Campylobacter hepaticus and Spotty Liver Disease in Poultry Open Access

This review summarizes the major research findings within the field of spotty liver disease (SLD) research over the last 5 yr. The topics covered include the cost of SLD, the epidemiology of SLD, intervention strategies to combat SLD, and insights into SLD pathogenesis including the molecular biology of the pathogen. The review does not aim to give a detailed description of the etiology or history of SLD because other articles and reviews have already done so in more than adequate detail (1,2,3,4).

Briefly, SLD is a bacterial disease of poultry that primarily affects cage-free reared layer chickens (5,6), often at or around the time of peak lay, although not exclusively (7); it is characterized by the emergence of white necrotic spots on the surface of the liver, decreased egg production, and increased mortality (8,9,10). The disease itself is reported as being of serious economic concern to the layer industry in affected countries (10,11,12) and has only emerged as such in recent decades as the industry has shifted away from cage production.

The causative agent of SLD was first definitively identified in 2015–16 as Campylobacter hepaticus (9,13), and then a second agent, Campylobacter bilis, was discovered in 2022–23 (14,15). While described as an emerging pathogen in most jurisdictions around the world, it is routinely described in Australia as the most serious or most prevalent bacteriologic threat to the layer industry (8,10,16), and so in the Australian context, the term “emerging” is, perhaps, no longer appropriate, as its presence and impact on birds and the industry are well established. Historically, the bulk of research into SLD has come from Australia, but over the last 5 yr, there has been a surge in research globally, but especially from the United States (17,18,19,20,21,22,23), and there are increasing numbers of reports from other regions around the world, including New Zealand (24), Costa Rica (25), Germany (26), and Jordan (27).

Given that the causative agents of SLD were only relatively recently identified, many questions remain to be addressed around both epidemiologic factors involved in SLD and the more fundamental nature of the virulence mechanisms by which the pathogen generates disease. Recent studies have gone some way in shedding light on these matters or, at the very least, have helped the field develop the tools necessary to pursue relevant investigations.

SLD is a serious economic burden on the layer industry. In Australia, it has been described as the most significant bacteriologic threat to that nation’s egg industry (1,5). More recently, many European countries, such as Germany, France, and the Netherlands, have begun to recognize SLD as having a serious economic impact on affected farms. It would be of interest to estimate the potential global economic impact of SLD to start to understand the industry-wide consequences of the disease. Economic modelling is difficult, due in part to the paucity of global research on SLD. Despite the increasing number of reports of C. hepaticus isolation and characterization from various countries and some studies of disease outbreaks in individual farms or groups of farms, there have been no published reports of the amount of SLD seen throughout any countries other than Australia (10). What can be asserted with confidence is that SLD outbreaks are mainly associated with cage-free production systems (1,28), and so localities that still primarily have caged egg production (such as China, the largest producer of eggs in the world) may, at this time, be minimally impacted by SLD.

In 2020, Courtice et al. reported some cost estimates for the impact of SLD in typical Australian flocks (16). They analyzed 55 disease outbreaks and calculated a cost of between AUD$0.50 and AUD$4.29 (US $0.32 to $2.71) per bird in affected flocks. To be able to extrapolate from these figures to an industry-wide cost, the rate of SLD incidence across the whole industry needs to be known. Due to a lack of published data on rates of SLD occurrence, we can only estimate the occurrence rate based on anecdotal reports from workers with deep industry experience. Thus, from multiple sources (producers, vets, industry bodies), it is suggested that within the Australian layer industry, at least 50% of all cage-free layer flocks are likely to be impacted by at least one SLD outbreak during their production cycle. Using this information from the Australian layer industry, an attempt can be made to estimate the cost of SLD in other areas of the globe. Using data from the Food and Agriculture Organization (29), where available, and only including production from Organisation for Economic Co-operation and Development nations with recent reported SLD outbreaks (either within the literature or directly reported to the authors via industry sources), as they will share somewhat similar animal husbandry practices, we estimate that within Australia, New Zealand, the United States, United Kingdom, Germany, Austria, Spain, and Italy, there is a combined layer flock of over 850 million birds per annum, of which more than 350 million (for sources, see Table 1) are raised in cage-free production systems. If an occurrence rate of 25% of cage-free flocks being affected is assumed at a median cost of AUD$2 (US $1.26) per bird, then a conservative estimate of the cost of SLD in these countries would be over AUD$180 million (US $113.5 million) per year. With the application of less conservative values, i.e., assuming similar occurrence rates as in Australia and the higher estimate of per-bird costs, the cost of SLD within the above countries could be as high as AUD$775 million per year (US$489 million). The economic impact is comprised not only of the loss of saleable eggs, but also includes mortality, treatment costs, and improvement of hygiene measures. This figure is only likely to grow as regulatory requirements and consumer sentiment drive demand for cage-free eggs, and this largely emerging disease becomes more prevalent and widespread. A breakdown of the predicted cost per region using these assumptions can be found in Table 1.

