Vaccination of chickens with the 34 kDa carboxy-terminus of Bpmp72 reduces colonization with Brachyspira pilosicoli following experimental infection

ABSTRACT

The anaerobic intestinal spirochaete Brachyspira pilosicoli colonizes the large intestine of a variety of species of mammals and birds, and may result in colitis, diarrhoea and reductions in growth rate. Naturally occurring infections in chickens are largely confined to adult laying and breeding birds. In this study, the 34 kD carboxy-terminus of the prominent outer membrane protein Bmp72 of B. pilosicoli was expressed as a histidine-tagged recombinant protein and used to immunize two groups (B and C) of 15 individually housed layer chickens. Vaccination was with either 100 μg (B) or 1 mg (C) protein emulsified with Freund’s incomplete adjuvant delivered into the pectoral muscles, followed three weeks later by 1 mg of protein in phosphate buffered saline delivered via crop tube. Two weeks later these and 15 non-vaccinated positive control birds (group A) housed in the same room were challenged via crop tube with B. pilosicoli avian strain CPS1. B. pilosicoli was detected in the faeces of all control birds and in 14 of the vaccinated birds in each vaccinated group at some point over the 30-day period following challenge. Colonization was delayed and the duration of excretion was significantly reduced (P = 0.0001) in both groups of vaccinated birds compared to the non-vaccinated control birds. Fewer immunized birds had abnormal caecal contents at post mortem examination compared to non-vaccinated birds, but the difference was not statistically significant. This study indicates that recombinant Bmp72 C-terminus has potential to be developed for use as a vaccine component to provide protection against B. pilosicoli infections.

RESEARCH HIGHLIGHTS

• Laying chickens were immunized with recombinant Brachyspira pilosicoli membrane protein Bpmp72.

• Immunized birds had a highly significant reduction in the duration of colonization.

• Fewer immunized than control birds had abnormal caecal contents after infection.

• Bpmp72 showed potential for use as a novel vaccine component for B. pilosicoli.

KEYWORDS:

• Brachyspira pilosicoli
• spirochaete
• recombinant vaccine
• chickens
• intestinal spirochaetosis
• protection

Introduction

The anaerobic intestinal spirochaete Brachyspira pilosicoli colonizes the large intestine of many species of birds and mammals, including humans. Similarities between strains recovered from different host species suggest that B. pilosicoli has zoonotic potential (Hampson et al., 2006a; Neo et al., 2013). The condition in individuals colonized with B. pilosicoli has been called “intestinal spirochaetosis” (Taylor et al., 1980; Esteve et al., 2006). Colonization may be associated with the occurrence of diarrhoea and reduced growth rates, most commonly and notably in young pigs and in adult poultry (Hampson, in press), but also in companion animals such as horses and dogs (Hampson et al., 2006b; Prapasarakul et al., 2011). In adult laying and breeding chickens B. pilosicoli is one of three species associated with a condition called “avian intestinal spirochaetosis” (Le Roy et al., 2015; Hampson, in press). Infections with B. pilosicoli may result in delays in the onset of egg production, a decrease in the number and quality of eggs produced, a reduced growth rate, and/or an increase in faecal water content causing wet litter (Trampel et al., 1994; Stephens & Hampson, 2002; Mappley et al., 2014). Viable B. pilosicoli cells have been isolated in substantial numbers from the carcasses of spent hens in supermarkets in Belgium (Verlinden et al., 2012), emphasizing the possibility of foodborne transmission of this spirochaete to humans, or to other species such as dogs that may be fed with raw chicken carcasses.

