary pathogen that is classified into type A, B, C,D, and E based on an isolate’s ability to producefour major toxins (alpha, beta, epsilon and iota tox-ins)(Rood and Cole, 1991; Songer, 1996). CertainC. perfringens also produce minor toxins such astheta toxin, and enteric toxins such as C. perfringensenterotoxin and beta 2 toxin, that are involved inhuman and animal enteric disease. C. perfringens

ATCC3624 used for the delivery vector develop-ment in these studies is classified as type A and isnegative for production of enteric toxins (C. perfrin-

gens enterotoxin and beta 2 toxin), but producesalpha toxin and theta toxin (Fisher et al., 2005;Smedley et al., 2004).

Mobile group II introns are site-specific retroele-ments that use retrohoming mechanism to directlyinsert the excised intron lariat RNA into a specificDNA target site. The inserted RNA is then reverse-transcribed by the associated intron-encoded enzymeprotein (IEP), producing a DNA insertion that inac-tivates the disrupted gene. Since the DNA target siteis recognized primarily by base pairing of intronRNA and a few IEP recognition spots, the intron tar-geting sequences can be modified to insert into a spe-cific DNA target site (Lambowitz and Zimmerly,2004; Perutka et al., 2004; Yao et al., 2006; Zhonget al., 2003). We recently developed a Clostridium-specific targetron donor plasmid and used it to suc-cessfully inactivate the plc gene in the C. perfringens’

chromosome(Chen et al., 2005).Theta toxin produced by C. perfringens type A is

a lethal toxin that contributes to virulence of C. per-

fringens (Awad et al., 1995). The pfoA gene ofC. perfringens ATCC3624 used for development oforal delivery vehicle is located on the bacterial chro-mosome and is 1500 bp nucleotides encoding the500 amino acid theta toxin. From the 10 targetroninsertion sites in pfoA gene of C. perfringens

Y. Chen et al. / Plasmid 58 (2007) 182–189 185

ATCC3624 predicted by the Sigma TargetronDesign web site (http://www.sigma-genosys.com/targetron/checksequence/), the insertion site in theantisense strand at position 32/33 from the initialATG in pfoA gene was chosen because a targetroninserted in the antisense orientation is transcribedinto the complement of the intron RNA sequence,cannot be spliced, and yields a permanent targetgene disruption. The pfoA targetron carrying donorplasmid pJIR750hi was constructed (Fig. 1) by mod-ifying the plc targetron-carrying donor plasmidpJIR750ai, as described in Section 2. Subsequently,pJIR750hi was electroporated into our previouslyconstructed plc knockout mutant of C. perfringens

ATCC3624. CM-resistant colonies were screenedto identify targetron insertions using PCR withone primer hybridizing to intron sequences andanother to pfoA sequences present near the insertionsite. 42 out of 44 colonies analyzed showed apositive PCR result (data not shown).

Fig. 1. Schematic diagram of plasmid constructions of a pfoA targetron(pJIR750hi/p27).

Because of a high chromosomal insertion fre-quency, the pfoA targetron in pJIR750hi was subse-quently engineered to carry SIV p27 sequences. Forthis purpose, the SIV p27 ORF was amplified fromplasmid cp27 (Chen et al., 2004) by PCR and clonedinto the MluI site in pfoA targetron domain IV, anon-essential region that extends outside theintron’s catalytic core. Once constructed (Fig. 1),the pfoA targetron carrying p27 sequence(pJIR750hi/p27) was electroporated into the plc

knockout mutant of C. perfringens ATCC 3624.CM-resistant colonies were screened for insertionof this complex targetron by PCR using one primerhybridizing to the p27 ORF sequence and anotherto pfoA sequence present near the insertion site.0.2% of the colonies screened showed PCR positive(data not shown). Furthermore, PCR amplificationof the putative plc/pfoA double knockout mutantcolonies with primers hybridizing to pfoA geneflanking the insertion site produced the expected

(pJIR750hi) and a pfoA targetron carrying the p27 gene cassette

186 Y. Chen et al. / Plasmid 58 (2007) 182–189

2.7 kb PCR product, while a 400-bp PCR productwas amplified from the parental plc knockoutmutant colonies using the same primers (Fig. 2).Sequencing of the 2.7-kb PCR product confirmedthe complex targetron had inserted at the predictedsite in the pfoA gene (data not shown). After curingthe input plasmid, Southern blot analysis using anintron-specific probe was performed using EcoRI-digested DNA isolated from the wild-type parentalC. pefringens ATCC3624, the plc mutant and theplc/pfoA double knock out mutant with p27sequence in order to confirm that the targetronhad only inserted into the pfoA gene. As expected,the intron-specific probe did not hybridize toDNA from the ATCC3624 wild-type parent, buthybridized as a single band to DNA from plc

mutant and as two bands to DNA from the doublemutant (Fig. 3). The putative plc/pfoA doubleknockout mutant’s phenotype was verified byanaerobic growth on sheep blood agar plates(Fig. 3). Wild-type ATCC3624 produced theexpected double zones of hemolysis mediated byalpha toxin and theta toxin activities, while the plc

mutant produced only a single zone of hemolysis.

