Abstract
Broiler chickens are widely considered an important source of human exposure to Campylobacter jejuni because of the high numbers found colonizing the chicken gut and the consequent contamination of processed carcasses. We hoped to intervene in C. jejuni gut colonization by using a defined probiotic. Chicken intestinal contents were screened for diverse bacterial isolates that manifested C. jejuni inhibition. These antagonistic bacteria were fed directly to chickens before or after C. jejuni challenge. The prophylactic probiotic treatments were effective only when very low challenge levels of C. jejuni were used. Otherwise, probiotic treatments failed to reduce C. jejuni colonization. Regardless of treatment, as birds further aged C. jejuni numbers soon approached the levels observed in the control birds. We sought an alternative approach, as commercial broilers may be exposed to infectious levels of C. jejuni at any time during production. Two of our most promising antagonistic isolates, Lactobacillus salivarius NRRL B-30514 and Paenibacillus polymyxa NRRL B-30509, were further studied for effectiveness in reducing C. jejuni in chickens. When 250 mg of purified bacteriocins (produced by these organisms)/kg feed were fed therapeutically to chickens colonized with C. jejuni, colonization was reduced by at least one million-fold. Treatments with viable probiotic bacterial cultures were ineffective in reducing C. jejuni in chickens, while bacteriocin treatment from these corresponding bacteria substantially reduced C. jejuni colonization in the live birds.
Introduction
Competitive exclusion (CE) has long been known to reduce Salmonella colonization in chickens Citation[1]. The mechanism by which CE reduces Salmonella has been a matter of scientific speculation. Some explanations include: (i) the antagonizing bacteria preferentially occupy Salmonella intestinal colonization niches and/or preferentially consume required substrates; (ii) the probiotic organisms have a shorter doubling time than Salmonella and outgrow the pathogen; (iii) antagonists produce volatile fatty acids associated with Salmonella killing; (iv) probiotic organisms elicit a host immune modulation response to clear Salmonella; or (v) the CE bacteria produce metabolites that interfere with or kill the target organism. In vivo data supporting these contentions have not definitively proven or disproven these as the central mechanism involved in demonstrable intestinal reduction of Salmonella via CE.
In an attempt to use a similar practical approach, we previously observed inconsistent Campylobacter jejuni reductions by administration of CE in chickens Citation[2]. On several occasions, reduction of C. jejuni intestinal colonization was demonstrated, while at other times the same CE flora did not provide campylobacter reduction. We wanted to determine whether a more systematic selection of diverse antagonistic enteric bacteria might yet provide consistent control of broiler intestinal colonization. Fuller and Gibson Citation[3] suggested that probiotic organisms benefit the health of the host animal by improving the balance of the intestinal microbial community. In the course of characterizing the in vitro antagonism of campylobacter by selected chicken gut flora, we gathered data to explain a mechanism by which in vivo antagonism does occur, and successfully applied these observations to dramatically reduce C. jejuni colonization in the broiler chickens.
