Following an imaginary Campylobacter population from farm to fork and beyond: a bacterial perspective

Campylobacter jejuni is one of the most common causes of food-borne bacterial gastroenteritis worldwide; in the European Union, it is the most common cause, before Salmonella (Bronzwaer et al. 2009). In the United States, 1Æ4 million yearly cases were estimated (Mead et al. 1999), though reported cases have since shown a decline (Anon 2010). Nevertheless, in comparison with the decline in the number of salmonellosis cases in many countries, far less success can be reported for campylobacteriosis. An infection with Campylobacter is usually selflimiting, but the economic costs of campylobacteriosis are substantial, and complications such as Guillain–Barre´ syndrome (GBS) can be life-threatening. One of the main sources for this zoonotic disease is consumption or handling of contaminated poultry meat.

 

Interventions to reduce the risk of poultry meat have been targeted at four levels: (1) to reduce the numbers of flocks colonized by Campylobacter; (2) to reduce the numbers of bacteria in the birds prior to slaughter; (3) to avoid cross-contamination during slaughter from positive to negative birds; and (4) to reserve contaminated meat for frozen products, which reduces the number of viable organisms. Despite multiple efforts, in most countries the poultry industry has not been able to reduce the prevalence, or levels of contamination, to acceptable levels. A few successes have been reported using strategies 3 and 4, mainly from Nordic countries with small-sized poultry industries (Jore et al. 2010). The measures that were necessary for these successes are difficult to implement in large-scale, low-cost production and may not be costeffective in highly competitive markets. Lately, promising results were reported to decrease the bacterial load by decontamination of end products using high-intensity light pulse (Haughton et al. 2011) or lactic acid washes (Rajkovic et al. 2010).

 

Several authors have reviewed possible targets to improve safety of the food chain or practicalities of individual mitigation measures (e.g. Newell and Fearnley 2003; de Zoete et al. 2007; Boysen and Rosenquist 2009; Bronzwaer et al. 2009; Lin 2009; Cox and Pavic 2010; Guerin et al. 2010). Here, an overview of the complete route from farm to fork is presented, completed with events during human infection, and possibilities to reduce the risk of poultry-associated campylobacteriosis cases are identified.

 

The virulence properties of these bacteria have mostly been investigated using in vitro models, in lack of a suitable animal model. The search of conserved virulence genes has mostly been in vain, with the exception of cytolethal distending toxin (CDT). Although this toxin is strongly conserved within Camp. jejuni and a mechanistic explanation of its toxicity is available (Smith and Bayles 2006), it has not been demonstrated that the toxin is directly responsible for (some of) the symptoms that are typical for campylobacteriosis. Recently, novel theories have been put forward that could explain how these bacteria cause disease in humans, and these will be reviewed here. Current insights from recently published date are summarized, and novel hypotheses can be proposed, for instance on the microheterogeneity of a clonal Campylobacter population. Data indicate that as yet unrecognized, indirect routes may exist by which Campylobacters, which multiplied in poultry, might reach humans, independently from food-related sources.

 

This story begins with hatching chicken eggs in a hatchery. As the little chicks leave their eggs, they will start to collect a microflora, which in most cases does not (yet) contain Camp. jejuni or Campylobacter coli. (the term ‘Campylobacter’ will be used here indiscriminately.) Chicks usually become colonized by these Gram-negative, motile and microaerophilic bacteria, belonging to the epsilon division of the Proteobacteria, a few weeks after hatch (Newell and Fearnley 2003). This delay might be due to a lack of exposure, but in addition, it appears that young chicks are temporarily resistant to colonization: 3-week-old birds are more susceptible to colonization upon challenge than birds that are 1–2 weeks old; confusingly, and in apparent contradiction to this, day-old chicks are also more susceptible than 1-week-old birds (Cawthraw and Newell 2010). It was also shown that 2-day-old chicks are more susceptible than 2-week-old birds (Ringoir et al. 2007). The age-related susceptibility has been attributed to maternal antibodies (Cawthraw and Newell 2010), but the efficiency of transmission between coprophagic birds is also age-dependent. When all these factors were taken into account in a mathematical model, age-dependent transmissibility was identified as the main factor explaining this so-called lag phase (Conlan et al. 2011). Proper biosecurity measures can extend this lag phase, up to the age that broilers are being slaughtered. More frequently, though, an unidentified breach of biosecurity is responsible for entry of Campylobacter in a flock. Thus, typically, Campylobacter enters a flock of growing broilers when the birds are 2–3 weeks of age or any time thereafter. The birds remain healthy during this colonization, so that Campylobacter remains unidentified, unless specifically tested for.

 

The relevant question to ask is where these bacteria come from. Risk assessment of chicken colonization on the farm has indicated various means of entry that may vary between countries and farm types, and these have been extensively reviewed elsewhere (Newell and Fearnley 2003). A few routes can be excluded. Chicks do not normally acquire Campylobacter from the mother hen, as horizontal transmission is rare. Feed is also an unusual contamination source, but drinking water can be a port of entry, depending on the water quality. Other risk factors are human activity, rodents, insects, other farm animals present on the farm and the environment outside the housing in general, as Campylobacter seems to be ubiquitous in the environment (Newell and Fearnley 2003). The organism can only grow inside a host (mostly birds and mammals) so that environmental loads are the result of animal faecal contamination. Humans can introduce the organisms into a farm via dirt on boots, hands or utensils. In particular, thinning (partial depopulation of the flock) poses a risk of introducing Campylobacter into the remaining bird population (Allen et al. 2008). Some studies found a higher incidence in flocks of larger size (Na¨ther et al. 2009), many publications report higher incidences in free-range or organic birds than in conventionally farmed animals (e.g. Bokkers and de Boer 2009; Na¨ther et al. 2009), and in nearly all studies, the incidence is higher in summer than in winter.

 

Once one or a few birds have been exposed to Campylobacter (from whatever source), the bacteria will colonize their crop and intestine, in particular the caeca, and multiply. The body temperature of chickens happens to be close to the optimal growth temperature of these thermophilic bacteria. The chicks do not suffer from this colonization in any way although bacteria can occasionally be identified in the liver and the reproductive tract (Cox et al. 2009; Alter et al. 2011). The bacteria leave the birds via faeces, and pecking and coprophagy ensure a rapid spread between birds. Once a few birds start secreting high numbers of bacteria, the complete flock will soon be positive within a few days, depending on the flock size (Newell and Fearnley 2003; Nauta et al. 2009a). Nevertheless, the numbers of Campylobacter found in the caeca of individual birds can vary considerably within a flock (Hansson et al. 2010). During their short life span, broilers do not eradicate the bacteria, though they do produce antibodies against the bacteria (Newell and Fearnley 2003). Campylobacter levels decrease in numbers over time in laying hens, but still roughly two-thirds of sampled laying hens can be found positive (Dipineto et al. 2011).

