Published Ahead of Print 18 July 2007. 10.1128/CVI.00162-07.

2007, 14(9):1070. DOI:Clin. Vaccine Immunol. R. R. Kulkarni, V. R. Parreira, S. Sharif and J. F. Prescott Enteritis

-Induced NecroticClostridium perfringensImmunization of Broiler Chickens against

http://cvi.asm.org/content/14/9/1070Updated information and services can be found at:

These include:

REFERENCEShttp://cvi.asm.org/content/14/9/1070#ref-list-1at:

This article cites 34 articles, 14 of which can be accessed free

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

on O

ctober 26, 2014 by HS

C LIB

RA

RY

http://cvi.asm.org/

Dow

nloaded from

CLINICAL AND VACCINE IMMUNOLOGY, Sept. 2007, p. 1070–1077 Vol. 14, No. 91556-6811/07/$08.00�0 doi:10.1128/CVI.00162-07Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Immunization of Broiler Chickens againstClostridium perfringens-Induced

Necrotic Enteritis�

R. R. Kulkarni, V. R. Parreira, S. Sharif, and J. F. Prescott*Department of Pathobiology, University of Guelph, Guelph, Ontario N1G 2W1, Canada

Received 6 April 2007/Returned for modification 29 May 2007/Accepted 3 July 2007

Necrotic enteritis (NE) in broiler chickens is caused by Clostridium perfringens. Currently, no vaccine againstNE is available and immunity to NE is not well characterized. Our previous studies showed that immunity toNE followed oral infection by virulent rather than avirulent C. perfringens strains and identified immunogenicsecreted proteins apparently uniquely produced by virulent C. perfringens isolates. These proteins were alpha-toxin, glyceraldehyde-3-phosphate dehydrogenase, pyruvate:ferredoxin oxidoreductase (PFOR), fructose 1,6-biphosphate aldolase, and a hypothetical protein (HP). The current study investigated the role of each of theseproteins in conferring protection to broiler chickens against oral infection challenges of different severities withvirulent C. perfringens. The genes encoding these proteins were cloned and purified as histidine-taggedrecombinant proteins from Escherichia coli and were used to immunize broiler chickens intramuscularly.Serum and intestinal antibody responses were assessed by enzyme-linked immunosorbent assay. All proteinssignificantly protected broiler chickens against a relatively mild challenge. In addition, immunization withalpha-toxin, HP, and PFOR also offered significant protection against a more severe challenge. When the birdswere primed with alpha-toxoid and boosted with active toxin, birds immunized with alpha-toxin were providedwith the greatest protection against a severe challenge. The serum and intestinal washings from protected birdshad high antigen-specific antibody titers. Thus, we conclude that there are certain secreted proteins, inaddition to alpha-toxin, that are involved in immunity to NE in broiler chickens.

Necrotic enteritis (NE) is an economically important entericdisease of chickens caused by Clostridium perfringens. The dis-ease is usually controlled by antimicrobial drugs administeredat prophylactic doses either in water or in feed. However, thereis concern about the routine prophylactic use of antimicrobialdrugs in food animal production because of their contributionto antimicrobial resistance problems. If antimicrobial drugswere banned for such purposes in North America, there mightbe an increase in NE in broiler flocks, as has happened inScandinavia (12).

Although vaccination offers an alternative approach to an-timicrobial drugs for control of the disease, very little is knownabout immunity to NE. However, there has been considerablework on immunity to C. perfringens in other circumstances,since it is a cause of gas gangrene in people. That work hasidentified the alpha-toxin, a phospholipase C exoenzyme, bothas a major virulence factor and as an important protectiveimmunogen (5, 30, 34). In addition, based on naturally occur-ring antibodies or maternal vaccination, some studies suggestthat antibodies to alpha-toxin are important in immunity to NEin chickens (10, 19). However, the role of alpha-toxin or anyother protein in immunity to NE in chickens remains to bedemonstrated, and one study has shown the immunizing effectsof alpha-toxin-negative mutants (32). A recent study also dem-onstrated that an alpha-toxin-negative mutant produced NE

experimentally in chickens, demonstrating that factors otherthan alpha-toxin are important in the pathogenesis of NE (14).

Recent studies from this laboratory showed that the immu-nizing ability for protection against NE was associated withinfection with virulent rather than with avirulent C. perfringensstrains (32). Several proteins apparently uniquely expressed byvirulent, protective C. perfringens strains that reacted to serumand intestinal antibodies from infection-immunized birds wereidentified by mass spectrometry (15). These secreted proteinswere alpha-toxin, glyceraldehyde-3-phosphate dehydrogenase(GPD), pyruvate: ferredoxin oxidoreductase (PFOR), fructose1,6-biphosphate aldolase (FBA), and a hypothetical protein(HP). On the basis of these findings, the objective of thecurrent study was to investigate the role of each of theseproteins in immunizing broiler chickens against experimentalchallenges of various severities with virulent C. perfringens.

MATERIALS AND METHODS

Cloning, overexpression, and purification. Escherichia coli strains were used toclone and express the genes of interest; E. coli DH5� (recA lacZ�M15) (Strata-gene, La Jolla, CA) was used as the host for plasmid construction, and E. coliBL21-Star(DE3) (F� ompT hsdSB (rB

� mB�) gal dcm rne-131) (Invitrogen,

Carlsbad, CA) was used for the overexpression of histidine-tagged fusion pro-teins. These strains were grown in Luria-Bertani medium at 37°C, and whenrequired, kanamycin was added to the medium at a concentration of 50 �g/ml.The chromosomal DNA of virulent, protective C. perfringens strain CP4 was usedas the source of DNA for the expression of the secreted antigens. PCR ampli-fications were performed with a Platinum PCR SuperMix high-fidelity kit (In-vitrogen, Burlington, ON, Canada) and the specific primers described in Table 1.After purification (PCR purification kit; QIAGEN, Mississauga, ON, Canada),the PCR products (alpha-toxin, HP, GPD, FBA, and truncated PFOR [tPFOR])were cloned into vector pET28a (N�- and/or C�-terminal His tag vector, Kmr;Novagen Inc., Madison, WI) to generate proteins fused with histidine residues

* Corresponding author. Mailing address: Department of Pathobi-ology, University of Guelph, Guelph, Ontario N1G 2W1, Canada.Phone: (519) 824-4120, ext. 54453. Fax: (519) 824-5930. E-mail:prescott@uoguelph.ca.

