Veterinary Microbiology 136 (2009) 202–205

Short communication

Prevalence of netB among some clinical isolates of Clostridium perfringensfrom animals in the United States

Thomas G. Martin, Joan A. Smyth *

Department of Pathobiology and Veterinary Science, University of Connecticut, 61 North Eagleville Road, Storrs, CT 06269, USA

A R T I C L E I N F O

Article history:

Received 22 September 2008

Received in revised form 22 October 2008

Accepted 24 October 2008

Keywords:

Clostridium perfringens

Necrotic enteritis

netB

Poultry

Chicken

Bovine

A B S T R A C T

A previously unknown pore forming toxin, called NetB toxin, which is produced by some

Australian strains of Clostridium perfringens has recently been reported. This toxin was

reported to be critical to the development of the disease necrotic enteritis, in chickens. To

investigate the occurrence of the toxin gene (netB) in non-Australian C. perfringens strains,

one hundred and six American isolates of C. perfringens were examined. Ninety-two

isolates were from chickens, and 14 were from cattle. The netB gene was found in 14

isolates from chickens (7 from chickens with necrotic enteritis, and 7 from unrelated

chickens with no evidence of necrotic enteritis). The netB gene was also detected in an

isolate recovered from a 3-year-old cow with liver abscesses. The products of all positive

netB PCR reactions were sequenced, and these showed 100% nucleotide identity to the netB

sequence published in GenBank. Five isolates which had been recovered from five chickens

with necrotic enteritis (from four flocks) were netB negative. An additional 24 isolates

recovered from one of these lesioned chickens were also netB negative. The present study

represents the first study of C. perfringens isolates outside Australia for netB, and the first

identification of netB in an isolate from a species other than chickens. The results indicate

that the role of NetB in the induction of necrotic enteritis needs to be further investigated,

by determining the disease producing capability of both netB positive strains recovered

from normal chickens, and netB negative strains recovered from chickens with necrotic

enteritis.

� 2008 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Veterinary Microbiology

journal homepage: www.elsev ier .com/ locate /vetmic

1. Introduction

Necrotic enteritis (NE), a serious disease of chickens andturkeys, was first described by Parrish in 1961 in England.It has since been reported in all poultry producingcountries as an important economic disease (Van Immer-seel et al., 2004). It has been estimated that this diseasecosts the world poultry industry roughly $2 billionannually (McReynolds et al., 2004).

Clostridium perfringens type A is the most commonlyrecovered clostridial type from cases of necrotic enteritis(Songer, 1996; Olkowski et al., 2008). Despite the disease

* Corresponding author. Tel.: +1 860 486 0820; fax: +1 860 486 2794.

E-mail address: joan.smyth@uconn.edu (J.A. Smyth).

0378-1135/$ – see front matter � 2008 Elsevier B.V. All rights reserved.

doi:10.1016/j.vetmic.2008.10.026

being first described almost 50 years ago, the virulencefactors which lead to the development of NE have yet to befully understood. Based on experimental studies, earlyinvestigators suggested that alpha toxin is the major toxininvolved in causing NE (Al-Sheikhly and Truscott, 1977a,b;Fukata et al., 1988), and there are recent reports thatvaccination against alpha toxin is at least partially protectiveagainst the disease (Lovland et al., 2004; Copper et al., 2009).However, Keyburn et al. (2006) have reported that alphatoxin knockout mutants of C. perfringens were able toproduce necrotic enteritis in an experimental model.

Earlier this year, Keyburn et al. (2008) reported thediscovery of a previously unidentified pore forming toxin ofC. perfringens which they named NetB, and the encodinggene, netB. netB was only identified in strains recovered fromchickens with NE. It was neither found in isolates from

Table 1

Origin and number of Clostridium perfringens isolates examined from

chickens. The number positive for netB is given in brackets.

Small intestine Cecum Liver Total

Birds with NE 12 (7a) 0 0 12 (7a)

Birds without NE 52 (4a) 27 (2a) 1 (1a) 80 (7a)

Total 64 (11a) 27 (2a) 1 (1a) 92 (14a)

NE, necrotic enteritis.a Number netB positive.

