Introduction

Since 2005, Campylobacter spp. have been reported as the most common gastrointestinal bacterial pathogen in the European Union (EU), accounting for 148,181 incidences of human campylobacteriosis in 2023 (European Food Safety Authority [EFSA] and European Centre for Disease Prevention and Control [ECDC], 2024). Campylobacter spp. affect more than 1.5 million people annually in the United States (Centers for Disease Control and Prevention [CDC], 2024). Campylobacter jejuni (C. jejuni) contributes around 90% of the human campylobacteriosis cases, while Campylobacter coli (C. coli) accounts for approximately 10% (EFSA and ECDC, 2024; Ministry of Health Singapore, 2025). Abdominal cramping, fever, and diarrhea with or without bloody stool are the clinical symptoms of campylobacteriosis, and commonly, these are indistinguishable from other bacterial gastrointestinal illnesses (Baek et al., 2024; Iversen et al., 2024; Kelly and Hodges, 2024). Symptoms may persist for 3–10 days, with an incubation period of 1–11 days (Myintzaw et al., 2023). Campylobacteriosis leads to sodium malabsorption syndrome, the severity of which depends on the type of strain and patient’s immune status (Bücker et al., 2018; Imbrea et al., 2024). However, on some occasions, post-infectious sequelae may develop, such as Miller Fisher syndrome (MFS), Guillain–Barré syndrome (GBS), reactive arthritis, and intestinal tract chronic inflammatory conditions, with a latent period of weeks or more (Backert et al., 2017; Heimesaat et al., 2023; Keithlin et al., 2014).

In an oxygen requirement and tolerant study conducted by Kaakoush et al. (2007), C. jejuni was classified as an obligate microaerophile. It was purported that the microaerophilic propensity of Campylobacter spp. is adapted to low oxygen concentration in the avian gut (Park, 2002). Campylobacter spp. grow in microaerobic conditions, with oxygen concentration ranging from 2.5% to 15%, compared to the atmospheric oxygen level of 21% (Haines et al., 2011; Kaakoush et al., 2007; Lynch et al., 2011). As Campylobacter spp. are transmitted from animals to humans zoonotically through food, the survivability of the pathogen outside host animal against environmental stress is crucial for its transmission to humans (Begley and Hill, 2015). Aerobic tolerance (aerotolerance) is a key survival mechanism for Campylobacter spp. in the food industry, where the oxygen level is high for the survival of pathogens (ca. 21%) (Kim et al., 2019). Aberrant aerotolerant Campylobacter spp. have been increasingly reported in various meats and viscera, including chicken meat, duck meat, turkey, pork, chicken livers, chicken gizzards, and beef livers (Guk et al., 2021; Karki et al., 2018; Oh et al., 2015a; Song et al., 2020). Aerotolerance levels of Campylobacter spp. can be classified into aero-sensitive (OS, did not survive after 12 h of aerobic incubation), aerotolerant (AT, survived after 12 h of aerobic incubation), and hyperaerotolerant (HAT, survived after 24 h of aerobic incubation) (Oh et al., 2015a). This review aims to compile information on AT Campylobacter spp. in terms of prevalence, environmental persistence, genetic relatedness, and the potential risks for humans.

Methodology

Search structure and strings applied

The search steps were completed using the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Page et al., 2021). The search was conducted in five electronic bibliographic databases, such as Scopus, Web of Science, PubMed, Scielo, and ProQuest, to increase the chance of identifying all potentially relevant articles. The date of the search was 10 January 2025. The search string used in Scopus, Web of Science, PubMed, and ProQuest was aerotoleran^* AND campylobacter, while the search string used in Scielo was (Aerotolerance) OR (Aerotolerant) AND (Campylobacter). Figure 1 depicts the scheme of the study design describing the systematic review and meta-analysis based on PRISMA. The systematic review and meta-analysis protocol is not registered in any platform.

Figure 1 The scheme of study design describes the systematic review based on the PRISMA guidelines.

Eligibility criteria

Inclusion criteria

The inclusion criteria considered articles linking Campylobacter spp. to aerotolerance. No limitations were imposed on geographic location and sample size. Only peer-reviewed research articles published in English were included.

Exclusion criteria

The exclusion criteria barred all articles published before 1991, as studies prior to this year might not have accurately reported the true Campylobacter spp. (Vandamme et al., 1991, 1992). Further, articles that lack full text were also excluded. In addition, despite aerotolerance closely pertaining to oxidative stress, studies related to oxidative stress without descriptions on the aerotolerance phenotype were outside the remit of this study. Moreover, articles that reported the prevalence of AT Campylobacter spp. were excluded when the sample size was less than 30 (Je et al., 2024).

Screening process

Records were imported into Rayyan, and duplicates were removed (Ouzzani et al., 2016). Articles were then screened by comparing the inclusion and exclusion criteria on both title and abstract. Next, full-text articles were assessed for relevance. After reviewing the bibliography of the articles included, additional eligible studies were included. Based on the content of the included articles, several subtopics that built the body of this manuscript were synthesized as shown in Figure 1.

Two co-authors screened the articles independently based on the aforementioned inclusion/exclusion criteria. The included and excluded articles were based on the agreement between the two co-authors. When the agreement was not met, the final decision was made by the third co-author.

Risk of Bias Assessment (quality assessment)

The quality assessment tool for in vitro studies (QUIN tool) was adapted from Sheth et al. 2024. The QUIN tool allows researchers to assess the quality and risk bias of the included studies. The QUIN tool in this study consists of eight applicable criteria, as shown in Table S1. Each criterion was scored as follows: two points when adequately specified, one point when inadequately specified, and zero point when not specified. The scoring guidelines grade the study into high risk of bias (<50%), medium risk of bias (50–70%), and low risk of bias (>70%), by the following formula:

Meta-analysis

A proportional meta-analysis on the prevalence of AT Campylobacter spp. isolates was conducted with the Comprehensive Meta-Analysis software version 3.7 (Biostat Inc., Englewood, NJ, USA) (Borenstein, 2022). The proportion of AT Campylobacter spp. in this meta-analysis comprised both HAT and AT isolates. Prevalence with a 95% confidence interval (95% CI) was quantified based on the total sample and the corresponding positive number. Forest plots were constructed to illustrate the prevalence of individual studies along with the pooled estimate within 95% CI. Random-effect model was employed to account for both sampling error (within-study variance) and statistical heterogeneity (between-study variance) (Kanters, 2022). Heterogeneity was assessed using Cochran’s Q statistic and I² inconsistency index. The estimated pooled prevalence can be inflated if many of the included studies have no events and have relatively low sample sizes (Churchill et al., 2019). Thus, meta-analyses were limited to the studies with a minimum sample size of 30 (Je et al., 2024).

