May 25, 2017
The designations employed and the presentation of material throughout this publication do not imply the expression of any opinion whatsoever on the part of UNESCO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The ideas and opinions expressed in this publication are those of the authors; they are not necessarily those of UNESCO and do not commit the Organization.
Banting, G. and Figueras Salvat, MJ. 2017. Arcobacter. In: J.B. Rose and B. Jiménez-Cisneros, (eds) Global Water Pathogen Project. http://www.waterpathogens.org ( A. Pruden, N. Ashbolt and J. Miller (eds) Part 3 Bacteria) http://www.waterpathogens.org/book/arcobacter
Acknowledgements: K.R.L. Young, Project Design editor; Website Design: Agroknow (http://www.agroknow.com)
|Last published: May 25, 2017|
The genus Arcobacter was created in 1991 to accommodate two aerotolerant Campylobacter species and since then has rapidly evolved. As of 2017, the genus constitutes >22 species (with several more in process of description), with some described species being able to produce human bacteremia and diarrhea. Arcobacter butzleri has been the most prevalent species in meat products (chicken, pork, beef, lamb), milk, cheese and shellfish. This has led to the inclusion of A. butzleri in the list of microbes considered a serious hazard to human health by the International Commission on Microbiological Specifications for Foods. Nevertheless other species like Arcobacter cryaerophilus, Arcobacter skirrowii, Arcobacter thereius and Arcobacter tropiarum have also been associated with disease in human and animals. In humans Arcobacter can produce bacteremia and isolation rates from patients with diarrhea range from 0.11% to 1.31%. Contaminated food or water is considered the likely route of transmission to human and animals. In three reported waterborne outbreaks, Arcobacter was recovered either from the drinking water or from the faces of the patients with diarrhea, and in all cases the drinking water was fecally contaminated. A clear correlation between the concentration of fecal indicator bacteria in the water and the presence of Arcobacter has been demonstrated and the persistence of these bacteria in wastewater indicates that this could be one ecological reservoir. Metagenomic studies have revealed a high prevalence (5 to 11%) and genetic diversity of arcobacters among the bacteria communities present in wastewater, and this has been associated with the growth ability of these bacteria within the sewage system. The dominant species in wastewater is A. cryaerophilus based on metagenomics, yet in studies that utilize culture enrichment A. butzleri is the prevailing species, suggesting a culture bias towards A. butzleri. Due to the requirement of selective culture to isolate Arcobacter, there may be an ascertainment bias regarding the true rates and species composition of Arcobacter from various environments including human stools. Conventional wastewater secondary treatment is not able to fully remove the bacteria of this genus but chlorination and UV treatment seems to be effective. Currently, Campylobacter spp. are seen as a suitable surrogate for treatment efficacy of arcobacters. However, caution should be taken as Arcobacter spp. are found at levels up to 4 log10 higher in raw wastewater than Campylobacter spp. and likely have the ability to replicate in wastewater. As Arcobacter is a relatively new genus, gaps exist in the understanding of its true prevalence, the genetics of its virulence and its infectious dose in humans. Therefore, the risks associated with exposure to Arcobacter spp. are still unclear.
The true burden of Arcobacter infection is currently unknown. The first cases date back to 1992 when an outbreak of recurrent abdominal cramps associated with Arcobacter butzleri was described in an Italian school (Vandamme et al., 1992b). Since then relatively few studies have been published investigating the presence of Arcobacter in humans (reviewed in Collado and Figueras, 2011; Figueras et al., 2014; Arguello et al., 2015). Evidence exists to suggest that Arcobacter induces gastroenteritis, however the biological nature of its enteropathogenicity has yet to be clearly determined (reviewed in Collado and Figueras, 2011). Reporting of this disease agent is limited and it will be some time before the full impact of Arcobacter infections on human health is known.
Arcobacter distribution in humans is currently unknown due to the emerging nature of reporting. However, it will likely have a broader distribution than Campylobacter due to its ability to survive and possibly replicate in water. Thus far the limited reported human cases have come from four continents, suggesting broad distribution. The information derived from culturing human stools from patients with diarrhea indicate a prevalence of Arcobacter ranging from 0.1% to 1.25%, while in studies that detected these bacteria in feces by PCR results were much higher ranging from 0.4% to 56.7% (Figueras et al., 2014; de Boer et al., 2013; Collado et al., 2013; Ferreira et al., 2014; Webb et al., 2016a). In a dedicated retrospective study, performed in Belgium from 2008 to 2013, Arcobacter spp. were the fourth most prevalent bacterial pathogen detected, with an incidence of 1.31% after studying 6,774 fecal samples from patients with enteritis (Van den Abeele et al., 2014). In various other studies, detection of Arcobacter by molecular methods showed vastly different rates of prevalence as compared to detection by culture, but the highest difference have been detected in a recent study ca. 60% vs. 0.8%, respectively (Webb et al. 2016a; and references therein). The introduction of MALDI-TOF as a fast identification tool in clinical laboratories will likely help with the future identification of additional clinical cases of Arcobacter infection (Figueras et al., 2014). It should be noted that culture enrichment using selective antibiotics might result in significant bias and underestimation of Arcobacter spp. other than A. butzleri. This is due to the fact that the antibiotics used in culture enrichment select for antibiotic-resistant strains, while preventing recovery of antibiotic-sensitive strains that may be of human health concern.
The majority of cases of enteritis and bacteremia caused by Arcobacter appear to be self-limiting and do not require antimicrobial treatment. However, the severity or prolongation of symptoms may justify the use of antibiotic treatment being fluoroquinolones and tetracycline suggested for the treatment of human and animal infections.
