Safety and efficacy of a maternal vaccine for the passive protection of broiler chicks against necrotic enteritis

Abstract

Necrotic enteritis is a potentially fatal multifactorial disease of chickens, which under commercial conditions is often associated with increased levels of mortality and reduced bird performance. The safety and efficacy of a Clostridium perfringens type A alpha-toxoid (Netvax™) formulated as an oil emulsion was investigated, following maternal immunization of broiler breeder hens, housed under commercial conditions, by the intramuscular route. A total of 11,234 hens were vaccinated across two integrated poultry sites. The vaccine was safe with no systemic reactions or adverse effects on bird performance detected. Vaccination resulted in a significant increase in anti-alpha toxin antibody in the hen that was maintained throughout the study, and subsequently transferred to their progeny throughout the laying period via egg yolk. Chicks hatched from eggs produced from vaccinated hens were shown to have reduced mortality specifically related to progeny flocks where gross gut lesions associated with necrotic enteritis were observed in control chicks. Further, whilst C. perfringens was isolated from control chicks with necrotic enteritis lesions, no such isolations were made at these time points from chicks from vaccinated hens. These results indicate that, under commercial conditions, maternal vaccination with Netvax™ can help to control losses related to necrotic enteritis.

Introduction

Necrotic enteritis (NE) is a complex enteric disease of poultry associated with Clostridium perfringens type A. Global costs to the industry have been estimated at up to $2 billion per annum (Kaldhusal & Lovland, 2000; McDevitt et al., 2006). Control of the disease has been achieved through the routine prophylactic administration of antimicrobial drugs either in-feed or in-water (Ficken & Wages, 1997); however, as a result of an increasing concern about the contribution of such practices to antimicrobial resistance problems, there is growing pressure to ban this practice. In consequence, interest in alternative approaches including vaccination has been increasing.

C. perfringens is a commensal in the gut of vertebrates including poultry (Songer, 1996; Porter, 1998); however, data suggest that only specific strains produce NE (reviewed by Cooper & Songer, 2009). Even when such strains are present, development of clinical NE appears to be at least partially dependent on the presence of pre-disposing factors such as prior damage to the mucosal surface of the intestine or a diet containing fishmeal or high levels of indigestible polysaccharides, as found in rye, wheat and barley (reviewed by Williams, 2005).

Intestinal necrosis has been induced through the inoculation of young birds with crude broth cultures of C. perfringens containing alpha-toxin (Al-Sheikhly & Truscott, 1977; Truscott & Al-Sheikhly, 1977), and in germ-free chickens the effect of a similar challenge was neutralized by prior treatment with anti-alpha-toxin antibodies (Fukata et al., 1988). In contrast, Keyburn et al. (2006) found that lesions typical of NE were still observed in conventional birds following challenge with mutants constructed with deletions in the alpha-toxin gene, raising questions around the role of alpha-toxin in the disease process. Nevertheless, progeny chicks from hens vaccinated with cell-free supernatant toxoids derived from C. perfringens type A or type C were shown to be protected against sub-clinical NE (Løvland et al., 2004); however, such preparations would contain antigens other than alpha-toxin. Partial protection of chicks against NE, however, has been demonstrated following vaccination with either recombinant alpha-toxin (Kulkarni et al., 2007; Cooper et al., 2009) or Salmonella vectored alpha-toxin (Kulkarni et al., 2008; Zekarias et al., 2008). On the contrary, Thompson et al. (2006) showed that oral immunization with live alpha-toxin-deficient mutants of C. perfringens can still protect against experimentally induced NE. Antigens other than alpha-toxin may therefore also be able to play a role as key immunogens in a vaccine (Kulkarni et al., 2008; Jiang et al., 2009).

The objective of the current study was to extend previous observations on the performance of cell-free supernatant toxoids under experimental conditions and to investigate the safety and efficacy of an experimental vaccine against NE for the passive protection of broiler chicks housed under commercial conditions.