� Published ahead of print on 18 July 2007.

1070

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

(six-His). The resulting plasmids were introduced into E. coli BL21-Star(DE3),following the manufacturer’s instructions. The nucleotide sequences of thecloned PCR products were verified by sequencing both strands. Expression ofrecombinant proteins by E. coli was induced by isopropyl-�-D-thiogalactopyrano-side at a final concentration of 1 mM. The histidine-tagged proteins were purifiedunder native conditions by affinity chromatography on nickel-nitrilotriacetic acid(Ni-NTA) agarose, following the manufacturer’s instructions (QIAGEN).Briefly, when the proteins were expressed as soluble proteins, the bacterialpellets were resuspended in a lysis/binding buffer (50 mM NaH2PO4, 300 mMNaCl, 10 mM imidazole) containing lyzozyme (1 mg/ml) and were incubated for60 min on ice. The bacterial cells were lysed by using a French pressure cell(three to four cycles of 1,000 lb/in2). The supernatant was collected by centrif-ugation and was added to the Ni-NTA agarose. The washing and elution stepswere performed with buffers containing increasing concentration of imidazole(20 mM to 250 mM). Finally, imidazole was removed from the eluted material bydialysis against phosphate-buffered saline (PBS; pH 7.2), the recombinant pro-teins were concentrated with an Amicon filter (pore size, 10 kDa; Millipore,Billerica, MA), and the protein concentration was determined by using aPlusOne 2-D Quant kit (Amersham Biosciences, San Francisco, CA).

SDS-PAGE and Western immunoblotting. Purified recombinant proteins wereseparated by one-dimensional sodium dodecyl sulfate (SDS)-polyacrylamide gelelectrophoresis (PAGE) in a 12.5% acrylamide gel under denaturing conditions,as described by Laemmli (16). The proteins were transferred to a nitrocellulosemembrane with a pore size of 0.45 �m by using a mini-gel transfer assembly(Bio-Rad Laboratories, Hercules, CA). After completion of the transfer, thenonspecific binding sites on the membranes were blocked for 1 h with blockingbuffer containing 1% casein (Bio-Rad Laboratories) and the membranes wereincubated with immune sera collected from infection-immunized birds in a pre-vious study (32) for 1 h at a 1:1,000 dilution. Goat anti-chicken immunoglobulinY (IgY; heavy and light chains; Cedarlane Laboratories, Hornby, ON, Canada)was used as the secondary antibody at a 1:2,000 dilution. The blots were devel-

oped, and specific immunoreactive protein bands were visualized by using analkaline phosphatase-conjugated substrate kit (Bio-Rad Laboratories).

Immunization and challenge. The experiments with chickens and the condi-tions for their use were approved by the University of Guelph Animal CareCommittee, in accordance with the Canadian Council on Animal Care’s guide-lines. Commercial 1-day-old male White Plymouth Rock broiler chickens (Bon-nie’s Chick Hatchery, Elmira, ON, Canada) were fed an antibiotic-free chickenstarter diet containing 20% protein for 13 days, followed by a formulated wheat-based grower feed containing 28% protein (Arkell Research Station, Universityof Guelph). The birds were immunized with purified recombinant proteins atdifferent dose levels in the pectoral muscle in a volume of 0.2 ml per bird two tothree times with an interval of 1 week and challenged a week after the lastimmunization, when the birds were 4 weeks old. For the experimental infection(challenge) of birds, virulent C. perfringens (CP4) was grown in cooked meatmedium (Difco) for 24 h at 37°C. Fluid thioglycollate medium (Difco) was theninoculated with a 3% (vol/vol) inoculum from the C. perfringens-infected cookedmeat medium and incubated at 37°C for 24 h. The growth at 24 h was 8.24 � 0.09C. perfringens log10 CFU/ml. The inoculated fluid thioglycollate medium wasthen mixed with feed at a ratio of 2:1 (vol/wt). The inoculated feed was freshlyprepared twice per day and fed to chickens that were fasted for 20 h prior tochallenge.

The general experimental design is summarized in Table 2. Quil-A (SuperfosBiosector, Vedbaek, Denmark) was used as an adjuvant to immunize the chick-ens (50 �g/bird/injection), and the unimmunized controls in each experimentreceived only Quil-A, followed by a challenge similar to that used for the im-munized groups. All the proteins were tested for their abilities to protect thebirds against a gradient of severity of challenge (mild-moderate-severe). A“mild” challenge (experiment 1) was produced by a duration of challenge of 3days; the mildness of the challenge was confirmed by the lesion scores for thenonimmunized birds. A “moderate” challenge (experiment 2) was produced by aduration of challenge of 5 days, in which the birds were fed a fixed amount of

TABLE 1. Primers used to amplify genes encoding proteins used in immunization experiments

Gene Direction Sequence (5�–3�) Amplicon size(bp)

Alpha-toxin Forward CCGCTCGAGTTGGGATGGAAAAATTGAT 1,100Reverse CCGGAATTCTTTATATTATAAGTTGAATTT