The genus Arcobacter along with the closely related genus Campylobacter, are members of the family Campylobacteraceae. Helicobacter is another closely related genus, within the family Helicobacteraceae. Both Campylobacteraceae and Helicobacteraceae are members of the order Campylobacterales. Arcobacter is a relatively new genus, being proposed in 1991 to accommodate two aerotolerant species previously considered atypical campylobacters (Vandamme et al., 1991). Since then the number of described species has increased progressively from six species in 2008 to more than twenty-two species as of 2017 (http://www.bacterio.net/arcobacter.html). The majority of species described to date have been isolated from shellfish, livestock, marine environments, and sewage. Arcobacter differ from Campylobacter by their aerotolerance and ability to grow at lower temperatures (15°C). Only four species have been identified in humans to date: A. butzleri, A. cryaerophilus, A. skirrowii and A. thereius. Of these, A. butzleri and A. cryaerophilus, are likely the most clinically significant on the basis of the detection frequencies in humans to date.
Of the species tested, A. butzleri appears to be the least fastidious showing growth on lactose, glucose and citrate with some strains displaying thermotolerance at 42°C. A. butzleri also displays tolerance to 1.5% NaCl and has the ability to reduce nitrate (Vandamme et al., 1992a; Levican et al., 2013a). The fast growing characteristics of A. butzleri in enrichment culture may mask the abundance of other species (as well as when enriching for campylobacters (Banting et al., 2016) such as A. cryaerophilus (Houf et al., 2002; Collado and Figueras, 2011; Levican, 2016; Levican et al., 2016). A. cryaerophilus was not recovered using medium supplemented with NaCl in the study of Salas-Massó et al. (2016), despite being the second most abundant species using the Arcobacter-CAT medium without salt, indicating that A. cryaerophilus cannot tolerate 2.5% NaCl. Therefore, the range of species recovered from the environment may depend upon the culture media and incubation conditions employed.
At least twelve new Arcobacter species have been described since 2009, making Arcobacter a rapidly expanding genus, due to the introduction of molecular identification methods. A further expansion of the genus is likely as these new methods are more routinely used. Some species like A. cloacae and A. defluvii have been isolated from sewage (Levican et al., 2013a) and recently A. faecalis and A. lanthieri have been recovered from a human waste septic tank and from pig and dairy cattle manure, respectively (Whiteduck-Léveillée et al., 2015a; 2015b). Others have been isolated from marine shellfish: A. bivalviorum (Levican et al., 2012), A. ebronensis (Levican et al., 2015), A. ellisii (Figueras et al., 2011a), A. molluscorum (Figueras et al., 2011b), A. mytili (Collado et al., 2009a), A. venerupis (Levican et al., 2012) and A. lekithochrous (Diéguez et al., 2017). From seawater, A. pacificus (Zhang et al., 2015) and A. aquimarinus (Levican et al., 2015) have recently been described. There will likely be further species described, as specific culture methods are refined to enable efficient culture of novel Arcobacters. In fact, the use of Arcobacter-CAT broth supplemented with 2.5% NaCl (w/v) with a subsequent culture on marine agar after passive filtration enabled the recovery of 52 new Arcobacter marinus and 6 Arcobacter halophilus isolates. These two species had previously only been known from single strains recovered the former from a mix of seaweeds, starfish and seawater from Korea and the latter from a hypersaline lagoon in Hawaii (Salas-Massó et al., 2016). The same approaches enabled recovery of several new species from shellfish and water, which have not yet been described (Salas-Massó et al., 2016). Additionally, a separate new species (A. porcinus) has been identified amongst strains included in the description of A. thereius (Figueras et al., 2017).
It is probable that the Arcobacter genus will be subdivided in the near future into two or more genera on the basis of: (1) 16S rRNA genetic differences between some type strains (<95%) and (2) their variable ecological niches (Figueras unpublished results).
Formal transmission linkages are currently speculative, but based on knowledge of Campylobacter it seems that likely sources for Arcobacter transmission are fecal contamination combined with food and water matrices. Recreational water and drinking water are both likely sources of exposure along with irrigation water, as Arcobacter appear to survive well in water (Van Driessche and Houf, 2008). Irrigation or washing of crops meant for raw consumption (vegetables) with water containing Arcobacter is a potential transmission route (Hausdorf et al., 2013; Fernandez-Cassi et al., 2016). Arcobacter has been found not only in raw food products but also in meals at popular restaurants in Bangkok where it was determined that the risk of exposure per consumed meal was 13%, and it was up to 75% in the case of 10 meals or more (Collado and Figueras, 2011; and references therein). Arcobacter, like Campylobacter, are also commonly found on poultry carcasses (Atanassova et al., 2008; Fallas-Padilla et al., 2014) and consumption of raw or undercooked poultry is likely a significant source of transmission. There is currently limited data on foodborne and waterborne Arcobacter transmission to determine its true importance in public health, as reviewed by Collado and Figueras (2011) and Hsu and Lee (2015).
Non-culture based methods used to screen for Arcobacter carriage in stool samples rely on the DNA extraction of fecal samples followed by PCR detection and/or quantification. In one study of ~500 random stool samples submitted for bacteriological testing, 0.4% tested positive for A. butzleri by PCR but subsequent recovery by culture was unsuccessful (de Boer et al., 2013). There are few studies for undeveloped regions, therefore rates of Arcobacter carriage are difficult to determine in these areas. One study performed in South Africa observed that ~13% of hospital patients were positive for Arcobacter carriage, using molecular detection methods, compared to ~3% in (healthy) school children (Samie et al., 2007). Further studies in under-developed regions will be necessary to determine if Arcobacter carriage rates are significantly different than those in developed countries studied to date.