Materials and Methods

The study comprised the immunization of broiler breeder hens at each of two commercial sites and the follow-up of progeny birds from three hatches at Site 1 and two hatches at Site 2. Both the hatcheries and the rearing farms were blinded to the immunization status of the flocks from which the chicks originated, as were the laboratories at which samples were analysed. Follow-up of progeny birds from additional hatches was restricted by the availability of suitable commercial broiler farms able to both meet the standards of good clinical practice and provide the support required within the study protocol. A summary of the study design is presented in Table 1. A total of 11,234 pullets received the test vaccine, with 9304 birds in the control group.

Table 1.  Summary of study design.

	Site 1	Site 2
Breed	ISA 957	Ross 508
Number	1644 (vaccinated)	9590 (vaccinated)
	1634 (control)	7670 (control)
Broiler breeder hens
Age at vaccination 1	11 weeks	11 weeks
Age at vaccination 2	19 weeks	18 weeks
Age at hatch 1	30 weeks	27 weeks
Age at hatch 2	35 weeks	32 weeks
Age at hatch 3	45 weeks	…
Broiler chickens
Slaughter	56 to 64 days	Female 37 to 38 days
		Male 57 days

Vaccine

The vaccine comprised an experimental batch of Netvax™ containing C. perfringens type A toxoid at a concentration of 3 total combining power units (with reference to the Fifth International Standard for Gas-Gangrene Antitoxin (C. perfringens alpha-antitoxin), Equine code PE) per dose emulsified in a light mineral oil. The vaccine (0.5 ml) was administered by intramuscular injection into the thickest part of the breast muscle, with the first dose administered into the left breast and the second into the right breast. Control birds did not receive any vaccine.

Husbandry

Two integrated poultry companies were used in the study, the first (Site 1) located in Germany and the second (Site 2) located in Italy.

At Site 1, 1-day-old ISA 957 broiler breeder chicks were supplied by the hatchery to the rearing farm. The entire flock was managed following the standard guidelines provided by the hatchery, including a routine vaccination programme. The pullets were initially housed at a density of 16 birds/m² in a building divided into two pens. Before vaccination at 11 weeks they were transferred to the laying house, and vaccinates and controls were kept in separate pens at a stocking density of 8 birds/m². Eggs from each treatment group were collected over 5-day to 7-day periods from hens at 30, 35 and 45 weeks of age and were transferred to the hatchery where they were set in separate trays and hatched under the same conditions. One-day-old broiler chickens from week 30, week 35 and week 45 hatches were placed at one of two different commercial farms. At each farm, treatment groups were located in the same house but were separated by nets. Birds were sent for slaughter at 56 to 64 days of age

At Site 2, 1-day-old Ross 508 broiler breeder chicks were supplied by the hatchery to the rearing farm. The entire flock was managed following the standard guidelines provided by the hatchery, including a routine vaccination programme. The pullets were initially housed at a density of 14 to 17 birds/m² in a three-floor building, with four pens per floor. At 21 weeks of age the birds were moved to the broiler breeder farm, where they were placed in two separated one-floor houses on the same site at a stocking density of 9 to 10 birds/m². Eggs from each treatment group were collected over a 7-day period from hens at 27 and 32 weeks of age and were transferred to the hatchery where they were set in separate trays and hatched under the same conditions. One-day-old broiler chickens from week 27 and 32 hatches were placed at one of two different commercial farms. At each farm, the two treatment groups were housed in separated houses in a multi-floor building. Females were sent for slaughter at 37 to 38 days of age, and males were sent for slaughter at 57 days of age.

At each site, standard coccidiostat-free diets (prestarter, Site 1 only; starter, grower and finisher, both sites) were fed to the progeny chicks from the two treatment groups.

Pre-vaccination examinations

Broiler breeder pullets were observed for general physical condition the day preceding each vaccination. Blood samples were collected from 25 pullets from each group before each vaccination.

Clinical observations

Clinical observations were conducted just before each vaccination, twice during the first 24 h following each vaccination and on a daily basis for 14 days after each vaccination. Following this initial 14-day daily observation period, the health status of the birds was recorded on a weekly basis until week 45 at Site 1 and week 32 at Site 2. Mortality was recorded on a daily basis during the whole study period starting from the day of first vaccination.