HP Forward CCGCTCGAGGAATAAGAGAAAAATAGCAG 5,400Reverse CCGGGTACCACGTTAAATAAATAGAACAT

GPD Forward CCGCTCGAGGGTAAAAGTAGCTATTAACGG 1,000Reverse CCGGGTACCTTAGAAACTAAGCATTTTAAA

FBA Forward CCGCGGATCCATGGCATTAGTTAACGCAAA 900Reverse CCGCCTCGAGAGCTCTGTTTACTGAACCGA

tPFOR Forward CCGCCTCGAGCACTTCATTAGAACCAGTTG 1,600Reverse CCGCGGATCCTAGCTAAGTAGTCTTGGTCTCCG

CGGATCCTAGCTAAGTAGTCTTGGTCT

TABLE 2. Summary of experimental design

Exptno. Immunization groupa Dosage (�g) of

vaccine/bird Frequency of administration No. of days (severity)of oral challenge

1 VC, Sup, alpha-toxoid, GPD, HP, FBA, and tPFOR 20 Three times; days 7, 14, and 21 3 (mild)2 VC, MC, GPD, HP, FBA, and tPFOR 40 Two times; days 7 and 14 5 (moderate)3 VC, alpha-toxoid/alpha-toxin,b GPD, HP, tPFOR, and

a combination of GPD and HP20 Three times; days 7, 14, and 21 5 (severe)

4A VC and FBA 20 Three times; days 7, 14, and 21 3 (mild-moderate)4B VC, alpha-toxin,c and FBA 20 Three times; days 7, 14, and 21 5 (severe)

a VC, vehicle-only controls; Sup, crude culture supernatant of virulent C. perfringens, in which the birds received 60 �g/injection of culture supernatant that wasprocessed and concentrated following a protocol described earlier (15); MC, mock-immunized controls, in which the birds were mock immunized with an unrelatedprotein that was cloned, expressed, and purified from E. coli in the same manner as the C. perfringens-related proteins.

b The birds received alpha-toxoid in the first two injections, followed by active alpha-toxin in the third.c The birds received three injections of alpha-toxin, in which the first and the third injections were with 20 �g and the second was reduced to 10 �g.

VOL. 14, 2007 IMMUNIZATION OF CHICKENS AGAINST C. PERFRINGENS 1071

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

feed that sometimes ran out before the next feeding (feedings were given every12 h); the moderateness of the challenge was confirmed by the lesion scores forthe nonimmunized birds. A “severe” challenge (experiment 3) was produced bychallenge for 5 days and by ensuring that the birds constantly had infected feedavailable; the severity of the challenge was confirmed by the lesion scores for thenonimmunized birds. In experiment 4A, the challenge was considered “mild-moderate,” since the birds in these groups that received virulent C. perfringens for3 days but were necropsied on day 6 were found to have lesion scores higher thanthose of birds that were challenged for 3 days and necropsied on day 4. Alpha-toxin was used in both an active and a toxoid form to immunize the birds, andtoxoid was prepared by a previously described protocol (11). Briefly, the purifiedtoxin was incubated with 0.2% formalin and 0.05 M L-lysine at 30°C until itsactivity was completely lost, as confirmed by a 5% egg yolk agar plate assay. Inall experiments, the number of birds in each group was between 10 and 20, andall birds were identified individually. Blood was collected from the wing veinfrom all the groups at three times: preimmunization (day 0), midexperiment (day10), and prechallenge (day 20). Intestinal washings were collected at necropsy byusing PBS.

Necropsy. The chickens were euthanized with carbon dioxide gas, and theirsmall intestines (duodenum to ileum) were examined for grossly visible lesions.Any chickens that had reached a predetermined severity of clinical illness priorto necropsy were euthanized and later necropsied. Intestinal lesions in the smallintestine (duodenum to ileum) were scored as follows: 0, no gross lesions; 1, thinor friable wall or very mild and superficial generalized inflammation; 2, focalnecrosis or ulceration; 3, large patches of necrosis; 4, severe extensive necrosis;5, death during the experiment with lesion scores of 4� (25). Blind scoring wasused to avoid scorer bias.

Measurement of antibody titers in chicken sera and intestinal washings. Thespecific antibody titers were determined by the end-point dilution method by useof an enzyme-linked immunosorbent assay (ELISA). Microtiter plates (Immu-lon-2; Dynatech Laboratories, Chantilly, VA) were coated with recombinantproteins (5 �g/ml in 0.1 M carbonate buffer, pH 9.6) for 60 min at 37°C, followedby an overnight incubation at 4°C. After the coated plates were blocked for 60min at 37°C with PBS containing 3% bovine serum albumin (BSA; Sigma), serafrom the immunized birds, along with sera from their concurrent controls, wereserially diluted in PBS containing 1% BSA and incubated for 2 h with recombi-nant protein-coated plates at room temperature. After the plates were washedwith PBS containing 0.1% Tween 20 (PBST), alkaline phosphatase-coupled goatanti-chicken IgY (heavy and light chains; diluted 1:5,000 in PBST–1% BSA) wasadded to the microplates and the mixture was incubated for 60 min at roomtemperature. After extensive washing of the plates with PBST, the color reactionwas developed by using a alkaline phosphate substrate kit (Bio-Rad Laborato-

ries), following the manufacturer’s instructions. The reaction was stopped byadding 0.4 M NaOH. The absorbance at 405 nm was measured in an ELISAspectrophotometer. The specific antibody titer of the immune serum was ex-pressed as the reciprocal of the serum dilution (log2 optical density) that gave anA405 value above the cutoff, which was defined as twice the absorbance value ofthe unimmunized and mock control wells run in duplicate.