 

Birds within a flock usually carry one or a few strains only, but with so many different possible sources for Campylobacter, there is extensive strain variation found in different flocks. One strain may colonize consecutive flocks or dominate a poultry house for some time, but in essence, it cannot be predicted which strain will colonize a next round of birds (Alter et al. 2011).

 

Even if all birds are contaminated with the same strain, there will be minor genetic variation in the bacterial population as a whole. Two mechanisms are mainly responsible for this microvariation, both depending on replication errors. (Less frequent variation as a result of DNA recombination is ignored here.) One mechanism is the introduction of variation in homopolymeric G-stretches found in a number of Campylobacter genes. The length of these stretches varies between cells of clonal descendance, resulting in genes being variably expressed (Wassenaar et al. 2002), causing detectable phenotypic variation, as reviewed by van Putten et al. (2009). None of these so-called contingency genes are essential, and few are conserved. Many homopolymeric stretches can be found in a variety of transferase enzymes, but the presence (of these genes and of the stretches that may or may not be present) varies widely between strains (T.M. Wassenaar, unpublished observations). The microheterogeneity of a clonal population owing to these polymorphisms improves avian colonization (Wilson et al. 2010). The second mechanism to introduce minor variation in a bacterial population is also the result of sloppy replication. DNA polymerase of Campylobacter makes mistakes even in the absence of a homonucleotide stretch, which remains uncorrected at an estimated rate of 1Æ9 · 10)6 per kilobase (Wilson et al. 2009). This would produce on average three mutations per 1000 cells. A typical chicken caecum can easily hold 108 cells at any given time, which could bear as many as 3 · 105 mutations (assuming these would be neutral, so that there is no selection). The true number of mutations will be lower owing to negative selection, but the population does not reproduce in one chicken – a flock may consist of 10 000 birds or more. It is therefore not unreasonable to expect thousands of random mutations in a population of Campylobacter that resides in a typical chicken flock by the time the birds are ready for slaughter. Very few of these mutations would be detectable by current genotyping methods, such as multilocus sequence typing or pulsed-field gel electrophoresis. The microheterogeneity of a Campylobacter population has to be kept in mind, but it is dwarfed by the wide genetic diversity of strains that can colonize different flocks.

Given the scenario described above, what can be done to prevent contamination of poultry on the farm? After all, the most effective way to avoid food contamination by zoonotic bacteria is to prevent the colonization of food animals in the first place.

 

Because chicks do not get Campylobacter from their mothers, a top-down approach, which has been shown to be very effective to reduce Salmonella enterica Enteritidis from poultry, will not work, though improved biosecurity measures will certainly be beneficial. Naturally, spread from one flock to another should be avoided, so the use of continuous production systems as opposed to allin- all-out systems, or reuse of litter, should be strongly discouraged. One of the most significant risks of introducing Campylobacter in a flock is human activity, and biosecurity, training of personnel, and decontamination of rooms, boots, equipment and vehicles can all contribute to reduce this risk. In Denmark, the role of flies as vectors to spread Campylobacter was demonstrated, and fly nets were shown to reduce introduction of the bacteria into broiler houses (Hald et al. 2007). Whether insect control is as effective in other countries remains to be established.

 

The birds themselves can also be targets to try and reduce the risk of colonization. Vaccination of birds, another strategy with proven success to combat Salmonella, has been tried for Campylobacter, but early attempts were mostly without success (Newell and Fearnley 2003; Lin 2009). Subunit vaccines based on flagellin fragments provided homologous protection in some instances, but the variation of this protein, in part caused by variation in its glycosylation states, limits cross-protection between strains. This work has been summarized elsewhere (Newell and Fearnley 2003; Layton et al. 2011). Lifeattenuated Salmonella vaccines constructed to express immunogenic Campylobacter peptides have also been tried, with more promising results. Vaccine constructs based on peptide fragments of proteins CjaA (Cj0982c) or PeB1 (Cj0921c) resulted in decreased colonization levels (Buckley et al. 2010), though the decrease was not as dramatic as originally described (Wyszyn´ ska et al. 2004). An attenuated Salmonella vaccine construct containing a peptide from protein Cj0113 (also called Omp18 or CjaD) resulted in complete protection against colonization upon heterologous challenge (Layton et al. 2011). Such experiments suggest that it may indeed be possible to produce an effective chicken vaccine that protects against all or most Campylobacter strains, though none of these attempts have yet been scaled up to production levels.

 

Competitive exclusion by defined or undefined bacterial mixtures, probiotics and prebiotics have also been tried and largely found ineffective to prevent Campylobacter carriage (reviewed by Lin 2009). However, mathematical modelling suggests that it may not be necessary to completely abolish contamination. For an effect on public health, it might be sufficient if contamination of poultry meat could be reduced by a 100- or 1000-fold (Nauta et al. 2009a). In this respect, prebiotics, probiotics or competitive exclusion may have been discarded too soon, as their effectiveness had been assessed mostly with the goal to eliminate Campylobacter colonization completely. For instance, reduction of one or two logs in Campylobacter carriage had been demonstrated with lactic acid bacteria (Morishita et al. 1997). Recently, the use of probiotic bacteria has been revisited (Santini et al. 2010), though the term ‘probiotic’ is inaccurate here, as the desired outcome is not healthier birds but a reduced Campylobacter load instead.

 

If contamination cannot be avoided, it may be possible to eliminate the bacteria or decrease their load prior to slaughter. That is the idea behind phage therapy. Lysogenic bacteriophages can be highly specific or have a broader bacterial host range. A narrow species range (Campylobacter only) but a broad strain range therein is desirable, so that the phages to be used are able to attack the wide range of Campylobacter strains that can colonize chickens. Bacteria and their infective phages usually reach an equilibrium in which neither is eliminated, but when a naive Campylobacter population is present in a bird, an initial infection with a high dose of phage can reduce bacterial loads with several orders of magnitude (Wagenaar et al. 2005). Thus, applying large numbers of phages shortly before slaughter could, in theory, reduce Campylobacter carriage (Connerton et al. 2011). The method is currently being tested under laboratory conditions, but it has not yet been shown to be effective in a commercial setting (Lin 2009). Bacteria develop resistance to their phages, and this could render the procedure ineffective, especially because it will be difficult to eradicate all phages from the rooms after removal of a treated flock. If residual phages would establish infection in a new round of flocks prematurely, there will be far less of a reduction when the aimed treatment is started prior to slaughter. In conclusion, phage therapy has not shown its worth yet, but more data are needed to decide whether this is the way to go.