Injection site reaction

The presence of injection site reactions was monitored individually in 50 randomly selected pullets from each treatment group, twice weekly for 3 weeks after each vaccination, by inspection of the body surface around the injection site. These observations were extended until the injection site reaction disappeared if the local signs persisted beyond the initial 3-week observation period. At the conclusion of the injection site observation period (week 30), the injection sites were examined in 50 randomly selected birds from each treatment group for signs of tissue fibrosis by incision of the breast musculature, following euthanasia. In addition, parent hens that died during the production period were examined for visible macroscopic reactions and tissue fibrosis at the site of injection.

The occurrence of systemic reactions or other adverse events were monitored during the entire study period.

Egg production and viability

Data were collected throughout the study period by the hatchery as per usual management practices.

Samples for evaluation of specific antibody titres

On weeks 30, 35 and 42 at Site 1 and on weeks 27 and 32 at Site 2, the following samples were collected: blood samples were collected from 25 randomly selected pullets/hens per treatment group; and 60 randomly selected eggs laid on a single day in that week were collected for the determination of IgY titres.

Broiler chickens

The general health of the broiler chickens was monitored from placement until slaughter. Weight was measured on a weekly basis, by weighing 100 chicks in groups of 10, from each treatment group using calibrated scales. Mortality was recorded on a daily basis except for chicken broilers from the week 30 hatch at Site 1, where dead birds were kept by the farmer and counted by the investigator during routine visits. Feed consumption was determined based on each site's routine procedures. Concomitant veterinary treatments administered to the chickens outside of routine farming practices for each site were recorded.

Clinical observations for signs of NE (depression, ruffled feathers, inappetance, closed eyes, immobility and dark-coloured diarrhoea) were performed on a daily basis for the first 4 weeks after hatch and thereafter on a weekly basis. In the event of any clinical signs being observed, five sick birds were necropsied and any lesions were examined and scored. A number of birds from the alternate flock (vaccinate or control) were also examined at this time. In addition, all birds that died during the study (either natural or euthanized) were necropsied and lesions scored, except where the cause of death was well established and not related to NE. Gut lesions were scored according to the following: 0 = no C. perfringens-associated gross gut lesions; 1 = less than 10 C. perfringens-associated gross gut lesions; 2 = at least 10 C. perfringens-associated gross gut lesions; 3 = presence of at least one C. perfringens-associated gut lesion with a maximum extension larger than the circumference of the gut mucosa.

On days where lesions associated with NE were observed in any bird, a bacteriological analysis of the gut contents was conducted in all birds examined on that day as follows. Two grams of caecal contents were weighed and diluted to 10 − 2, 10 − 4 and 10 − 6. Then 100 µl aliquots of these dilutions were plated on blood agar plates. The plates were incubated anaerobically at 37°C for 18 to 24 h, and then colonies typical of C. perfringens were counted.

Blood samples were collected from 25 randomly selected chicks from each treatment group at 7 days of age.

Blood samples for serology

A minimum of 2 ml blood was collected in 4 ml serum separator tube vacutainers from the brachial vein. Blood samples were allowed to clot at ambient temperature for a minimum of 2 h prior to separation of the serum following centrifugation. Serum was stored frozen until testing for the presence of specific antibody by enzyme-linked immunosorbent assay (ELISA).

Egg collection

Eggs were collected and stored at 2 to 8°C before processing. For each treatment, eggs were pooled into groups of 10 and the IgY extracted from the egg yolk using a chicken IgY purification kit (Pierce Eggcellent Chicken IgY purification Kit). Purified IgY was stored below –15°C until testing for the presence of specific antibody by ELISA and haemolysis inhibition assay (HIA).

Enzyme-linked immunosorbent assay

A modification of the procedure described by Heier et al. (2001) was followed. Briefly, ELISA plates were coated overnight at 2 to 8°C with 1 µg/ml phospholipase C type XIV (Sigma). Plates were blocked by incubation with phosphate-buffered saline containing 3% bovine serum albumin and 1% Tween 20 for 30 min at 37°C. Serial two-fold dilutions of test samples were added to the plate from an initial 1-in-10 dilution and incubated at 37°C for 1 h. At the end of the incubation, plates were washed and an optimal dilution of an anti-chicken IgG peroxidase conjugate (Sigma) was added to each well. Plates were incubated at 37°C for 30 min and bound conjugate was detected using TMB (SureBlue Reserve^®TMB); optical densities were determined at 450 nm (OD_{450 nm}). The cut-off value for the assay was determined as the mean OD_{450 nm}+12 standard deviations for a panel of 50 sera from specific pathogen free birds reared under barrier containment, and a cut-off reference control developed to give an equivalent OD_{450 nm} value. Appropriate positive, negative and cut-off reference controls were included on each plate. End points for control and test samples were calculated relative to the mean OD_{450 nm} for the cut-off reference control.