T.G. Martin, J.A. Smyth / Veterinary Microbiology 136 (2009) 202–205 203

chickens which did not have NE, nor in 32 C. perfringens

isolates from mammals. Keyburn et al. (2008) furtherreported that netB knockout mutants failed to produce NE inchickens, while such mutants complemented with the wildtype netB gene did cause NE. Clearly, the identification of thetoxin responsible for NE would represent a major advance inknowledge, but there remains a need to examine isolatesand cases of disease in other countries for the presence ofnetB. Here we report an investigation of the occurrence ofnetB among isolates of C. perfringens from chickens andcattle, and in particular, we examine its occurrence withrespect to the disease NE in chickens.

2. Materials and methods

2.1. Source of isolates

Isolates for the present study had been obtained fromchickens and cattle by two strategies: (1) swabbing oftissues from carcasses submitted to the ConnecticutVeterinary Medical Diagnostic Laboratory (CVMDL) fornecropsy examination (these animals originated mainly inConnecticut, but also in other New England states, NewYork and Pennsylvania) and (2) swabbing of the cecum andsmall intestine of all birds which died during rear on abroiler farm in Connecticut (25 isolates). For all chickensswabbed, a gross examination for evidence of NE wasperformed by experienced observers. Histological exam-ination was also performed on most birds.

All swabs were streaked onto 5% sheep blood agar andincubated overnight at 37 8C under anaerobic conditions.Resultant colonies which exhibited a double zone ofhemolysis, and which failed to grow under aerobicconditions, were tentatively identified as C. perfringens.Isolates were stored at �80 8C until further testing.

C. perfringens reference strains for genotyping werekindly provided by Dr. J. Glenn Songer (University ofArizona, Tucson, AZ).

2.2. Selection criteria for chicken isolates to be examined

Twelve isolates recovered from the small intestine of 12chickens with NE were examined. These birds hadoriginated from seven separate farms which were experi-encing an increased mortality rate. Later, an additional 24isolates from lesioned intestine in one of these birds werealso examined.

From chickens without evidence of NE, 70 isolates from61 chickens on 24 farms were initially examined. Theyrepresented 33 isolates from broilers and 37 isolates fromadult birds. They also represented 19 isolates from cecum,50 from small intestine, and 1 from a lesioned liver. For 7 ofthe 61 birds, an isolate from both small intestine andcecum was tested, and in the case of 2 birds, twopresumptive C. perfringens isolates which had differentcolonial morphologies but were from the same tissuelocation, were examined.

A further 10 isolates from chickens without evidence ofNE were subsequently selected for examination based onthe detection of netB among the original selection of 70.Thus, if netB was found in an isolate from a bird, then all C.

perfringens isolates recovered from that bird and itsflockmates were targeted for testing. This accounted for8 additional isolates from the cecum and 2 from the smallintestine (see Table 1 for details of tissue of origin of all 80isolates from chickens with no evidence of NE).

2.3. Selection criteria for bovine isolates

Fourteen isolates recovered from cattle that had diedfrom gastrointestinal disease on 13 different farms wereexamined.

2.4. Toxinotyping for major toxins, beta2 toxin and

enterotoxin

All isolates which had been tentatively identified as C.

perfringens as described above, were tested for thepresence of cpa, cpb, iA, etx, cpb2 and cpe which encoderespectively for alpha toxin, beta toxin, iota toxin, epsilontoxin, beta2 toxin and enterotoxin of C. perfringens, byusing a multiplex PCR method. The primer, reagentconcentrations and reaction conditions were as previouslypublished (Songer and Bueschel, 1999) except that TaqBuffer Advance (5 PRIME, Gaithersburg, MD) was usedinstead of Fisher Assay Buffer A. DNA templates wereprepared as described by Songer and Bueschel (1999). PCRproducts were visualized by UV transillumination, follow-ing electrophoresis through a 1.5% agarose gel containing1.0 mg/ml of ethidium bromide.

2.5. Detection of netB

All isolates were examined for the presence of the netB

gene as described by Keyburn et al. (2008) except that theprimers were added to a final concentration of 0.1 mM and5 ml of DNA template prepared as above for the multiplexPCR was used. PCR products were visualized as describedabove.