Meta-regression and a priori subgroup meta-analysis were conducted to assess the potential reasons for between-study heterogeneity. Prevalence estimates were compared with three potential moderator variables (subgroups): source of AT Campylobacter spp. isolates (chicken, duck, and humans), species of AT Campylobacter spp. (C. jejuni and C. coli), and aerotolerance level of Campylobacter spp. (HAT and AT). Subgroup meta-analysis was performed if more than two studies were obtained in each subgroup (Plishka et al., 2021). In the subgroup of the source of AT Campylobacter spp. isolates, organs, feces, and meat samples were regrouped and analyzed based on their respective animal sources. In addition to the above-mentioned moderators, the potential covariates, including the publication year and location, were assessed with meta-regression.

Publication bias was assessed with the funnel plot, Egger’s weighted regression test, and fail-safe N test (Egger et al., 1997; Rothstein et al., 2005). An asymmetric funnel plot provides visual judgment for publication bias (Borenstein et al., 2009). Egger’s test is a statistical test to discover funnel plot asymmetry (van Enst et al., 2014). The fail-safe N test computes the number of missing studies (N) required to incorporate meta-analysis before the P value becomes nonsignificant (Rosenthal, 1979). If the N > 5 k +10, where k is the number of studies incorporated in the meta-analysis, the robustness of the meta-analysis can be assured (Rosenthal, 1979).

Statistical analysis

An alluvial plot illustrated the overview of Campylobacter spp. isolates with different levels of aerotolerance from several sources studied against environmental stresses was generated using the Origin 2023b (OriginLab Corporation, MA, USA). Ordinal logistic regression was performed to investigate the relationship between Campylobacter spp. sequence type (ST) and aerotolerance level with the Minitab 19 statistical software (Minitab, LLC, PA, USA). Radial graphs of clonal complexes (CC) and ST of C. jejuni and C. coli with different aerotolerance levels were illustrated with the Flourish platform (flourish.studio/).

Results and Discussion

Characteristics of articles and risk of bias

In all, 1,222 articles were retrieved from SCOPUS, Web of Science, Scielo, ProQuest, and PubMed. After screening, 39 peer-reviewed articles were included in this review. In the screening of the title and abstract, 989 articles were excluded (758 studies were not relevant, 191 studies were in a foreign language, 23 studies were pre-1991, and 19 studies were not research journal articles). During full-text assessment, 7 articles were excluded due to the lack of full text, and 20 articles that were not relevant were eliminated. Next, 10 studies were included after an additional search based on the reference list. All studies included in the pooled analysis reported positive samples for AT Campylobacter spp.

The included studies were synthesized into several sub-topics: 16 studies for Section “Campylobacter spp. aerotolerance classification”, 16 studies for Section “Prevalence and source of aerotolerant Campylobacter spp.”, 17 studies for Section “Survivability of aerotolerant Campylobacter spp. against environmental stress”, 8 studies for Section “Genetic relatedness of aerotolerant Campylobacter spp. strains”, and 11 studies for Section “Risk of aerotolerant Campylobacter spp. to humans”, as demonstrated in Figure 1. As shown in Table S1, most studies were considered to have a low risk of bias (>70%), and only one study was identified with a medium risk of bias (68.75%).

Aerotolerant Campylobacter spp. were first reported in the 1970s (Hanna et al., 1983; Neill et al., 1979). The genus Arcobacter was first proposed in 1991 based on nucleic acid hybridization and other molecular studies of AT Campylobacter (Vandamme et al., 1991). Arcobacter species were distinguished from the true Campylobacter spp. by their capability to grow between 15°C and 30°C and their aerotolerance properties (Vandamme et al., 1991, 1992). From that, Campylobacter cryaerophila and Campylobacter butzleri were proposed to be classified under arcobacter (Vandamme et al., 1991, 1992).

Campylobacter spp. aerotolerance classification

In total, 16 studies classified Campylobacter spp. isolates based on the level of aerotolerance, while 14 studies investigated C. jejuni, and 5 studies investigated C. coli as shown in Table 1. The majority of studies referenced Oh et al. (2015a), probably because of its simplicity and clear-cut aerotolerance classification. From that, aerotolerance levels of C. jejuni were classified after aerobic incubation in Mueller Hinton (MH) broth with shaking at 200 revolutions per minute (rpm) at 42°C into OS (did not survive after 12 h), AT (survived after 12 h), and HAT (survived after 24 h) (Oh et al., 2015a). The term ‘intermediate aerotolerant’ was adopted by Pokhrel et al. (2023) for AT. In comparison, Lee et al. (2019) classified AT C. jejuni with survival after 120 h under 500 rpm aerobic shaking at 37°C. Furthermore, Jones et al. (1993) and Shagieva et al. (2021) classified C. jejuni as aerotolerant if colonies grew after aerobic incubation at 37°C or 42°C, respectively. Although Chynoweth et al. (1998) did not classify aerotolerance in their study, the ability of C. jejuni to grow aerobically on an agar plate was reported and was regarded as aerotolerant in this review.

While the majority of the researchers adopted the method reported by Oh et al. (2015a), the protocol did not specify details, such as the surface-to-volume ratio of MH broth during aerobic incubation. Aerotolerance study in Campylobacter spp. is relatively new. Thus, a standardized method for determining the level of aerotolerance is crucial for inter-laboratory comparison, which may consider the following aspects: (1) growth phase of the inoculating culture (after certain hours of incubation); (2) cell concentration; (3) type of inoculating media; (4) aerobic stress of incubation, such as the surface-to-volume ratio of the media; (5) incubation flask/tube is protected from light/unprotected; (6) incubation flask/tube is tightened/cracked/loosened/left open; (7) incubation temperature; (8) degree and type of agitation during incubation; (9) cut-off points for aerobic incubation duration for aerotolerance level classification; and (10) enumeration method. The schematic diagram for the standardization protocol is shown in Figure S1.

Prevalence and source of aerotolerant Campylobacter spp.

Studies on the prevalence of AT Campylobacter spp. are limited, and the 16 published studies are summarized in Table S2. The prevalence of AT strains varies considerably between studies, ranging from 0% to 100% in C. jejuni and 41.9% to 100% in C. coli. Shagieva et al. (2021) reported an inconsistent AT profile of C. jejuni over three replicates. Thus, the three strains that consistently survived aerobically in all three replicates are labeled as AT in Table S2.

The Campylobacter spp. reported in studies stated in Table S2 are sorted by the source of the strains and presented in Table 2. Pond water and the outlet of a wastewater treatment plant from Shagieva et al. (2021) and udder cloth from Jaakkonen et al. (2020) were regrouped into the classification of environment, as shown in Table 2. Moreover, C. jejuni (697 strains) were characterized for aerotolerance level than C. coli (300 strains). Among them, chicken and pork were the most studied sources for aerotolerance in C. jejuni and C. coli, respectively. In C. jejuni, dairy product had the highest prevalence of HAT (69.5%), while duck had the highest prevalence of AT (62.2%). In C. coli, humans were found to have the highest prevalence of HAT (60%), and duck was reported to have the highest prevalence of AT (38.7%). In both C. jejuni and C. coli, AT (39.9% in C. jejuni and 35.2% in C. coli) and HAT (32.9% in C. jejuni and 44.3% in C. coli) were more prevalent than the OS counterparts. These estimates indicated that the AT strains of Campylobacter spp. could be an emerging phenomenon.