The isolation and identification of the causative agent of SLD in 2015–16 (9,13) represented an important and foundational breakthrough in the study of SLD. So, it was with some excitement, and perhaps surprise, when reports emerged in 2022–23 of the identification of a new bacterium, first associated with, and then shown to be a causative agent, of SLD.

In 2022, Phung et al. (39), while attempting to obtain further C. hepaticus isolates from outbreaks of SLD in commercial layer birds from Australia, found that six of their isolates did not present typical features of C. hepaticus, despite also being Gram-negative vibrio-shaped cells. These samples were obtained from the bile of birds clinically confirmed to be suffering from SLD. Multiple analyses were applied, including investigation into the phylogeny of key genetic features used to differentiate bacterial species (16S RNA and hsp60 genes), whole-genome sequencing followed by multiple comparative analyses, matrix-assisted laser desorption/ionization–time of flight analysis, and a suite of biochemical assays used to establish criteria for Campylobacteraceae species differentiation (40). It was found that the novel isolate VicNov18 (which would go on to become the type strain for the species) and a small group of phylogenetically related isolates formed an independent clade, distinct from, but closely related to, C. hepaticus, and this new clade was named Campylobacter bilis. Biochemically, C. bilis could be differentiated from C. hepaticus by its reduced ability to hydrolyze hippurate, reduced ability to reduce nitrate, increased growth on MacConkey agar, reduced growth in the presence of 2,3,5-triphenyltetrazolium chloride and metronidazole, increased growth in the presence of cefoperazone, and reduced resistance to cephalothin and nalidixic acid. The disease induction model previously developed for C. hepaticus was applied to C. bilis, and it was definitively demonstrated that C. bilis could induce liver lesions that were indistinguishable from those induced by C. hepaticus (15). The ability of more than one bacterial species to cause very similar pathology and disease is not unusual; even within the campylobacters, human diarrheal disease can be caused by a variety of Campylobacter species, predominantly Campylobacter jejuni and Campylobacter coli, but also others including Campylobacter lari and Campylobacter upsaliensis (41).

A C. bilis–specific PCR method for rapid identification and detection of C. bilis DNA has been developed (15). A diagnostic PCR method that was already established for C. hepaticus (42) was found to detect both C. hepaticus and C. bilis, but it could not differentiate the two. The glycerol kinase gene, which was used as the PCR target, is highly conserved between C. hepaticus and C. bilis, but it is less conserved in other bacterial species, and thus it provided a diagnostic target. The variation in sequence between C. hepaticus and C. bilis glycerol kinase genes was leveraged to design primers that were specific for C. bilis.

Recently, genomic sequences of C. bilis have been submitted to the GenBank database (BioProject accession no.: PRJNA1138730), indicating that this second causative agent of SLD has been identified in the United States as well as Australia.

The lack of tools for the rapid and accurate detection and isolation of C. hepaticus and C. bilis from field samples represents an important hindrance for its further study and, more importantly, long-term amelioration of disease. A 2021 study by Young et al. developed a high-resolution melt (HRM) PCR assay for the detection and differentiation of C. hepaticus from C. jejuni and C. coli (43). The HRM assay targeted the gene hsp60, which is widely conserved. The assay proved to be sensitive to an initial amount of 10⁻³ ng of C. hepaticus genomic DNA and capable of differentiating among C. hepaticus, C. jejuni, and C. coli. The main advantage of this technique is the rapidity of results when compared to end-point PCRs established for C. hepaticus detection by Van et al. in 2017 and Courtice et al. in 2023 (11,42). However, it is worth noting that this study was undertaken prior to the identification and reporting of C. bilis, so it is unclear how effective this assay would be at the detection and differentiation of this second causative agent.