Despite the spirochaete’s wide distribution and association with disease in many species, control of B. pilosicoli infections is still difficult and largely relies on the use of antimicrobial treatments. Unfortunately, strains of the species, including those recovered from chickens, increasingly are being found to be resistant to commonly available antimicrobials (Herbst et al., 2018), so other methods for control are required. There is evidence to show that causing reductions in the viscosity of the colonic contents through dietary manipulation can reduce colonization with B. pilosicoli in experimentally infected pigs (Hopwood et al., 2002; Lindecrona et al., 2004), whilst treatments with probiotic bacteria have provided some protection in experimentally infected laying chickens (Mappley et al., 2013). To our knowledge, there are no commercially available vaccines for use in controlling B. pilosicoli infections. In an early study, intramuscular immunization of weaner pigs with an autogenous B. pilosicoli bacterin did not prevent colonization following experimental infection (Hampson et al., 2000). On the other hand, experimental immunization of mice with either of two histidine-tagged recombinant oligopeptide-binding proteins of B. pilosicoli significantly reduced cumulative days of colonization following challenge with B. pilosicoli and, for one protein, significantly fewer mice were colonized compared with the non-vaccinated control group (Movahedi & Hampson, 2010). These results suggest that vaccination with recombinant proteins may have potential to help control infections with B. pilosicoli, although they have not been tested for efficacy in other relevant animal species where natural infections occur that impact on health.

A prominent 72 kDa outer membrane protein that is recognized by immune sera has been demonstrated on the surface of the outer envelope of B. pilosicoli and has been confirmed to be widely distributed amongst strains of the species (Tenaya et al., 1998). This molecule has been named Bpmb72 in accordance with the recommended nomenclature for Brachyspira species membrane proteins (Hampson et al., 2006c). The purpose of the described study was to investigate whether the carboxy-terminus (C-terminus) of recombinant Bpmp72 could be used as a subunit vaccine component to help control infections with B. pilosicoli. The findings have previously been disclosed as part of a patent relating to Bpmp72 (Hampson & La, 2004).

Materials and methods

Permissions

The experiments were approved by the Murdoch University Animal Ethics Committee (approval number 887R/01).

Identification of the Bpmp72 gene sequence

The gene encoding a 72 kDa outer membrane protein of B. pilosicoli, designated Bpmp72, was identified by immuno-screening a lambda bacteriophage library constructed using genome fragments of B. pilosicoli strain P43/6/78^T, as previously described (Lee et al., 2000). Absorbed hyperimmunized pig serum raised against B. pilosicoli strain 1648 (Tenaya et al., 1998) was used for the immune-screening.

PCR amplification and sequence comparison of Bpmp72 in Brachyspira strains

Two pairs of primers that annealed to regions internal and external to the bpmp72 coding sequence were designed and optimized for PCR detection of bpmp72 using chromosomal DNA from the 82 strains of Brachyspira species obtained from the reference collection at Murdoch University (48 strains of B. hyodysenteriae, 18 strains of B. pilosicoli, 12 strains of B. intermedia, eight strains of B. murdochii, four strains of B. innocens, two strains of “Brachyspira canis”, and one strain each of B. alvinipulli and B. aalborgi). The external primer set consisted of Bpmp72-US (5′-CGTTTAGCTGAACTTGAAGCTATG-3′) and Bpmp72-DS (5′-GTAATGCTCTGTCTTAATCAT-3′) that anneal to complementary sequences 78 base pairs (bp) upstream and 178 bp downstream to the coding region of bpmp72, respectively. The internal primer set consisted of Bpmp72-F913 (5′-CAAGTAATAGCTAAAGGTGATG-3′) and Bpmp72-R1692 (5′-TTACTGTTGTGCTTGAGTAGTG-3′) which amplified a 780 bp region within the coding region of the Bpmp72 gene. PCRs were performed using the HotStarTaq Master Mix Kit (Qiagen, Germantown, MD) according to manufacturer’s instructions. An annealing temperature of 50°C was used for both PCRs.

PCR products from six B. pilosicoli strains 1404/6A, 3295/60B, Will7, 9J-0438, Q98.0078.38 and HB60-1 were purified using the UltraClean PCR Clean-up Kit (Qiagen, Germantown, MD) according to the manufacturer’s instructions. Sequencing of the PCR product was performed using the ABI PRISM™ Dye Terminator Cycle Sequencing Ready Reaction Mix (PE Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Sequencing products were analysed using the ABI 373A DNA Sequencer. Sequence results were edited, compiled and compared using SeqEd v1.0.3. The genomes of the four B. pilosicoli strains P43/6/78^T, 95/1000, B8044 and WesB that are held in NCBI were searched for the presence of bpmp72. The sequences from all 10 strains were compared using ClustalX.