Materials and methods
Isolation of antagonistic bacteria
To assure diversity of antagonistic bacteria, intestinal materials from hundreds of healthy birds, from dozens of processed Russian broiler flocks, were used as donor source material. Both broilers that were colonized with Campylobacter and non-colonized birds were used as bacterial donors. Birds were commercially processed and after evisceration ceca were aseptically excised, individually placed in sterile plastic bags, and returned on ice to the laboratory within 4 h. Three bacterial taxa were designated for further study: 1) lactic acid bacteria (LAB), 2) facultative spore-forming bacteria, and 3) Escherichia coli. Correspondingly, isolates were obtained by 1) directly streaking the cecal material onto MRS (deMan, Rogosa, Sharpe) agar and incubating at 37°C for 24 h; 2) diluting cecal materials in buffer (1:10), heating at 8°C for 15 min, and streaking the remaining viable organisms on either Plate Count or Starch agar and incubating at 30°C for 24 and 48 h; and 3) directly streaking cecal content onto EMB (eosin methylene blue) agar and incubating plates at 37°C for 24 h. Isolates were re-streaked for purity and identified using API test strip procedures. About 107 cfu of each candidate antagonistic isolate was surface plated onto agar and grown overnight on each respective agar. Subsequently, ∼0.5 cm3 agar plugs were aseptically excised, inverted, and transferred onto Brucella agar plates seeded with ∼107 cfu of C. jejuni NCTC 11168, and the cultures were incubated overnight at 42°C for 36–48 h under microaerobic conditions (5% O2, 10% CO2, 85% N2). Antagonistic isolates were initially selected based on the size of the zone of inhibition. Antagonists producing larger zones were saved as most desirable. Antagonistic bacteria isolated and used for subsequent chicken treatments included Bacillus subtilis 1316, E. coli BS10, E. coli BS230, Enterococcus faecium LWP 26, Ent. faecium LWP 408, Ent. faecium 21, Ent. durans 26, Eubacteria saburreum LWP-130, Lactobacillus acidophilus PVD-26, Lact. acidophilus LWP-320, Lact. acidophilus VF-1, Lact. salivarius PVD-32 (NRRL 30514), Lact. salivarius PVD-279, Lact. salivarius LWP-163, Lact. salivarius LWP-287, Lact. salivarius PVD-10-5, Lact. salivarius PVD-68-6, Mitsuokella multiacidus LWP-92, Paenibacillus polymyxa 37, Paen. polymyxa 119, and Paen. polymyxa 602 (NRRL B-30509).
Chicken handling and administration of bacteria
Approval for the conduct of the following animal experiments was provided by the Institutional Animal Care and Use Committee (PMS-03-03, ‘Control of Campylobacter in Poultry Production’). Healthy day-of-hatch chicks were obtained from commercial hatcheries and returned to the laboratory facilities within 4 h of procurement. Birds were randomly distributed into groups of 5–20 chicks each. Individual birds were gavaged with selected probiotic strains of bacteria (indicated above), using inoculation volumes of 0.2–0.5 ml per bird and times of administration that ranged from day of hatch through 10 days post-hatch. Using selected strains of C. jejuniCitation[4], mid-log cultures were grown on Campylobacter blood agar plates at 42°C for 24 h, and incubated under microaerobic conditions (5% O2, 10% CO2, 85% N2). The growth was resuspended and diluted in saline buffer to provide challenges ranging from 103 to 108 cfu/chick and times from day of hatch to day 4 post-hatch. Probiotic organisms were provided both by prophylactic (before campylobacter challenge) and therapeutic (post campylobacter colonization) administrations. At designated times, chickens were sacrificed, ceca were aseptically dissected, and contents were plated onto Campylobacter selective media. Plates containing serial dilutions of cecal contents were incubated at 42°C for 36–48 h, under microaerobic conditions. Representative colonies were confirmed as C. jejuni by both phase-contrast microscopy and latex agglutination tests. Enumeration was reported after log10 conversion. Mean values and standard deviations were reported for each group.
Bacteriocin production and chicken treatment
Bacteriocin from Paen. polymyxa B-30509 was produced and purified as described previously Citation[5]. In brief, the culture was grown for 40 h at 32°C in Kugler's broth, then cells were removed by centrifugation and filter-sterilized. The cell-free supernatant was exposed to 80% ammonium sulfate and the protein precipitate was redissolved and dialyzed before being further purified by gel filtration and ion exchange chromatography. Chicken challenge and bacteriocin treatments have been described elsewhere Citation[6]. In brief, the bacteriocin obtained as described above was micro-encapsulated in PVP (polyvinylpyrrolidone) and incorporated into broiler chicken feed at 250 mg bacteriocin/kg feed. Selection of Lact. salivarius B-30514, characterization and production of its bacteriocin, and use for chicken treatments are described elsewhere Citation[7]. We precisely followed these protocols for bacteriocin production and distribution into feeds.