Currently, unlike that used for C. jejuni, there is no established selective medium for C. hepaticus isolation. This lack impedes the rapid isolation of C. hepaticus from infected flocks, necessitating isolation from biologic fluids such as the bile, which requires euthanatization of the animal in question. The establishment of a C. hepaticus selective medium would be of value to the SLD field for the purpose of rapid isolation from heterologous samples (i.e., cloacal swabs, feces, environmental, etc.). Phung et al. (2020) successfully used a motile-filter method that relies on the ability of motile bacteria to penetrate a 0.65-μm filter, which can be used to isolate C. hepaticus from complex microbial samples such as cecal content (44). The method worked well on fresh samples when C. hepaticus was not overwhelmed by other motile and faster-growing bacteria. Currently, this physical barrier method is the most effective means by which to isolate C. hepaticus from complex microbiologic samples. Sakur et al. (2024) attempted to devise a selective medium for C. hepaticus (45). They tested the effect of multiple parameters, including media type, the absence or presence of defibrinated horse blood and bile acids, and the presence of 14 different antibiotics, on the growth of five different C. hepaticus isolates. They found that the addition of bile acids generally decreased growth of C. hepaticus, and the presence of defibrinated horse blood generally increased C. hepaticus growth. The conclusion was that Bolton plus blood media supplemented with vancomycin and trimethoprim represented the most appropriate medium for growth of C. hepaticus. However, the medium was of limited selectivity because it also supported the growth of C. jejuni. That study did not determine performance in primary isolation of C. hepaticus from infected birds or environmental samples, either from microbiologically simple samples such as bile or from complex samples such as cecal content. Specific growth media for C. hepaticus and C. bilis remain to be developed.

Much of the research into SLD has centered around the epidemiology of the disease as well as various attempts at mitigation interventions (see below for further detail). In contrast, research into the fundamental pathogenesis of SLD, such as understanding the molecular basis of disease or the behavior of the pathogen, has remained understudied. Work conducted over the last 5 yr has begun to shed light on the SLD pathogenesis process, and these discoveries may provide important contributions to future intervention strategies, i.e., vaccines.

The DNA of C. hepaticus can be detected in environmental samples such as dust and mud as well as from fauna such as rats, wild birds, and insects (11,46). These environmental reservoirs have been suggested as potential sources of introduction of the pathogen into commercial flocks, but live C. hepaticus bacteria have not been isolated from these sources. The lack of isolation from such samples may be due to the technical difficulties of C. hepaticus isolation without a specific selective medium but may also be because any C. hepaticus present is in a nonculturable state. Phung et al. (2022) found that the incubation of C. hepaticus in either isotonic Ringer’s solution or in sterile water resulted in a gradual reduction of normal culturable cells (47). The time it took for there to be no culturable cells left ranged between 5 and 65 days depending on the solution, strain used, and temperature at which the cells were incubated. Differential fluorescence microscopy showed that even in the absence of culturable bacteria, apparently intact cells were still present, and these had changed to a coccoid form from their usual spiral form. The cells could be resuscitated by the addition of a combination of supplements (Vitox and FBP [both Oxoid™ products], and additional L-cysteine) to media routinely used in the growth of fastidious microorganisms, thus demonstrating that the stressed cultures contained viable but nonculturable (VBNC) cells. Finally, they demonstrated that VBNC C. hepaticus cells spiked into chicken feces could subsequently be reisolated and resuscitated using the above media formulation, demonstrating its utility for resuscitation of VBNC from microbially complex samples. The study reinforces the potential importance of environmental transfer of C. hepaticus in SLD even in the absence of culturable cells, and it is suggested that VBNC cells in the environment, but particularly those in standing water in the production system, are an important reservoir for infection, as is suggested in the case of C. jejuni in poultry (48).