Preparation of recombinant protein

The bpmp72 gene was cloned into the Xpress Protein Expression System (Invitrogen, Carlsbad, CA), expressed in E. coli BL21 and purified by affinity chromatography as described previously (La et al., 2004). Due to the size of the gene product, the gene was cloned as a C-terminus portion (polypeptide residues 304–563).

Experimental chickens

Forty-five ISA Brown layer pullets of approximately 18 weeks of age were purchased from a local hatchery. On arrival, faecal swabs were taken to check for the absence of Brachyspira species. The birds were weighed and assigned to three groups of 15 that were randomly dispersed in a bank of cages in one room of an isolation animal house. Each bird was housed in an individual cage that was located adjacent to and separated from neighbouring birds by open wire-mesh partitions. All the birds were fed ad libidum on a custom-made layer diet that contained 50 ppm zinc bacitracin, which has been shown to enhance colonization with B. pilosicoli in chickens (Jamshidi & Hampson, 2002). The birds in groups B and C were vaccinated, and subsequently all groups were challenged orally with a strain of B. pilosicoli and monitored for 30 days.

Immunization protocol

On the day after arrival, the birds in groups B and C received their first vaccination. This consisted of an intramuscular injection into the pectoral muscle with either 100 μg (group B) or 1 mg (group C) recombinant protein emulsified in Freund’s incomplete adjuvant to a 1 ml volume. Three weeks later each was given an oral boost with 1 mg recombinant protein in 2 ml phosphate buffered saline administered by crop tube.

Experimental infection

Two weeks after the oral vaccination all the birds were challenged via a crop tube with 4 ml of exponential log-phase (∼10⁹/ml) Australian avian B. pilosicoli strain CPS1 (Stephens & Hampson, 2002), grown in Kunkle’s pre-reduced anaerobic broth (Kunkle et al., 1986). Challenge was repeated over three consecutive days.

Monitoring faecal excretion of B. pilosicoli

Faecal swabs were collected from the cloaca of each bird on arrival at the animal house, and then three times per week following experimental challenge. The swabs were cultured anaerobically on selective agar as described below.

Post mortem examination

All birds were removed for post mortem examination 30 days after the last day of the experimental inoculation and were killed by cervical dislocation. The caecal contents were examined from all birds, and they were swabbed and cultured for B. pilosicoli in the same manner as for faeces. Swabs were streaked onto Trypticase Soy agar (BBL) plates containing 5% (v/v) defibrinated sheep blood, 400 μg/ml spectinomycin and 25 μg/ml each of colistin and vancomycin (Jenkinson & Wingar, 1981). The plates were incubated for seven days at 37°C in anaerobic jars with an anaerobic environment of 94% H₂ and 6% CO₂ generated with anaerobic Gaspak plus sachets (BBL). The spirochaetes formed a characteristic low flat spreading growth surrounded by weak beta-haemolysis, and they had typical spirochaete morphology as observed under a phase contrast microscope. They were confirmed as B. pilosicoli using a species-specific PCR (La et al., 2003).

Comparisons between groups

The number of sampling days that faeces were positive for B. pilosicoli, and the presence of abnormal caecal contents were compared between groups using Fisher’s exact test (two-tailed).

Results

Genome sequence analysis

Analysis of the B. pilosicoli P43/6/78^T genome sequence (GenBank accession number CP002873) identified a 1692 bp open reading frame with a potential Shine-Dalgarno ribosome binding site, and putative −10 and −35 promoter regions immediately upstream from the ATG start codon. The gene sequence encoding the 72 kDa outer membrane protein was designated bpmp72 (B. pilosicoli membrane protein of 72 kDa molecular weight). The deduced translated polypeptide consisted of 563 amino acid residues with a predicted molecular weight of 62.1 kDa.