As shown in Figure 2 and Table 3, the pooled prevalence of AT Campylobacter spp. from 10 studies in a meta-analysis was 75.9% (95% CI: 59.6% to 87.1%, P = 0.003). The outputs of the subgroup analysis are shown in Figure 3 and Table 3; the pooled prevalence of AT Campylobacter spp. in chicken was 68.2% (95% CI: 46.5% to 84.1%, P = 0.098). As a comparison, the pooled prevalence of AT Campylobacter spp. in duck and humans was statistically significant (P < 0.001), with a pooled prevalence of 88.1% (95% CI: 80.1% to 93.1%) and 87.9% (95% CI: 75.8% to 94.4%), respectively. Based on meta-regression, as shown in Table 4, no statistically significant difference exists in the prevalence of AT Campylobacter spp. between chicken and duck, with P = 0.1226, and between chicken and humans, with P = 0.1165. Owing to the limitation of a 30-sample size in the respective studies, the source of AT Campylobacter spp., such as beef, dairy products, the environment, and turkey, were not included in the meta-analysis. This implies that further studies involving larger sample sizes from different sources are noteworthy for meta-analysis in the future.

Figure 2 Forest plot of the estimated prevalence of aerotolerant Campylobacter spp.

Figure 3 Forest plot of the estimated prevalence of aerotolerant Campylobacter spp. by subgroup of the source of isolates.

As shown in Figure 4 and Table 3, the pooled prevalence of AT C. jejuni was 73.7% (95% CI: 50.1% to 88.7%, P = 0.049). In comparison, the pooled prevalence of AT C. coli was statistically insignificant (P = 0.174), with an estimate of 73.7% (95% CI: 38.8% to 92.5%). Meta-regression in Table 4 reveals no significant difference between AT C. jejuni and AT C. coli, with P = 0.9929. However, C. jejuni was reported to have a higher survival bacterial concentration than C. coli after aerobic incubation (Mouftah et al., 2021).

Figure 4 Forest plot of the estimated prevalence of aerotolerant Campylobacter spp. by subgroup of the species.

As shown in Figure 5 and Table 3, the pooled prevalence of HAT Campylobacter spp. was 32.9% (95% CI: 22.6% to 45.1%, P = 0.007), with a Q statistic value of 100.20 and I² of 91.02%. Similarly, the pooled prevalence of AT Campylobacter spp. was statistically significant (P = 0.028), with a pooled prevalence of 38.4% (95% CI: 29.0% to 48.7%). Meta-regression in Table 4 shows no significant difference between HAT and AT, with P = 0.4996. However, some studies reported a higher prevalence of HAT than AT isolated from animals, such as broilers and cattle, suggesting that HAT strains have an advantage of colonizing in the gastrointestinal tracts (GIT) of animals (Guk et al., 2019; Jaakkonen et al., 2020; Mouftah et al., 2021).

Figure 5 Forest plot of the estimated prevalence of aerotolerant Campylobacter spp. by the subgroup of aerotolerance level.

An I² value of more than 75% indicated considerable heterogeneity (Higgins et al., 2003). Most of the heterogeneities in subgroup analysis and residual heterogeneities of all moderators in meta-regression were high (I² > 88%), as shown in Table 4, which indicates the data synthesis should be interpreted with caution. These could be due to unmeasured covariates, such as the sampling and diagnostic techniques, season, temperature, and other underlying factors (Imrey et al., 2020). This highlights the need for future studies with a sample size (>30) that adopt a standardized aerotolerance assay. Further, there was a weak relationship between the prevalence and the source of isolates (between-study variability (R²) = 35.0%, P = 0.1497). In addition, no relationships were discovered regarding the publication year (R² = 0% and P = 0.5653), location (R² = 0% and P = 0.7839), species (R² = 0% and P = 0.9929), and aerotolerance level (R² = 0% and P = 0.4996) on the pooled Campylobacter spp. prevalence. Funnel test asymmetry (Figure S2) was confirmed by Egger’s test (P = 0.043), as shown in Table 5, revealing significant publication bias in this study. However, the Fail-safe N test found that the number of studies needed to revert the significance was 252. This value was much larger than the number of studies of 60, which was obtained from the formula of 5 k + 10. Thus, the result of the meta-analysis can be considered robust to publication bias (Rosenthal, 1979).

The available prevalence data for AT Campylobacter spp., at the recent time point, are limited to C. jejuni and C. coli. Studies on the prevalence of other AT Campylobacter spp., such as C. concisus, C. upsaliensis, C. fetus, C. ureolyticus, and C. hyointestinalis, are the research gaps worth investigating, as these species are clinically important (Platts-Mills and Kosek, 2014). The prevalence of AT Campylobacter spp. in other retail foods is valuable, as Campylobacter spp. have been found in milk, eggs, fresh products, ready-to-eat food, and processed meat (Chai et al., 2007; Jaakkonen et al., 2020; Parisi et al., 2015; Toyomaki et al., 2012; Trimoulinard et al., 2017).

Survivability of aerotolerant Campylobacter spp. against environmental stress

The survivability of AT Campylobacter spp. against environmental stresses is one of the forefront traits for the pathogen to induce campylobacteriosis in humans. Figure 6 displays the alluvial plot of the overview of the Campylobacter spp. isolates with different levels of aerotolerance studied against environmental stresses, as summarized in 17 studies (Chynoweth et al., 1998; Hur et al., 2024; Jaakkonen et al., 2020; Jones et al., 1993; Karki et al., 2018, 2019; Lee et al., 2019; Mouftah et al., 2021; O’Kane and Connerton, 2017; Oh et al., 2015a, 2017, 2018, 2019; Ortega-Sanz et al., 2024; Pokhrel et al., 2023; Rodrigues et al., 2015; Shagieva et al., 2021).

Figure 6 Alluvial plot of the overview of the isolates of Campylobacter spp. with different levels of aerotolerance from different sources studied against environmental stresses. The taller nodes indicate a larger number of isolates studied, with the number indicating the number of isolates.

The alluvial plot in Figure 6 demonstrates that AT C. jejuni (1,051 strains) was the most widely studied isolate against environmental stresses. Each of the environmental stresses are discussed further in the following subtopics. The alluvial plot displays that HAT Campylobacter spp. (566/1,127) were the most widely distributed strains compared to AT (362/1,127) and OS (199/1,127). Thus, it was postulated that aerotolerant (HAT and AT) Campylobacter spp. are becoming more ubiquitous than the OS strains.