The intestinal antibody response was also measured by ELISA by the proce-dure described above for serum. Intestinal washings from at least 10 chickens pergroup were pooled, and the total protein content was measured by using aPlusOne 2-D Quant kit (Amersham Biosciences). The total protein was used asthe source of the primary antibody after the protein content of the initial dilution(1:10) was kept constant across all the groups. Alkaline phosphatase-conjugatedgoat anti-chicken IgY and IgA were used as the secondary antibodies and wereused at dilutions of 1:4,000 and 1:2,000, respectively. The end-point titers weredetermined as described above for the serum ELISA.

Statistical analysis. Statistical analysis was performed to determine whetherthere was a significant difference between the numbers of birds with lesions fromthe immunized groups and the numbers of birds with lesions from the unimmu-nized, vehicle-only controls. A two-tailed Fisher’s exact test was used to deter-mine whether the two groups differed in the proportions that fell into the twoclassifications of either lesions or no lesions, under the null hypothesis that theproportions were the same. The data were analyzed by use of a two-by-twocontingency table, with the data for the unimmunized control group in onecolumn and the data for the immunized groups in the other column. The lesionscores were ranked from 0 to 5�; however, a “protective” response was given tobirds with lesions scores of �1�. The null hypothesis was rejected at � 0.05.For the serum ELISA, a one-way analysis of variance was used to determinesignificant (P � 0.01) differences in the antibody titers between preimmunizedand immunized birds across all the groups. Statistical analysis could not beperformed for the intestinal antibody response determined by ELISA, since thewashings collected were pooled for each group.

RESULTS

Cloning, expression, and purification. All five genes selectedfor the immunization study were successfully cloned, ex-pressed, and purified to homogeneity. However, alpha-toxinappeared to be toxic to the host E. coli cells, such that it couldnot be obtained in a quantity sufficient for the immunizationstudies. Hence, commercially available purified alpha-toxin

FIG. 1. Recombinant Clostridium perfringens histidine-tagged proteins purified from Escherichia coli cells. (A) Coomassie-stained purifiedproteins; (B) reactivities of purified proteins to immune serum from chickens immune to NE. Lanes 1, alpha-toxin (45 kDa); lanes 2, GPD (40kDa); lanes 3, FBA (35 kDa); lanes 4, tPFOR (67 kDa); lanes 5, HP (90 to 100 kDa); lanes M, molecular mass standards.

1072 KULKARNI ET AL. CLIN. VACCINE IMMUNOL.

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

(Sigma Laboratories) (and alpha-toxoid) was used to the im-munize the birds. HP (190 kDa) was found to be cleaved uponexpression into two bands of 90 to 100 kDa. Attempts toexpress the entire protein by using different E. coli expressionhosts were unsuccessful. Since both bands reacted strongly toantihistidine antibodies as well as to immune sera collectedfrom infection-immunized birds (Fig. 1) from a previous study(32), both bands were further purified in large quantities andused for immunization.

Cloning of the pfor gene was not successful, despite severalattempts. However, a portion of the gene that encoded a trun-cated protein (tPFOR) of 67 kDa that contained the iron-sulfur active sites of this enzyme was successfully cloned andpurified in large quantities.

All the recombinant proteins purified from E. coli werevisualized by Coomassie staining, and their reactivities toantihistidine antibodies as well as immune serum were con-firmed by Western blotting in at least three separate exper-iments (Fig. 1).

Immunization experiments. In experiment 1, HP, GPD,tPFOR, and FBA offered significant protection against a mildchallenge (Table 3), with HP offering the greatest protection.Immunization with crude culture supernatant that containedall secreted proteins, including those that were purified, alsooffered significant protection. Alpha-toxoid did not protect thebirds against challenge.

In experiment 2, HP alone offered significant protectionagainst a moderate challenge, whereas GPD, tPFOR, and FBAdid not (Table 4). Mock immunization with an unrelated pu-rified recombinant fusion protein resulted in a mean lesionscore similar to that for the unimmunized controls. The in-creased mean lesion score for the controls compared to thosefor the birds in experiment 1 appeared to reflect the increasedduration of challenge.

In experiment 3, birds that received two initial injections ofalpha-toxoid but a third injection with active alpha-toxin hadthe greatest protection against a heavy challenge (Table 5).Birds immunized with either HP or tPFOR also had significantprotection against severe challenge. Although birds immunizedwith GPD, FBA, and the combination of GPD and HP hadmean lesion scores less than those for the nonimmunized con-trols, no statistically significant protection was observed.

In experiment 4A, birds immunized with FBA had signifi-

cant protection against a mild-moderate challenge comparedto the level of protection for the unimmunized controls (Table6). The mean lesion scores for the unimmunized controls inexperiment 4B were comparable to the scores for the unim-munized controls in experiment 3 that also received a severechallenge.

A visual summary of the mean lesion scores for the birdsfrom all immunized groups that received different doses ofantigens and challenges across different experiments, togetherwith those for the concurrent unimmunized controls, is shownin Fig. 2.

Antibody titers in chicken sera and intestinal washings. Allthe proteins used to immunize birds in the immunization ex-periments described produced significant antigen-specific se-rum antibody titers (Fig. 3) in comparison to the preimmuni-zation titers. Birds immunized twice with a larger antigenquantity had lower titers than birds immunized three timeswith a smaller quantity. The protection provided in experiment3 was not as marked as that provided in experiment 1, eventhough the antibody titers were generally similar. This differ-ence was attributed to the difference in the severity of thechallenge. There was a discrepancy between the titers of anti-body to alpha-toxin and protection, since either alpha-toxoid-and alpha-toxin-immunized but nonprotected birds in experi-ments 1 and 4, respectively, had higher titers than the alphatoxoid- and alpha-toxin-immunized birds that were signifi-cantly protected in experiment 3.