BlastX analysis of Bpmp72 identified it as a hypothetical protein with no significant overall similarity to other proteins in the GenBank databases. The protein had an approximate 5.7% homology with Treponemal membrane protein B (TmpB) of Treponema phagedenis, with a 60% similarity in an 80 amino acid region (14% coverage over 563 amino acids), and a 56% similarity in the same region of TmpB in Treponema pallidum subsp. pallidum strain Nichols. Amino acid sequence analysis of Bmbp72 revealed the presence of a 118-residue conserved lysine motif (LysM) peptidoglycan-binding domain between residues 466–526. This domain is a widespread protein module that was originally identified in enzymes that degrade bacterial cell walls. Other bacterial proteins possessing the LysM domain, such as Staphylococci IgG binding proteins and E. coli intimin, are involved in bacterial pathogenesis (Bateman & Bycroft, 2000). Sequence analysis of the bpmp72 gene from 10 reference strains of B. pilosicoli isolated from different species showed that it was highly conserved amongst the strains, with 98–100% similarity at the nucleotide level and 99–100% similarity at the amino acid level.

Bird health

The birds were healthy on arrival and remained so throughout the course of the experiment. Production data were not recorded because of the small size of the groups.

Colonization

Faecal swabs taken from the birds on their arrival at the animal house were negative for Brachyspira species. Cumulative colonization rates for B. pilosicoli in the three groups of birds following experimental challenge are shown in Table 1. Of the 45 birds, only the faeces of two vaccinated birds (one each from groups B and C) were negative for B. pilosicoli at some sampling point following experimental challenge. By nine days post-infection (dpi), the control group had developed an 80% colonization rate, compared to 7% and 0% in the two vaccinated groups. Subsequently, the colonization rate in the control group declined, although it remained at an average rate of 45% over the 30-day period. In contrast, the colonization rates in both vaccinated groups tended to increase with time, with a maximum colonization rate of 60% in group B at 30 dpi, and 40% in group C at 25 dpi. Colonization rates in the three groups were similar at 30 dpi. For the positive control group (A), positive swabs were obtained on between 3 and 6 samplings over the experimental period (mean of 4.53 sampling days, of a possible 11). For group B, the 14 birds were colonized for between 1 and 3 samplings (mean of 1.87 sampling days), and for group C between 1 and 5 samplings were positive (mean of 1.6 samplings positive). When cumulative totals of days that the birds were culture positive over all sampling times were compared, the non-vaccinated birds had 68 of 150 sampling days when their faeces were culture positive compared to 28 of 150 for the vaccinated birds in group B and 24 of 150 in group C. The differences between the non-vaccinated and vaccinated birds were highly significant (P < 0.0001).

Table 1. Brachyspira pilosicoli colonization rates as a percentage (and number) of birds for the non-vaccinated group and two vaccinated groups following experimental challenge.

	Days post-infection
Group^a	0	7	9	11	14	18	21	23	25	28	30	Cumulative total
Group A (non-vaccinated)	0% (0)	33% (5)	80% (12)	40% (6)	47% (7)	27% (4)	33% (5)	33% (5)	60% (9)	60% (9)	40% (6)	45% (68)
Group B (100 μg vaccine)	0% (0)	7% (1)	7% (1)	7% (1)	13% (2)	13% (2)	0% (0)	27% (4)	27% (4)	27% (4)	60% (9)	19% (28)
Group C (1 mg vaccine)	0% (0)	13% (2)	0% (0)	7% (1)	7% (1)	13% (2)	7% (1)	20% (3)	40% (6)	20% (3)	33% (5)	16% (24)

^aEach group consisted of 15 chickens.

Caecal abnormalities

At post mortem examination, 11 of the non-vaccinated birds had abnormal caecal contents (foamy, gassy, and/or sticky contents), whereas only five in group B and seven in group C had abnormal contents. Despite the trend, the different rates in the three groups failed to reach statistical significance.