Many species of Campylobacter undertake posttranslational modifications of a subset of their proteins via the addition of an N-linked heptasaccharide (49). The cellular machinery for this glycosylation process is encoded by a single protein glycosylation (pgl) gene locus generally containing 10 genes (50). In C. jejuni, this protein glycosylation plays a role in epithelial cell adherence and invasion (51), with strains lacking protein glycosylation greatly impaired in their ability to colonize chickens (52). Despite the potential importance of protein glycosylation, it was, until recently, unclear if C. hepaticus encoded a functional pgl locus, or if it performed protein glycosylation like many other Campylobacter species. Analysis has now shown that C. hepaticus does encode a pgl locus (53). The synteny of the genes in the locus was highly conserved across numerous Campylobacter species, and the amino acid sequence of the predicted proteins of the locus in C. hepaticus shared between 75.8% and 88.8% identity with their homologs in C. jejuni. Soybean agglutinin lectin blotting was used to demonstrate protein glycosylation and a 15-kDa lipooligosaccharide. Mass spectroscopy was then used to identify 35 glycosylated proteins and demonstrate that the heptasaccharide involved in C. hepaticus protein glycosylation was the same as the well-characterized heptasaccharide of C. jejuni. Further evidence that the C. hepaticus pgl locus behaves similarly to that of C. jejuni and other Campylobacter species was seen when a targeted pglB (a gene of central importance in the glycosylation process in most Campylobacter species) mutant was produced and shown to result in abrogated protein glycosylation in the strain (12).

Of the 35 glycoproteins identified in C. hepaticus, nine may be involved in niche adaptation and virulence, of which six have no homolog in C. jejuni (4,53). The identification of glycosylated proteins possibly involved in niche adaptation in C. hepaticus is important given the central role that glycosylation seems to play in host colonization of chickens for other campylobacters (54). It is interesting to speculate whether glycosylated proteins that are unique to C. hepaticus may have some involvement in determining the unusual traits that differentiate this pathogen from closely related pathogens such C. jejuni and C. coli, for example, the ready ability to colonize the harsh environment of the gall bladder. Identifying factors that play a role in adaptation to this environment would be of great interest to the field, as it may identify important virulence factors and possibly vaccine targets. One gene encoding a glycoprotein, a nitrate reductase, was found to be upregulated in the gall bladder environment (4), and it was speculated that this gene may play a role in adaptation of C. hepaticus to the low-oxygen environment of the gall bladder.

The genomes of C. hepaticus and C. bilis share an average nucleotide identity of 85.28%, and both have smaller genome sizes compared to other members of the Campylobacter genus. The genome sizes of C. hepaticus isolates range from 1.47 to 1.56 Mb (4,17,21,23,28,55), while C. bilis genome sizes range from 1.45 to 1.48 Mb (39). In contrast, C. jejuni genomes range from 1.62 to 1.80 Mb, and C. coli genome sizes range from 1.67 to 2.03 Mb (4). Campylobacter hepaticus has a reduced guanine-cytosine (GC) content, which is 2%–3.5% lower than that of C. jejuni and C. coli. The GC content of C. hepaticus strains ranges from 28.0% to 28.5%, while the GC content of C. bilis is 30.6%–30.8%, closer to that of C. jejuni and C. coli. Phylogenetically, C. hepaticus is most closely related to C. coli and C. jejuni (4,23,55). However, Ienes-Lima et al. found that Campylobacter fetus genomes (Campylobacter fetus subspecies fetus and Campylobacter fetus subspecies venerealis) are more closely related to C. hepaticus than C. coli and C. jejuni when phylogenetic analysis was performed based on concatenated single nucleotide polymorphism alignments (21). The authors also found that C. hepaticus genomes were divided into two distinct groups: the Australia and U.S. cluster and the U.K. cluster, and the type strain, C. bilis VicNov18^T, was identified as an intermediate group between them.

Campylobacter hepaticus contains more genes involved in carbohydrate utilization pathways than C. jejuni, which may help C. hepaticus survive in the carbohydrate-rich environment of the chicken liver (4,21). There is also a reduction in the number of genes associated with iron acquisition systems in C. hepaticus, while C. bilis genomes completely lack these systems (21,55). This characteristic likely aids in their survival and persistence in the chicken’s bile and liver niche.

Both C. hepaticus and C. bilis strains possess numerous genes associated with colonization and infection, such as those related to chemotaxis, motility, adherence, and antigen presentation (4,21). The cytolethal distending toxin (CdtA, CdtB, CdtC), a major virulence factor in C. jejuni (56), is absent in C. hepaticus genomes. Interestingly, the C. bilis type strain VicNov18^T and the currently available C. bilis genomes in the National Center for Biotechnology Information database are predicted to contain these cytolethal distending toxin subunits, which are most similar to those of Campylobacter volucris. Despite this, the toxin likely does not play a role in liver damage in SLD-infected birds, as it is absent in C. hepaticus genomes. No other candidate toxin genes have been identified through genomic analysis thus far. Van et al. (2019) observed that genes involved in glucose utilization, stress response, hydrogen metabolism, and sialic acid modification were upregulated in the bile environment (4). These genes may play a crucial role in the pathogenesis of C. hepaticus.