Studies of traveller’s diarrhea patients using PCR on fecal DNA extracts have also identified Arcobacter in 8% of diarrheal stools (Jiang et al., 2010). Hopefully, with the inclusion of Arcobacter in routine pathogen screening for traveller’s diarrhea more accurate data will become available to determine its true impact. Arcobacter has also been cultured from the feces of healthy individuals involved with the handling, slaughter and transportation of animal products. Arcobacter was isolated from 1.4% of healthy donors, but required enrichment for isolation (Houf and Stephan, 2007). This was likely due to the observation that these individuals were shedding less than 100 CFU of arcobacters/g of feces. Only A. cryaerophilus was recovered in this study confirming that low-level Arcobacter carriage in humans may be asymptomatic, which likely warrants further study in determining true Arcobacter carriage rates across healthy individuals. A recent study performed in Chile found that A. butzleri was the 4th most common amongst the detected campylobacters with an incidence of 1.4% in patients with diarrhea, but was not detected in the healthy controls (Collado et al., 2013). In a recent study performed in Canada based on A. butzleri PCR-detection a similar incidence (56.7% vs. 45.5%) between diarrheic and non-diarrheic individuals was observed, but DNA concentrations in the diarrheic stools were significantly higher (p<0.007) (Webb et al., 2016a). The Arcobacter incidence in non-diarrheic stools in this study is much higher than that reported in other studies described above.
There are still knowledge gaps regarding the virulence potential of Arcobacter infections. However, work reported to date is suggestive of Arcobacter spp. having the potential to adhere to and invade several cell lines including human intestinal epithelial cells causing cytotoxicity (Collado and Figueras, 2011; Levican et al., 2013c; Ferreira et al., 2015; Karadas et al., 2016). Putative virulence genes have been identified in various Arcobacter species/strains in numerous studies (Douidah et al., 2012; Levican et al., 2013c; Ferreira et al., 2015; Girbau et al., 2015; Piva et al., 2017). Further work is required to better understand the relationship between these putative virulence markers and human clinical outcomes, though some tantalizing evidence suggests that Arcobacter has the potential to induce tight junction dysfunction (Bücker et al., 2009; Karadas et al., 2016), which may lead to diarrhea (Collado and Figueras, 2011). This is similar to the effect observed with enteroaggregative E. coli (EAEC) infection (Strauman et al., 2010).
Arcobacter has repeatedly been identified in the feces of healthy livestock animals. Cattle consistently appear to carry A. butzleri, A. cryaerophilus and A. skirrowii (Wesley et al., 2000; van Driessche et al., 2005; Merga et al., 2011; Shah et al., 2013; Grove-White et al., 2014). Isolation rates vary between studies, but young cattle/calves were consistently shown to have higher Arcobacter carriage rates. Co-colonization with different Arcobacter species is also common. In these farming operations Arcobacter is also commonly detected in water samples and on the house floor of dairy operations. Having individual watering stations for dairy cows was shown to be protective against Arcobacter transmission, suggesting that water in dairy operations is a likely source of cross-infection (Wesley et al., 2000). Herd size is not likely related to infection rates, but source of feed may be. Additionally, housing of herds versus pasturing may lead to increased rates of Arcobacter carriage within herds (Grove-White et al., 2014), presumably due to the close proximity of the animals. Animals may carry multiple species, but detection of co-colonization may be difficult if performing culture enrichment. Direct plating from feces has been reported to have higher success at detecting co-colonization whilst lowering overall detection rates (Van Driessche et al., 2003). In the same study, post-enrichment rates of Arcobacter detection in livestock animals ranged from equine (15.4%) to porcine (43.9%), with presence also observed in ovines (16.1%) and bovines (39.2%).
Dairy farms have been shown to have a high probability of Arcobacter contamination. Due to the complexity of dairy operations and the focus on a liquid product (milk), there is a multitude of locations for potential Arcobacter contamination. In multiple studies Arcobacter has been identified in bulk milk tanks, milking apparatus, animal-watering stations, animal feed, barn floors, cattle feces (Wesley et al., 2000; Yesilmen et al., 2014; Giacometti et al., 2015a) and inline filters in milking apparatus (Serraino et al., 2013a). Arcobacter has also been identified in water buffalo milk (Giacometti et al., 2015b) and sheep dairy operations (Scarano et al., 2014). High rates of isolation of Arcobacter in dairy operations may be of significant concern to people due to the increasing popularity of artisanal dairy products (milk/cheese) that utilize raw milk. Arcobacter has been recovered from retail cheese (Scarano et al., 2014; Yesilmen et al., 2014) and raw milk (Revez et al., 2013; Yesilmen et al., 2014; Giacometti et al., 2015a), highlighting a direct linkage to human infection through consumption of contaminated foodstuffs.
Pigs are also a common source of Arcobacter, including several novel species. A. lanthieri (Whiteduck-Léveillée et al., 2015b) and A. trophiarum (De Smet et al., 2011) have been isolated from pig manure, A. suis from pig meat (Levican et al., 2013a), A. cibarius from piggery effluent (Chinivasagam et al., 2007) and A. thereius from the organs of spontaneous pig abortions (Houf et al., 2009). Several previously identified strains of A. thereius isolated from pork have recently been recognized as the novel species A. porcinus (Figueras et al., 2017). The zoonotic species A. butzleri, A. cryaerophilus and A. skirrowii have also been found in the digestive tracts and cloacae of healthy pigs (Van Driessche et al., 2004; Ho et al., 2006; De Smet et al., 2011; De Smet et al., 2012). From research to date, pigs appear to be the livestock animal most consistently associated with Arcobacter.
