The effect of environmental conditions on the occurrence of Campylobacter jejuni and Campylobacter coli in wastewater and surface waters

Abstract Aims The purpose of the study was to evaluate the occurrence of Campylobacter jejuni and Campylobacter coli in the aquatic environment based on the water origin, seasonality and physico‐chemical properties. Methods and Results The occurrence of C. jejuni and C. coli was determined in waste (29) or surface (56) waters in four different seasons. The air and water temperatures were measured during sampling and chemical analyses of water samples for ammonium, chloride, chlorine, nitrite, nitrate, phosphate and iron were performed. The thermotolerant Campylobacter spp. were more frequently detected in wastewater (59%; 17 positive samples) compared to surface water (38%; 21 positive samples), with the highest rate in autumn (67% of samples positive) and with a higher C. coli occurrence than C. jejuni (31% vs. 26%). Ammonium (above 0.2 mg/L) and chloride ion concentrations (above 60 mg/L) favour C. jejuni. Similarly, C. coli occurrence in water was supported by ammonium (above 0.2 mg/L), chloride (above 60 mg/L) and in addition by phosphate ion concentrations (below 0.7 mg/L). Conclusions Campylobacter presence in water is influenced by physico‐chemical parameters such as concentrations of ammonium and chloride ions. Significance and Impact of the Study Water environment is an alternative source of Campylobacter. The concentration of ammonium and chloride ions can be used as a basis for successful prediction of the potential occurrence of C. jejuni and C. coli in wastewater and surface water in future.

The environmental conditions, including the properties of water and weather conditions, can influence the presence, survival, shape or even internal factors of Campylobacter spp. The most common currently discussed internal factor is biofilm formation (Chmielewski & Frank, 2003;Donlan & Costerton, 2002;Teren et al., 2019). Currently, the effects of external conditions on internal factors were described in C. jejuni and C. coli isolates obtained from a wastewater treatment plant. These isolates formed more compact biofilms than isolates from surface waters or isolates from samples of human, food and animal origin (Shagieva et al., 2020).
The main reservoirs of Campylobacter spp. are considered poultry and wild birds. Campylobacter jejuni and C. coli are frequently detected as commensal bacteria in broilers, hens (Berndtson et al., 1996) and the prevalence of C. jejuni in poultry flocks can rise from less than 5% to more than 95% within 7 days (Hartnett et al., 2001). Once poultry are colonized with Campylobacter spp., they excrete large numbers of bacteria in their faeces. In this regard, young birds are more susceptible to Campylobacter infection. Foodproducing animals, such as pigs and cows, and wildlife such as deer plus other ruminants, may also act as reservoirs of C. jejuni and C. coli (Pattis et al., 2017). Campylobacteriosis accounts for an 160 million human cases around the world per year (Kirk et al., 2015). Therefore, C. jejuni and C. coli may also be present in other reservoirs.
Currently, there is limited knowledge about Campylobacter spp. survival in the aquatic environment. There are several studies that confirmed that water has frequently been identified as Campylobacter sources and are often associated with heavy rainfall and intrusion of contaminated waters (Bartholomew et al., 2014;Braeye et al., 2015;Gaardbo Kuhn et al., 2017;Gilpin et al., 2020;Hyllestad et al., 2020;Mouly et al., 2016). What is certain is that C. jejuni and C. coli can be found in water springs, rivers, ponds and lakes (Nachamkin et al., 2008). Therefore, human infections might just as well be caused by accidental ingestion of untreated surface water, e.g., during swimming (Sales-Ortells et al., 2015;Schönberg-Norio et al., 2004).
Hence, our focus here has been to investigate the occurrence of C. jejuni and C. coli in wastewater and surface water, then correlate our findings with environmental conditions, and eventually to determine factors that might influence Campylobacter spp. viability in water.