Results
Providing chicks with combinations of either Lactobacillus spp. or Paen. polymyxa probiotic antagonists before C. jejuni challenge did not alter the resulting level of cecal colonization. C. jejuni colonization ranged from an average of 3×107 to 3×108 per g of cecal material. This colonization occurred despite the prophylactic administration of large numbers (107–108 cfu/chick) of probiotic organisms, which had created zones of inhibition of C. jejuni in vitro. Additional, more extensive sets of multi-genera antagonistic probiotic bacterial strains, likewise, did not reduce C. jejuni colonization in chicks. Treated birds were given 108 cfu probiotics per chick, and yet C. jejuni were colonized with approximately the same numbers (∼109 cfu per g) in cecal materials as found in the untreated birds. Only when chicks were gavaged daily with defined probiotic antagonists, and then only when these birds were challenged with very low levels (∼103 cfu per chick) of C. jejuni, was colonization eliminated in the animals. Treated and control groups challenged with higher levels of C. jejuni were colonized at approximately 3×108 cfu per g. However, this prophylactic treatment by the probiotic bacteria was inconsistently effective, as even the treated group of chicks challenged with only ∼103 cfu C. jejuni per chick were subsequently colonized in excess of 109 cfu per g cecal contents. Even when a reduced level of C. jejuni colonization was noted following Lact. salivarius NRRL B-30514 treatment at 7 days of age, this significant difference (p≤0.017) as compared with the control birds was lost by day 10 of colonization. No reduction was seen when combinations of high levels of prophylactic probiotic bacteria were given via daily repeated feeding or by incorporating probiotics in the animal feed (). Daily feeding of very high numbers of probiotic bacteria, likewise, did not reduce cecal colonization by C. jejuni.
Table I. Lack of prophylactic activity of bacterial antagonists in chicks challenged at day 5 post-hatch with ∼105 cfu C.jejuni.
It was only after colonized chicks were fed bacteriocin on a daily basis that reduction of C. jejuni was manifested (p≤0.0001) among the treated birds as compared with the control group (). Bacteriocins used as therapy for this study were produced by Paen. polymyxa B-30509 and Lact. salivarius B-30514. These were bacterial isolates that were also used in the above probiotic studies. Complete elimination of C. jejuni from the 10 chick ceca was achieved among the group given 250 mg bacteriocin derived from Paen. polymyxa NRRL B-30509/kg feed. This was compared to an average of ∼109 cfu C. jejuni/g cecal materials among eight non-treated control chicks. The difference represented greater than a billion-fold reduction in colonization among the treated and control groups. Although total elimination was not achieved with 250 mg bacteriocin derived from Lact. salivarius NRRL B-30514/kg feed, dramatic reductions in C. jejuni numbers were observed as compared with the levels seen in the control birds. Eight of the nine treated chicks had no detectable levels of C. jejuni, while the remaining treated bird had only ∼103 cfu/g cecal materials.
Table II. In vivo anti-C. jejuni activity of bacteriocins produced by Paenibacillus polymyxa NRRL B-30509 and by Lact. salivarius NRRL B-30514 administered in emended feeds.
Table III. Influence of selected antagonistic strains given to chicks on C. jejuni colonization.
Table IV. Influence of selected antagonistic strains on C. jejuni colonization in chicks.
Table V. Day-of-hatch chicks initially provided with 2.0×108 Lact. salivarius NRRL 30514/chick and ∼106 Lact. salivarius/g feed for 10 days.
Table VI. Lack of prophylactic activity of antagonists in chicks challenged at 5 days of age with ∼105 cfu C.jejuni; birds were sacrificed and sampled at 13 days of age.
Table VII. Lack of prophylactic or therapeutic activity of antagonistic Lact. salivarius isolates against C. jejuni-colonized chicks.