Arcobacter has been identified in a wide range of non-livestock mammals including alpaca, gazelle, rhinoceros and gorilla (Wesley and Schroeder-Tucker, 2011). A. butzleri has been isolated from a colony of rhesus macaques that displayed recurrent watery diarrhea (Higgins et al., 1999). Dogs have also been shown to carry human-associated species A. butzleri and A. cryaerophilus (Houf et al., 2008). This is potential concern due to the close proximity of companion animals and previous reports detailing an increased risk of Campylobacteriosis amongst dog owners (Mughini Gras et al., 2013). Reptiles have also recently been shown to carry Campylobacteracea, including Arcobacter. Lacertilia (lizards), Serpentes (snakes) and Testudines (chelonians) have all been shown to carry the human-associated species A. butzleri, A. cryaerophilus and A. skirrowii (Gilbert et al., 2014). These animals, like dogs, are often kept as companion animals and fecal-oral transmission is possible if proper hygiene is not followed after handling. There are likely many additional animals that carry Arcobacter that have not yet been described.
Poultry produced for human consumption is likely to be a significant source of Arcobacter, much like Campylobacter. Several studies have shown rates of Arcobacter isolation from processed chicken meat/carcasses at rates of up to 100% (Atabay et al., 1998; Son et al., 2007; Rahimi, 2014). Rates of Arcobacter isolation are variable as there is currently no consistent isolation method adopted for culture/isolation. Based on current reports, the Arcobacter species commonly isolated from chicken are A. butzleri, A. cryaerophilus and A. skirrowii, in decreasing frequency of detection. This frequency of detection is similar to the species detected from human stool samples (see above). For many years it was unclear as to the source of Arcobacter during poultry slaughter. There has been a debate as to whether or not Arcobacter is carried in the gut of chickens, as in several studies researchers were unable to culture/identify Arcobacter from fresh chicken ceca and suggested that process water may be the source of Arcobacter contamination on chicken carcasses (Van Driessche and Houf, 2007a). However, subsequent studies demonstrated that these microbes can inhabit the chicken intestine, but that the age of the sampled animals and the method used for recovery and identification influence the prevalence of Arcobacter (Ho et al., 2008). Other studies utilizing both PCR and culture confirmation have been able to identify Arcobacter in a large percentage of chicken intestines, though the rates of detection vary significantly between flocks (Lipman et al., 2008). Arcobacter has also been identified in poultry house litter at high levels, but was not ubiquitous across all houses (Dumas et al., 2011), consistent with other reports of sporadic detection in chicken feces.
Geese (Atabay et al., 2008), ducks and turkeys (Collado et al., 2009b) have also been shown to carry the three human-associated Arcobacter species (A. butzleri, A. cryaerophilus, A. skirrowii). Like chicken, the risk of infection is due to improper hygiene after handing of raw meat and/or consumption of undercooked meat. A foodborne outbreak in a wedding reception was associated with consumption of broasted chicken contaminated with Arcobacter, demonstrating direct linkage between poultry consumption and human GI disease (Lappi et al., 2013). As geese and ducks are waterfowl, they may represent a significant source of Arcobacter contamination in still bodies of fresh water.
Water is a likely key component to Arcobacter transmission and in most studies looking at carriage rates in animals the associated water supplies commonly had Arcobacter present. This is particularly common in intensive farming operations where water is consumed by the animals or used to wash animal carcasses or vegetable produce (Hausdorf et al., 2013).
Arcobacter has been shown to possess the ability to form biofilms, which greatly impacts its ability to survive on abiotic surfaces (Houf et al., 2002; Ferreira et al., 2013). Arcobacter have been identified in water distribution pipe biofilms (Assanta et al., 2002) and on slaughter equipment in poultry abattoirs (Houf et al., 2002; Ferreira et al., 2013). This, coupled with the observation that Arcobacter can survive long periods in water (Van Driessche and Houf, 2008), or even replicate at slaughter and refrigeration temperatures 4 to 10 °C (Kjeldgaard et al., 2009) could place it in a very different level of risk compared to the other campylobacters.
Arcobacter in soil is most likely the result of fecal deposition from animals, watering with contaminated water, spreading of manure or release of sewage/septage from damaged or cross-connected systems and open defecation. Following rainfall this material may make its way into the aquifers or nearby waterways leading to potential human exposure. An overview of the global prevalence of Arcobacter in water has been presented by Hsu and Lee (2015). In addition, the association of Arcobacter with unicellular protists and plankton of various sizes has also been described (Maugeri et al., 2004; Hamann et al., 2016).
Arcobacter has been associated with at least four waterborne outbreaks. Two outbreaks were linked to contaminated well water with one of them being the first U.S. report of A. butzleri isolation from groundwater (Rice et al., 1999; Fong et al., 2007). A subsequent report described an outbreak of gastroenteritis that occurred after a drinking water network was been connected to a new building (Kopilovic et al., 2008) and another with a water distribution pipe breakage (Jalava et al., 2014). In three of these outbreaks multiple different pathogens were isolated from the tested water and/or the collected stool samples. This is consistent with the fact that the water was likely contaminated with sewage containing various pathogens (Fong et al., 2007; Kopilovic et al., 2008; Jalava et al., 2014). Failures with the chlorination process used for disinfection of drinking water has been considered the cause of the presence Arcobacter in the water and the cause of at least one outbreak (Rice et al., 1999).