Discussion

This work was intended to determine whether immunization with a portion of the prominent outer membrane protein Bmpb72 of B. pilosicoli, prepared as a histidine tagged recombinant protein, could provide subsequent protection from colonization with B. pilosicoli. Laying chickens were selected as the target species in which to test the recombinant protein as they can become naturally infected with B. pilosicoli, and subsequently may suffer egg production losses and/or wet faeces (Trampel et al., 1994; Stephens & Hampson, 2002). Hence the availability of a vaccine would have direct benefits to the industry. No attempt was made to monitor production in the experimental birds because the group sizes were relatively small and it is difficult to detect significant differences in these experimental circumstances (Stephens & Hampson, 2002). The birds were individually housed, and not subjected to the stresses and crowding that may occur in commercial laying flocks where avian intestinal spirochaetosis occurs. It has been shown that exposure to the stress hormone noradrenaline increases the ability of B. pilosicoli to grow and to colonize cells in vitro (Naresh & Hampson, 2011), and the same may apply in vivo. Housing the birds individually in banks of cages with close access in the same air space increased the opportunity for B. pilosicoli to be transmitted between the birds either by direct contact or by aerosol. Similarly, the inclusion of zinc bacitracin in the diet was intended to increase the susceptibility of the birds to colonization with B. pilosicoli (Stephens & Hampson, 2002; Jamshidi & Hampson 2002). As this experiment was intended to compare colonization rates in vaccinated with non-vaccinated (positive control) birds following experimental infection, a negative control group (non-vaccinated, non-infected) housed in a separate room was not included. Nevertheless, birds accessed from the same source, fed the same diet and not experimentally infected were regularly housed in the same isolation house: routine sampling had not found any to be colonized with B. pilosicoli or to have had abnormal caecal contents on post mortem examination (Jamshidi & Hampson, 2002).

The carboxy-terminus of Bmp72 was used as the vaccine candidate in this study because of its ease of production. It remains unclear as to whether immunization with the whole molecule or the N-terminus would offer similar or better levels of protection. The vaccine dose, adjuvant used, presentation routes and timing were not optimized, and it is likely that greater efficacy could be achieved by manipulating these parameters. The use of a priming intramuscular vaccination followed by an oral presentation of the antigen was intended to help stimulate mucosal protection (Keren et al., 1988; McCluskie et al., 2002). Protection might be further improved by incorporating heterologous prime-boost vaccination (Wang et al., 2017). In this experiment there was no difference in outcome when administering either 100 μg or 1 mg of the recombinant protein intramuscularly. Clearly it would be more economical to use the lower dose in a commercial vaccine as long as it was effective.

Although the vaccination procedure did not prevent colonization following experimental infection, it did significantly reduce its duration. It was interesting that the colonization rate in the vaccinated birds tended to increase towards the end of the experiment, and this could indicate a decline in immunity at a time when the infectious dose circulating between birds in the room was very high. An additional boosting dose of the vaccine might have increased the duration of protection.

The tendency for fewer of the vaccinated birds to have abnormal caecal contents at post mortem examination compared to the non-vaccinated birds was consistent with the less persistent colonization with B. pilosicoli recorded in the vaccinated groups. The findings also support previous observations that colonization with B. pilosicoli can cause sticky and foamy caecal contents (Trampel et al., 1994; Stephens & Hampson, 2002; Le Roy et al., 2015).

Together, these results suggest that Bpmp72 could be developed as a vaccine component to help protect birds (and possibly other animal species) from colonization by B. pilosicoli. The vaccine dose rate and administration route(s) require optimization in order to improve its efficacy. The availability of an effective vaccine would be very beneficial, including reducing the need for antimicrobial use to control infections with B. pilosicoli in birds and other species.

Acknowledgement

TL and DJH are authors on a patent relating to the diagnostic and therapeutic use of Bpmp72.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was funded by a project grant entitled “A study on the development of vaccines to control enteric bacterial infections of pigs” from the former Novartis Animal Vaccines Limited.