2.6. DNA sequencing

All products obtained following PCR for netB werepurified using the QIAquick PCR Purification Kit (Qiagen,Valencia, CA) according to the manufacturer’s instructions.The DNA concentration of the purified products wasdetermined using a spectrophotometer (NanoDrop 1000,Thermo Scientific, Waltham, MA) and sequencing wasperformed in both the forward and reverse direction at theUniversity of Connecticut’s Biotechnology BioserviceCenter. Sequences were assembled and analyzed usingDNAMAN (Lynnon Biosoft, Quebec, Canada). They were

T.G. Martin, J.A. Smyth / Veterinary Microbiology 136 (2009) 202–205204

also compared to the published sequence for netB

(GenBank accession no. EU143239) (Keyburn et al., 2008).

3. Results

3.1. Toxinotyping (including netB)

All presumptive C. perfringens isolates were confirmedas C. perfringens by genotyping. The 25 isolates examinedfrom NE lesions in one chicken were all the sametoxinotype (type A, cpb2 negative, cpe negative, netB

negative), and so, with respect to the following prevalencedata, they are considered as one isolate.

Of 106 isolates, 104 were type A (98.1%) and 2 were typeE (1.9%). Of the 104 type A isolates, 15 were netB positive(14.4%), 39 were cpb2 positive (37.5%) and 2 (1.9%) werecpe positive. Fourteen of the netB positive C. perfringens

isolates had been recovered from chickens (Table 1), andone had been recovered from the liver of a 3-year-old cowwith liver abscesses and enteritis.

All C. perfringens isolates recovered from chickens weretype A. The two C. perfringens type E isolates were negativefor netB and are not considered further.

3.2. Sequence analysis of netB PCR products

The 15 amplicons obtained in the netB PCR reaction wereall 384 base pairs in length. Comparison of these sequencesto each other and to the netB sequence published in GenBankrevealed 100% identity at the nucleotide level. This includesthe netB sequence from the bovine C. perfringens isolatewhich was positive for netB.

3.3. Analysis of the occurrence of netB among chicken isolates

Of the 14 netB positive isolates from chickens, 7 hadbeen recovered from chickens with NE, while 7 wererecovered from chickens which did not have evidence ofNE at necropsy examination by experienced personnel(Table 1). Further analysis of the occurrence of the netB

isolates is provided below.

3.4. Distribution of netB positive isolates among chickens

with necrotic enteritis

Of the 12 C. perfringens isolates which had beenrecovered from chickens with NE, 7 (58.3%) were netB

positive (Table 1). These netB positive isolates had beenrecovered from lesioned small intestine of seven differentbirds, which were derived from three different flocks.

The remaining five C. perfringens isolates (41.7%) werenetB negative. These had been recovered from five chickenswith NE lesions from four flocks. The 24 additional isolatesexamined from lesioned intestine of one of these birdswere also netB negative.

3.5. Distribution of netB positive isolates among chickens

with no evidence of necrotic enteritis

Of the initial 70 C. perfringens isolates from chickenswith no evidence of NE, 6 (8.6%) were netB positive.

Examination of the 10 additional C. perfringens isolatesfrom the same or related birds yielded 1 further netB

positive isolate. Thus in total, 7 out of 80 (8.8%) C.

perfringens isolates from chickens without evidence of NE,were positive for netB (Table 1). These seven netB positiveisolates had been recovered from two broiler birds and fouradult birds, on four unrelated farms. One adult bird hadnetB positive isolates in both the small intestine andcecum. Isolates from the small intestine and cecum of thebird in which a netB positive isolate was found in the liver,were negative for netB. The 25 isolates from the broilerfarm studied throughout rear (Strategy 2, Section 2) wereall netB negative.

Chickens without NE originated on 17 different farms; 4farms (23.5%) harbored birds that were netB positive. Oneof these flocks was receiving an antibiotic growthpromoter (AGP) in the diet.