There are limited data available regarding Arcobacter infections making it difficult to make definitive conclusions regarding incubation periods. However, one report of a putative Arcobacter outbreak at a wedding reception suggested that symptoms may be observed as little as 6 hours and as long as three days post ingestion (Lappi et al., 2013). The mean incubation period for this outbreak was approximately 31 hours.
As Arcobacter culture from stool is still an immature procedure there is limited data available on the levels shed from humans with Arcobacter infections. Asymptomatic A. cryaerophilus shedding has been reported to be <100 CFU/g of feces in seven individuals associated with animal slaughtering, meat handling and/or transportation (Houf and Stephan, 2007). A report from an Italian elementary school suggested the likelihood of person-to-person transmission of A. butzleri over a period of approximately three weeks with 10 cases in total (Vandamme et al., 1992b). Little else has been reported regarding person-to-person transmission amongst people, though it is likely analogous to Campylobacter in this regard. The infectious dose of Arcobacter spp. in humans is currently unknown, though is likely variable based upon species and/or strain.
In developing countries where sanitation and hygiene are poor and interaction with animals is frequent, person-to-person transfer between people is significant, particularly in children. Campylobacter carriage rates may be as high as 25% in children less than five years old, who may be asymptomatic (Coker et al., 2002). Diarrhea associated with A. butzleri seems more persistent and watery, but less acute and possibly more asymptomatic than associated effects from C. jejuni infections (Vandenberg et al., 2004). In fact recurrent episodes with abdominal pain seems to be the typical clinical presentation for Arcobacter (Figueras et al., 2014).
There are reports describing both vertical (mother to infant) and horizontal (animal to animal) transmission of Arcobacter between sows and piglets. Pregnant sows have been shown to carry A. cryaerophilus in their amniotic fluid, which is transmitted to their newborn piglets (Ho et al., 2006). However, sows were also shown to carry A. skirrowii or A. butzleri in their rectum and within three weeks the piglets had generally shed the A. cryaerophilus signature in favor of A. skirrowii and/or A. butzleri presumably acquired by the fecal-oral route due to proximity to the sows.
There are currently no reports describing susceptibility to Arcobacter infection in people. However, Arcobacter spp. together with other Campylobacter related organisms are considered important pathogens associated with diarrhea among HIV positive individuals from the Venda region, Limpopo, South Africa (Samie et al., 2007). Another HIV-positive study group had an Arcobacter incidence by culture of 2.67% (Patyal et al., 2011). However, repeated exposure to Arcobacter may contribute to immunity in healthy individuals. This may be possible considering that the phenomenon has been described for Campylobacter immunity, particularly in developing regions where sanitation and hygiene are poor and in developed regions amongst farm workers (Coker et al., 2002; Forbes et al., 2009). A report describing asymptomatic carriage of A. cryaerophilus in workers associated with meat handling supports this hypothesis (Houf and Stephan, 2007). Further studies are required to determine accurate asymptomatic carriage rates for Arcobacter in humans.
In Chile carriage rates of ~5% of Campylobacter-related organisms in non-diarrheic patients have been reported (Collado et al., 2013). However, Arcobacter was not detected in any of the controls, and was only found by molecular methods in two patients with diarrhea (1.4%) and in only one (0.7%) by culture. Conversely, in a Canadian study using molecular detection only, Arcobacter prevalence was very high, and there was little difference (57% vs. 46%) in Arcobacter detection rates between diarrheic and non-diarrheic patients (Webb et al., 2016a). The highest reported rate of Arcobacter detection to date is 9% in a group of Type 2 diabetic patients when using molecular methods (Fera et al., 2010b). In these same patients culture was only positive in 8%, reinforcing the idea that different Arcobacter detection methods may produce vastly different results, even in the same patient group. As such, care must be taken when comparing Arcobacter prevalence data between studies using different detection methods. Further studies, and more standardized methods will be required to better determine Arcobacter carriage rates in symptomatic vs. asymptomatic people.
To date there have been no reports regarding vaccine development targeting Arcobacter.
Arcobacter was originally isolated from livestock abortions through the use of Leptospira culture methods and were described as aerotolerant Campylobacter (Neill et al., 1978). Later it was determined that due to their aerotolerance and lower optimal growth temperatures that these isolates should form their own genus, Arcobacter (Vandamme, 1991). Since that time various methods have been developed to isolate Arcobacter. All are relatively similar to Campylobacter isolation methods, with the exception of temperature (Merga et al., 2011). Media bases such as Mueller-Hinton broth, Brucella broth and Bolton broth have been used. Arcobacter-specific broth (ASB) has also been marketed and is coupled with the CAT (Cefoperazone, Amphotericin B, Teicoplainin) selective supplement (Collado and Figueras, 2011). Antibiotics used in Arcobacter broths are often overlapping with those used in Campylobacter isolation. Freezing of fecal specimens is not recommended as it may produce a reduction of ~50% in the recovery of Arcobacter species (Merga et al., 2011) Sequencing of the A. butzleri genome revealed the presence of a large number of antibiotic resistance genes (Miller et al., 2007). In fact, A. butzleri RM4018 displayed resistance to 42 of 65 antibiotics tested, more than Campylobacter jejuni, C. coli or C. lari. It is therefore likely that temperature acts as the most important selective pressure in Arcobacter vs. Campylobacter culture. On this basis, recovery of Arcobacter from traditional Campylobacter media (Bolton and Preston broths) is not surprising and has been reported (Diergaardt et al., 2004; Merga et al., 2011; Figueras et al., 2014; Banting et al., 2016). Arcobacter is routinely cultured at 20 to 37°C under aerobic or microaerophilic conditions, with the most frequent temperature being 30°C, while Campylobacter spp. will not be effectively cultured at 30°C or below. Interestingly, an incubation temperature of 25 °C was used in a large Belgian study (Van den Abeele et al., 2014), which found an Arcobacter spp. incidence of 1.3% in cases with diarrhea. This incubation temperature was used, as it was the only incubator available with a temperature lower than 37°C (Van den Abeele personal communication). Thermotolerant Campylobacter are typically cultured at 42°C, but only a handful of Arcobacter spp. have the ability to grow at this temperature and this may even be strain dependent (Levican et al., 2013a). The majority of Arcobacter spp. will also grow in the presence of atmospheric oxygen, which will suppress Campylobacter growth.