Haemolysis inhibition assay

Test samples were heat treated at 56°C for 30 min. Serial two-fold dilutions of test samples in phosphate-buffered saline (pH 7.3) containing 1% bovine serum albumin were performed in V-bottomed 96-well plates, to give a final volume of 50 µl/well. An equal volume of C. perfringens alpha-toxin (selected as the highest dilution of toxin giving complete haemolysis of sheep red blood cells) was then added to each well and the plates incubated with shaking for 1 h at 37°C. Then 25 µl of 0.5% suspension of washed sheep red blood cells were added to each well and, following incubation at 37°C for 3 h (no shaking), the end point for each sample was calculated as the reciprocal of the highest dilution scored positive for the inhibition of haemolysis.

Data analysis

ELISA data were transformed by log₂ (titre/5) prior to analysis. The value 5 was used for titres reported as <10. Separate analyses were performed for each time point. For HIA results, data were transformed by log₂.

Statistical analysis was performed using SAS version 8.2 (SAS Institute, Cary, North Carolina, USA). Wilcoxon exact rank-sum tests were performed to compare differences in titres for chicks in control and vaccinated groups. Differences in mortality percentages of chicks in vaccinated and control groups were analysed by site and hatch using StatXact 6. P values were determined using the GLIMMIX SAS macro with binomial error and the logit link using SAS version 8.2. Weights for chicks were analysed using SAS PROC MIXED. Separate analyses were performed for site, hatch, and day. Statistical significance was declared for P<0.05.

Results

Safety

Local injection site reactions comprising generalized swelling, induration and or discolouration of the muscle tissue were observed in up to 12% of the birds at Site 1 (maximum size 60x20 mm²) and in up to 2% of the birds at Site 2 (maximum size 3x12 mm²) examined at any time point in the first 3 weeks following the initial vaccination. Following the second vaccination, up to 4% of the birds examined at Site 1 at any time point in the first 3 weeks also showed similar reactions (maximum size 25x20 mm²), but no reactions were observed over this time period at Site 2. At both sites tissue fibrosis was observed in a small number of birds 32 to 35 weeks after initial vaccination, but at Site 1 where the hens were kept to 42 weeks after initial vaccination no reactions were observed in the 50 birds examined at this time point. Further, no macroscopic lesions were observed following incision of the injection site in the breast musculature in birds that died or were euthanized during the laying period. No systemic reactions were observed in any of the hens at either site.

Egg production and hatchability throughout the laying cycle were comparable between vaccinated and control hens on each site, and in general were within the norms expected for each hatchery (Table 2).

Table 2.  Egg production and hatchability.

		27 weeks		30^a/32 weeks		35^a/37 weeks		40^a/42 weeks		45^a/47 weeks		50^a/52 weeks		60 weeks
		V	C	V	C	V	C	V	C	V	C	V	C	V	C
Egg production/100 hens	Site 1	642	665	635	639	451	468	552	566	578	569	537	530	494	493
	Site 2	573	527	582	584	561	550	527	508	451	465	390	388	–	–
Hatchability (%)	Site 1	25.1^b	18.8^b	69.1	62.0	85.2	80.6	85.0	84.5	85.0	83.9	78.4	84.3	76.6	80.6
	Site 2	82.5	86.0	87.6	87.3	87.4	88.6	86.2	87.2	85.4	85.4	–	–	–	–
V, vaccinated; C, control; –, not recorded. ^aHen age for egg production at Site 1 only. ^bHatchability was below the norm up to week 32 due to poor acclimatization by the cockerels when placed in the pens.