Sample collection
Water sampling was performed according to the EN ISO 19458 standard procedure (Mughini-Gras et al., 2016). One litre of water was collected and weather conditions (sunny, rainy, etc.) during sampling were recorded. Of 100 samples, 15 duplicates were excluded and the remaining 85 samples were used for analysis, of which 29 samples originated from two different municipal wastewater treatment plants and 56 samples from surface waters, mainly ponds and lakes in 37 different locations of the Czech Republic (Table S1). Wastewater samples (n = 29) were collected from the final output of municipal wastewater treatment plants in two locations (A and B) with 16 and 13 samples, respectively ( Figure  S1). All samples were divided into four groups based on 4 year seasons. Samples were collected in spring (n = 30), summer (n = 30), autumn (n = 18) and winter (n = 7). Spring was defined as the period from March 21st till June 20th, summer from June 21st till September 22nd, autumn from September 23rd till December 20th and winter from December 21st till March 20th. Surface water samples were collected from 37 locations ( Figure S1). From these, eight locations were selected for repetitive sampling (2-6 times) to cover all seasons. Together with water sampling, the air and water temperatures were measured and the samples were transferred to the laboratory in a cooling box for additional analyses. The temperature of air at the collection sites varied from 2 to 30℃, with the average yearly temperature 15.5℃. Detailed average temperatures of air and water in all seasons are presented in Supplementary data (Table S1). The temperature of all water samples varied from 5 to 30℃ with the average yearly temperature 16.5℃.

Chemical parameters of water samples
Sample pH values were measured by pH-Fix PT 4.5-10 (Macherey Nagel). Chemical analyses of water samples for ammonium, chloride, chlorine, nitrite, nitrate, phosphate and iron were performed semi-quantitatively by Visocolor Eco kits

Detection of Campylobacter spp. in poorly filterable water samples by the standard cultivation method
The standard ISO 17995 (Water quality-Detection and enumeration of thermotolerant Campylobacter spp.) (International Organization for Standardisation, 2019) was used for the detection of thermotolerant Campylobacter spp. with a slight modification (Strakova et al., 2021). Briefly, prefilters with a pore size of 1.4 µm (glass filter; Macherey Nagel) were used to remove mechanical particles from water before sample filtration. Thereafter, water samples (500 ml) were filtered (0.22 µm, mixed cellulose ester filter; Millipore Sigma), and the filters were transferred into two selective broths (Preston and Bolton broth) for enrichment and incubated at 42℃ in an anaerostat (AnaeroJar; Oxoid) under a microaerobic atmosphere (CampyGen 3.5 L; Oxoid). After 44 ± 4 h of incubation, inoculi were cultivated on Campylobacter blood-free selective agar (mCCDA; Modified charcoal-cefoperazone-deoxycholate agar) and incubated for another 44 ± 4 h under a microaerobic atmosphere at 42℃, followed by the isolation of  presumptive colonies on a nonselective blood agar and species level identification.

Campylobacter species identification
Suspect Campylobacter spp. colonies (n = 3) on blood agar were identified by multiplex PCR and matrix-assisted laser desorption/ionisation time of flight mass spectrometry (MALDI-TOF/MS) (Strakova et al., 2021). Briefly, bacterial DNA was extracted by thermal lysis and PCR was performed using the PPP master mix (Top-Bio) with primers (Generi Biotech) previously described for the detection of C. jejuni and C. coli (Bang et al., 2002;Linton et al., 1997;Winters et al., 1998). MALDI-TOF/MS (Autoflex speed TOF/TOF; Bruker) was used for confirmation of C. jejuni and C. coli identification by spectral comparison with the MALDI Biotyper library (MBT 8468; Bruker).

Statistical analysis
For statistical analysis, absolute and relative frequencies, median, and percentiles were used to describe categorical and continuous parameters (e.g., place and season sampling, water type, weather conditions, air and water temperature, pH, and 7 ion concentrations), respectively. Crude odds ratios (OR) accompanied by 95% confidence intervals (CI) were obtained by univariate logistic regression models with the aim of assessing the risk of C. jejuni and C. coli occurrence in water samples, given different weather conditions (e.g., sunny, rainy, after rain, cloudy) and ion concentrations. Univariate analyses were followed by multivariate logistic regression with backward elimination to obtain adjusted ORs. Decision trees for the best classification and prediction of C. jejuni and C. coli occurrence were built using the CHAID growing method. Due to the relatively low number of samples, the data were not divided into training and testing datasets. All data of collected samples were used for creation of trees and crossover validation was applied.