Many of the reports describing Arcobacter detection (Table 2) are presence/absence only, through the use of culture enrichment or qPCR (Fera et al., 2010a; Collado et al., 2010; Ertas et al., 2010). A few quantitative studies have been described including detection of Arcobacter in the feces of healthy cattle (van Driessche et al., 2005), from piggery effluent (Chinivasagam et al., 2007) and from irrigation and wastewater (Banting et al., 2016; Fernandez-Cassi et al., 2016; Webb et al., 2016b). In piggery effluent Arcobacter levels ranged between 105 to 108 MPN/100mL (Chinivasagam et al., 2007). In irrigation water levels averaged ~33 MPN/100 mL, while in wastewater A. butzleri was detected at levels up to 105 MPN/100 mL (Banting et al., 2016; Fernandez-Cassi et al., 2016; Webb et al., 2016b). Using the enumeration method described by the Chinivasagam, others have found concentrations of 3.7x105 MPN/100 mL of Arcobacter in a fecally contaminated freshwater stream (Collado et al., 2008). More recently using the same protocol, the Arcobacter concentration in secondary and tertiary (after a lagooning storage process) treated wastewater was 7.5x106 MPN/100 mL and 4.6x102 MPN/100 mL in outlet water, respectively (Fernandez-Cassi et al., 2016). The high concentrations of Arcobacter found in raw sewage or secondary treated wastewater enable the direct recovery of the bacteria (by passive filtration through 0.45 μm filters on blood agar) without an enrichment step, which will be a more objective approach to determine the dominant species (Levican et al., 2016). A quantitative miniaturized MPN-qPCR assay has been described for Campylobacter enumeration from surface and wastewater samples that also proved to be effective for recovery/enumeration of A. butzleri (Banting et al., 2016). Further optimization of this assay may lead to effective methods for quantitation of Arcobacter spp. from water matrices. A qPCR approach using PMA to inhibit amplification of DNA coming from dead bacteria has been developed (Salas-Massó et al. in preparation) which may allow more accurate quantitation of live Arcobacter by qPCR, as this bacterium, like others, is known to enter into a viable but non-culturable (VNBC) state when under stress (Fera et al., 2008). A. butzleri has been shown to be capable of resuscitation in rich media after up to 270 days storage in seawater at 4°C (Fera et al., 2008). As new Arcobacter species are identified/characterized, more optimized culture conditions are likely to be defined. However, growth requirements are unlikely to be standardized across the entire genus, as shown by specific growth requirements such as the salt requirements of A. halophilus, or the anerobiosis of A. anaerophilus.
Biochemical assays, PCR and/or DNA sequencing are typically performed to confirm strains as belonging to the genus Arcobacter. Typical biochemical responses for the genus are: positive catalase and urease activities, along with the ability to reduce nitrate and with a few exceptions the ability to hydrolyze indoxyl acetate. In addition, isolates are commonly tested for aerotolerance, resistance to NaCl, surfactants, tetrazolium chloride and various antibiotics. There are a variety of PCR assays described for the identification and speciation of Arcobacter targeting the 16S, 23S, gyrA and hsp60 genes (Collado and Figueras, 2011; Levican and Figueras, 2013b). The resolving power of these assays is variable and none of the assays reported to date have shown the ability to correctly speciate all Arcobacter strains (Levican and Figueras, 2013b). In fact, the majority of studies use methods that purportedly target only three species (A. butzleri, A. cryaerophilus and A. skirrowii); however, using these methods other species have been also been identified (Levican and Figueras, 2013b). The 16S rRNA-RFLP method described by Figueras et al. (2008) and updated in 2012, is the only method other than DNA sequencing able to differentiate the majority of the species of the genus (Figueras et al., 2012). However, with the rapidly expanding nature of the Arcobacter genus, all methods will require further evaluation to determine their efficacy. Currently, DNA sequence analysis of selected genes (rpoB, gyrB, cpn60 etc.) is the most accurate method for Arcobacter species assignment. This analysis can be done by BLAST comparison with sequences available in GenBank, but the construction of a phylogenetic tree that includes the new sequences along with those of the type strains of accepted species is a more reliable approach.