Survivability of aerotolerant Campylobacter spp. against atmospheric and oxidative stresses

One important question that remains to be solved is how this ubiquitous yet microaerophilic and fastidious pathogen can thrive in aerobic environments and subsequently infect humans. Despite exposure to the aerobic atmosphere being closely associated with oxidative stress, the stress induced by aerobic incubation was referred to as aerobic stress in this study. In contrast, the stress caused by oxidative inducers was considered oxidative stress. Oxidative inducers included, such as hydrogen peroxide (H₂O₂), cumene hydroperoxide (CHP), menadione (MND), paraquat (PQ), are stated in Table S3. Tables S3 and S4 summarize the survival of Campylobacter spp. with different levels of aerotolerance under atmospheric and oxidative stress.

Hyperaerotolerant C. jejuni survived better than the AT strains under aerobic stress (Jaakkonen et al., 2020; Oh et al., 2017). Further, HAT C. jejuni were found to survive better than the AT strains, whereas AT survived better than the OS strains under different atmospheric conditions (CO₂ and N₂) in poultry meat. Thus, both HAT and AT strains might be able to thrive on the modified atmospheric packaging, further compromising food safety.

The most common Campylobacter strain used for studying atmospheric survivability was C. jejuni NCTC 11168 (Atack et al., 2008; Jaakkonen et al., 2020; Karki et al., 2019; O’Kane and Connerton, 2017; Oh et al., 2015b; Rodrigues et al., 2015). Jaakkonen et al. (2020) and Karki et al. (2019) reported contradicting results, as C. jejuni NCTC 11168 was reported as HAT and OS, respectively, based on the aerotolerance assay proposed by Oh et al. (2015a). However, several researchers investigated the aerotolerance of C. jejuni NCTC 11168 without concluding the AT classification (Atack et al., 2008; Gundogdu et al., 2015; Oh et al., 2015b; Rodrigues et al., 2015). Thus, the inconsistency of result and the lack of clarity of aerotolerance classification reaffirm the importance of standardized protocol.

Karki et al. (2019) found that retail beef meat juice, chicken liver juice, and beef liver juice contain factors that enhance the aerobic survival of C. jejuni and C. coli (Karki et al., 2019). This is because the heme-containing proteins in meat and liver juice act as cofactors for catalase and superoxide dismutase (Pretorius et al., 2016; van Vliet et al., 2002). In addition, iron in meat and liver juice mediates genes (fdxA and ahpC) and regulatory proteins (PerR and Fur) that are involved in aerotolerance mechanisms (Baillon et al., 1999; Butcher et al., 2015; van Vliet et al., 1999, 2001). Thus, C. jejuni and C. coli might survive better in situ than in vitro, and the actual prevalence of AT Campylobacter spp. could be higher before being detected in the laboratory.

The aerobic survival of C. jejuni cocktail (mixture of four strains) was investigated on different food matrices (chicken nuggets and minced meat) (Chynoweth et al., 1998). In sterile minced chicken, incubation at 25°C showed better survival than at 37°C. The higher survival at 25°C could be due to the pathogen being less metabolically active than at 37°C and thus less susceptible to reactive oxidative species (ROS) (Chynoweth et al., 1998). Further, the survival of C. jejuni in chicken nuggets was lower than in minced meat, possibly because of the seasonings and food additives in the nuggets that inhibit the growth of C. jejuni (Chynoweth et al., 1998). Thus, the survival of single and multiple strains of Campylobacter spp. in different food matrices is worth elucidating.

Hwang et al. (2014) was the only study to report the survival of non-jejuni/coli Campylobacter isolates under aerobic incubation. The researchers reported that C. fetus was able to survive in brain heart infusion after 12 h of aerobic incubation. However, the researchers did not state the incubation temperature and the classification for aerotolerance. This study demonstrated that Campylobacter spp. other than C. jejuni and C. coli might have aerotolerance properties.

Campylobacter spp. is normally exposed to ROS during colonization on a host, thriving in the environment, and within its cellular metabolism (Gundogdu et al., 2015). Several researchers investigated the effect of ROS-promoting agents (oxidative inducers) on the viability of AT Campylobacter spp., as shown in Table S4. Oh et al. (2015a) reported that aerotolerant C. jejuni (HAT and AT strains) survived significantly better than the non-AT C. jejuni (OS strains) against 1-mM H₂O₂, 100-µM CHP, and 100-µM MND, respectively. However, results from Ortega-Sanz et al. (2024) did not show any trend in aerotolerance and survival against oxidative stress. In comparison between C. jejuni and C. coli, Karki et al. (2018) reported that there were significant differences in H₂O₂ sensitivity between strains, while there was no correlation between H₂O₂ sensitivity and the level of aerotolerance (five C. jejuni and eight C. coli strains).

As shown in Table S4, Rodrigues et al. (2015) conducted an oxidative stress assay under MAC, while Karki et al. (2018) and Oh et al. (2015a) did not mention incubation atmosphere. It is essential to note that aerobic incubation might have acted synergistically with oxidative inducers on the survival of Campylobacter spp., potentially interfering with the observations.

Results from several research groups demonstrated that AT Campylobacter spp. had better survival and growth under atmospheric and oxidative stress. However, based on the paucity of studies investigating the relationship between aerotolerance, atmospheric stress, and oxidative stress defenses, the interaction between phenotypes remains inconclusive.

Aerobically acclimatized Campylobacter spp.: Aerobic acclimatization refers to the incubation of Campylobacter spp. under aerobic conditions fo r a defined period, allowing them to adapt to aerobic stress. Researchers conducted aerobic acclimatization differently, with incubation temperatures of 37°C or 42°C under aerobic conditions for 1–7 days and ranging from 1 to 15 aerobic passages on various agars (Chynoweth et al., 1998; Jones et al., 1993; O’Kane and Connerton, 2017; Rodrigues et al., 2015; Shagieva et al., 2021). It was observed that aerobically acclimatized C. jejuni survived longer on blood agar than non-aerobically acclimatized C. jejuni at 4°C under aerobic incubation (Jones et al., 1993). Similarly, O’Kane and Connerton (2017) reported that aerobic acclimatization improved the aerotolerance of C. coli OR12. Further, Shagieva et al. (2021) reported that aerobically acclimatized C. jejuni demonstrated aerobic growth on agar. Contrarily, aerobic acclimatization did not influence the growth and survivability of both C. jejuni Bf (Bf is a C. jejuni strain named by Rodrigues et al., 2015) and NCTC 11351 under aerobic conditions (Chynoweth et al., 1998; Rodrigues et al., 2015). These conflicting results underscore the importance of a standardized method for aerobic acclimatization.

It has been postulated that the aerotolerance of C. jejuni could be due to transcriptional and translational changes (Bronnec et al., 2016a, 2016b). Microaerobic passage did not affect C. coli OR12’s ability to grow in aerobic conditions, suggesting that aerotolerance is stable and is due to genetic modification in lieu of temporary physiological adaptation (O’Kane and Connerton, 2017). However, Nennig et al. (2022) reported that more than half (43/80) of C. jejuni were able to be aerobically acclimatized, and whole genome sequencing data suggested that genetic sequence does not govern this trait. Thus, the molecular mechanism of aerobic acclimatization and aerotolerance in Campylobacter spp. could be an interesting research question.