The occurrence of C. jejuni and C. coli in wastewater
Seventeen (58.6%) wastewater samples were positive for Campylobacter spp. (Figure 1), and positive detection in samples from both localities was similar, 56% in location A and 62% in location B (Table S1). In detail, 10 (12%) C. jejuni and 15 (18%) C. coli strains were isolated from 17 samples. Moreover, 7 (8%) were found double positive for both Campylobacter species (Figure 1B-dark column, Table 1upper part).

The occurrence of C. jejuni and C. coli in surface water
In total, 56 surface water samples were collected from 37 locations ( Figure S1). to cover all seasons. From these, several locations were selected for repetitive sampling. Samples from 14 locations (38%) were positive for Campylobacter spp. Almost 38% of all surface water samples (n = 21) were positive for Campylobacter spp. (Figure 1). In general, the presence of Campylobacter spp. was decreased in surface water when compared to wastewater samples (even 58.6%) (Figure 1a). In detail, 12 (14%) of the surface water samples were positive for C. jejuni and similarly 11 (13%) were  Table 1).

Seasonality of thermotolerant Campylobacter spp. detected in the water environment
In general, the highest air and water temperatures were in summer ( Figure S1). Campylobacter detection was maximal in surface water in autumn. Two thirds (67%) of autumn samples were Campylobacter positive, while in the other seasons only about 40% were positive (Figure 2a). Indiscriminately of water origin our results have shown that the maximal occurrence of C. jejuni was in spring and winter, while the maximal occurrence of C. coli was in autumn (Figure 2b). The air and water temperature in autumn had no effect on the presence of C. jejuni and C. coli in autumn ( Figure S1). Thereafter, we monitored C. jejuni and C. coli over four seasons studying wastewater and surface water samples separately (Figure 3). The acquired data were compared by statistical analysis.

Wastewater seasonality
Wastewater samples were collected during the whole year with a minimum of sample in winter (4) and a maximum in autumn (10). The highest and the lowest temperature of air was reported in summer and in winter, respectively ( Figure  S1). The temperature of wastewater decreased from summer to autumn, followed by spring and winter. About 75% of summer samples and 70% of autumn-collected wastewater samples were found C. jejuni or C. coli positive. The lowest number of positive samples was detected in spring (28.6%) (Table 1-lower part).

Surface water seasonality
Unlike wastewater sampling, the number of collected surface water samples differed through the year given that surface water freezes in winter. Surface water samples taken in autumn were most likely to be positive (62.5%), followed by spring (43.5%) and winter (33.3%), while the lowest percentage of positive samples was detected in summer (22.7%) ( Table 1).

Seasonality of the particular species C. jejuni and C. coli
A general comparison of two Campylobacter species occurrence in the wastewater and surface water samples was then made (Figure 3). Both Campylobacter spp. were more often present in wastewater (Figure 3; dark column) than in surface water (Figure 3; light column). Campylobacter coli was generally more abundant in wastewater than C. jejuni. C. coli was also detected in about 60% of samples collected in autumn. In addition, C. jejuni strains were only isolated from surface water samples in spring and in summer.

The influence of water properties on the occurrence of C. jejuni and C. coli
Finally, the occurrence of C. jejuni and C. coli in wastewater and surface water was studied in relationship to weather, including two physical parameters (temperature of air and water) and nine environmental chemical properties of all water samples (Table S1). Univariate analyses showed that the risk of Campylobacter spp. occurrence was higher in sunny weather with increasing air temperature and with increasing concentrations of ammonium (NH 4 + ), nitrite (NO 2 − ) or iron (Table 2). Moreover, multivariate analyses showed that the risk of Campylobacter occurrence was increased 4.7 times in wastewater compared to surface water (OR 4.68, 95% CI 1.13-19.41, p = 0.034) and that Campylobacter risk was higher in water with higher ammonium (NH 4 + ) concentrations (OR 1.47, 95% CI 1.01-2.15, p = 0.044) ( Table 2). Enhanced phosphate concentrations (OR 0.2, 95% CI 0.08-0.74, p = 0.013) by 1 mg/L decreased the risk of Campylobacter spp. occurrence by 76% (Table 2). Other parameters were statistically nonsignificant in the multivariate analysis.  (Table  S2). Otherwise, univariate and multivariate analyses showed that physico-chemical parameters remained statistically nonsignificant for the risk of C. coli occurrence (Table S2).