There is an increasing body of literature detailing that Arcobacter is a consistent component of human sewage systems (Stampi et al., 1993, 1999; Moreno et al., 2003; González et al., 2007, González et al., 2010; Collado et al., 2008, 2010; McLellan et al., 2010; Fisher et al., 2014; Merga et al., 2014; Banting et al., 2016; Fernandez-Cassi et al., 2016; Webb et al., 2016b), likely able to replicate outside of a vertebrate host within the sewage infrastructure system and wastewater treatment plants (McLellan et al., 2010; Shanks et al., 2013, Fisher et al., 2014). This supports the observation of the levels of Arcobacter found in wastewater being much higher than expected based on human carriage rates. Human-associated Arcobacter species A. butzleri, A. cryaerophilus and A. skirrowii have all be isolated from sewage samples by culture based methods (Stampi et al., 1993, 1999; Moreno et al., 2003; González et al., 2007, 2010; Collado et al., 2008; Levican et al., 2013a; Merga et al., 2014; Levican et al., 2016), and presumably could be present in human excreta in waterless sanitation or open defecation situations. As noted above, sewage has been the origin of some of the recently described new species in the genera. Utilization of metagenomic sequencing of sewage DNA extracts using pyrosequencing or next-generation sequencing technologies has revealed that Arcobacter is highly prevalent in sewage (McLellan et al., 2010; Van de Walle et al., 2012; Shanks et al., 2013; Cai et al., 2014; Fisher et al., 2014). Arcobacter has been identified in wastewater treatment plants globally and is reported to represent from 2% of total sewage partial 16S RNA gene sequences (Cai et al., 2014) up to 85% in some samples when the target region of the 16S gene was V4-V5 (Fisher et al., 2014). This proportion will be influenced by the region targeted within the 16S RNA gene (V1 to V9), which may explain the variable rates/concentrations observed in different metagenomic studies. Due to the nature of metagenomic sequencing using short fragments of the 16S RNA gene, speciation as operational taxonomic units (OTUs) is not always possible because of the lack of sequence diversity between closely related species. However, based on oligotyping sequence analysis of wastewater samples, it has been observed that two oligotypes that matched 100% to two different subgroups of the species A. cryaerophilus (subgroups 1B and 1A, respectively) were dominant, while the oligotype that corresponded with A. butzleri was only infrequently observed (Fisher et al., 2014). These results are in agreement with those reported when using culture by direct plating without enrichment (Figueras, unpublished results). Fisher et al. (2014) also reported a number of oligotypes that did not much with any of the existing species and could represent additional, un-described species in the genus Arcobacter from sewage.
Arcobacter has been isolated from pig and cattle manure from which the recently proposed new species Arcobacter lanthieri has been described (Whiteduck-Léveillée et al., 2015b). Effluent from piggery operations has been shown to carry high levels (up to 108 MPN/100 mL) (Chinivasagam et al., 2007). Piggery effluent is typically held in retention ponds, which may subsequently be applied to pastures nearby. Soil collected from these irrigated pastures also contained culturable Arcobacter at levels up to 104 MPN/g of soil. The two most prevalent Arcobacter species identified were known to be zoonotic (A. butzleri and A. cryaerophilus) indicating a possible risk associated with dispersal of lagoon waste from piggery operations (see section III below).
Groundwater contamination with Arcobacter has been reported on several occasions, being associated with two gastroenteritis outbreaks: one from a child summer camp (Idaho, USA) and the other from South Bass Island, Ohio, USA (Rice et al., 1999; Fong et al., 2007). In the latter case, an exceptionally high amount of rainfall over a period of three months resulted in massive groundwater contamination, likely from local sewage and septic systems in the area. Some of the contaminated wells were auxiliary connected with the distribution system and Arcobacter was isolated from 7/16 of these wells via a Campylobacter culture isolation method and PCR identification. Each of these wells was also positive for, viruses, E. coli and total coliforms supporting the hypothesis that the source of the bacteriological contamination was sewage. Arcobacter was recovered from the wells that that had higher levels of fecal contamination.
In the summer camp outbreak in Idaho, the chlorination system failed, which was the single treatment barrier for the well water at the camp (Rice et al., 1999). In the outbreak >80% of the individuals staying at the camp became ill with gastroenteritis. The single well on site tested positive for both total and fecal coliforms and A. butzleri was isolated from the well. Two A. butzleri isolates from the well showed the ability to survive for at least 16 days at 5°C with less than 0.5 log10 loss of viability, but cells were sensitive to chlorination (Rice et al., 1999). In both outbreaks, it was suspected that contamination of the wells may have been present for some time prior to the detection and recognition of the outbreaks.
Similar to groundwater, drinking water can be a source of Arcobacter as several reports detail the isolation of Arcobacter spp. from treated water (Jacob et al., 1993; Jacob et al.,1998; Ertas et al., 2010). In Turkey 3% of drinking water samples (chlorine treated) tested contained A. butzleri with 2% of the samples having culturable A. butzleri and A. skirrowii (Ertas et al., 2010). Another report of Arcobacter in drinking water described a waterborne outbreak in rural Finland in which the main drinking water pipe was broken during road construction and contaminated groundwater entered into the water line and made its way to a water storage reservoir, which was not disinfected (Jalava et al., 2014). Subsequent to the event Arcobacter was detected by PCR at two distribution system locations (one of which was a biofilm) and also by 16S rRNA gene sequencing of a water concentrate from the storage reservoir. However, no Arcobacter was recovered by culture-based methods.
A waterborne outbreak similar to the one described in Finland (Jalava et al., 2014) was reported in Slovenia after the drinking water system in a new building was connected. A. cryaerophilus was isolated from the feces of one (3.2%) of the 43 reported cases with acute gastroenteritis (Kopilović et al., 2008). In the latter outbreak as in the one of South Bass Island the gastroenteritis cases had multiple etiologies, due to the different pathogens present in the sewage assumed to have contaminated the drinking water, combined by different susceptibility of the exposed population on the basis of the immunity etc.