Growth of aerotolerant Campylobacter spp. under aerobic conditions: To our knowledge, Chynoweth et al. (1998) first reported a qualitative investigation on C. jejuni to grow on solid agar, and Rodrigues et al. (2015) were the first to report this quantitatively. C. jejuni Bf exhibited a stationary phase of 10 h, which could be regarded as the adaptive period for aerobic stress, followed by a growth of 2.5 log₁₀ Colony Forming Units per milliliter (CFU/mL) from 10 h to 48 h on Kamali agar under aerobic incubation (Rodrigues et al., 2015). To date, no Campylobacter spp. other than C. jejuni and C. coli, have been reporte d to have the ability to grow under aerobic conditions, as shown in Table S5.

Aerobic growth of AT Campylobacter spp. is facilitated in liquid media (k _H = 9.28 x 10^-4 mol L^-1 atm^-1 in water at 42°C) because of lower oxygen transfer, compared to solid agar (Rodrigues et al., 2015). Next, a cell density of more than 10⁴ CFU/mL could grant better aerobic stress tolerance in AT Campylobacter spp. because the oxygen demand was higher than that of low cell densities (lower than 10⁴ CFU/mL) (Rodrigues et al., 2015). These two factors reduced the available oxygen and facilitated the aerobic growth of AT Campylobacter spp. (Rodrigues et al., 2015).

Shagieva et al. (2021), O’Kane and Connerton (2017), and Rodrigues et al. (2015) reported that Campylobacter spp. grew on Kamali, blood, and charcoal cefoperazone deoxycholate agars (CCD), respectively. These agar plates contain substances that neutralize toxic substances from oxygen (Gharst et al., 2013). However, the aerobic growth of C. jejuni on nutrient agar, as demonstrated by Chynoweth et al. (1998), might have greater aerobic growth potential because the agar did not contain any additives that mitigate aerobic stress.

Although the AT C. coli OR12 Aer P32 strain demonstrated aerobic growth on blood agar, it did not exhibit superior survival in liquid medium under aerobic conditions, unlike other C. jejuni and C. coli strains (O’Kane and Connerton, 2017). This finding did not justify such coherence, and the researchers did not explain it in their studies. Thus, this finding marks the importance of comparing the aerotolerance levels of Campylobacter spp. both in vitro (in broth and on agar) and in situ (food matrices).

Survivability of aerotolerant Campylobacter spp. against chemical antimicrobial agent

The only reported chemical antimicrobial agent used to control Campylobacter spp. in relation to aerotolerance is peracetic or peroxyacetic acid (PAA). PAA is the most commonly used antimicrobial in pre- and post-chill stages in poultry processing (Kataria et al., 2020). PAA offers a wide range of antimicrobial activities because of the combined action of acidic and oxidizing properties (Bauermeister et al., 2008). PAA is normally applied at main poultry chiller for up to 220 ppm, and for the post-chill immersion tank, the maximum permissible concentration is 2,000 ppm (United States Department of Agriculture-Food Safety and Inspection Service, 2019). PAA is an effective antimicrobial agent that reduces the population of Campylobacter spp. with a reduction of up to 1.3 log₁₀ CFU (Chen et al., 2014). Only three studies were available on raw chicken skin, as summarized in Table S6. These studies found the same trend, with HAT strains survived better than the AT strains, whereas both HAT and AT strains thrived better than OS strains (Mouftah et al., 2021; Oh et al., 2018, 2019).

Most of C. jejuni isolates resisted PAA better than C. coli, with a higher final mean cell number (Mouftah et al., 2021). However, this is the only study that compared the survival of C. jejuni and C. coli. Since the application of PAA is in the equilibrium of H₂O₂ and acetic acid, the survival of HAT and AT strains in the presence of PAA could be due to the augmented oxidative stress defense (Oh et al., 2019; Yuan et al., 1997). Studies on the survivability of AT Campylobacter spp. against other chemical decontamination agents are the knowledge gaps that remain to be explored. Examples of other chemical decontamination agents used in Campylobacter spp. decontamination include chlorine, chlorine dioxide, acidified sodium chlorite, trisodium phosphate, and peroxy acid (Hansson et al., 2018).

Survivability of aerotolerant Campylobacter spp. against temperature stress

Temperature treatments, such as refrigeration, freezing, and pasteurization, are vital for food preservation. The survival of AT Campylobacter spp. against temperature stress is shown in Table S7. In situ studies reported that HAT strains survived better under refrigeration (4°C) than AT strains, whereas AT strains survived better than OS strains (Jaakkonen et al., 2020; Mouftah et al., 2021; Oh et al., 2017, 2018). These studies reported a consistent trend between aerotolerance levels and refrigeration survival. In addition, OS strains have a significantly higher proportion in the cold-sensitive group compared to the cold-tolerant group (Hur et al., 2024). On the contrary, Pokhrel et al. (2023) reported that AT strains survived better than HAT and OS strains in refrigerated chicken drumsticks. In comparison, in vitro studies produced mixed results. Lee et al. (2019) and Shagieva et al. (2021) found no difference between aerotolerance and refrigeration survival in MH broth and brain heart infusion broth, respectively. In contrast, C. jejuni OS strains reduced significantly more AT and HAT strains in both MH broth and chicken meat (Oh et al., 2017, 2019). Interestingly, Jaakkonen et al. (2020) reported variability in the refrigerated raw milk survival of C. jejuni within CC-21, even within the same ST, implying unconserved traits in survivability.

Similar to refrigeration, HAT Campylobacter spp. isolates portrayed significantly better survival on chicken skin than AT strains under freeze-thaw stress, whereas AT strains outperformed OS strains (Mouftah et al., 2021; Oh et al., 2018). Further, freeze-thaw stress in MH broth also showed a cognate trend as on chicken skin (Oh et al., 2019). On the contrary, Pokhrel et al. (2023) reported that AT strains survived better than HAT and OS strains in frozen chicken drumsticks. The discrepancy between studies could be due to the low number of HAT strains (five strains) in the study conducted by Pokhrel et al. (2023).

Under pasteurization (72°C, 15 s) and the extended heat-treatment (72°C, 30 s) in milk, HAT C. jejuni and C. coli strains survived better than AT strains, while OS survived the least (Mouftah et al., 2021; Oh et al., 2018). On the contrary, C. jejuni isolated from retail chicken were sensitive to heat (72°C, 30 s), regardless of AT properties (Oh et al., 2019). Varied survival trends of Campylobacter spp. with different aerotolerance levels against refrigeration, freeze-thaw, and heat treatment affirm the need for future study.

Similar to the aerobic stress trend described in the previous section, meat and liver juice significantly improved the survival of AT C. coli and non-AT C. jejuni at 4°C, compared to MH broth (Karki et al., 2019). It is conjectured that retail meat and liver juice from both chicken and beef had a significant influence, but not the origin of the juice. As a result, there could be better low-temperature survival of Campylobacter spp. under in situ conditions than under laboratory conditions.