Antibody responses

In the broiler breeder hens at Site 1, at the time of initial vaccination (11 weeks of age) mean serum antibody titres (ELISA) against C. perfringens alpha-toxin were at equivalent low levels in both vaccinated and control birds (Figure 1). Following vaccination, mean specific antibody titres increased, reaching a peak (6.3 log₂ (titre/5) 6 weeks after second vaccination (25 weeks of age)), after which the titres slowly fell throughout the rest of the laying cycle to approximately 4.0 log₂ (titre/5) at 45 weeks of age. Mean specific antibody titres in control birds remained low (<2.0 log₂ (titre/5)) throughout the study, with the highest levels (approximately 1.0 log₂ (titre/5)) being recorded in the late phase of the study at 35 and 45 weeks of age.

Figure 1.  Serum antibody titres against C. perfringens type A alpha-toxin determined by ELISA following vaccination of broiler breeder hens at Site 1.

Mean specific antibody titres (ELISA) in pools of IgY from eggs collected from hens at 25, 30, 35 or 45 weeks of age at Site 1 showed an eight-fold decrease in antibody levels over time (Figure 2a). Measure of antibody functionality by HIA, however, showed an approximate two-fold reduction in specific titre between eggs collected from hens at 25 and 30 weeks of age, titres remaining stationary at approximately 8.0 log₂ in eggs collected at the remaining time points (Figure 2b). As indicated by non-overlapping 95% confidence intervals, specific antibody levels in eggs from vaccinated birds were consistently significantly higher than those from control birds.

Figure 2.  IgY titres against C. perfringens type A alpha-toxin determined by ELISA and haemolysis inhibition from eggs collected from vaccinated or control hens at 25, 30, 35 and 45 weeks of age at Site 1.

Mean antibody levels in chicks, 7 days after hatching from eggs laid by vaccinated hens, showed a two-fold reduction in titre between eggs laid by 30-week-old hens and those laid by 35-week-old hens, although no further decline was seen up to 45 weeks of age (Figure 3). Mean antibody levels remained significantly higher than that detected in chicks hatched from eggs laid by control hens at each of the time points investigated. Similar results for hen serum, egg IgY and chick serum were also seen on Site 2 (data not shown).

Figure 3.  Serum antibody titres against C. perfringens type A alpha-toxin in 7-day-old progeny chicks from vaccinated or control hens at Site 1.

Concomitant treatments

At Site 1, Baytril (10 mg/kg) was orally administered in drinking water to the chicks for the week 35 hatch from vaccinated hens at 6, 7 and 8 days old only, in order to treat a yolk sac infection. At Site 2, amoxicillin (30 g/100 l) was orally administered in drinking water to the chicks from control hens as therapy for the onset of NE, for the week 27 hatch at 24, 25 and 26 days of age and for the week 32 hatch at 25, 26, 27 and 28 days of age

Clinical efficacy

With the exception of the 35-week hatch at Site 1, where an increase in mortality associated with a yolk sac infection was reported, chick mortality by week was observed to decrease over time to a low plateau (Figure 4). Progeny from the control broiler breeder flocks at the week 45 hatch at Site 1 and the week 27 hatch at Site 2, however, showed a small increase in this trend during week 4 (Figure 4c, d). The pattern of accumulated mortality remained similar between progeny from vaccinated and control broiler breeder flocks, although, apart from the 35-week hatch at Site 1, overall mortality was consistently lower in the broilers from the vaccinated hens than in the control hens for all hatches.

Figure 4.  Percentage mortality by week (bars) or accumulated throughout the study (lines) in progeny from vaccinated and control broiler breeder flocks at Site 1 (4a: 30-week hatch, 4b: 35-week hatch, 4c: 45-week hatch) and at Site 2 (4d: 27-week hatch, 4e: 32-week hatch).

Differences in percentage chick mortality from day of birth until sent for slaughter (Table 3) were statistically significant at Site 1 for the 45-week hatch (P = 0.0037) and at Site 2 for the 27-week hatch (P = 0.0082). The percentage chick mortality from day of birth until day 21 (Table 3) was also statistically significant at Site 1 for the 45-week hatch (P = 0.0456). This significant reduction in mortalities at Site 1 was also accompanied by a significantly higher mean weight at the end of the study in chicks from vaccinated hens compared with those from control hens (P = 0.0006).