3.6. Occurrence of netB compared to the occurrence

of cpb2 and cpe

All seven netB positive isolates from cases of NE werealso cpb2 positive, while the five isolates from NE cases thatwere netB negative were also cpb2 negative. Of the sevennetB positive isolates from chickens without NE, four(57.1%) isolates were also cpb2 positive. The netB positiveisolate from a bovine was also cpb2 positive. Thus in total,12 out of the 15 (80%) netB positive isolates were cpb2

positive.Of the 39 cpb2 positive type A isolates from all animal

sources, 12 (30.8%) were netB positive, while from the 67cpb2 negative type A isolates only, 3 (4.8%) were netB

positive. No isolates were positive for both netB and cpe.

4. Discussion

It has recently been reported by Keyburn et al. (2008)that NetB, a newly discovered pore forming toxin, is criticalto the development of NE in chickens. In very strongsupport of this, they reported that netB knockout mutantsfailed to produce NE in chickens, while such mutantscomplemented with the wild type netB gene, caused NE.They also reported that they recovered netB positivestrains only from chickens with NE. The definitivediscovery of the toxin which enables C. perfringens toproduce NE obviously would represent a major break-through in the understanding of the pathogenesis of thisserious and economically important disease. The presentstudy aimed to extend the findings of Keyburn et al. (2008)by investigating the occurrence of the netB gene amongnon-Australian C. perfringens isolates and to examine for anassociation with NE. In the present study, isolates fromchickens suffering from NE, and isolates from chickenswith no evidence of NE, as well as isolates from cattle wereexamined.

Keyburn et al. (2008) did not find netB among isolatesfrom chickens that did not have NE. However, in thepresent study, netB was found in 7 out of 80 (8.8%) of suchisolates. One of the farms of origin was feeding AGPs,indicating that this practice did not eliminate netB positivestrains. It would be interesting to determine if the netB

T.G. Martin, J.A. Smyth / Veterinary Microbiology 136 (2009) 202–205 205

positive strains recovered from the non-lesioned birdswere capable of producing NE in a disease model.

Examination of the occurrence of netB among isolatesfrom chickens with NE showed that, while the netB genewas present in some of the isolates recovered from cases ofNE, it was not detected in 5 of 12 (41.7%). This finding is notconsistent with the hypothesis that NetB is the cause of NE.To exclude the possibility that the netB negative findingsrepresented inadvertent selection of a negative strain fromamong a mixed population of negative and positive strains,24 additional C. perfringens isolates from a lesioned area inone of these diseased birds were tested. All were netB

negative, providing strong evidence that the C. perfringens

involved in this case, were genuinely netB negative. In thiscontext, it is worth noting that Keyburn et al. (2008) statedin discussion that they also found four netB negativestrains from cases of NE, and furthermore, these netB

negative isolates caused NE when administered in adisease model. This finding does not support theirsuggestion that NetB is the critical factor in diseasedevelopment. Taken together, all of the preceding findingssuggest that either NetB is not essential to the develop-ment of NE in all cases, or that netB is very readily lost bythe organism, or that netB was not detected due to possiblenucleotide variation in the primer binding region. Thelatter perhaps seems less likely considering that all 15 ofour netB amplicons showed 100% nucleotide identity witheach other, and with the Australian netB sequence.

Twelve out of the 15 netB positive isolates also harboredthe plasmid borne cpb2 gene, raising the possibility of anassociation. While not a major focus of this study, it wasalso interesting that cpe was found in two isolates fromchickens. The gene has only been reported in a smallnumber of studies involving C. perfringens from chickens(Gholamiandekhordi et al., 2006; Miwa et al., 1997).

Finally, the finding of netB in an isolate from a cowrepresents the first report of this gene in a non-chicken C.

perfringens isolate. This isolate was recovered from a cowwith multiple liver abscesses and enteritis. Knownpyogenic bacteria were also recovered from the abscesses,and the C. perfringens organisms were considered anincidental finding at the time.

In conclusion, netB has now been confirmed to occur innon-Australian strains of C. perfringens. It was found innormal chickens at a low incidence (8.8%) and in 58.3% ofchickens with necrotic enteritis. Experimental studies in adisease model are necessary to investigate the diseaseproducing capabilities of both the netB negative strainsrecovered from cases of NE, and also the netB positivestrains recovered from normal chickens.

netB has also been identified for the first time in a C.

perfringens isolate from a species other than chicken, i.e.

cattle. It was not considered responsible for disease in thatanimal.