 

The use of specific bacteriocins, compounds produced by other bacterial species that are toxic to Campylobacter, is also currently experimentally being tested (Svetoch and Stern 2010). Its effectiveness also relies on avoidance of resistance, as well as broad-range toxicity to combat all possible Campylobacter strains, which poses difficulties similar to those mentioned for phage therapy. Whether phages or bacteriocins can be produced and applied in sufficient quantities economically has not been proven, but models suggest that phage therapy might be costeffective (Mangen et al., 2007).

 

In conclusion, not one method is likely to eliminate Campylobacter carriage in broilers, but a number of methods can, single or in combination, reduce the bacterial load sufficiently to have an effect on human health. Biosecurity, including strict hygiene measures, control of insects, vermin and rodents, could effectively be combined with vaccination, probiotics, phage therapy or bacteriocin treatment. Maybe particular effective combinations of measures have to be developed for local settings to optimally suit the type of production and to counter differences in farm size and density, climate, flock size, housing facilities and other variables. Campylobacter will not be eradicated from poultry farms, but more can be done to reduce their populations.

 

This story continues with a population of Campylobacter that lives in a flock of broilers by the time the birds are transported to the abattoir. This transport will bring not only the birds but also their bacteria to the slaughterhouse. The transportation crates and vehicles used and reused for such transport received quite a bit of attention as a possible source of contamination – in this case, from birds of positive flocks to negative flocks. Such contamination occurs (Hansson et al. 2005), but it is of limited relevance. Because Campylobacter cannot replicate outside a host, negative birds that do get contaminated en route to the slaughterhouse will not live long enough to produce significant numbers of bacteria. The contamination is considerable, though, if these crates, vehicles and workers enter a next production farm to collect a new round of birds; those loaded for slaughter will end their short lives soon, but all other flocks living on that farm may now be exposed to Campylobacter that is brought in by vehicle tyres, crates, boots or hands.

 

Once the birds enter the slaughter process, their bacteria will be spread, and Campylobacter can be easily detected, in large quantities and a wide variety of strains, in slaughterhouses when broilers are being processed. Laying hens produce lower levels of contamination (Johnsen et al. 2007). The next line of defence against food-borne zoonoses should be to avoid contamination of carcasses and meat, but that miserably fails in the case of highly automated, high-throughput poultry processing. Especially, bacteria living in the crop of birds are able to spread as a result of rupture, which is more common than rupture of intestines (van Gerwe et al. 2010). Even birds that enter the slaughterhouse without Campylobacter are likely to leave the premise as contaminated meat (Ellerbroek et al. 2010). Much work has been carried out to analyse the individual steps and identify critical points, with variable results. In a systematic review of these results, the authors identified that  publications reported a consistent decrease in bacterial load (in terms of both prevalence and concentration) following scalding and an increase following defeathering; variable results were reported for the chilling, evisceration and washing steps (Guerin et al. 2010). When individual strains were identified along the processing line, instead of determining numbers of bacteria only, cross-contamination between birds originating from different flocks became apparent (Newell et al. 2001). It was subsequently shown that persistence in the abattoir varies, and some strains are able to survive standard cleaning and decontamination practices (Peyrat et al. 2008). It would be of interest to see whether persistence is variable between not only strains with a different genetic repertoire but also cells that vary owing to contingency gene variation.

 

It has been suggested that slaughter of negative flocks separate from and prior to positive flocks (scheduled processing) could avoid contamination between flocks (Wagenaar et al. 2006). This requires reliable testing of the birds prior to slaughter, which is usually carried out at the farm. Testing should not be carried out too soon, to avoid negatively tested flocks becoming colonized by the time they are slaughtered. This, and unreliable test results, could explain why scheduled processing produced disappointing results when tested in the Netherlands (Nauta et al. 2009b). Inadequate disinfection of the slaughter facilities at the end of the day may also limit the effect of scheduled processing. A slightly different approach is followed in Denmark, where the separation of negative and positive flocks determines the fate of the meat: negative flocks are used to produce fresh poultry meat, while meat of positive flocks is frozen (Wegener 2010). Freezing decreases the Campylobacter load to very low levels and is thus an effective measure to inactivate the organisms. However, the demand for fresh meat exceeds the production of negative birds in Denmark, so that the system is currently not working.

 

Better results were obtained in Iceland. From 1995 to 1998, the island saw an increase in campylobacteriosis cases, accumulating to a peak in 1999 (Jore et al. 2010). To counter the trend, preharvest testing was introduced for the relatively small-sized local poultry industry, and only flocks shown negative 2–4 days prior to slaughter were used to produce fresh meat; positively tested flocks were deemed for production of less-valued frozen or cooked meat. This resulted in a sharp decrease in human domestically acquired cases in 2000 and a steady decrease since (Jore et al. 2010). When the subtypes of Campylobacter in positive flocks (as determined by flaA-SVR typing) were compared to those of human isolates with an onset of disease within 2 weeks of processing of the positive flock, it was observed that meat with higher numbers of Campylobacter was significantly more often implicated in human cases (Callicott et al. 2008). This suggests that meat at retail with higher contamination levels provide a higher the risk, a correlation that had also been predicted by risk assessment (Rosenquist et al. 2003) and by mathematical modelling (Bronzwaer et al. 2009). Although the intervention results reported from Iceland are impressive, it should be noted that their poultry industry is of a modest scale, and implementation of the same testing and processing scheme would be impractical in countries with larger industries.

 

When all else fails, decontamination of carcasses can reduce the contamination load of meat. Physical decontamination, by means of forced air chilling, crust freezing or treatment with steam and ultrasound, has been shown to reduce bacterial loads, though none of these methods were as effective as freezing the meat completely (Boysen and Rosenquist 2009). Whereas the cost-effectiveness of these measures has not been calculated, carcass decontamination with electrolysed oxidizing water was predicted to be highly cost-effective (Gellynck et al. 2008). None of these measures are yet implicated on a routine basis.