Although no C. bilis strains have been reported to harbor plasmids, plasmids containing tetracycline resistance genes were detected in C. hepaticus isolates from Australia and the U.K. (4,55). These tetracycline resistance–encoding plasmids are similar to C. jejuni pCJDM210L-like plasmids and C. coli pCC31-like plasmids, respectively. The presence of such antibiotic resistance plasmids raises concerns about potential horizontal gene transfer to human and veterinary pathogens, not only among Campylobacter species, but also to other pathogens, which could compromise the treatment of various diseases. Alternative treatments, other than antibiotics, are urgently needed for the control of SLD.

In recent years, the molecular tools available for SLD, C. hepaticus, and C. bilis study have proliferated. Important progress has been made in specific PCRs, enzyme-linked immunosorbent assays (ELISAs), and tools and methods for targeted genetic manipulation.

While there are several specific PCRs for detection of the pathogen’s genetic material, an alternative to the direct detection of the pathogen is the detection of anti–C. hepaticus/bilis circulating antibodies within poultry flocks. Whereas PCRs detect current infections, ELISAs can detect previous infection, which may have been cleared but that leave the evidence of chicken immune responses to the pathogen. A series of three studies by Muralidharan et al. (7,57,58) developed and used ELISAs for the detection of IgY antibody responses directed against C. hepaticus.

An initial ELISA method, known as SLD-ELISA1, used wells coated with total C. hepaticus protein to capture circulating anti–C. hepaticus antibodies from serum collected from a combination of SLD-free birds, naturally and experimentally C. hepaticus infected chickens, and birds vaccinated using formalin-killed C. hepaticus cells (57). As expected, because of high genetic homology, the assay was found to be highly cross-reactive with C. jejuni and C. coli circulating antibodies. This cross-reactivity could be ameliorated by pre-absorption of the serum against C. jejuni total protein extract and thus produce a signal that was mainly driven by anti–C. hepaticus antibodies. Because of the often high incidences of C. jejuni/C. coli colonization in commercial chickens, it is essential that any test to identify anti–C. hepaticus antibodies has a high degree of specificity. The SLD-ELISA1 was found to have a specificity of 95.5% and sensitivity of 97.6%.

Despite having established an assay capable of detecting and, to an extent, quantifying circulating anti–C. hepaticus antibodies in chickens, Muralidharan et al. recognized several limitations of the ELISA flowing from the use of total protein as the capture antigen, namely, that it was laborious to prepare, prone to batch to batch variation, and required pre-absorption of the sera. To address these limitations, they developed SLD-ELISA2, which utilized a recombinant fragment of an immunogenic C. hepaticus protein as the capture antigen (58). A combination of approaches, including immunoproteomic, bioinformatic, western blotting, and liquid chromatography–tandem mass spectrometry profiling, was used to identify a fragment of the C. hepaticus filamentous hemagglutinin adhesin protein (amino acids 1628–1899) as a suitable antigen. This fragment is highly conserved among C. hepaticus isolates and showed no evidence of cross-reactivity with C. coli or C. jejuni reactive sera. The resultant ELISA was found to have a specificity of 94.7% and sensitivity of 93.1% and could be adequately quality controlled and produced.

To demonstrate the applicability of SLD-ELISA2 in an industry context, Muralidharan et al. used SLD-ELISA2 to survey for the presence of C. hepaticus antibodies in 11 Australian free-range layer farms, both with and, importantly, without histories of SLD outbreaks (seven and five flocks, respectively) (7). These surveys were coupled with a PCR analysis of cloacal swabs for the presence of C. hepaticus/C. bilis genomic DNA. All seven farms with a history of SLD outbreaks had sera-positive birds with a prevalence ranging from 2% to 64% (average 29.6%) among tested samples, with five of the seven flocks producing positive PCR results for the presence of C. hepaticus/C. bilis genomic DNA. Perhaps unexpectedly, of the five farms with no history of SLD outbreaks, four had sera-positive samples ranging in prevalence, giving an average of 10.8% prevalence (although this was heavily biased due to a single flock with 41% sera-positivity). Three of the five farms, all of which had sera-positive samples, produced positive PCR results for the presence of C. hepaticus/C. bilis genomic DNA. It was found that the results of the PCR and SLD-ELISA2 strongly correlated.