Table VIII. In vivo anti-C. jejuni activity of bacteriocins produced by Paenibacillus polymyxa NRRL B-30509 and by Lact. salivarius NRRL B-30514 administered in emended feeds.
Discussion
Directly fed probiotic bacterial treatments had limited influence on broiler chick colonization by C. jejuni (), while purified bacteriocins provided dramatic reductions over that commensal pathogen (). In 1984, Goren et al. Citation[8] were among the first to report on Salmonella reductions in a large European field trial in which CE was administered by spray application to chicks. Intestinal carriage of Salmonella was reduced among the treated groups of chickens. The poultry industry in Europe, primarily Finland and Sweden, has continued to routinely use CE prophylactically to treat flocks deemed to be at risk for Salmonella infection Citation[9]. On the other hand, in the United States, the FDA has been reluctant to provide approval for administration of undefined microbial preparations. The FDA position has been that starting CE material would have to be comprehensively screened for all pathogens that may be present and then disseminated in such a treatment product. Consistent product efficacy would also need to be assured among CE batches.
Previous studies have indicated that standard preparations of CE, even when effective against Salmonella spp., offered no resistance to colonization by Campylobacter spp. Citation[10], Citation[11]. On the other hand, Morishita et al. Citation[12] reported reductions of C. jejuni among colonized chickens after treatment with avian-isolated cultures of Lact. acidophilus and Streptococcus faecium. Likewise, Hakkinen and Schneitz Citation[13] reported that a non-defined CE product (Broilact) reduced colonization by C. jejuni in a chick challenge model. These approaches have not yet been reproduced and are not yet employed by the poultry industry to reduce the target organism. Chaveerach et al. Citation[14] reported isolation of a Lactobacillus sp. strain P93 that showed bactericidal activity against Campylobacter spp. They suggested that the bactericidal effect was due to the production of organic acids in combination with an anti-Campylobacter protein. They did not, however, report on purification of the protein or on testing the efficacy in a chicken model.
In the United States, colonization of the gastrointestinal (GI) tract in commercial chickens occurs in a non-predictable manner and, in part, is dependent on the chance distribution of the organism in the chicken's environment. In the present study, Campylobacter-antagonistic probiotics were provided to alter the GI tract flora but these alterations did not reduce C. jejuni in the chicken gut. Metabolic by-products such as peroxide compounds and volatile fatty acids are recognized antimicrobials that have been attributed to the dominance of a CE flora Citation[15–17].
We had hoped to systematically employ CE to reduce C. jejuni in chicks. After constructing a diverse panel of bacterial isolates antagonistic to C. jejuni, we wished to determine whether prophylactic or therapeutic administration of probiotics might provide reduction of pathogen colonization in chickens. It was only after repeated ‘failures’ in vivo that we observed the consistent benefit of administering bacteriocin produced by these isolates. Reporting a similar failure, Barnes et al. Citation[15] had isolated Bacteroides hypermegas and Bifidobacterium sp., antagonistic to Salmonellae in vitro, but these organisms showed no reductions of Salmonella typhimurium in a chick model. The success in bacteriocin feeding reported in this manuscript corroborates results published previously Citation[6], Citation[7]. Although CE has proven effective against Salmonella spp., our results do not support the thesis that administering bacteriocin-producing bacteria would form an effective CE preparation for reduction of C. jejuniCitation[18].
Bacteriocins are known to disrupt the integrity of the cellular membrane among the target organisms by causing pore formation, cell lysis, and subsequent death Citation[19]. Bacteriocins consist of bactericidal polypeptides and the mechanism is thought to be similar to that expressed by ionophore antibiotics. The current study provided our developmental rationale for an overall approach toward bacteriocin application to reduce poultry-borne pathogens. Our completely defined bacteriocin treatment would preclude inadvertent administration of pathogens and would yield predictable and reproducible pathogen killing.