 

For the story to continue, we presume that a Campylobacter population is present on a piece of poultry meat that is ready to be packed. Package and storage conditions are optimized to delay spoilage, but the reduced oxygen atmosphere in fact increases the chance of survival of Campylobacter. A gas mixture containing oxygen would be better in this case, as it would inactivate the majority of Campylobacter cells (Boysen et al. 2007; Rajkovic et al. 2010). Multiple studies have shown that frozen meat contains fewer viable bacteria than fresh meat so that it poses a reduced risk. This is countered by the presence or addition of chicken skin as it increases the bacterial load, even when the product is frozen (Sampers et al. 2008). Even the consumer’s practice during shopping can have an influence. It was shown that infants or children riding in a shopping cart that contained raw meat were at risk of getting Campylobacter (Patrick et al. 2010). This demonstrates that leak-proof packaging, or providing extra plastic bags to customers for transportation of packed meat, could reduce the risk of exposure.

 

The way poultry meat is treated in the kitchen is also an important factor that determines whether a meal might result in an episode of diarrhoea. Consumers are advised to cook the meat thoroughly, as this will kill all bacteria being present, but this may in fact not be the most effective advise. The majority of Campylobacter cells are found at the surface of meat, not deeply inside (Luber and Bartelt 2007), and even moderate heating will already kill most surface bacteria. In contrast, cross-contamination to food stuffs that are not heated (salads, for instance), by improper hand hygiene or reuse of utensils such as cutting board and knives, is possibly a more common cause of infection than consumption of undercooked meat (Luber 2009). The link between campylobacteriosis and summer barbeques (e.g. Allerberger et al. 2003) may be not be due to consumption of undercooked meat, but the shared use of plates and utensils for raw and cooked meat. Of particular concern is chicken liver, which frequently contains internalized Campylobacters, and this should always be heated through (Whyte et al. 2006). Indeed, liver pate´ produced with undercooked liver provides a high risk for consumers and has been a source of several outbreaks (Forbes et al. 2009; Inns et al. 2010).

 

Eventually, to let this story end in disease, live bacteria are consumed by an unfortunate individual. The dose of Campylobacter that is required for disease is not well known, as only few dose–response experiments have been performed on human volunteers. A recent study showed that half of the dosed volunteers became ill after swallowing 105 bacteria (Tribble et al. 2010), but because the used bacteria were adapted to growing in a laboratory, the infective dose of wild-type strains may in fact be lower.

 

The mechanism by which Campylobacter causes disease is still not well understood, and the search for virulence genes has resulted in a number of candidates (recently reviewed by Zilbauer et al. 2008 and Dasti et al. 2009), though no ‘smoking gun’ was found. A number of Campylobacter proteins have been described as ‘virulence factors’ in the literature, but this term suffers from overuse (Wassenaar and Gaastra 2001). The so-called virulence factors of Campylobacter are often not conserved between strains that can nevertheless cause disease (for example, the Vir plasmid is infrequently present, Louwen et al. 2006), or their presumed role in virulence is solely based on in vitro models. The results obtained with these models poorly correlate with symptoms observed in patients (Fearnley et al. 2008; Law et al. 2009). The best-characterized virulence gene, CDT, which is found to be expressed by most Campylobacter strains, has been proposed as one of the key factors responsible for disease (reviewed in Zilbauer et al. 2008). Nevertheless, strains that are not able to express the protein, owing to a deleterious mutation, can cause disease with symptoms that are typical for campylobacteriosis (Abuoun et al. 2005). The toxin is antigenic during human infection, but not during infection in chickens, although it is expressed in the avian host (Abuoun et al. 2005); despite this expression, chickens do not display symptoms. These observations are hard to explain in a model of disease that depends on toxicity of CDT.

 

The volunteer study by Tribble et al. (2010) showed that, as a result of an immune response, the individuals were protected against reinfection within 2 months, but this protection waned after 1 year. Serological evidence suggests that human exposure (to an extent that results in antibody production) is far more common than disease, so that temporary immunity may in fact be common. Indeed, in developing countries, most patients are infants and children, presumably because immunity resulting from high-level, frequent exposure protects adults against disease (Havelaar et al. 2009).

 

However, the human immune system seems to act antithetical against Campylobacter. Current insights suggest that the symptoms of enteritis are not caused by bacterial toxins produced by the bacteria but result from a local over-reaction of the intestinal innate immune system.

 

Toll-like receptors (TLRs), which recognize conserved patterns in bacterial surface and internal structures (LPS, lipoproteins, proteins and DNA), are key mediators of gut innate immunity. Following their activation, a cascade of events eventually results in the production of cytokines, notably IL-8, which consequently recruits neutrophils. Details of this process have been recently reviewed (van Putten et al. 2009). When invasive and motile Campylobacter cells penetrate the intestinal mucus layer, they are engulfed by the intestinal cells, possibly as a result of proteins that the bacteria secrete by means of their flagella. This is thought to lead to disruption of the integrity of the epithelial lining, sensed by TLRs, and results in the release of cytokines. These cytokines are mostly responsible for the symptoms of diarrhoea (van Putten et al. 2009).

 

Several previous observations fit this model. For instance, bacterial adherence and invasion as well as CDT expression have been known to mediate IL-8 production (Hickey et al. 2000), and this may explain how these properties contribute to virulence. A gnotobiotic mouse model was used to demonstrate that Campylobacter induces the TLR ⁄ NF-kB pathway, with inflammatory diarrhoea as a result, when knockout mice deficient for IL10 were used (Lippert et al. 2009). It can now be hypothesized that frequent exposure might dampen this innate immune response, resulting in tolerance such as seen in the adult population in developing countries. In developed countries where people are less frequently exposed, this hypothesized ‘protection’ would be shortlived, in support of the observations from human volunteers (Tribble et al. 2010). The model could further explain the different outcome of colonization between humans and animals. Differences in TLR responses between chicken and human cells are now being studied and may help to explain the asymptomatic colonization of this avian host (de Zoete et al. 2010). The residual microflora most likely also contributes to the outcome of colonization: using mice with a reconstituted intestinal microflora, it was shown that human, but not murine microflora resulted in TLR4-dependent diarrhoea (Bereswill et al. 2011). This could explain why colonization of wildtype mice does not result in disease.

 

This model of campylobacteriosis as an immunopathological disease would explain why the search of conserved virulence genes has been unsatisfying. Instead of explaining the pathology solely by classical virulence mechanisms such as toxin production or bacterial invasion, it seems that human diarrhoea is the result of the wrong bacteria in the wrong host, resulting in an excessive and damaging immune response. If this model proves to be correct, most Campylobacter strains might have the potential to cause disease, irrespective of their genetic repertoire, because the innate immune system recognizes conserved patterns, rather than specific antigens. Whether exposure results in disease and how disease outcome varies would be (in part or at large) determined by the intestinal innate immune response of the individual. This response would be dictated by previous (dose- and frequencydependent) exposure, as well as the received dose. This would obviously have consequences for risk assessment models.