TABLE 3. Intestinal lesion scores of birds immunized with threeinjections intramuscularly and then infected with a mild

challenge with C. perfringens

Protein No. ofchickens

No. of chickens with thefollowing lesion scores:

Meanno. of

chickens0 1� 2� 3� 4� 5�

Vehicle-only controls 10 1 3 4 1 1 0 1.55Culture supernatanta 10 8 1 1 0 0 0 0.4Alpha-toxoid 12 3 4 3 1 1 0 1.41HPa 12 10 2 0 0 0 0 0.16GPDa 10 7 2 1 0 0 0 0.4tPFORa 10 4 5 1 0 0 0 0.7FBAa 10 4 6 0 0 0 0 0.6

a Immunized groups that had significantly fewer chickens with lesions than theunimmunized vehicle-only control group (Fisher’s exact test, P � 0.05).

TABLE 4. Intestinal lesion scores of birds immunized with twoinjections intramuscularly and then infected with a moderate

challenge with C. perfringens

Protein No. ofchickens

No. of chickens with thefollowing lesion scores:

Meanno. of

chickens0 1� 2� 3� 4� 5�

Vehicle-only controls 18 1 4 9 2 1 1 2.05Mock controls 10 0 1 6 3 0 0 2.20HPa 17 10 3 4 0 0 0 0.64GPD 17 3 8 6 0 0 0 1.17tPFOR 18 7 4 3 3 0 1 1.33FBA 18 2 7 6 2 0 1 1.66

a The immunized group had significantly fewer chickens with lesions then thevehicle-only control group (Fisher’s exact test, P � 0.05).

TABLE 5. Intestinal lesion scores of birds immunized with threeinjections intramuscularly and then infected with a severe

challenge with C. perfringens

Protein No. ofchickens

No. of chickens with thefollowing lesion scores:

Meanno. of

chickens0 1� 2� 3� 4� 5�

Vehicle-only controls 22 0 5 5 6 4 2 2.68Alpha toxoid/toxina,b 19 10 8 1 0 0 0 0.53HPa 20 8 6 4 2 0 0 1.0GPD 18 4 4 6 1 1 1 1.64tPFORa 19 9 2 6 2 0 0 1.05GPD � HP 19 5 5 7 1 1 0 1.36

a Immunized groups that have significantly fewer chickens with lesions thanthe unimmunized vehicle-only control group (Fisher’s exact test, P � 0.05).

b The birds in this group received alpha-toxoid in the first two injections andalpha-toxin in the third.

VOL. 14, 2007 IMMUNIZATION OF CHICKENS AGAINST C. PERFRINGENS 1073

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

The intestinal antibody responses to all the proteins used forimmunization showed higher IgY titers than IgA titers (Fig. 4),but the titers of both isotypes were markedly lower than thoseobserved in the sera of the birds. However, the IgY and IgAtiters were generally similar in birds immunized with alpha-toxoid and alpha-toxin (experiments 1, 3, and 4).

DISCUSSION

This study has shown, for the first time, that a degree ofimmunity to NE in broiler chickens can be produced by im-munization with several different secreted C. perfringens pro-teins and that the degree of protection is a function of theseverity of the challenge. All the proteins used in immuniza-

FIG. 2. Summary of mean lesion scores for birds from all immunized groups across different experiments, together with those for the concurrentunimmunized controls. VC, vehicle-only controls, A-tox, alpha-toxin; Sup, culture supernatant of C. perfringens; G�H, combination of GPD andHP; Exp, experiment; �, the birds in this group were challenged for 3 days and necropsied on day 6; ��, the birds in this group were given a severechallenge, like the one used in experiment 3; *, the immunized group had significantly fewer chickens with lesions than the unimmunizedvehicle-only control group (Fisher’s exact test, P � 0.05).

FIG. 3. Serum IgY ELISA titers of broiler chickens immunized intramuscularly with Clostridium perfringens purified proteins. Serum wascollected at three time points: day 0, for determination of the preimmunization titer; day 10 (midexperiment); and day 20, for determination ofthe prechallenge titer. Exp, experiment. In experiment 4A, the birds were challenged for 3 days and necropsied on day 6. In experiment 4B, thebirds were given a severe challenge, like the one used in experiment 3. *, significant titer values compared to the preimmunization titers (P � 0.01).

1074 KULKARNI ET AL. CLIN. VACCINE IMMUNOL.

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

tion, including alpha-toxin, offered significant protection, de-pending on the severity of the challenge. It seems that protec-tion against NE lies in the secreted component of C.perfringens, since immunization with crude culture supernatantthat included all proteins tested in the current study largelyprotected the birds against challenge. Of the five secretedproteins used for immunization, three proteins, namely, alpha-toxin, HP, and tPFOR, significantly protected the chickensagainst a heavy challenge, whereas the other two proteins,GPD and FBA, significantly protected the birds only againstmild challenge. Nevertheless, a degree of protection againsteven severe challenge was apparent with the latter proteins.