The prediction of the occurrence of Campylobacter spp. in water samples
Based on the analysis of ion concentrations, we attempted to predict the presence of Campylobacter spp. in water.
Our results showed that the concentration of ammonium salts increased above 0.2 mg/L supported the occurrence of Campylobacter spp. On the other hand, if the concentration of ammonium salts was below 0.2 mg/L, then the air temperature below 18℃ could support the occurrence of Campylobacter spp. Based on these two parameters, the total prediction success was 71.8% (Figure 4). Similarly, the parameters were used for prediction of C. jejuni ( Figure 5). Higher concentrations of chloride over 60 mg/L, together with higher concentrations of ammonium salts over 0.2 mg/L, increased the risk of C. jejuni occurrence. The total success of prediction of C. jejuni occurrence in water was 80.0%, based on the presence of chloride and ammonium salts ( Figure 5).
For the prediction of C. coli presence ( Figure 6) in water, the spectrum of important ions was similar. Lower concentrations of ammonium ions resulted in C. coli absence. Higher concentrations of chloride over 60 mg/L, together with higher concentrations of ammonium ions over 0.2 mg/L, and lower concentrations of phosphate ions below 0.7 mg/L, increased the risk of C. coli (Figure 6) occurrence. The total prediction success of C. coli occurrence was 85.9% based on the chloride, ammonium and phosphate concentration in water samples ( Figure 6).

DISCUSSION
Campylobacteriosis is usually connected to the consumption of undercooked poultry meat or raw milk. The aquatic environment is less well-described as an alternative source of C. jejuni and C. coli. Therefore, our approach here consisted of determining how the occurrence of C. jejuni and C. coli varied in both wastewater and surface water samples, according to origin and seasonal variability. The two thermotolerant Campylobacter spp. were present in almost 60% of the wastewater samples despite the fact that all tested samples were collected at the final output of municipal wastewater treatment plants. In general, surface water samples were less positive for C. jejuni and C. coli than wastewater samples, in accordance with previous findings (Mulder et al., 2020). The strains isolated from wastewater might well originate from the human population because of their excretion from Campylobacter-positive people, especially during late spring, summer and early autumn. On the other hand, surface waters can be contaminated through wild birds or via sewage from farms. When looking at the particular Campylobacter species, C. coli was more abundant than C. jejuni in water of both origins. These results correlate with the results of a current study from the Netherlands (Mulder et al., 2020). This outcome may be due to its generally higher prevalence in the environment, or due to the higher sensitivity of C. jejuni to the physico-chemical characteristics of the environment (Obiri-Danso et al., 2001).
Several factors, such as the temperature of the water or air, that might contribute to the increased occurrence of Campylobacter spp. in water have been considered previously (Cook & Bolster, 2007;Rollins & Colwell, 1986;Thomas et al., 2002). According to our data, we conclude that autumn is the major season for C. jejuni and C. coli occurrence in water. Such Campylobacter seasonality is also in agreement with the reports of others (Mulder et al., 2020;Nygård et al., 2004). Campylobacter spp. can survive for longer periods in water sources during winter, because they are sensitive to UV F I G U R E 5 Decision tree for the occurrence of Campylobacter jejuni based on ion concentrations light and higher temperatures, so bacterial populations may be decimated during the summer (Obiri-Danso et al., 2001). On the other hand, during the summer period, animals and birds may act as reservoirs for recontamination of water sources, which correlates with the increased incidence of campylobacters between June and November (Nygård et al., 2004). Indeed, autumn but also summer have been shown to be seasons with a higher prevalence of human campylobacteriosis (The European Union One Health 2018 Zoonoses Report, 2019). The enhancement of human campylobacteriosis might also be due to increased consumption of cross-contaminated food in the summer and autumn during frequent barbecuing, picnics, etc. (Gölz et al., 2014). Therefore, another future study should be performed to compare Campylobacter tolerance and sensitivity to temperature, pH, light and oxygen to explain Campylobacter water seasonality. We hypothesized that water composition such as ion concentrations might affect the occurrence of C. jejuni and C. coli. They are able to use nitrate (NO 2 − ) and nitrite (NO 3 − ) as terminal electron acceptors for growth under severe oxygen-limited conditions (Pittman et al., 2007). Therefore, Campylobacter spp. produce different oxidases and reductases, particularly fumarate reductase, DMSO reductase, nitrite reductase and nitrate reductase, which allow bacteria to grow under low oxygen conditions (Smith et al., 2000). The nitric oxide radical (NO) can arise from nitrogenous species such as nitrite (NO 2 − ), nitrate (NO 3 − ) or from ammonium ions (NH 4 + ) (Pittman et al., 2007), that are the subjects of our study. Ammonium can be later degraded by oxygenation to nitrite (NO 2 − ) and further to nitrate (NO 3 − ). A nitrate-rich diet significantly elevates the exhaled nitric oxide radical (Olin et al., 2001), which is encountered by Campylobacter spp. and may influence the outcome of infections. According to univariate analyses of nitrogen-containing salts, increased levels of ammonium (NH 4 + ) and nitrite (NO 2 − ) ions support the presence of Campylobacter spp. in water. Moreover, nitrite (NO 2 − ) increased the risk of C. jejuni contamination confirming the results of another study where a high concentration of dissolved nitrogen increased the survival of C. jejuni (Cook & Bolster, 2007). In accordance with the literature, seasonal variabilities in concentrations of nitrite (NO 2 − ) salts were not observed (Schullehner et al., 2017). However, our multivariate analyses that compared ions mutually supported the proposition that ammonium salts significantly increase the risk of Campylobacter contamination in water.