Neither one of the immune assays (SLD-ELISA1, SLD-ELISA2) has been tested against the serum of C. bilis–infected animals, and we acknowledge that it is unclear how much cross-reactivity may occur between C. hepaticus– and C. bilis–positive serum samples. It is currently unclear how much of the burden of SLD in the field is caused by C. bilis, and so understanding if these ELISAs are useful assays with respect to C. bilis should be a priority of future research.

A major impediment to the study of SLD pathogenesis had been the lack of any genetic tools for the manipulation of C. hepaticus or C. bilis. The elucidation of virulence determinants for C. hepaticus has been greatly hindered by the inability to generate site-directed mutants of the pathogen in those genes speculated to be involved in disease pathogenesis or niche adaptation. The techniques for generating such mutants have been used in the past to investigate bacterial pathogens of chickens (54,59) and allow for the identification of novel virulence factors and careful dissection of the molecular basis of the pathogenesis process. Recently, tools and methods for the genetic manipulation of C. hepaticus have been developed (12). This represents a significant step forward for the study of the genetic basis of virulence and pathogenesis of the bacteria that cause SLD.

In 2024 (12), small C. coli cryptic plasmids were identified and cloned into an Escherichia coli plasmid, pMW2, which encodes a kanamycin resistance gene known to function in both E. coli and Campylobacter (60) to produce shuttle plasmids capable of replication in both E. coli and C. hepaticus. The new shuttle plasmids, pJBM2 and pJBM3, were used to refine a transformation method to maximize the efficiency of electroporation into C. hepaticus type strain HV10^T. It was then shown that all C. hepaticus strains tested could be transformed, although at widely varying frequencies. Importantly, the type strain of C. bilis could also be transformed.

The investigators then used a suicide plasmid vector for targeted mutagenesis via homologous recombination. The pglB gene, a gene essential for protein glycosylation, was deleted and replaced with an antibiotic resistance marker. As expected, the resultant mutant strain was defective in protein glycosylation. Finally, the shuttle plasmid pJBM3 was engineered to complement the mutation and partially restore the glycosylation phenotype.

This work has provided a suite of tools and methods that enable the further investigation of the genetic and molecular basis of C. hepaticus biological functions, including motility, invasion, survival in challenging environments, and other aspects of virulence. Such studies may provide clues about how SLD could be controlled, including identification of potentially useful vaccine antigens.

SLD has been reported primarily in laying hens, especially free-range birds (1,20,24,25,27,28,61,62), while pullets may be colonized with the respective pathogen but do not seem to be affected clinically (5,44). There are many risk factors that have been suggested to predispose birds to the development of SLD (Fig. 1). Female birds in other housing systems can also show symptoms, and SLD has been occasionally described in broilers and layer breeders (1,28). The incidence of SLD has been increasing, which is associated with the expansion of cage-free and free-range housing systems in many countries. In some regions, C. hepaticus may be endemic in free-range layer production (19). The availability of a scratch area was identified as one important risk factor for SLD outbreaks and led, during a descriptive survey of layer farms in Australia, to an SLD prevalence of about 40% to 70% (5,63,64). The causative agent has been detected in the soil surrounding poultry houses, dust, and mud, as well as in the feces of wild birds (44) and in insects and wild fauna (46). An experimental study provided evidence that finches may be infected with C. hepaticus and develop specific lesions (65), and hence it is possible to speculate that other wild birds may be infected by C. hepaticus and could be a source of bacterial shedding in the housing environment. The contamination of the environment, especially repeated detection of C. hepaticus in stagnant water, seems to be one of the most important factors increasing the risk of infection of consecutive flocks (19). As C. hepaticus may survive at 4 C up to 21 days in water, and even up to 65 days in Ringer’s solution, we may speculate that the pathogen may persist in water on the premises (47). Even after cleaning and disinfection, the pathogen may not be eliminated, and subsequent flocks may become colonized. In the United States and Australia, but also Germany, SLD is seen more frequently during the wet and hot seasons, suggesting a weather impact (1,19,27).