The role of alpha-toxin in immunity to NE has been sus-pected but not previously clearly demonstrated. Priming withalpha-toxoid and boosting with active toxin offered the bestprotection, whereas immunization with three injections of ei-ther alpha-toxoid or of active toxin offered no protection (Ta-bles 3 and 6). The failure of active toxin to protect birds againsta heavy challenge may have resulted from the toxin’s activityon immune system cells. The failure of alpha-toxoid to offerprotection in experiment 1 may be the result of a degradationeffect from creation of the toxoid on the protein that wasobserved on the SDS-polyacrylamide gel (data not shown).However, it is clear from Fig. 3 that use of the toxoid wasadequate to induce antibodies sufficient for the birds to toler-ate the active toxin given as a booster in experiment 3. Thefindings from the present study and an earlier study (15) sug-gest that antibodies to conformational (rather than linear)epitopes of alpha-toxin are critical for protection against NE.Achieving conformational but nontoxic epitopes in a vaccinemay prove challenging (1, 7, 33). Although many studies haveemphasized the importance of the nontoxic C-terminal domainin offering protection against experimental gas gangrene (5, 30,34), some have shown that the neutralizing epitopes are on the

N terminus (18). It seems likely that the positioning of theprotective, neutralizing, conformational epitopes of alpha-toxin is subtle. For this reason, other immunogens such asthose identified here may be more feasible candidates for usefor immunization.

Perfringolysin O is a potent hemolytic cytolysin that medi-ates necrosis in the pathogenesis of clostridial gas gangrene(29) and is an important protective immunogen in mouse andguinea pig gas gangrene models (8). Our previous study sug-gested its possible role in NE immunity in broiler chickens(15). In the current study, the purified alpha-toxin (SigmaLaboratories) used to immunize birds had traces of perfringo-

FIG. 4. Intestinal IgY and IgA ELISA titers of broiler chickens immunized intramuscularly with Clostridium perfringens purified proteins. Thesamples analyzed were from pooled intestines collected from at least 10 chickens in each group. Exp, experiment. In experiment 4A, the birds werechallenged for 3 days and necropsied on day 6. In experiment 4B, the birds were given a severe challenge, like the one used in experiment 3.

TABLE 6. Intestinal lesion scores of birds immunized with threeinjections intramuscularly and then infected with a mild-moderate

or a severe challenge with C. perfringens

Group and protein No. ofchickens

No. of chickens with thefollowing lesion scores:

Meanno. of

chickens0 1� 2� 3� 4� 5�

Chickens infected withmild-moderatechallenge

Vehicle-only controls 10 0 5 2 0 1 2 2.3FBAa 13 4 6 1 1 0 1 1.23

Chickens infected withsevere challenge

Vehicle-only controls 11 0 3 2 3 1 2 2.72Alpha-toxinb 10 0 2 4 4 0 0 2.2FBA 14 1 4 9 0 0 0 1.57

a The immunized group had significantly fewer chickens with lesions than theunimmunized vehicle-only control group (Fisher’s exact test, P � 0.05).

b The birds in this group received three injections of alpha-toxin, in whichthe first and the third injections were with 20 �g and the second was reducedto 10 �g.

VOL. 14, 2007 IMMUNIZATION OF CHICKENS AGAINST C. PERFRINGENS 1075

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

lysin O, which we identified using mass spectrometry (data notshown). However, the relative amounts in the otherwise ap-parently pure toxin preparation (as assessed by SDS-PAGE)were not quantified. It is possible that the protection observedin alpha-toxoid- and alpha-toxin-immunized birds (Table 5)can be partly attributed to perfringolysin O or even to traces ofother but undetected immunogenic proteins and that a syner-gistic effect on the induction of neutralizing antibodies againstboth toxins may have contributed to better protection (2).

The observation that immunization with secreted proteinsother than alpha-toxin provides to birds some immunityagainst NE highlights the likely involvement of several proteinsin the pathogenesis of this infection. Both alpha-toxin andperfringolysin O are regulated in C. perfringens by the VirR-VirS two-component regulon (4, 26), a regulon that also con-trols the genes involved in energy metabolism, such as FBA, aswell as others that may be indirectly involved in bacterial vir-ulence (3, 13, 28). There is growing evidence that certain en-zymes such as GPD and FBA, which are conventionally re-garded as metabolic or “housekeeping” enzymes, may have a“dual role” in both the pathogenesis of and the immunity toother infections (6, 9, 17, 20, 22–24). Interestingly, a recentstudy showed that antibodies to FBA and GPD of Streptococ-cus pneumoniae showed age-dependent increased titers in thesera of children of different ages. Immunization of mice withrecombinant GPD and FBA offered significant protectionagainst respiratory challenge with virulent S. pneumoniae (17).A role for FBA in immunity to Onchocerca volvulus has alsobeen suggested (21). Similarly, PFOR, an enzyme crucial foranaerobic energy metabolism, has been suggested to have arole in immunity to invasive amoebiasis (31). HP is a novelprotein of C. perfringens of unknown function that has beenidentified in its genome (27) and that may have protease ac-tivity (zinc metallopeptidase), based on the analysis of its pro-tein structure (15). It will be of interest to determine whetherHP is a virulence determinant. It is apparent from the presentimmunization study that, besides alpha-toxin, other proteins(HP, GPD, tPFOR, and FBA) are important in some aspectsof the host-pathogen interaction during the disease process.The recent demonstration that alpha-toxin is apparently notessential in the pathogenesis of NE (15) supports the sugges-tion that other proteins are involved.

Alpha-toxoid- and alpha-toxin-immunized, protected birdshad lower antibody titers than toxoid-immunized birds thatwere not protected, suggesting the importance of conforma-tional epitope-specific neutralizing antibodies in mounting aprotective immune response. This implies the importance ofthe quality of the response in providing protection. The intes-tinal antibody response, as expected, was mainly dominated byIgY, since systemic immunization resulted in more antigen-specific IgY than antigen-specific IgA (Fig. 4) that reached themucosal surfaces under inflammatory or necrotic conditions ofthe gut, allowing the seepage of serum IgY at the site ofinfection. It is also possible that a mucosal IgY response ismore important in immunity to C. perfringens-induced NE,since a previous study showed that C. perfringens proteins tomucosal IgA in the intestinal washings collected from orallyinfection-immunized birds had weak reactivities (15). Immu-nization with HP, which significantly protected the birdsagainst challenge doses of all severities, in all three experi-

ments produced IgA titers higher than those in the other im-munized groups. However, this association of IgA titers withprotection was not evident in the groups immunized with ei-ther the alpha-toxoid and alpha-toxin or tPFOR, which alsosignificantly protected the birds against a heavy challenge inexperiment 3.