Genes encoding cold shock proteins are not present in Campylobacter spp. (Hazeleger et al., 1998). Thus, Campylobacter spp. may harbor another tolerance mechanism in response to cold shock (Mouftah et al., 2021). Since some studies found a relation between aerotolerance and temperature stress survival, the mechanism behind this could be the existence and expression of certain genes that contribute to both phenotypes.

Survivability of aerotolerant Campylobacter spp. against osmotic stress

Campylobacter spp. are sensitive to hyperosmotic conditions, with inhibition reported at more than 2% NaCl, a common preservative used in food against pathogens (Doyle and Glass, 2010; Park, 2002). Only three studies have linked the aerotolerance of Campylobacter spp. to osmotic stress resistance, and all the studies were conducted in MH agar or MH broth, as shown in Table S8. Mouftah et al. (2021) and Oh et al. (2018) reported higher survival in HAT and AT C. jejuni or C. coli strains than OS strains under hyperosmotic stress (4% NaCl). Cross-protection might exist between aerotolerance and osmotic stress resistance, and vice versa (Mouftah et al., 2021). Conversely, the degree of hyper-osmotolerance was found to be greatly variable between strains and did not pertain to aerotolerance properties (Oh et al., 2019).

Half of the C. jejuni survived at 4% NaCl, whereas none of the C. coli survived, and difference in the final cell concentration was significant between the two species (Mouftah et al., 2021). However, only one study compared the two species. The relationship between aerotolerance and hyper-osmotolerance remains to be clarified through additional research, such as investigations of different food matrices and other Campylobacter spp.

Survivability of aerotolerant Campylobacter spp. against acidic stress

To the best of our knowledge, no study has elucidated the relationship between aerotolerance of Campylobacter spp. and acidic stress tolerance. Compared to other foodborne pathogens, C. jejuni is more sensitive to acid (Birk et al., 2010). This sensitivity is due to the lack of an acid resistance system (Birk et al., 2012; Merrell and Camilli, 1999; Park et al., 1996; Richard and Foster, 2003). However, Campylobacter spp. proteins, such as SodB, AhpC, and Dps, observed during acid stress, were associated with oxidative stress (Baillon et al., 1999; Birk et al., 2012; Ishikawa et al., 2003; Pesci et al., 1994; Purdy et al., 1999). Thus, C. jejuni might normally be in an oxygen-alert state, allowing it to remove ROS and undesirable components from acid stress (Afriliana et al., 2018; Birk et al., 2012).

The pH of certain fermented foods is in the range of 4.5–5.5 (Nout, 1994). In addition, the pH of coffee ranges from 5.4 to 6.2, while the pH of milk is 6.5–6.7 (Afriliana et al., 2018; Renan et al., 2006; Sikand et al., 2010). Another example of acidic food is marinated meat (Birk et al., 2010). AT Campylobacter spp. could survive better in acidic food, which poses a food safety issue. The prevalence and survivability of AT Campylobacter spp. in the aforementioned acidic food are worth investigating.

Influence of atmospheric conditions on the survivability of aerotolerant Campylobacter spp. against environmental stresses

Certain survival studies against oxidative, temperature, and osmotic stresses, as shown in Tables S4, S7, and S8, respectively, did not provide information on the atmospheric incubation conditions. Atmospheric conditions could influence the survival mechanism of Campylobacter spp. against environmental stresses. For instance, AT and HAT C. jejuni showed better resistance to different atmospheric stresses (aerobic, N₂, and CO₂) than AT, whereas AT strains had better resistance than OS strains at 4°C (Oh et al., 2017). Furthermore, C. jejuni (with all levels of aerotolerance) survived significantly better in refrigeration under aerobic conditions than microaerobic conditions (Hur et al., 2024). Thus, atmospheric conditions, such as aerobic and modified atmospheric packaging, on the survival of AT Campylobacter spp. against environmental stresses, is an interesting research gap.

Inactivation dynamics of aerotolerant Campylobacter spp.

Different experimental settings of the included studies in Section “Survivability of aerotolerant Campylobacter spp. against environmental stress” render data harmonization difficult. Some studies reported the results in duration and percentage of survival, while others presented the results in charts without numeric figures. Thus, based on the available numeric data, the inactivation dynamics are summarized as follows:

• Under aerobic stress, HAT, AT, and OS C. jejuni had 2, 3, and 4 log₁₀ CFU reduction, respectively, after 3 days of incubation in MH broth (Oh et al., 2017). Further, AT C. coli declined to below the limit of detection (1.3 log₁₀ CFU/mL) from inoculation of 6 log₁₀ CFU/mL after 24 h aerobic incubation in MH broth (O’Kane and Connerton, 2017).

• Under oxidative stress, 3.35 µM CHP exposure for 60 min reduced viability of AT and HAT C. jejuni > 6.25 ± 0.32 log₁₀ CFU/mL, while 5 mM H₂O₂ reduced 5.89 ± 0.60 log₁₀ CFU/mL (Ortega-Sanz et al., 2024). Under 0.5 mM PQ exposure for 60 min, AT C. jejuni reduced 1.4–1.5 log₁₀ CFU/mL (Rodrigues et al., 2015).

• Under chemical antimicrobial stress, PAA reduced HAT and AT C. jejuni and C. coli equally by ca. 4.5 log₁₀ CFU/mL, while OS strains were reduced by ca. 7 log₁₀ CFU/mL (Mouftah et al., 2021). Furthermore, PAA inactivated HAT and AT C. jejuni by ca. 2.1 and 2.5 log₁₀ CFU/mL, respectively, while OS strains were reduced by 3.7 log₁₀ CFU/mL (Oh et al., 2019).

• Under chilling (4°C) stress, storing HAT, AT, and OS C. jejuni for 7 days reduced 4.9, 5.1, and 7.4 log₁₀ CFU/mL, respectively (Oh et al., 2019). On contrary, at the same condition as Oh et al. (2019), greater reduction was reported in HAT C. jejuni (0.27 log₁₀ CFU/mL) than OS and AT C. jejuni with 0.14 log₁₀ CFU/mL and 0.08 log₁₀ CFU/mL, respectively. After 5 days of chilling, OS C. jejuni reduced 1.5 log₁₀ CFU/g and AT C. jejuni reduced 2.1–3.0 log₁₀ CFU/g (Lee et al., 2019).

• Under freezing (-20°C) stress, HAT, OS, and AT C. jejuni reduced 1.12, 0.90, and 0.60 log₁₀ CFU/mL, respectively, regardless of the storage period (Pokhrel et al., 2023).

Gene regulation of aerotolerant Campylobacter spp.