SLD can be considered to be a multifactorial disease, as triggering factors are of relevance to induce disease in C. hepaticus colonized birds (7,44). Campylobacter hepaticus may circulate for some time in the flock without the development of clinical symptoms (44,64). It has been speculated that asymptomatically infected birds may be undetected carriers of the pathogen (11). The concept of such asymptomatic carriage is supported by the work of Muralidharan et al., discussed above, where serum was found to be positive for anti–C. hepaticus antibodies in flocks with no history of SLD (7). The possibility that the presence of C. hepaticus–positive sera in what were believed to be flocks with no history of SLD outbreaks may be due to undetected outbreaks cannot be ruled out. In fact, the authors hypothesize that in the single farm with a 41% sera-positivity, this likely was the case, as evidenced by an observed drop in egg production after a mass smothering event. However, for the other three farms with no history of SLD outbreaks that had sera-positive samples, it was not suggested that there was any undetected SLD outbreak; of these farms, two were PCR positive for the presence of C. hepaticus/C. bilis genomic DNA from cloacal swabs, which is highly suggestive of the presence of the live organism in the animals. In addition to this, in the farms with a known history of SLD outbreaks, C. hepaticus/C. bilis genomic DNA was often detected in the cloaca of birds well after disease had resolved—in one farm, over a year after the outbreak had occurred (7).

These results point to asymptomatic carriage of C. hepaticus/C. bilis both in flocks without disease and in flocks postdisease. The work of Courtice et al. provides further evidence of asymptomatic carriage in flocks after SLD outbreak, where once again, flocks were found to be positive for C. hepaticus either via PCR or culturing, sometimes months after SLD had occurred (11). The role of asymptomatic carriage of C. hepaticus/C. bilis as a reservoir of either initial disease outbreak or repeated disease outbreak or contamination of the housing system, and thus as a trigger for future outbreaks, remains understudied, but these reports emphasize why future research on the topic is warranted.

The disease occurs on average at the peak of lay around 23–32 wk. It was suggested that hormonal changes associated with suppression of the cell-mediated immunity and a change in serum iron levels support the colonization of the gall bladder by C. hepaticus (2,66) and favor SLD development in birds at that age. If newly commissioned free-range flocks are affected, the age range may be larger, varying between 23 and 75 wk at least (11). This is suggested to coincide with stress due to transfer of the pullets to the layer farm, contributing to immunosuppression and subsequently a diminished defense in the chicken host (8,67). However, other stress insults such as predator attacks, mite infestations, heat, and failure of drinking water or feeding systems may contribute to the clinical development of SLD (Table 2). SLD can persist for weeks (68), or it can be observed more than once within the same flock (26,44,64). Reoccurrence rates may reach up to more than 30%, as observed in an Australian study (64), and could happen multiple times. More than one C. hepaticus strain may circulate within one flock, suggesting poor cross-immunity between isolates (26). Interestingly, SLD is often associated with other health problems such as increased worm burden or other bacterial and viral infections (26). Novel chaphamaparvoviruses have been found in clinically diagnosed SLD cases, but there is currently no evidence that they have a role in causing or exacerbating SLD (69). No exacerbating impact was observed for factors such as probiotic use or other feed additives, water supply disinfection, or water source in layer flocks (5). The genotype of the chickens, their flock size, and individual nest space may affect their susceptibility to clinical SLD, but the data and reports are controversial, and, therefore, these possible risk factors still need to be investigated further (1,26,63,64).

Biosecurity and the hygiene of the chicken’s environment can be considered as the most important interfering measures to control SLD outbreaks, though difficult to implement depending on the conditions in the hen’s environment. These measures are especially challenging in the extensive free-range systems. Sufficient down time after cleaning and disinfection should be pursued, as C. hepaticus may be detected in the feces and environment even weeks after the first SLD case (19). Separation of hens from their excrements by slatted floors is suggested to contribute to better biosecurity and subsequently reduce the risk of SLD (5). A major challenge for the field is the persistence of the pathogens in the environment (63,72), leading to their transfer to replacement flocks and an endemic situation on the premises (2). Campylobacter hepaticus may persist especially in mud and puddles, which may require additional efforts to decontaminate the pasture. Efforts should be directed towards separation of birds from these wet areas (20). Multiple-age flocks should be avoided as these may support the persistence of the pathogen in the environment. Warm temperatures are suggested to favor the development of the disease (11); therefore, heat stress, but also other stress factors, needs to be controlled (5).