In conclusion, this is the first report that has demonstratedthe immunizing ability of C. perfringens secreted proteins, in-cluding alpha-toxin, in protecting broiler chickens against NE.It seems likely that some of the secreted proteins that appearto be important in NE immunity also play a previously unsus-pected role in the pathogenesis of the disease. This study alsosuggests that conformational epitopes of alpha-toxin are im-portant in immunity and that antibody to alpha-toxin providesbirds with better protection. Nevertheless, there are other pro-teins that might be suitable candidates for use in vaccines forthe prevention of this important disease.

ACKNOWLEDGMENTS

We thank the staff of the OMAFRA Isolation Facility, University ofGuelph, for the housing and care of the broiler chickens.

This work was supported by the Ontario Ministry of Agriculture,Food, and Rural Affairs; by the Poultry Industry Council, Ontario,Canada; and by the Saskatchewan Chicken Industry DevelopmentFund. We also thank the Natural Sciences and Engineering ResearchCouncil of Canada, Agriculture and Agri-Food Canada, and theCanadian Poultry Research Council of Canada for funding.

REFERENCES

1. Alape-Giron, A., M. Flores-Diaz, I. Guillouard, C. E. Naylor, R. W. Titball,A. Rucavado, B. Lomonte, A. K. Basak, J. M. Gutierrez, S. T. Cole, and M.Thelestam. 2000. Identification of residues critical for toxicity in Clostridiumperfringens phospholipase C, the key toxin in gas gangrene. Eur. J. Biochem.267:5191–5197.

2. Awad, M. M., D. M. Ellemor, R. L. Boyd, J. J. Emmins, and J. I. Rood. 2001.Synergistic effects of alpha-toxin and perfringolysin O in Clostridium perfrin-gens-mediated gas gangrene. Infect. Immun. 69:7904–7910.

3. Banu, S., K. Ohtani, H. Yaguchi, T. Swe, S. T. Cole, H. Hayashi, and T.Shimizu. 2000. Identification of novel VirR/VirS-regulated genes in Clos-tridium perfringens. Mol. Microbiol. 35:854–864.

4. Ba-Thein, W., M. Lyristis, K. Ohtani, I. T. Nisbet, H. Hayashi, J. I. Rood,and T. Shimizu. 1996. The virR/virS locus regulates the transcription of genesencoding extracellular toxin production in Clostridium perfringens. J. Bacte-riol. 178:2514–2520.

5. Bennett, A. M., T. Lescott, R. J. Phillpotts, M. Mackett, and R. W. Titball.1999. Recombinant vaccinia viruses protect against Clostridium perfringensalpha-toxin. Viral Immunol. 12:97–105.

6. Boel, G., H. Jin, and V. Pancholi. 2005. Inhibition of cell surface export ofgroup A streptococcal anchorless surface dehydrogenase affects bacterialadherence and antiphagocytic properties. Infect. Immun. 73:6237–6248.

7. Eaton, J. T., C. E. Naylor, A. M. Howells, D. S. Moss, R. W. Titball, and A. K.Basak. 2002. Crystal structure of the C. perfringens alpha-toxin with theactive site closed by a flexible loop region. J. Mol. Biol. 319:275–281.

8. Efimova, M. G., V. A. Blagoveshchenskii, and B. V. Khatuntseva. 1982.Protective properties of theta-hemolysin obtained by affinity chromatogra-phy. Zh. Mikrobiol. Epidemiol. Immunobiol. 12:87–92.

9. Grifantini, R., E. Bartolini, A. Muzzi, M. Draghi, E. Frigimelica, J. Berger,F. Randazzo, and G. Grandi. 2002. Gene expression profile in Neisseriameningitidis and Neisseria lactamica upon host-cell contact: from basic re-search to vaccine development. Ann. N. Y. Acad. Sci. 975:202–216.

10. Heier, B. T., A. Lovland, K. B. Soleim, M. Kaldhusdal, and J. Jarp. 2001. Afield study of naturally occurring specific antibodies against Clostridium per-fringens alpha-toxin in Norwegian broiler flocks. Avian Dis. 45:724–732.

11. Ito, A. 1968. Alpha-toxoid of Clostridium perfringens. I. Purification andtoxoiding of alpha-toxin of C. perfringens. Jpn. J. Med. Sci. Biol. 21:379–391.

12. Kaldhusdal, M., and A. Lovland. 2000. Necrotic enteritis: the economicalimpact of Clostridium perfringens is greater than anticipated. World Poultry16:50–51.

13. Kawsar, H. I., K. Ohtani, K. Okumura, H. Hayashi, and T. Shimizu. 2004.Organization and transcriptional regulation of myo-inositol operon in Clos-tridium perfringens. FEMS Microbiol. Lett. 235:289–295.

14. Keyburn, A. L., S. A. Sheedy, M. E. Ford, M. M. Williamson, M. M. Awad,J. I. Rood, and R. J. Moore. 2006. The alpha-toxin of Clostridium perfringens

1076 KULKARNI ET AL. CLIN. VACCINE IMMUNOL.

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from

is not an essential virulence factor in necrotic enteritis in chickens. Infect.Immun. 74:6496–6500.