Table 3.  Comparison of mortality, gut lesions and bird performance in progeny from vaccinated and non-vaccinated broiler breeders.

	Site 1						Site 2
	30		35		45		27		32
Hatch (week)	Vaccinated	Control	Vaccinated	Control	Vaccinated	Control	Vaccinated	Control	Vaccinated	Control
Mortality (%) Day 0 to day 21	0.97	1.10	2. 09^a	1.60	0.40	0.72	1.97	2.23	1.94	2.04
Mortality (%) Day 0 to end	1.16	1.39	2.38^a	1.93	0.77	1.41	3.31	3.91	2.76	2.76
% Incidence gut lesions (NE	0	0	0	0	0	66.6 (day 26)	0	92 (days 25/56)	0	81.8 (days 25/26)
Isolation of Clostridium perfringens from possible NE lesions	…	…	0/4	…	0/1	1/4day 26 (1.2×10^7 cfu/g)	0/10	25/25days 25/26 (median count 2.2×10^3 cfu/g)	0/5	10/10days 25/26 (median count 2.6×10^3 cfu/g)
Final Weight (Kg)	1.819	1.824	1.896	1.889	2.053	1.889	3.424^b (1.430)	3.390^b (1.475)	3.560^b (1.610)	3.525^b (1.588)
FCR	–	–	2.402	2.336	2.449	2.377	1.891	1.899	1.880	1.885
^aHigher mortality associated with an egg yolk infection
^bMales slaughtered at 57 days of age, figures in brackets females slaughtered at 37 to 38 days of age

The numbers of birds examined for the presence of gut lesions associated with NE, for all sites throughout the study, are presented in Table 4. No gross gut lesions were observed in any of the birds from vaccinated hens at either Site 1 or Site 2 (Table 3 and 4). Similarly, no C. perfringens-associated gross gut lesions were observed in birds hatched to control hens in weeks 30 or 35 at Site 1. In birds hatched to control hens at week 45, however, clinical signs of disease (ruffled feathers) were observed in a proportion of birds at 25 to 26 days of age and 66.6% (4/6) of birds examined post-mortem on day 26 had positive gut scores of either 2 (three birds) or 3 (one bird), giving a mean lesion score of 1.5 for all six birds. At Site 2, clinical signs of disease (depression, ruffled feathers and inappetance) were again observed in birds hatched to control hens, at around 24 to 26 days of age in flocks from both hatches. For the 27-week hatch, 92.3% (12/13) of birds examined on day 25 were found to have C. perfringens-associated gross gut lesions (mean lesion score 1.15) and 91.7% (11/12) on day 26 (mean lesion score 0.92), giving an overall mean lesion score of 1.1 (20 x score of 1, three x score of 2). For the 32-week hatch, 100% (12/12) of birds examined on day 25 were found to have C. perfringens-associated gross gut lesions (mean lesion score 1.58) and 60.0% (6/10) on day 26 (mean lesion score 0.7), giving an overall mean lesion score of 1.1 (11 x score of 1, seven x score of 2).

Table 4.  Numbers of birds examined for the presence of NE gut lesions per week.

	Site 1						Site 2
	Week 30 hatch		Week 35 hatch		Week 45 hatch		Week 27 hatch		Week 32 hatch
Week	C	V	C	V	C	V	C	V	C	V
1	31	34	25	36	18	11	22	20	25	15
2	10	6	28	29	14	4	13	16	21	16
3	7	4	5	5	1	4	11	14	22	18
4	1	1	1	1	18 ^a	3	34 ^b	20	33 ^c	19
5	1	0	6	1	8	3	12	11	10	11
6	1	0	0	1	6	1	13	10	16	17
7	8	6	1	2	4	5	12	11	15	20
8	0	0	5	5	2	1	11	9	11	12
V, vaccinated; C, control. Bold text denotes when specific lesions were observed. ^a4/6 birds examined on day 26 were diagnosed as having NE (mean lesion score = 1.5). ^b12/13 birds examined on day 25 (mean lesion score = 1.15) and 11/12 birds examined on day 26 (mean lesion score = 0.92) were diagnosed as having NE. ^c12/12 birds examined on day 25 (mean lesion score = 1.58) and 6/10 birds examined on day 26 (mean lesion score = 0.7) were diagnosed as having NE.