CE failed to exclude C. jejuni in this study, although CE has been repeatedly demonstrated to be effective against Salmonella spp. C. jejuni reside in the crypts of Lieberkuhn, deep within the intestinal villi. Limited numbers of competing flora can be found in that niche. This is in part due to the unique nutrients exploited there Citation[20] and the microaerobic atmosphere, to which Campylobacter spp. are adapted. Thus, competing flora were excluded from that niche and probiotic therapy appeared to have limited influence on C. jejuni colonization. However, exogenous bacteriocins, which are aqueous soluble, were able to diffuse into this unique niche and destroy Campylobacter spp. in vivo. We suggest that those intestinal niche differences manifested by Salmonella spp. and C. jejuni in gut colonization are likely responsible for the difference in CE outcomes. Salmonella spp. are believed to adhere to the intestinal linings Citation[21], while C. jejuni is thought to be free swimming within the mucin layer, deep within the luminal crypts Citation[22]. In the crypts of Lieberkuhn there is a comparative lack of diversity in bacterial flora present and C. jejuni appears to be the dominant organism. Thus, our selected antagonistic probiotics did not, apparently, come in direct contact with the target organism as they do not occupy this portion of the intestine. It is likely that bacteriocins are produced in situ by bacteria in limited quantities. This quantity will effectively compete with proximally colonizing target bacteria by killing its host competition. It is unlikely that bacteriocin produced in vivo by the producer organism would be over-produced, as the anabolic costs to the organism are too great to be wasted. Because bacteriocins are aqueous soluble, the quantity we provided to the animals was adequate to reach the crypts containing the C.jejuni and killed that target. Further dose-response studies will be needed to establish requisite quantities of bacteriocins needed to kill C. jejuni in the chicken gut. These antagonistic strains were selected by virtue of our in vitro testing for zones of inhibition against C. jejuni. The zones of inhibition were due to the presence of bacteriocins Citation[5], Citation[7].
Although production of bacteriocins by intestinal bacteria has been recognized, its prominent role within gut ecology has not been appreciated. In part, this is due to the high metabolic costs expended by bacteria to elaborate and secrete these non-structural polypeptides. It is likely that bacteriocins play additional roles in regulating the intestinal flora, such as signaling within and among microbial species Citation[23].
Bacteriocins represent a novel therapeutic approach to address intestinal infections. These natural antimicrobial polypeptides have proven effective in reducing a previously rampant infection. Although bacteriocins are routinely produced by these same probiotic intestinal bacteria, those bacteriocin concentrations achieved in vivo were inadequate to kill Campylobacter spp. under in vivo conditions. The present study does not allow us to conclude that bacteriocins are the most important mechanism involved in CE, but our data do support bacteriocins as a significant contributor to the benefits derived from probiotics Citation[24]. By providing these polypeptides as a therapeutic, an alternative treatment has been offered.
Acknowledgements
Funds for the conduct of this research were provided by the US State Department (ISTC #1720p), the US Department of Agriculture (CRIS 6612–32000–034–00), and the State Research Center for Applied Microbiology and Biotechnology.