| 733
CAMPYLOBACTERS AND CONDITIONS OF WATER ENVIRONMENT increased ammonium salt concentrations above 0.2 mg/L promoted Campylobacter occurrence. It is assumed that iron is an essential element for Campylobacter spp. and therefore that iron level could influence Campylobacter in water. Surprisingly, the impact of iron concentration on the presence of C. jejuni and C. coli in water was not detected either.
Low air temperatures (below 18℃) were also shown to support the occurrence of Campylobacter spp. in water when concentrations of ammonium salts were below 0.2 mg/L. Moreover, in the case of C. coli, lower concentrations of phosphate salts also correlate with the occurrence of these species in water. The total prediction success for C. jejuni and C. coli was 80% and 86%, respectively. In view of this fact, further studies will be needed for validation of our data and their wider implementation.
Despite the fact that Campylobacter spp. are among the most demanding bacteria requiring specific conditions when cultivated in the laboratory, they are able to survive and multiply in the aquatic environment. We can summarize that the occurrence of C. jejuni and C. coli in the aquatic environment was influenced by the origin, seasonality and physico-chemical properties of waters. In detail, our study showed that wastewaters, ponds and lakes are the reservoirs of C. jejuni and C. coli. In addition, 9 (11%) samples were double positive for both species but the occurrence of C. coli was higher (31%) than C. jejuni (26%). Despite globally, human campylobacteriosis is caused by thermophilic C. jejuni (>75%) and, to a lesser extent, C. coli (Igwaran & Okoh, 2019). It can be explained thus that C. coli is more commonly associated with waterfowl and environmental sources that have, presumably, been contaminated by them (Sheppard et al., 2009;Sheppard & Maiden, 2015). The yearround occurrence of Campylobacter spp. in water peaking in the autumn demonstrated that C. jejuni and C. coli are able to survive temperatures around 8℃ and to aerobic conditions. One health approach suggests that the implementation of efficient regular Campylobacter control measures not only in food production but even in water environment, such as in wastewaster treatment plants and surface waters, can help to minimize the occurrence of Campylobacter spp. and thereby it can enhance public health safety.
In conclusion, the occurrence of Campylobacter in water is influenced by physico-chemical parameters. Some of these, e.g., the concentration of ammonium and chloride ions, can be used as a basis for successful prediction of the potential occurrence of C. jejuni and C. coli in wastewater and surface water in future.