Gene regulation underlying AT Campylobacter spp. for environmental persistence is intricate and remains poorly understood. AT C. jejuni under aerobic incubation and after aerobic acclimatization was reported to have an increase in the transcription levels of genes involved in aerobic detoxification, which include sodB, ahpC, katA, tpx, and trxB (Rodrigues et al., 2016). In relation to the elevated transcription levels, the corresponding proteins (AhpC, KatA, Tpx, and TrxB) were reported to have higher abundance after aerobic incubation (Rodrigues et al., 2016). Similarly, katA, ahpC, and trxB were reported to have higher expression in aerobic incubation than microaerobic conditions in HAT C. coli, whereas their expression decreased in the OS C. coli (Guk et al., 2022). Further, ahpC mutation in HAT C. jejuni reduced aerotolerance level to AT (Oh et al., 2015a). AT C. jejuni was reported with higher expression in cosR (oxidative stress response), oorDABC (oxygen metabolism), and mreB (cell morphology) after aerobic incubation (Bronnec et al., 2016b; Rodrigues et al., 2015). Under aerobic incubations at 37°C and 4°C, the expression of htrB (stress response) gene was higher in AT C. jejuni than in OS C. jejuni (Lee et al., 2019). Further studies are warranted to elucidate the gene regulatory mechanisms of AT Campylobacter spp. in response to environmental stress.

Mechanism of persistence of aerotolerant Campylobacter spp.

C. jejuni devolves into viable but non-culturable (VNBC) stage and forms biofilms upon aerobic stress (Mouftah et al., 2021; Oh et al., 2015b; Yagi et al., 2022). Yagi et al. (2022) reported that the AT C. jejuni strain remained in VBNC stage for a longer period than the aero-sensitive strains under aerobic stress. The researchers also postulated that AT strains might enter the VBNC stage more slowly than the aero-sensitive strains under aerobic conditions (Yagi et al., 2022). This might be due to the AT strains resisting aerobic stress more, and the longer VBNC stage implies greater environmental persistence. In addition, Oh et al. (2015b) postulated that AT C. jejuni enters the VBNC stage after 12 h of aerobic incubation. However, the research did not compare to the aero-sensitive strain.

Aerotolerance was found to have a relationship with biofilm formation potential in C. jejuni and C. coli, with HAT isolates showing significantly greater biofilm formation potential than AT and OS isolates at 42°C under microaerobic conditions (Mouftah et al., 2021). On the contrary, Pokhrel et al. (2024) reported that different aerotolerance levels yielded no significant difference in the number of C. jejuni biofilms attached to stainless steel coupons at 42°C, regardless of the microaerobic and aerobic conditions. However, at room temperature, a significantly lower number of C. jejuni biofilm attached to stainless steel coupons was observed in HAT C. jejuni than AT and OS counterparts (Pokhrel et al., 2024). This suggests that incubation temperature could influence the formation of biofilms. In addition, Ortega-Sanz et al. (2024) reported that significantly higher biofilm formation was observed in HAT C. jejuni than in AT strains under aerobic conditions on polystyrene, but no significant difference was observed on stainless steel coupons. Thus, the type of material might influence the formation of biofilm. Further, AT C. jejuni Bf strain developed similar biofilm volume and thickness under both aerobic and microaerobic conditions (Bronnec et al., 2016b). However, under aerobic incubation, a more compact and structured biofilm was formed, creating a microaerobic niche that supported the growth (Bronnec et al., 2016b).

HAT and AT C. jejuni are found to have higher superoxide dismutase and catalase activities (the two important enzymes in oxidative stress defense) than OS strains (Oh et al., 2019). C. jejuni possesses both surface polysaccharides, capsular polysaccharide (CPS) and lipopolysaccharide (LPS), on the outer membrane (Jeon et al., 2009). Under aerobic conditions, the carbon metabolism of C. jejuni increased, leading to elevated amino acid uptake (Kim et al., 2023). This stimulated the formation of thicker CPS and LPS, which potentially function as a permeability barrier, protecting the pathogen from excess oxygen in aerobic atmosphere (Kim et al., 2023). Thus, the comparison of the structure of CPS and LPS among aerotolerance levels could be interesting. Further, the relationship between aerotolerance, VBNC, biofilm, oxidative stress defense enzymes, and protective polysaccharides of Campylobacter spp. remains scanty.

Genetic relatedness of aerotolerant Campylobacter spp. strains

Discriminatory typing methods for Campylobacter spp. are important for improving the understanding of epidemiology and genetic background (Nielsen et al., 2010). Multilocus sequence typing (MLST) is a valuable typing tool for discriminating Campylobacter spp. isolates and defining population structure (Dingle et al., 2001, 2005). The ST obtained from MLST can be grouped into CC based on relatedness (Nielsen et al., 2010). Table S9 shows the summary of eight studies on the classification of CC and ST based on aerotolerance level in C. jejuni and C. coli. The aerotolerance level comparison between ST was not made for Kiatsomphob et al. (2019), as the study did not disclose ST for each CC. Data from Table S9 are regrouped based on the respective CC and ST of C. jejuni and C. coli with aerotolerance levels, and are shown in Figure 7. Data from certain studies that did not classify CC to aerotolerance were further analyzed and classified based on the data from their published articles and supplementary documents (Guk et al., 2019, 2021; Jaakkonen et al., 2020; Kiatsomphob et al., 2019).

Figure 7 Radial graphs of CC and ST of (A) OS C. jejuni, (B) AT C. jejuni, (C) HAT C. jejuni, (D) OS C. coli, (E) AT C. coli, and (F) HAT C. coli. The frequency of ST is in bold face below the respective ST. CC: clonal complex; ST: sequence type; UA: unassigned to any CC; ND: not determined; OS: aero-sensitive; AT: aerotolerant; HAT: hyperaerotolerant.

Ordinal logistic regression analysis of 364 Campylobacter spp. isolates (input data from Table S9) showed that ST had a statistically significant (P < 0.001) influence on aerotolerance level. It has been postulated that the ST of Campylobacter spp. is related to aerotolerance levels. As shown in Figure 7, CC-353 showed the highest prevalence in HAT C. jejuni, while CC-21 showed the highest prevalence in AT C. jejuni. CC-353 was associated with a very high level of resistance to erythromycin (Jehanne et al., 2025). Further, CC-21 was the most prevalent in human and poultry slaughterhouse (Jeong et al., 2025).

In C. coli, CC-828 was the most prevalent CC in all three classes of aerotolerance (OS, AT, and HAT) strains, as shown in Figure 7. The most prevalent C. coli ST among both HAT and AT in the study of Guk et al. (2019) was ST-855, and this ST was not detected in OS strains. In contrast, the highest prevalent ST (ST-1068) among HAT strains in the study conducted by Guk et al. (2021) was detected in OS strains. Further, MLST data of C. coli isolates from ducks were compared to genetic relatedness to human sources in PubMLST and NCBI databases (Guk et al., 2019). Eight ST out of 18 from duck isolates were identified and found to be shared with human sources (Guk et al., 2019). HAT C. coli had a significantly higher proportion among shared ST than non-shared ST, while OS showed the opposite trend (Guk et al., 2019). In addition, a higher proportion of genetic relatedness was also reported between swine-derived HAT C. coli and human C. coli isolates, than OS strains (Guk et al., 2021).