Confirmation of the diagnosis of NE was obtained by culturing gut or hepatic tissues from the birds in both vaccinate and control groups necropsied on the days when specific gut lesions where observed (i.e. Site 1, week 45 hatch, day 26; Site 2, week 27 and week 32 hatches, days 25 and 26) in order to identify the presence of C. perfringens or other pathogens. At both Sites 1 and 2 no C. perfringens was isolated from any of the birds hatched from vaccinated hens. In contrast, C. perfringens was isolated from gut samples of 1/4 birds from the week 45 hatch at Site 1, hatched to control hens at 26 days of age when lesions of NE were observed, and in 100% (25/25 Site 2, week 27 hatch and 10/10 Site 2, week 32 hatch) hatched to control hens at 26 days of age when lesions of NE were observed (Table 3).

Bird performance

Feed conversion ratios (Table 3) were comparable between progeny from either vaccinated or control hens, and for each site were within the norms accepted by that producer.

For Site 1 the birds were slaughtered between 56 and 64 days of age. There were no significant differences in terms of rejects for broilers hatched from either vaccinated or control hens. Rejects before slaughter due to either low bodyweight (<800 g) or death during transport were as follows: week 30 hatch, controls 0.16%, vaccinates 0.15%; week 35 hatch, controls 0.16%, vaccinates 0.07%; and week 45 hatch, controls 0.29%, vaccinates 0.09%. Rejects after slaughter because of a poor-quality carcass were as follows: week 30 hatch, controls 0.81%, vaccinates 1.14%; week 35 hatch, controls 0.39%, vaccinates 0.34%; and week 45 hatch, controls 0.38%, vaccinates 0.41%. These figures were comparable with the values previously established for birds from this producer. At Site 2, males and females were slaughtered separately (females at 37 to 38 days of age, and males at 57 days of age). Yields (as defined by carcass weight divided by live body weight) were as follows: week 27 hatch, male controls 63.1%, male vaccinates 63.2%; female controls 64.0%, female vaccinates 64.0%; and week 32 hatch, male controls 63.3%, male vaccinates 63.5%; female controls 64.2%, female vaccinates 64.3%—these were comparable with the established norms for this producer.

Discussion

The present study investigated the safety and efficacy of a novel vaccine based on C. perfringens type A alpha-toxoid for the passive protection of broiler chickens against NE. The study was conducted at two commercial sites and involved a total of 11,234 vaccinated and 9304 control broiler breeders and the clinical observation of five batches of the broiler progeny.

Overall the vaccine appeared to be safe, with no systemic reactions or adverse effects on bird performance or reproduction observed. Differences in the incidence and size of local injection site reactions were noted between the two sites, and this may reflect differences in the breed of the birds. In general, the reactions were regarded as mild and appeared to resolve over time. Tissue fibrosis was observed in a small number of birds 30 to 35 weeks after initial vaccination, but it is impossible to correlate this with the use of Netvax™, since the birds received other injectable vaccines as part of the standard procedures at both locations.

As reported by other workers in laboratory studies (Løvland et al., 2004), we have confirmed that, under field conditions, vaccination of broiler breeder hens with a conventional C. perfringens alpha-toxoid vaccine induces the production of specific anti-alpha-toxin antibodies in the circulation of the hen, which remain at significant levels throughout the laying cycle. We have further demonstrated that specific antibodies are transferred from the circulation of the hen to egg yolk, resulting in elevated levels of such antibodies in the circulation of 7-day-old chicks hatched from eggs laid by broiler hens up to at least 45 weeks of age.