Fig. 2. (a) Schematic diagram showing the pfoA inactivation by insertitargetron insertion in the pfoA mutant by colony PCR using primer 1insertion: 400 bp.

As expected, the plc/pfoA double knockout mutantcarrying the p27 gene sequence was completelynon-hemolytic (Fig. 3). The effect was reversedwhen the plc/pfoA double knockout mutant carryingp27 gene was complemented by a recombinant plas-mid that carried the pfoA gene. A single zone ofhemolysis, as observed in the plc mutant, wasrestored in the plc/pfoA double knockout mutantcarrying p27 gene sequences and transformed withthe pfoA-carrying plasmid (data not shown).

Production of SIV p27 by the plc/pfoA doublemutant carrying p27 gene sequences was evaluatedby Western blotting using anti-SIV polyclonal anti-body. As shown in Fig. 4a cell lysates fromATCC3624 wild-type parent, plc mutant and plc/

pfoA double mutant carrying p27 gene sequences,p27 protein was detected only in sporulating cultureof the plc/pfoA double mutant carrying p27 genesequences, but not in vegetative culture, since p27expression is driven by the sporulation-dependentcpe promoter.

Stability of the toxin gene inactivation and p27production are particularly important for our goalof constructing C. perfringens based vectors, where

on of a targetron carrying the p27 gene cassette. (b) Detection ofand 2 shown in a. PCR product with insertion: 2.7 kb; without

Fig. 3. (a) Genotypic confirmation of targetron insertions by Southern blot using an intron-specific probe. (b) Phenotypic confirmation ofthe plc/pfoA inactivation by anaerobic growth of the mutant and related ATCC3624 derivatives (ATCC3624 wt parent, plc mutant, andplc/pfoA mutant carrying p27 ORF).

Fig. 4. (a) Western blot analysis of bacterial lysates from the ATCC3624 wild-type parent, plc mutant, and plc/pfoA double mutantcarrying p27 ORF grown as sporulating (spores) or vegetative (veg.) cultures. (b) Western blot analysis for p27 of small intestinal contentextract following oral feeding of plc/pfoA mutant carrying p27 gene of one mouse. Segment 1 to 4 are jejunum, segment 5 and 6 are ileum.

Y. Chen et al. / Plasmid 58 (2007) 182–189 187

reversion to toxin expression is not acceptable. Thestability of the plc/pfoA double knockout mutant car-rying p27 gene was evaluated after an additional 20rounds of vegetative culture and more than 10 roundsof sporulation, as assessed by three measurements: (i)

performing PCR with primers hybridizing to pfoA

gene flanking the insertion site, (ii) monitoring thetatoxin activity with anaerobic growth on sheep bloodagar plates and (iii) performing Western blot of spor-ulating bacterial mutants to examine p27 expression.

188 Y. Chen et al. / Plasmid 58 (2007) 182–189

The results demonstrate that the plc/pfoA doubleknockout mutants carrying p27 gene sequences main-tained their intron-inserted pfoA gene, growth prop-erty on sheep blood agar plates under anaerobicconditions and p27 production without any indica-tion of reversion to the wild-type pfoA gene (datanot shown). These three approaches confirmed thatthe chromosomal insertion of the complex targetroninto pfoA and expression of p27 from bacterial chro-mosome are very stable.

It is possible that disruption of one bacterialchromosomal gene may influence the functions ofother chromosomal genes. To address this question,the growth kinetics, spore formation and suscepti-bility to electroporation of the plc/pfoA doubleknockout mutant carrying p27 gene were evaluated.Compared with plc knockout mutant of C. perfrin-

gens ATCC3624, the mutant maintained all theproperties of the parental bacteria. Preservation ofthese bacterial properties controlled by numerousbacterial genes indicates that the targetron insertioninto theta gene does not grossly alter these traits ofthe mutant.

An important aspect of an oral delivery vehicle isits ability to deliver a large amount of intact pro-teins or therapeutic agents to the lower gastro-intes-tines (GI) tract, where the major GALTs arelocated. To assess whether the p27 protein producedby the plc/pfoA double mutant carrying p27 genesequences could survive the harsh conditions ofthe upper GI tract and be delivered to the lowerGI tract, mice were fed with 2 · 1010 CFU of spor-ulating plc/pfoA double mutants carrying p27 geneseuqences, which contained 400 lg of p27 protein.Ninety minutes after this feeding, the contents ofthe small intestines were extracted and subjectedto Western blot analysis for p27 protein. Fig. 4bshows that large amount of intact p27 protein wasdetected in terminal ileum.

p27 expression in the C. perfringens mutant isunder the control of sporulation-induced promoter.During sporulation, a certain amount of p27 formsan inclusion body in the cytoplasm of sporulatingcell due to the high expression and is released in sol-uble and particulate forms when the mother celllyses at the completion of sporulation. Therefore,this C. perfringens based oral delivery vehicle couldbe used to deliver intact soluble and particulate pro-teins to small as well as large intestines since a largeamount of intact p27 protein was detected in mousecolon at later time points after oral feeding (datanot shown).