 

Most cases of campylobacteriosis are self-limiting and do not require antibiotic medication, though complications such as bacteraemia may require antibiotic intervention. (Postinfectious sequels such as GBS cannot be prevented by antibiotic treatment.) The increase in fluoroquinolone resistance in the past two decades has been a reason of concern, as these antibiotics are frequently used for empiric treatment of undefined gastroenteritis cases. Fluoroquinolones were also used in the poultry industry in significant quantities, which has contributed to the observed increase in resistance incidence. The use of enrofloxacin has subsequently been banned in poultry production in the United States, and growth-promoting use of antibiotics is discouraged or banned in many countries. Because of the increased prevalence of fluoroquinolone resistance, the first choice of treatment for diagnosed campylobacteriosis is erythromycin (Allos 2001), a macrolide to which Camp. jejuni remains mostly susceptible at present. An increase in the prevalence of macrolide resistance in Camp. coli, especially in isolates from tylosin-treated pigs, is of concern. The effect of fluoroquinolone treatment during an acute human infection is limited, even when the bacteria causing the infection are susceptible (Wassenaar et al. 2007). An explanation may be the microheterogeneity discussed above that results from mutations introduced during replication. A relatively frequent single-nucleotide mutation in the gyrase gene of Campylobacter is sufficient to result in resistance. This mutation can occur at any time in a growing population, but when that population is exposed to fluoroquinolone, only the mutants survive and the population shifts from susceptible to resistant. This can happen in humans and animals alike.

 

This review started with bacteria colonizing chickens on a farm, and these were followed on their way, using meat as the vehicle, to reach a person who subsequently fell ill. These are only a minute fraction of all the bacteria that had multiplied in the birds. The vast majority of the cells living in the chicken flock will be secreted in faeces and die, while some may make it to a next animal host in which they can multiply.

 

One of the enigmas of Campylobacter is its wide host range, despite its small genome size and, related to this, its limited genetic repertoire. How do the bacteria manage to adapt to a novel host, which frequently will be a different animal species compared to the one they have just left? Classical genetic selection would predict that a growing population adapts its gene expression to the host they colonize, but how do the cells adjust quickly to a novel host, frequently having to switch between avian and mammalian hosts? Possibly, the microheterogeneity attributable to polymorphic genes is a key to this adaptation. When a strain contains ten genes with variable G-stretches (some strains contain over 20 of these variable genes) that independently either allow or abort expression, this would result in 1024 (210) possible combinations, with maybe as many possible phenotypes as the outcome. Such microheterogeneity may prepare a bacterial population for the variety of challenges they encounter before they can colonize a next host. If this would indeed be the function of these polymorphic genes, it would be advantageous for a growing population to increase heterogeneity, instead of Darwinistic selection of a particular suitable phenotype that is most adequately equipped for a given host. It seems that the population acts as a kind of ‘super-organism’ where there is a cooperative benefit of introducing genetic heterogeneity. Indeed, recently it was shown that a population multiplying in a chicken increases microheterogeneity owing to contingency gene variation (Wilson et al. 2010). The adaptation to murine colonization also resulted in shifts of the lengths of G- or C-stretches and did not select for a homogeneous population (Jerome et al. 2011).

 

The numbers of Campylobacter produced by chickens are enormous, and as chicken production has increased worldwide over the past decades, so has the production of Campylobacter populations. These bacteria may enter the environment when waste from poultry production and processing is disposed. Feathers, slaughter waste and faeces can all be highly contaminated with Campylobacter. The cells may not survive for long outside a host, as they are sensitive to oxygen, UV and desiccation, but only a few cells need to make it to a next animal host. This may pose indirect routes by which humans can eventually get contaminated from Campylobacter strains that originally multiplied in farmed chickens, through routes independent of the food chain. Very little research has been carried out to determine the fate of bacteria from waste products of the poultry industry. It may, however, explain why there is a broader overlap in strain identity between humans and chickens (estimated between 40 and 78%, Sheppard et al. 2009) than the (up to) 40% of cases that can be attributed to consumption of poultry meat (Vellinga and van Loock 2002; Friedman et al., 2004).

 

There are far more Campylobacter cells living in chickens, at any given time, than in diseased humans. This has consequences to the selective pressures to which the bacteria have to respond. Although we regard Campylobacter as a pathogen (which it is, to humans), the species has a commensal lifestyle in birds and may colonize a wide variety of mammals asymptomatically. This is where the population that we are aware of multiplies most of the time. It means that the bacteria are optimally equipped to live in an avian or mammalian intestine without causing disease. There is little selective pressure to evolve virulence genes in such a setting, and humans are an insignificant host to maintain a viable population of Campylobacter on a global scale. As pointed out in the previous section, the disease that results from colonization in humans is probably the result of an unfortunate host–microbe combination and can be considered as collateral damage. This view does not make the disease less severe, but it changes the perspective on the selective pressures that shape the bacterial population.

 

In the near future, Campylobacter will remain to cause disease in humans, and a significant proportion of these cases will remain to be, directly or indirectly, attributable to poultry production. To reduce this proportion, efforts must be taken by the industry to apply the valuable insights that research have provided. A number of critical control points are known and can be acted on, while knowledge gaps that still exist have been identified and need to be further explored. It is time for a joined action to try and reduce the Campylobacter numbers in chickens, to improve the safety of poultry consumption worldwide. Increased awareness is needed that bacteria are constantly being released in the environment via waste products of farmed animals, especially poultry, and these bacteria may reach the human host via unknown routes. This provides further challenges for control and mitigation. From novel insights into the pathogenicity of this organism, it might be proposed that all strains have the potential to cause disease, but the individual host determines the outcome of infection. This has consequences to the risk models currently being used. Finally, even clonal offspring of a Campylobacter cell is not completely homogeneous, and genetic variation within such populations should be studied in more detail to establish their potential role in the life cycle of this organism.

 

1. Abuoun, M., Manning, G., Cawthraw, S.A., Ridley, A., Ahmed, I.H., Wassenaar, T.M. and Newell, D.G. (2005) Cytolethal distending toxin (CDT)-negative Campylobacter jejuni strains and anti-CDT neutralizing antibodies are induced during human infection but not during colonization in chickens. Infect Immun 73, 3053– 3062.