Acknowledgements

The authors gratefully acknowledge partial fundingfrom USDA Animal Health Programme (CONS00805) andthe remaining funding from the University of Connecticut’sStart-up Programme.

References

Al-Sheikhly, F., Truscott, R.B., 1977a. The interaction of Clostridium per-fringens and its toxins in the production of necrotic enteritis ofchickens. Avian Dis. 21, 256–263.

Al-Sheikhly, F., Truscott, R.B., 1977b. The pathology of necrotic enteritis ofchickens following infusion of broth cultures of Clostridium perfrin-gens into the duodenum. Avian Dis. 21, 30–240.

Copper, K.K., Trinh, H.T., Songer, J.G., 2009. Immunization with re-combinant alpha toxin partially protects broiler chicks against experi-mental challenge with Clostridium perfringens. Vet. Microbiol. 133,92–97.

Fukata, T., Hadate, Y., Baba, E., Uemura, T., Arakawa, A., 1988. Influence ofClostridium perfringens and its toxin in germ-free chickens. Res. Vet.Sci. 44, 68–70.

Gholamiandekhordi, A.R., Ducatelle, R., Heyndrickx, M., Haesebrouck, F.,Van Immerseel, F., 2006. Molecular and phenotypical characterizationof Clostridium perfringens isolates from poultry flocks with differentdisease status. Vet. Microbiol. 113, 143–152.

Keyburn, A.L., Sheedy, S.A., Ford, M.E., Williamson, M.M., Awad, M.M.,Rood, J.I., Moore, R.J., 2006. Alpha-toxin of Clostridium perfringens isnot an essential virulence factor in necrotic enteritis in chickens.Infect. Immun. 74, 6496–6500.

Keyburn, A.L., Boyce, J.D., Vaz, P., Bannam, T.L., Ford, M.E., Parker, D., DiRubbo, A., Rood, J.I., Moore, R.J., 2008. NetB, a new toxin that isassociated with avian necrotic enteritis caused by Clostridium per-fringens. PLoS Pathog. 4, e26.

Lovland, A., Kaldhusdal, M., Redhead, K., Skjerve, E., Lillehaug, A., 2004.Maternal vaccination against subclinical necrotic enteritis in broilers.Avian Pathol. 33, 81–92.

McReynolds, J.L., Byrd, J.A., Anderson, R.C., Moore, R.W., Edrington, T.S.,Genovese, K.J., Poole, T.L., Kubena, L.F., Nisbet, D.J., 2004.Evaluation of immunosuppressants and dietary mechanisms inan experimental disease model for necrotic enteritis. Poult. Sci.83, 1948–1952.

Miwa, N., Nishina, T., Kubo, S., Honda, H., 1997. Most probable numbers ofenterotoxigenic Clostridium perfringens in intestinal contents ofdomestic livestock detected by nested PCR. J. Vet. Med. Sci. 59,557–560.

Olkowski, A.A., Wojnarowicz, C., Chirino-Trejo, M., Laarveld, B., Sawicki,G., 2008. Sub-clinical necrotic enteritis in broiler chickens: noveletiological consideration based on ultra-structural and molecularchanges in the intestinal tissue. Res. Vet. Sci. 85, 543–553.

Parish, W.E., 1961. Necrotic enteritis in the fowl (Gallus gallus domesticus).I. Histopathology of the disease and isolation of a strain of Clostridiumwelchii. J. Comp. Pathol. 71, 377–393.

Songer, J.G., 1996. Clostridial enteric diseases of domestic animals. Clin.Microbiol. Rev. 9, 216–234.

Songer, J.G., Bueschel, D., 1999. Multiplex PCR Procedure for GenotypingClostridium perfringens. http://www.microvet.arizona.edu/Faculty/songer/multiplexprocedure.pdf.

Van Immerseel, F., De Buck, J., Pasmans, F., Huyghebaert, G., Haesebrouck,F., Ducatelle, R., 2004. Clostridium perfringens in poultry: an emergingthreat for animal and public health. Avian Pathol. 33, 537–549.