Antibiotic therapy was shown to be associated with reduced duration of clinical disease, mortality, and drop in egg production by almost half (64). However, the disease could reoccur after treatment due to insufficient clearance of C. hepaticus colonization, and therefore, repeated treatment may be necessary. Antibiotic choice should consider the withholding period, which may vary between countries, and antibiotic resistance profiles of the isolates, as sensitivity of bacterial strains for antibiotic groups differs (1). Linco-Spectin, neomycin, amoxicillin, or chlortetracycline have been used with variable success, and, especially after amoxicillin and chlortetracycline treatment, reoccurrence of SLD was described (1,64). The presence of a plasmid carrying a tetracycline resistance gene, tet(O), is associated with the reported chlortetracycline resistance of C. hepaticus (12,55).

Experimentally, bacterin-based, inactivated, and live vaccines have been tested (18,73). So far, no commercial vaccines are available to control SLD, but autogenous vaccines have been used in the field with variable success rates (74). Autogenous vaccines did induce C. hepaticus–specific antibodies. Depending on the number of booster applications, antibody titers were variable, and duration increased with the third immunization. However, SLD was not fully prevented after vaccination, although the severity of lesions decreased compared to nonvaccinated control birds after C. hepaticus challenge (74).

Vitamins and short- and medium-chain organic acids supplementation through the drinking water may support the recovery of birds from SLD. The application of apple cider vinegar and citric acid to the drinking water failed to clear C. hepaticus from challenged birds, and a slight reduction of about 20% in prevalence was observed after oregano supplementation (22). Others have reported that a combination of oregano and sanguinarine feed additives can reduce the incidence and severity of SLD outbreaks (75). Avoiding biofilm development in the drinker line may reduce the risk of SLD development (26). Non-antibiotic products such as mushroom extract as well as other phytogenic products composed of essential oils, bitter substances, and saponins may improve productivity and clinical health during an SLD outbreak (1,76). The application of 2% w/w dietary biochar from eucalyptus hardwood reduced the number of C. hepaticus–positive chicken within a challenged flock (77). A recent study by Quinteros et al. demonstrated partial protection of laying hens against the impact of C. hepaticus infection on lesion development and drop in egg production by the immunomodulatory effects of isoquinoline alkaloids (sanguinarine) (78). Additionally, the expression of proinflammatory cytokines was reduced in C. hepaticus–colonized chickens after treatment compared to the infected untreated controls (78). As SLD was suggested to impact gut microbiota adversely, measures to improve gut health by high-quality diets, and possibly using probiotics, could contribute to the recovery process of affected chickens and reduce the chance of reoccurring outbreaks (68). Bacillus-based probiotic products showed effective in vitro inhibition against C. hepaticus isolates (79).

The last 5 yr have represented a period of increasing research in the field of SLD, as this review demonstrates. This impetus is reflective of the increasing impact and importance of this disease. Despite these new insights, SLD remains understudied. We suggest that future research should concentrate on certain fundamental questions:

1. A better understanding of the global spread and prevalence of SLD:

   • SLD is probably underreported, and only by knowing its true spread can research into it be properly conducted. Routine surveillance of laying flocks must be adapted to include C. hepaticus/C. bilis screening.

2. Translocation of the bacterium from the gastrointestinal tract to the gall bladder:

   • While colonization of the gall bladder is not unique to the bacteria that cause SLD, delineation of the mechanism by which this occurs will help shed light on the SLD pathogenesis process, as it is hypothesized to be an important step in disease progression.

3. Identification of key genes associated with pathogenesis in C. hepaticus/C. bilis:

   • No toxin or specific genetic factor has yet been shown to be involved in/essential for the pathogenesis process of SLD. The development of new molecular tools for the analysis of such factors will greatly aid in the identification of such elements.

   • Key studies using the newly identified pathogen C. bilis need to be replicated, protein glycosylation needs to be confirmed, and ELISAs need to be tested.

4. Investigation of cofactors and their impacts:

   • Explain the tipping point from subclinical infection to disease development.

5. Development of suitable prophylactic approaches:

   • Suitable vaccines (i.e., homologous vs. heterologous) need to be developed.

   • The impact of strain variation on protection needs to be investigated and characterized.

6. Identification of suitable treatments:

   • Alternatives to antibiotics need to be identified and developed.

Taken together, these approaches should help with the long-term development of effective vaccine/intervention strategies to help alleviate the significant global cost of SLD.

GC =

guanine-cytosine content

HRM =

high-resolution melt;

SLD =

spotty liver disease;

VBNC =

viable but nonculturable