2. Allen, V.M., Weaver, H., Ridley, A.M., Harris, J.A., Sharma, M., Emery, J., Sparks, N., Lewis, M. et al. (2008) Sources and spread of thermophilic Campylobacter spp. during partial depopulation of broiler chicken flocks. J Food Prot 71, 264–270.

3. Allerberger, F., Al-Jazrawi, N., Kreidl, P., Dierich, M.P., Feierl, G., Hein, I. and Wagner, M. (2003) Barbecued chicken causing a multi-state outbreak of Campylobacter jejuni enteritis. Infection 31, 19–23.

4. Allos, B.M. (2001) Campylobacter jejuni infections: update on emerging issues and trends. Clin Infect Dis 32, 1201–1206.

5. Alter, T., Weber, R.M., Hamedy, A. and Glu¨nder, G. (2011) Carry-over of thermophilic Campylobacter spp. between sequential and adjacent poultry flocks. Vet Microbiol 147, 90–95. 

6. Anon (2010) Preliminary FoodNet data on the incidence of infection with pathogens transmitted commonly through food – 10 states, 2009. Morb Mortal Wkly Rep 59, 418–422.

7. Bereswill, S., Fischer, A., Plickert, R., Haag, L.M., Otto, B., Ku¨hl, A.A., Dashti, J.I., Zautner, A.E. et al. (2011) Novel murine infection models provide deep insights into the ‘‘me´nage a` trois’’ of Campylobacter jejuni, microbiota and host innate immunity. PLoS ONE 6, e20953.

8. Bokkers, E.A. and de Boer, I.J. (2009) Economic, ecological, and social performance of conventional and organic broiler production in the Netherlands. Br Poult Sci 50, 546– 557.

9. Boysen, L. and Rosenquist, H. (2009) Reduction of thermotolerant Campylobacter species on broiler carcasses following physical decontamination at slaughter. J Food Prot 72, 497–502.

10. Boysen, L., Knøchel, S. and Rosenquist, H. (2007) Survival of Campylobacter jejuni in different gas mixtures. FEMS Microbiol Lett 266, 152–157.

11. Bronzwaer, S., Hugas, M., Collins, J.D., Newell, D.G., Robinson, T., Ma¨kela¨, P. and Havelaar, A. (2009) EFSA’s 12th Scientific Colloquium – assessing health benefits of controlling Campylobacter in the food chain. Int J Food Microbiol 131, 284–285.

12. Buckley, A.M., Wang, J., Hudson, D.L., Grant, A.J., Jones, M.A., Maskell, D.J. and Stevens, M.P. (2010) Evaluation of live-attenuated Salmonella vaccines expressing Campylobacter antigens for control of C. jejuni in poultry. Vaccine 28, 1094–1105.

13. Callicott, K.A., Harg?ardo´ ttir, H., Georgsson, F., Reiersen, J., Frig?riksdo´ ttir, V., Gunnarsson, E., Michel, P., Bisaillon, J.R. et al. (2008) Broiler Campylobacter contamination and human campylobacteriosis in Iceland. Appl Environ Microbiol 74, 6483–6494.

14. Cawthraw, S.A. and Newell, D.G. (2010) Investigation of the presence and protective effects of maternal antibodies against Campylobacter jejuni in chickens. Avian Dis 54, 86–93.

15. Conlan, A.J., Line, J.E., Hiett, K., Coward, C., Van Diemen, P.M., Stevens, M.P., Jones, M.A., Gog, J.R. et al. (2011) Transmission and dose-response experiments for social animals: a reappraisal of the colonization biology of Campylobacter jejuni in chickens. J R Soc Interface May 18. [Epub ahead of print].

16. Connerton, P.L., Timms, A.R. and Connerton, I.F. (2011) Campylobacter bacteriophages and bacteriophage therapy. J Appl Microbiol doi: 10.1111/j.1365-2672.2011.05012.x. [Epub ahead of print].

17. Cox, J.M. and Pavic, A. (2010) Advances in enteropathogen control in poultry production. J Appl Microbiol 108, 745– 755.

18. Cox, N.A., Richardson, L.J., Buhr, R.J. and Fedorka-Cray, P.J. (2009) Campylobacter species occurrence within internal organs and tissues of commercial caged Leghorn laying hens. Poult Sci 88, 2449–2456.

19. Dasti, J.I., Tareen, A.M., Lugert, R., Zautner, A.E. and Gross, U. (2009) Campylobacter jejuni: a brief overview on pathogenicity- associated factors and disease-mediating mechanisms. Int J Med Microbiol 300, 205–211.

20. Dipineto, L., Gargiulo, A., Russo, T.P., de Luca Bossa, L.M., Borrelli, L., Menna, L.F. and Fioretti, A. (2011) Campylobacter jejuni, Campylobacter coli, and cytolethal distending toxin genes in laying hens. Avian Dis 55, 103–105.

21. Ellerbroek, L.I., Lienau, J.A. and Klein, G. (2010) Campylobacter spp. in broiler flocks at farm level and the potential for cross-contamination during slaughter. Zoonoses Public Health 57, e81–e88.

22. Fearnley, C., Manning, G., Bagnall, M., Javed, M.A., Wassenaar, T.M. and Newell, D.G. (2008) Identification of hyperinvasive Campylobacter jejuni strains isolated from poultry and human clinical sources. J Med Microbiol 57, 570–580.

23. Forbes, K.J., Gormley, F.J., Dallas, J.F., Labovitiadi, O., Mac- Rae, M., Owen, R.J., Richardson, J., Strachan, N.J. et al. (2009) Campylobacter immunity and coinfection following a large outbreak in a farming community. J Clin Microbiol 47, 111–116.

24. Friedman, C.R., Hoekstra, R.M., Samuel, M., Marcus, R., Bender, J., Shiferaw, B., Reddy, S., Ahuja, S.D. et al. (2004) Risk factors for sporadic Campylobacter infection in the United States: a case–control study in FoodNet sites. Clin Infect Dis 38(Suppl. 3), S285–S296.

25. Gellynck, X., Messens, W., Halet, D., Grijspeerdt, K., Hartnett, E. and Viaene, J. (2008) Economics of reducing Campylobacter at different levels within the Belgian poultry meat chain. J Food Prot 71, 479–485.

26. Van Gerwe, T., Bouma, A., Wagenaar, J.A., Jacobs-Reitsma, W.F. and Stegeman, A. (2010) Comparison of Campylobacter levels in crops and ceca of broilers at slaughter. Avian Dis 54, 1072–1074.