Passively acquired antibody is critical for the protection of newly hatched chicks, since during the first few weeks of life their immune system is not fully developed. Thus, although the first surface IgY-positive lymphocytes can be detected around the time of hatch (Kincade & Cooper, 1971; Lawton et al., 1975), Lawrence et al. (1981) did not detect IgY secreting B cells in chick plasma until 6 days after hatching, and maturation of T and B lymphocytes in the gut-associated lymphoid tissue is not complete until about 2 weeks of age (Bar-Shira et al., 2003). It has been recognized that the principal mechanism for the transfer of IgY (IgG) from the hen to the circulation of the chick is via egg yolk and that the total levels observed in egg yolk are directly proportional to those seen in the hens’ serum (Loeken & Roth, 1983; Al-Natour et al., 2004). Alpha-toxin-specific IgY in egg yolk as measured by ELISA was found to demonstrate a declining trend over time, which was not observed when measured by HIA. This may reflect a difference in the relative sensitivities of the two test procedures. Alternatively, since the relationship between total IgY in hen serum and egg IgY appears to extend to the relative levels of antibodies against specific antigens (Hamal et al., 2006), this could reflect a maturation in the hens’ antibody response favouring functional (HIA) antibody over total anti-alpha-toxin antibody.

In poultry, C. perfringens infections may cause either a clinical or a sub-clinical form of disease. The clinical form of the disease results in increased mortality associated with gross lesions occurring predominantly in the jejunum (Wages & Opengart, 2003). It has been proposed that clinical disease results from the adhesion of C. perfringens to damaged mucosa facilitating bacterial proliferation and toxin production (Kageyama et al., 1987; Baba et al., 1992), and Si et al. (2007) demonstrated a positive correlation between the incidence of NE and both the level of alpha-toxin expression and increased C. perfringens proliferation. An average C. perfringens count of 5 log₁₀ colony-forming units (CFU)/g appeared to be associated with a threshold for the development of NE with a lesion score of 2. In this study, in the control birds for both hatches at Site 2, clinical signs of NE were consistently observed during week 4 with median counts of approximately 3 log₁₀ CFU/g associated with mean NE lesion scores of approximately 1.5. In both instances, however, treatment with amoxicillin was initiated immediately on the appearance of clinical signs and this would have controlled the proliferation of the organism and consequently minimized the development of intestinal lesions. No gross lesions associated with NE or clinical signs, however, were observed in the progeny of vaccinated hens. In addition, these chicks were also found to have an overall lower rate of mortality, with a statistically significant reduction associated with those time points at which gross gut lesions were also observed, thus providing strong evidence of efficacy against clinical disease under commercial conditions. Further evidence of clinical efficacy is provided by the failure to isolate detectable numbers of C. perfringens at any time from chicks from vaccinated hens, although this is in contrast to the findings of Løvland et al. (2004), where C. perfringens was isolated from the progeny of alpha-toxoid vaccinated hens with a median count of at least 4 log₁₀ CFU/g from 21 days of age. The antigen content of this vaccine was, however, only 0.25 total combining power units/dose and the adjuvant was aluminium hydroxide gel rather than an oil emulsion, possibly resulting in the lower level of efficacy observed against clinical disease. In contrast, the sub-clinical form of the disease results in a reduced ability to digest and absorb food leading to lower body weight gain and increased feed conversion ratio (Elwinger et al., 1992; Løvland & Kaldhusdal, 2001). Whilst overall, final body weights were favourable for those birds hatched from eggs laid by vaccinated hens, there was no trend to improved feed conversion ratio (FCR). It is therefore likely that larger-scale trials may be required to confirm any effect on subclinical disease.

The results of this study support other evidence suggesting that alpha-toxin can act as a key protective immunogen for the prevention of NE in chickens (Kulkarni et al., 2007; Zekerias et al., 2008; Cooper et al., 2009), although it should be recognized that alternative immunogens have also been demonstrated to have a potential role in protection (Kulkarni et al., 2008; Jiang et al., 2009; Kulkarni et al., 2010) and that these may also be present in a cell-free supernatant toxoid.

Vaccination of broiler breeder hens with Netvax™ and the subsequent transfer of maternal immunity has thus been shown, under commercial conditions, to provide a safe and efficacious mechanism to reduce both the mortality and the severity of gut lesions associated with NE in their progeny.

Acknowledgements

The authors thank Beth Cook and David Lane for their technical assistance in the preparation and testing of serum and egg IgY samples, and Dr Dianne Sweeney for performing the statistical analysis.