References
- Nurmi E, Rantala M. New aspects of Salmonella infection in broiler production. Nature 1973; 241: 210–11
- Stern NJ. Mucosal competitive exclusion to diminish colonization of chickens by Campylobacter jejuni. Poultry Sci 1994; 73: 402–7
- Fuller R, Gibson GR. Modification of the intestinal microflora using probiotics and prebiotics. Scand J Gastroenterol 1997; 222: 28–31
- Stern NJ, Bannov VA, Svetoch EA, Mitsevich IP, Volozhantsev NV, Gusev VV, et al. Distribution and characterization of Campylobacter spp. from Russian poultry. J Food Prot 2004; 67: 239–45
- Svetoch EA, Stern NJ, Eruslanov BV, Kovalev YN, Volodina LI, Perelygin VV, et al. Isolation of Bacillus circulans and Paenibacillus polymyxa strains inhibitory to Campylobacter jejuni and characterization of associated bacteriocins. J Food Prot 2005; 68: 11–17
- Stern NJ, Svetoch EA, Eruslanov BV, Kovalev YN, Volodina LI, Perelygin VV, et al. Paenibacillus polymyxa purified bacteriocin to control Campylobacter jejuni in chickens. J Food Prot 2005; 68: 1450–3
- Stern NJ, Svetoch EA, Eruslanov BV, Perelygin VV, Mitsevich EV, Mitsevich IP, et al. Isolation of a Lactobacillus salivarius and purification of its bacteriocin that is inhibitory to Campylobacter jejuni in the chicken gastrointestinal system. Antimicrob Agents Chemother 2006; 50: 3111–16
- Goren E, de Jong WA, Doornebal P, Koopman JP, Kennis HM. Protection of chicks against Salmonella infection induced by spray application of intestinal microflora in the hatchery. Vet Q 1984; 6: 73–9
- Mead GC. Prospects for “competitive exclusion” treatments to control Salmonellas and other foodborne pathogens in poultry. Vet J 2000; 159: 111–23
- Stern NJ, Bailey JS, Blankenship LC, Cox NS, McHan F. Colonization characteristics of Campylobacter jejuni in chick ceca. Avian Dis 1988; 32: 330–4
- Shanker S, Lee A, Sorrell TC. Horizontal transmission of Campylobacter jejuni amongst broiler chicks: experimental studies. Epidemiol Infect 1990; 104: 101–10
- Morishita TY, Aye PP, Harr BS, Cobb CW, Clifford JR. Evaluation of an avian-specific probiotic to reduce the colonization and shedding of Campylobacter jejuni in broilers. Avian Dis 1997; 41: 850–5
- Hakkinen M, Schneitz C. Efficacy of a commercial competitive exclusion product against Campylobacter jejuni. Br Poult Sci 1999; 40: 619–21
- Chaveerach P, Lipman LJA, van Knapen F. Antagonistic activities of several bacteria on in vitro growth of 10 strains of Campylobacter jejuni/coli. Int J Food Microbiol 2004; 90: 43–50
- Barnes EM, Impey CS, Stevens BJH. Factors affecting the incidence and anti-salmonella activity of the anaerobic caecal flora of the young chick. J Hyg (Camb) 1979; 82: 263–83
- Nisbet DJ, Corrier DE, Scanlan CM, Hollister AG, Reier RC, DeLoach JR. Effect of defined continuous-flow derived bacterial culture and dietary lactose on Salmonella typhimurium colonization in broiler chickens. Avian Dis 1993; 37: 1017–25
- Corrier DE, Nisbet DJ, Scanlan CM, Hollister AG, Caldwell DJ, Thomas LA, et al. Treatment of commercial broiler chickens with a characterized culture of cecal bacteria to reduce salmonellae colonization. Poultry Sci 1995; 74: 1093–101
- Joerger RD. Alternatives to antibiotics: bacteriocins, antimicrobial peptides and bacteriophages. Poultry Sci 2003; 82: 640–7
- Tagg JR, Dajani AS, Wannamaker LW. Bacteriocins of Gram-positive bacteria. Bacteriol Rev 1976; 40: 722–56
- Hugdahl MB, Beery JT, Doyle MP. Chemotactic behavior of Campylobacter jejuni. Infect Immun 1988; 56: 1560–6
- Finlay BB, Heffron F, Falkow S. Epithelial cell surfaces induce Salmonella proteins required for bacterial adherence and invasion. Science 1989; 243: 940–3
- Beery JT, Hugdahl MB, Doyle MP. Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Appl Environ Microbiol 1988; 54: 2365–70
- Hoffman L, D'Argenio D, Bader M, Miller S. Microbial recognition of antibiotics: ecological, physiological, and therapeutic implications. Microbe 2007; 2: 175–7
- Atlas CM. Probiotics – snake oil for the new millennium?. Environ Microbiol 1999; 1: 375–82