27. Guerin, M.T., Sir, C., Sargeant, J.M., Waddell, L., O’Connor, A.M., Wills, R.W., Bailey, R.H. and Byrd, J.A. (2010) The change in prevalence of Campylobacter on chicken carcasses during processing: a systematic review. Poult Sci 89, 1070–1084.

28. Hald, B., Sommer, H.M. and Skovga°rd, H. (2007) Use of fly screens to reduce Campylobacter spp. introduction in broiler houses. Emerg Infect Dis 13, 1951–1953.

29. Hansson, I., Ederoth, M., Andersson, L., Va°gsholm, I. and Olsson Engvall, E. (2005) Transmission of Campylobacter spp. to chickens during transport to slaughter. J Appl Microbiol 99, 1149–1157.

30. Hansson, I., Pudas, N., Harbom, B. and Olsson Engvall, E. (2010) Within-flock variations of Campylobacter loads in caeca and on carcasses from broilers. Int J Food Microbiol 141, 51–55.

31. Haughton, P.N., Lyng, J.G., Morgan, D.J., Cronin, D.A., Fanning, S. and Whyte, P. (2011) Efficacy of high-intensity pulsed light for the microbiological decontamination of chicken, associated packaging, and contact surfaces. Foodborne Pathog Dis 8, 109–117.

32. Havelaar, A.H., van Pelt, W., Ang, C.W., Wagenaar, J.A., van Putten, J.P., Gross, U. and Newell, D.G. (2009) Immunity to Campylobacter: its role in risk assessment and epidemiology. Crit Rev Microbiol 35, 1–22.

33. Hickey, T.E., McVeigh, A.L., Scott, D.A., Michielutti, R.E., Bixby, A., Carroll, S.A., Bourgeois, A.L. and Guerry, P. (2000) Campylobacter jejuni cytolethal distending toxin mediates release of interleukin-8 from intestinal epithelial cells. Infect Immun 68, 6535–6541.

34. Inns, T., Foster, K. and Gorton, R. (2010) Cohort study of a campylobacteriosis outbreak associated with chicken liver parfait, United Kingdom, June 2010. Euro Surveill 15, pii: 19704.

35. Jerome, J.P., Bell, J.A., Plovanich-Jones, A.E., Barrick, J.E., Brown, C.T. and Mansfield, L.S. (2011) Standing genetic variation in contingency loci drives the rapid adaptation of Campylobacter jejuni to a novel host. PLoS ONE 6, e16399.

36. Johnsen, G., Kruse, H. and Hofshagen, M. (2007) Genotyping of thermotolerant Campylobacter from poultry slaughterhouse by amplified fragment length polymorphism. J Appl Microbiol 103, 271–279.

37. Jore, S., Viljugrein, H., Brun, E., Heier, B.T., Borck, B., Ethelberg, S., Hakkinen, M., Kuusi, M. et al. (2010) Trends in Campylobacter incidence in broilers and humans in six European countries, 1997–2007. Prev Vet Med 93, 33–41.

38. Law, B.F., Adriance, S.M. and Joens, L.A. (2009) Comparison of in vitro virulence factors of Campylobacter jejuni to in vivo lesion production. Foodborne Pathog Dis 6, 377–385.

39. Layton, S.L., Morgan, M.J., Cole, K., Kwon, Y.M., Donoghue, D.J., Hargis, B.M. and Pumford, N.R. (2011) Evaluation of Salmonella-vectored Campylobacter peptide epitopes for reduction of Campylobacter jejuni in broiler chickens. Clin Vaccine Immunol 18, 449–454.

40. Lin, J. (2009) Novel approaches for Campylobacter control in poultry. Foodborne Pathog Dis 6, 755–765.

41. Lippert, E., Karrasch, T., Sun, X., Allard, B., Herfarth, H.H., Threadgill, D. and Jobin, C. (2009) Gnotobiotic IL-10; NF-kappaB mice develop rapid and severe colitis following Campylobacter jejuni infection. PLoS ONE 4, e7413.

42. Louwen, R.P., van Belkum, A., Wagenaar, J.A., Doorduyn, Y., Achterberg, R. and Endtz, H.P. (2006) Lack of association between the presence of the pVir plasmid and bloody diarrhea in Campylobacter jejuni enteritis. J Clin Microbiol 44, 1867–1868.

43. Luber, P. (2009) Cross-contamination versus undercooking of poultry meat or eggs – which risks need to be managed first? Int J Food Microbiol 134, 21–28.

44. Luber, P. and Bartelt, E. (2007) Enumeration of Campylobacter spp. on the surface and within chicken breast fillets. J Appl Microbiol 102, 313–318.

45. Mangen, M.J., Havelaar, A.H., Poppe, K.P. and de Wit, G.A. and CARMA Project Team. (2007) Cost-utility analysis to control Campylobacter on chicken meat: dealing with data limitations. Risk Anal 27, 815–830.

46. Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M. and Tauxe, R.V. (1999) Foodrelated illness and death in the United States. Emerg InfectDis 5, 607–625.

47. Morishita, T.Y., Aye, P.P., Harr, B.S., Cobb, C.W., Clifford, J.R., Na¨ther, G., Alter, T. and Martin, A. (1997) Evaluation of an avian-specific probiotic to reduce the colonization and shedding of Campylobacter jejuni in broilers. Avian Dis 41, 850–855.

48. Na¨ther, G., Alter, T., Martin, A. and Ellerbroek, L. (2009) Analysis of risk factors for Campylobacter species infection in broiler flocks. Poult Sci 88, 1299–1305.

49. Nauta, M., Hill, A., Rosenquist, H., Brynestad, S., Fetsch, A., van der Logt, P., Fazil, A., Christensen, B. et al. (2009a) A comparison of risk assessments on Campylobacter in broiler meat. Int J Food Microbiol 129, 107–123.

50. Nauta, M.J., van der Wal, F.J., Putirulan, F.F., Post, J., van de Kassteele, J. and Bolder, N.M. (2009b) Evaluation of the ‘‘testing and scheduling’’ strategy for control of Campylobacter in broiler meat in The Netherlands. Int J Food Microbiol 134, 216–222.

51. Newell, D.G. and Fearnley, C. (2003) Sources of Campylobacter colonization in broiler chickens. Appl Environ Microbiol 69, 4343–4351.

52. Newell, D.G., Shreeve, J.E., Toszeghy, M., Domingue, G., Bull, S., Humphrey, T. and Mead, G. (2001) Changes in the carriage of Campylobacter strains by poultry carcasses during processing in abattoirs. Appl Environ Microbiol 67, 2636– 2640.