In summary, using a Clostridium-specific-target-ron donor plasmid, we simultaneously inactivateda chromosomal gene and introduced a foreign geneonto the chromosome of C. perfringens. This engi-neered Clostridium does not harbor any input plas-mid or antibiotic genes. Furthermore, this bacterialmutant stably expresses a large amount of heterolo-gous protein and delivers the intact protein to theGALT after oral administration. Since the mutantC. perfringens strain carries inactivated alpha toxinand theta toxin genes and also does not produceany enteric toxins, it provides a safe delivery systemfor antigens or therapeutic compounds to GALT.More importantly, due to its high frequency andsite-specific insertion, our targetron approach is apromising tool to facilitate construction of C. per-

fringens mutants or other bacterial mutants carryingmultiple heterologous DNA insertions for expres-sion from a single bacterial chromosome. Thisopens the possibility of using this technology toco-express antigens and mucosal adjuvant in thesame bacterial cell, allowing their co-delivery intoGALT.

Acknowledgments

We thank Samera Sayeed, Jihong Li, and MilkaRodriguez for technical support and helpful discus-sions. This work was funded by a Grant R03AI067515-01(YC), R21 AI065352-01(PG), andRO1 056177-05 (BMc) from the National Instituteof Allergy and Infectious Diseases.

References

Awad, M.M., Bryant, A.E., Stevens, D.L., Rood, J.I., 1995.Virulence studies on chromosomal alpha-toxin and theta-toxin mutants constructed by allelic exchange provide geneticevidence for the essential role of alpha-toxin in Clostridium

perfringens-mediated gas gangrene. Mol. Microbiol. 15 (2),191–202.

Chen, Y., Helmus, R., McClane, B., Hoffman, R., Watkins, S.,Wehrli, T., Gupta, P., 2004. Use of a Clostridium perfringens

vector to express high levels of SIV p27 protein for thedevelopment of an oral SIV vaccine. Virology 329 (2),226–233.

Chen, Y., McClane, B.A., Fisher, D.J., Rood, J.I., Gupta, P.,2005. Construction of an alpha toxin gene knockout mutantof Clostridium perfringens type A by use of a mobile group IIintron. Appl. Environ. Microbiol. 71 (11), 7542–7547.

Detmer, A., Glenting, J., 2006. Live bacterial vaccines—a reviewand identification of potential hazards. Microb. Cell Fact. 5, 23.

Fischetti, V.A., Medaglini, D., Pozzi, G., 1996. Gram-positivecommensal bacteria for mucosal vaccine delivery. Curr. Opin.Biotechnol. 7 (6), 659–666.

Y. Chen et al. / Plasmid 58 (2007) 182–189 189

Fisher, D.J., Miyamoto, K., Harrison, B., Akimoto, S., Sarker,M.R., McClane, B.A., 2005. Association of beta2 toxinproduction with Clostridium perfringens type A humangastrointestinal disease isolates carrying a plasmid entero-toxin gene. Mol. Microbiol. 56 (3), 747–762.

Kokai-Kun, J.F., Songer, J.G., Czeczulin, J.R., Chen, F.,McClane, B.A., 1994. Comparison of Western immunoblotsand gene detection assays for identification of potentiallyenterotoxigenic isolates of Clostridium perfringens. J. Clin.Microbiol. 32 (10), 2533–2539.

Lambowitz, A.M., Zimmerly, S., 2004. Mobile group II introns.Annu. Rev. Genet. 38, 1–35.

Perutka, J., Wang, W., Goerlitz, D., Lambowitz, A.M., 2004. Useof computer-designed group II introns to disrupt Escherichiacoli DExH/D-box protein and DNA helicase genes. J. Mol.Biol. 336 (2), 421–439.

Rood, J.I., Cole, S.T., 1991. Molecular genetics and pathogenesisof Clostridium perfringens. Microbiol. Rev. 55 (4), 621–648.

Smedley 3rd, J.G., Fisher, D.J., Sayeed, S., Chakrabarti, G.,McClane, B.A., 2004. The enteric toxins of Clostridium

perfringens. Rev. Physiol. Biochem. Pharmacol. 152, 183–204.Songer, J.G., 1996. Clostridial enteric diseases of domestic

animals. Clin. Microbiol. Rev. 9 (2), 216–234.Yao, J., Zhong, J., Fang, Y., Geisinger, E., Novick, R.P.,

Lambowitz, A.M., 2006. Use of targetrons to disrupt essentialand nonessential genes in Staphylococcus aureus revealstemperature sensitivity of Ll. LtrB group II intron splicing.RNA 12 (7), 1271–1281.

Zhong, J., Karberg, M., Lambowitz, A.M., 2003. Targeted andrandom bacterial gene disruption using a group II intron(targetron) vector containing a retrotransposition-activatedselectable marker. Nucleic Acids Res. 31 (6), 1656–1664.