Pyrosequencing of 16S rDNA from seawater DNA extracts in Korea revealed the presence of large amounts of Arcobacter, but this was observed in the summer season only (Suh et al., 2015). Using a qPCR approach, abundance of Arcobacter in Lake Erie beach water was negatively correlated with water temperature in agreement with previous research reporting that they survive better at lower water temperatures (Lee et al., 2012; Salas-Massó et al., 2016). Campylobacter has been observed to have a seasonal occurrence in environmental surface waters in summer/fall that were identified as being from a bovine or wild bird host (Strachan et al., 2013). As Arcobacter has a similar host distribution as Campylobacter it is possible that Arcobacter follows a similar temporal distribution. A tendency for certain species to show a seasonality pattern has been observed (Levican et al., 2014; Salas-Massó et al., 2016), as well as a peak in detection rates of arcobacters in diarrheic stools between June and October in Alberta, Canada (Webb et al., 2016a). Further surveillance data is required to determine the nature of such seasonality in Arcobacter recovery.
Various species of Arcobacter have been detected in a spinach processing plant at multiple locations (Hausdorf et al., 2013). Arcobacter was detected in the process water and wash basins and the known potential pathogenic species A. butzleri was cultured from washed and blanched spinach, showing the potential for leafy green vegetables to be implicated in gastrointestinal disease. Arcobacter m-PCR detection after enrichment culture revealed that vegetables (tomatoes, parsley and lettuces) irrigated with reclaimed secondary treated wastewater disinfected with chlorine and UV were positive for Arcobacter in 14.3% of the samples from Spain, but the bacteria was not detected by culture (Figueras, personal communication).
Recovery of Arcobacter from marine environments (Collado et al., 2008; Collado et al., 2009a; Levican et al., 2014; Salas-Massó et al., 2016) also supports the idea that its environmental persistence is greater than that of Campylobacter spp. Type strains of A. butzleri could not be grown in seawater nor bio-accumulated in mussels, under the tested conditions used by Ottaviani et al. (2013). However, the capacity of this species and others to bio-accumulate in shellfish after exposure to heavily contaminated water has been demonstrated in other studies (mentioned above in sections 184.108.40.206, 2.2.7 and 2.2.10).
Arcobacter is, however, generally considered to be quite susceptible to traditional wastewater treatment (see Section 3.0; Table 5), though exceptions have been noted (Cai et al., 2014; Webb et al., 2016b). In addition, chlorine is highly effective at inactivation of Arcobacter. As little as 5 minutes in chlorinated drinking water can reduce culturable A. butzleri levels by >5 log10 (Table 4; Moreno et al., 2004). Traditional drinking water treatment typically involves chlorination, and removal of Arcobacter by traditional drinking water treatment/disinfection has been shown to be highly effective (as assayed by culture) (Collado et al., 2010). Despite treatment efficacy for removal of Arcobacter, recontamination may occur and result in outbreaks as noted above (Ertas et al., 2010; Collado and Figueras, 2011; Jalava et al., 2014).
Soil treatment units (STU) receiving effluent from private septic tanks have been reported to be highly effective at removal of culturable Arcobacter (Tomaras et al., 2009).
The average reduction of Arcobacter after secondary treatment via lagooning has been reported to be ~99.99% (Fernandez-Cassi et al., 2016; Table 5).
The efficacy of treatment works can also be monitored using NGS approaches, by observing the removal of certain bacterial genera. For example, the Shatin WWTP in Hong Kong was shown to be highly effective at removing certain genera (Streptococcus, Enterococcus, Blautia), while others, including Arcobacter, were less susceptible to the biological treatment (Cai et al., 2014). In this study, Arcobacter was found to be the 4th most prevalent genus in treated effluent, though NGS technology does not offer insight into the viability of these bacteria. In a recent study, Arcobacter was reported as one of the most prevalent genera in the influent samples to the WWTPs studied (Gonzalez-Martinez et al., 2016). Arcobacter was the most abundant genus in the bioreactor (28%) and together with Bacteroides (25%) represented more than 50% of the total bacterial population. The potential ecological role attributed to Arcobacter in the bioreactor was BOD removal and de-nitrification (Gonzalez-Martinez et al., 2016).
Non-quantitative studies using culture enrichment and PCR have been able to isolate different Arcobacter spp. from secondary-treated wastewater effluent (Moreno et al., 2003; González et al., 2007, 2010; Collado et al., 2010; Levican et al., 2016). Average density of Arcobacter in a secondary treated water effluent entering a tertiary lagoon was 7.51x106 MPN/100 mL (Fernandez-Cassi et al., 2016) but higher values have also been reported (Levican et al., 2016). Despite the fact that significant numbers of viable Arcobacter may be released in secondary treated wastewater, drinking water treatment has been shown to be effective at removal of Arcobacter (Collado et al., 2010). For example, secondary treated wastewater containing live Arcobacter supplied to the intake of a drinking water treatment plant (DWTP), resulted in non-detects following flocculation, filtration, ozonation and chlorination treatment (Collado et al., 2010).
Activated sludge treatment is effective at removal of Campylobacter, but Arcobacter only displayed a 2 log10 removal following clarification, while actually increasing by 2 log10 during the activation stage (Stampi et al., 1999). An earlier report looking specifically at A. cryaerophilus only showed a 1 log10 removal of culturable cells (Stampi et al., 1993). As there is a variety of Arcobacter spp. in sewage, this data may not be representative activated sludge treatment efficacy on different arcobacters.