53. Patrick, M.E., Mahon, B.E., Zansky, S.M., Hurd, S. and Scallan, E. (2010) Riding in shopping carts and exposure to raw meat and poultry products: prevalence of, and factors associated with, this risk factor for salmonella and campylobacter infection in children younger than 3 years. J Food Prot 73, 1097–1100.

54. Peyrat, M.B., Soumet, C., Maris, P. and Sanders, P. (2008) Recovery of Campylobacter jejuni from surfaces of poultry slaughterhouses after cleaning and disinfection procedures: analysis of a potential source of carcass contamination. Int J Food Microbiol 124, 188–194.

55. Van Putten, J.P., van Alphen, L.B., Wo¨sten, M.M. and de Zoete, M.R. (2009) Molecular mechanisms of Campylobacter infection. Curr Top Microbiol Immunol 337, 197–229.

56. Rajkovic, A., Tomic, N., Smigic, N., Uyttendaele, M., Ragaert, P. and Devlieghere, F. (2010) Survival of Campylobacter jejuni on raw chicken legs packed in high-oxygen or highcarbon dioxide atmosphere after the decontamination with lactic acid ⁄ sodium lactate buffer. Int J Food Microbiol 140, 201–206.

57. Ringoir, D.D., Szylo, D. and Korolik, V. (2007) Comparison of 2-day-old and 14-day-old chicken colonization models for Campylobacter jejuni. FEMS Immunol Med Microbiol 49, 155–158.

58. Rosenquist, H., Nielsen, N.L., Sommer, H.M., Nørrun, B. and Christensen, B.B. (2003) Quantitative risk assessment of human campylobacteriosis associated with thermophilic Campylobacter species in chicken. Int J Food Microbiol 83, 87–103.

59. Sampers, I., Habib, I., Berkvens, D., Dumoulin, A., Zutter, L.D. and Uyttendaele, M. (2008) Processing practices contributing to Campylobacter contamination in Belgian chicken meat preparations. Int J Food Microbiol 128, 297– 303.

60. Santini, C., Baffoni, L., Gaggia, F., Granata, M., Gasbarri, R., Di Gioia, D. and Biavati, B. (2010) Characterization of probiotic strains: an application as feed additives in poultry against Campylobacter jejuni. Int J Food Microbiol 141, S98–S108.

61. Sheppard, S.K., Dallas, J.F., Strachan, N.J., MacRae, M., McCarthy, N.D., Wilson, D.J., Gormley, F.J., Falush, D. et al. (2009) Campylobacter genotyping to determine the source of human infection. Clin Infect Dis 48, 1072–1078.

62. Smith, J.L. and Bayles, D.O. (2006) The contribution of cytolethal distending toxin to bacterial pathogenesis. Crit Rev Microbiol 32, 227–248.

63. Svetoch, E.A. and Stern, N.J. (2010) Bacteriocins to control Campylobacter spp. in poultry – a review. Poult Sci 89, 1763–1768.

64. Tribble, D.R., Baqar, S., Scott, D.A., Oplinger, M.L., Trespalacios, F., Rollins, D., Walker, R.I., Clements, J.D. et al. (2010) Assessment of the duration of protection in Campylobacter jejuni experimental infection in humans. Infect Immun 78, 1750–1759.

65. Vellinga, A. and van Loock, F. (2002) The dioxin crisis as experiment to determine poultry-related Campylobacter enteritis. Emerg Infect Dis 8, 19–22.

66. Wagenaar, J.A., Van Bergen, M.A., Mueller, M.A., Wassenaar, T.M. and Carlton, R.M. (2005) Phage therapy reduces Campylobacter jejuni colonization in broilers. Vet Microbiol 109, 275–283.

67. Wagenaar, J.A., Mevius, D.J. and Havelaar, A.H. (2006) Campylobacter in primary animal production and control strategies to reduce the burden of human campylobacteriosis. Rev Sci Tech 25, 581–594.

68. Wassenaar, T.M. and Gaastra, W. (2001) Bacterial virulence: can we draw the line? FEMS Microbiol Lett 201, 1–7.

69. Wassenaar, T.M., Wagenaar, J.A., Rigter, A., Fearnley, C., Newell, D.G. and Duim, B. (2002) Homonucleotide stretches in chromosomal DNA of Campylobacter jejuni display high frequency polymorphism as detected by direct PCR analysis. FEMS Microbiol Lett 212, 77–85.

70. Wassenaar, T.M., Kist, M. and de Jong, A. (2007) Re-analysis of the risks attributed to ciprofloxacin-resistant Campylobacter jejuni infections. Int J Antimicrob Agents 30, 195– 201.

71. Wegener, H.C. (2010) Danish initiatives to improve the safety of meat products. Meat Sci 84, 276–283.

72. Whyte, R., Hudson, J.A. and Graham, C. (2006) Campylobacter in chicken livers and their destruction by pan frying. Lett Appl Microbiol 43, 591–595.

73. Wilson, D.J., Gabriel, E., Leatherbarrow, A.J., Cheesbrough, J., Gee, S., Bolton, E., Fox, A., Hart, C.A. et al. (2009) Rapid evolution and the importance of recombination to the gastroenteric pathogen Campylobacter jejuni. Mol Biol Evol 26, 385–397.

74. Wilson, D.L., Rathinam, V.A., Qi, W., Wick, L.M., Landgraf, J., Bell, J.A., Plovanich-Jones, A., Parrish, J. et al. (2010) Genetic diversity in Campylobacter jejuni is associated with differential colonization of broiler chickens and C57BL ⁄ 6J IL10-deficient mice. Microbiology 156, 2046–2057.

75. Wyszyn´ ska, A., Raczko, A., Lis, M. and Jagusztyn-Krynicka, E.K. (2004) Oral immunization of chickens with avirulent Salmonella vaccine strain carrying C. jejuni 72Dz ⁄ 92 cjaA gene elicits specific humoral immune response associated with protection against challenge with wild-type Campylobacter. Vaccine 22, 1379–1389.

76. Zilbauer, M., Dorrell, N., Wren, B.W. and Bajaj-Elliott, M. (2008) Campylobacter jejuni-mediated disease pathogenesis: an update. Trans R Soc Trop Med Hyg 102, 123–129.

77. de Zoete, M.R., van Putten, J.P. and Wagenaar, J.A. (2007) Vaccination of chickens against Campylobacter. Vaccine 25, 5548–5557.

78. de Zoete, M.R., Keestra, A.M., Roszczenko, P. and van Putten, J.P. (2010) Activation of human and chicken toll-like receptors by Campylobacter spp. Infect Immun 78, 1229–1238.