Several studies reported a high proportion of shared STs in Campylobacter spp. between food-producing animals and human clinical cases (Asakura et al., 2019; Litrup et al., 2007; Ramonaite et al., 2017). Findings of Ortega-Sanz et al. (2024), Kiatsomphob et al. (2019), and Rodrigues et al. (2015), as summarized in Table S9, showed that HAT and AT strains were highly prevalent in human campylobacteriosis cases. Thus, the relationship between aerotolerance level and the overlap ST between food-producing animals and human clinical cases will be an interesting research field.

Risk of aerotolerant Campylobacter spp. to humans

In view of the enhanced resistance of AT Campylobacter spp. against multiple environmental stresses, such as atmospheric stress, disinfectant exposure, freeze-thaw, refrigeration, heat treatment, and osmotic stress, the AT Campylobacter spp. strains might survive better throughout the food chain. This would, in turn, augment foodborne transmission to humans (Guk et al., 2019; Mouftah et al., 2021).

Mihaljevic et al. (2007) found that 5 h aerobic exposure of C. jejuni significantly elevated invasion capacity in the Caco-2 cells model, suggesting that normal atmospheric conditions might be associated with the pathogenicity of pathogen. The AT Campylobacter spp. might have better oxidative stress resistance and could become more pathogenic, as the pathogen might be less susceptible to the free radicals produced by host immune response (Atack and Kelly, 2009). Furthermore, the aerotolerance of Campylobacter spp. is to some extent related to oxidative stress resistance and colonization factors in chickens, which contribute to their high prevalence in food as well as their elevated pathogenicity in human infections (Bolton, 2015; Flint et al., 2016; Hermans et al., 2011).

The risk of AT Campylobacter spp. to human health can be higher than that of the normal Campylobacter spp. evaluated in the past, as AT strains are commonly present in food. Thus, stricter regulations could be implemented for AT Campylobacter spp., as in the case of E. coli O157:H7 in the pathogenic E. coli strain.

Antibiotic resistance in aerotolerant Campylobacter spp.

To date, studies relating the aerotolerance profile of Campylobacter spp. to the antibiotic resistance profile are scarce, with no study conducted on C. jejuni. No significant differences were found between the antibiotic resistance rates and aerotolerance profiles in 56 C. coli isolates (Guk et al., 2019). However, 25% (7/28) of HAT and 9.1% (2/22) of AT C. coli isolates were highly resistant to ciprofloxacin (≥32 µg/mL), while none was from OS isolates (Guk et al., 2019). Similarly, in another study conducted by the same research group on pig feces, HAT C. coli showed the highest proportion (18.8%) in high-level ciprofloxacin resistance (Guk et al., 2021). Thus, AT strains might be more difficult to treat and place a greater medical burden because of higher antibiotic resistance. However, more studies are warranted to elucidate the relationship between aerotolerance and antibiotic resistance profiles.

Prevalence of virulence genes in aerotolerant Campylobacter spp.

Similar to the antibiotic-resistant profile, only a few studies linked virulence genes to the aerotolerance of Campylobacter spp. HAT C. coli isolates were reported to possess all 10 virulence genes analyzed, such as flaA (motility), flhB (motility), cadF (adhesion), pldA (adhesion), iamA (invasion), ceuE (invasion), cdtA (cytoxin), wlaN (Guillain–Barré syndrome), hcp (type VI secretion system), and virB11 (type IV secretion system), with the prevalence ranging from 3.6% to 100% (Guk et al., 2019, 2021). As compared to HAT C. coli, only six virulence genes were reported in OS C. coli, whereas seven virulence genes were found in AT C. coli (Guk et al., 2019). However, the prevalence of virulence genes was not significantly different between OS, AT, and HAT isolates (Guk et al., 2019, 2021).

Conversely, among 70 C. jejuni isolates, HAT C. jejuni strains depicted significantly higher frequencies of virulence genes than OS strains, suggesting HAT strains could be more pathogenic to humans than their OS counterparts (Oh et al., 2017). Similarly, Rodrigues et al. (2018) reported that AT C. jejuni Bf strain is one of the most virulent strains among the 10 strains tested. However, the study did not relate the aerotolerance levels of strains to the level of virulence.

The gene involved in the aerobic survival of Campylobacter spp. can also contribute to virulence. For instance, sodB, which is important for C. coli survival under aerobic conditions, plays an important role in colonizing the chick (Purdy et al., 1999). In the context of the virulence genes expression, such as cadF, cdtB (cytotoxin), ciaB (invasion), and clpP (stress tolerance), no significant difference was discovered in the expression between two AT C. jejuni strains and one OS C. jejuni strain (Lee et al., 2019). Therefore, the prevalence and expression of virulence genes in relation to aerotolerance level warrants further study to improve our understanding. AT strains with higher virulence genes might increase pathogenicity and thus contribute to higher campylobacteriosis cases.

Conclusion and the Way Forward

Research has increasingly demonstrated that AT Campylobacter spp. are prevalent in retail food, exhibiting increased pathogenicity and resistance to multiple environmental stresses. AT Campylobacter spp. might bring immense burdens to the economy and health care. The persistence of AT Campylobacter spp. is associated with a battery of survival mechanisms against various environmental stresses. Based on the current findings, it will not be surprising that the term ‘aerobic microorganism’ could be used to describe Campylobacter spp. in near future. This demonstrated the dire need to bolster our understanding of the relationship between survivability and the associated mechanisms in order to develop a holistic preventive measure to curtail this emerging pathogen.

To translate the available findings into real-world food safety practices, a surveillance program for AT Campylobacter spp. in the food chain is pivotal, and policymakers should urgently evaluate and revise the current protocol for public health. From that, modified atmospheric packaging, better hygienic practices, and greater coverage of cold chain transportation could be corrective measures to be considered.

In order to better contribute to future research in AT Campylobacter spp., several research gaps could be underpinned as illustrated in Figure S3 and stated as follows:

• Standardized protocols for aerotolerance level classification and aerobic acclimatization.

• Prevalence study of aerotolerance level and aerobic growth involving a larger sample size, a variety of foods, and different geographical regions.

• Elucidation on other clinically important Campylobacter spp.

• Survivability of single-strain, multiple-strain of the same species and multiple species against environmental stresses in vitro and in situ.

• Identification of ST of the strains and relating it to the source of the isolate and aerotolerance level.

• Phenotypic mechanism for aerotolerance, aerobic acclimatization, and aerobic growth.

• Antibiotic resistance profile on a variety of antibiotics with different doses and the prevalence and expression of virulence genes.

• Gene regulatory mechanism between aerotolerance, antibiotic resistance, and virulence.

• Characterization and assessment of health burden via risk assessment for better policy implementation.