15. Kulkarni, R. R., V. R. Parreira, S. Sharif, and J. F. Prescott. 2006. Clostrid-ium perfringens antigens recognized by broiler chickens immune to necroticenteritis. Clin. Vaccine Immunol. 13:1358–1362.

16. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly ofthe head of bacteriophage T4. Nature 227:680–685.

17. Ling, E., G. Feldman, M. Portnoi, R. Dagan, K. Overweg, F. Mulholland, V.Chalifa-Caspi, J. Wells, and Y. Mizrachi-Nebenzahl. 2004. Glycolytic en-zymes associated with the cell surface of Streptococcus pneumoniae are an-tigenic in humans and elicit protective immune responses in the mouse. Clin.Exp. Immunol. 138:290–298.

18. Logan, A. J., E. D. Williamson, R. W. Titball, D. A. Percival, A. D. Shuttle-worth, J. W. Conlan, and D. C. Kelly. 1991. Epitope mapping of the alpha-toxin of Clostridium perfringens. Infect. Immun. 59:4338–4342.

19. Lovland, A., M. Kaldhusdal, K. Redhead, E. Skjerve, and A. Lillehaug. 2004.Maternal vaccination against subclinical necrotic enteritis in broilers. AvianPathol. 33:83–92.

20. Madureira, P., M. Baptista, M. Vieira, V. Magalhaes, A. Camelo, L. Oliveira,A. Ribeiro, D. Tavares, P. Trieu-Cuot, M. Vilanova, and P. Ferreira. 2007.Streptococcus agalactiae GAPDH is a virulence-associated immunomodula-tory protein. J. Immunol. 178:1379–1387.

21. McCarthy, J. S., M. Wieseman, J. Tropea, D. Kaslow, D. Abraham, S.Lustigman, R. Tuan, R. H. Guderian, and T. B. Nutman. 2002. Onchocercavolvulus glycolytic enzyme fructose-1,6-bisphosphate aldolase as a target fora protective immune response in humans. Infect. Immun. 70:851–858.

22. Modun, B., and P. Williams. 1999. The staphylococcal transferrin-bindingprotein is a cell wall glyceraldehyde-3-phosphate dehydrogenase. Infect.Immun. 67:1086–1092.

23. Pancholi, V., and G. S. Chhatwal. 2003. Housekeeping enzymes as virulencefactors for pathogens. Int. J. Med. Microbiol. 293:391–401.

24. Pancholi, V., and V. A. Fischetti. 1997. Regulation of the phosphorylation ofhuman pharyngeal cell proteins by group A streptococcal surface dehydro-genase: signal transduction between streptococci and pharyngeal cells. J.Exp. Med. 186:1633–1643.

25. Prescott, J. F. 1979. The prevention of experimentally induced necroticenteritis in chickens by avoparcin. Avian Dis. 23:1072–1074.

26. Shimizu, T., W. Ba-Thein, M. Tamaki, and H. Hayashi. 1994. The virR gene,a member of a class of two-component response regulators, regulates theproduction of perfringolysin O, collagenase, and hemagglutinin in Clostrid-ium perfringens. J. Bacteriol. 176:1616–1623.

27. Shimizu, T., K. Ohtani, H. Hirakawa, K. Ohshima, A. Yamashita, T. Shiba,N. Ogasawara, M. Hattori, S. Kuhara, and H. Hayashi. 2002. Completegenome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc.Natl. Acad. Sci. USA 99:996–1001.

28. Shimizu, T., K. Shima, K. Yoshino, K. Yonezawa, T. Shimizu, and H.Hayashi. 2002. Proteome and transcriptome analysis of the virulence genesregulated by the VirR/VirS system in Clostridium perfringens. J. Bacteriol.184:2587–2594.

29. Stevens, D. L., and A. E. Bryant. 1993. Role of theta toxin, a sulfhydryl-activated cytolysin, in the pathogenesis of clostridial gas gangrene. Clin.Infect. Dis. 16(Suppl. 4):S195–S199.

30. Stevens, D. L., R. W. Titball, M. Jepson, C. R. Bayer, S. M. Hayes-Schroer,and A. E. Bryant. 2004. Immunization with the C-domain of alpha-toxinprevents lethal infection, localizes tissue injury, and promotes host responseto challenge with Clostridium perfringens. J. Infect. Dis. 190:767–773.

31. Thammapalerd, N., D. Kotimanusvanij, M. Duchene, J. A. Upcroft, R.Mitchell, A. Healey, N. Samarawickrema, S. Tharavanij, G. Wiedermann,and P. Upcroft. 1996. Pyruvate: ferredoxin oxidoreductase from Entamoebahistolytica recognized by a monoclonal antibody. Southeast Asian J. Trop.Med. Public Health 27:63–70.

32. Thompson, D. R., V. R. Parreira, R. R. Kulkarni, and J. F. Prescott. 2006.Live attenuated vaccine-based control of necrotic enteritis of broiler chick-ens. Vet. Microbiol. 113:25–34.

33. Titball, R. W., A. M. Fearn, and E. D. Williamson. 1993. Biochemical andimmunological properties of the C-terminal domain of the alpha-toxin ofClostridium perfringens. FEMS Microbiol. Lett. 110:45–50.

34. Williamson, E. D., and R. W. Titball. 1993. A genetically engineered vaccineagainst the alpha-toxin of Clostridium perfringens protects mice against ex-perimental gas gangrene. Vaccine 11:1253–1258.

VOL. 14, 2007 IMMUNIZATION OF CHICKENS AGAINST C. PERFRINGENS 1077

on October 26, 2014 by H

SC

LIBR

AR

Yhttp://cvi.asm

.org/D

ownloaded from