Reduction of Arcobacter at Two Conventional Wastewater Treatment Plants in Southern Arizona, USA

This study aimed to identify the bacterial community in two wastewater treatment plants (WWTPs) and to determine the occurrence and reduction of Arcobacter, along with virulence genes (ciaB and pldA). A total of 48 samples (24 influent and 24 effluent) were collected at two WWTPs in southern Arizona in the United States, monthly from August 2011 to July 2012. Bacterial DNA extract was utilized for 16S rRNA metagenomic sequencing. Quantification of Arcobacter 16S rRNA gene was conducted using a recently developed SYBR Green-based quantitative PCR assay. Among 847 genera identified, 113 (13%) were identified as potentially pathogenic bacteria. Arcobacter 16S rRNA gene was detected in all influent samples and ten (83%) and nine (75%) effluent samples at each plant, respectively. Log reduction ratios of Arcobacter 16S rRNA gene in Plant A and Plant B were 1.7 ± 0.9 (n = 10) and 2.3 ± 1.5 (n = 9), respectively. The ciaB gene was detected by quantitative PCR in eleven (92%) and twelve (100%) of 12 influent samples from Plant A and Plant B, respectively, while the pldA gene was detected in eight (67%) and six (50%) influent samples from Plant A and Plant B, respectively. The prevalence of potentially pathogenic bacteria in WWTP effluent indicated the need for disinfection before discharge into the environment.


Introduction
Wastewater treatment plants (WWTPs) collect and treat wastewater. Water reclaimed after treatment can be utilized for various purposes, including agricultural irrigation [1,2], recreational purposes [3], to reduce pollution in rivers [4,5], and as a drinking water source [6], as a part of integrated and sustainable water resource management.
Arcobacter spp. were first detected in 1991 and are gram-negative, non-spore forming curved or helical rod-shaped bacteria of the family Campylobacteraceae [7,8]. Arcobacter has been detected at WWTPs in influent samples, whereas in effluent samples Bacillus, Pseudomonas, and Acinetobacter were observed to be in high abundances (33.2 ± 25.3%, 5.1 ± 17.3%, and 4.0 ± 11.8%, respectively). There was no significant difference in abundances of most of the potentially pathogenic bacteria between influent and effluent water samples. In Plant A and Plant B, 29 and 34 out of 42 potentially pathogenic bacteria did not show any significant differences in abundance between influent and effluent water samples, respectively (t-test, p > 0.05).

Occurrence of Total Bacteria, Arcobacter and Associated Virulence Genes in Wastewater Samples
The concentrations of total bacterial 16S rRNA gene in influent samples from both Plant A and Plant B were 9.7 ± 0.3 and 9.7 ± 0.4 log copies L −1 , respectively, and in effluent samples, the concentrations of total bacterial 16S rRNA gene were 8.0 ± 0.4 and 8.3 ± 0.2 log copies L −1 in Plant A and Plant B, respectively. The concentrations of Arcobacter 16S rRNA gene in wastewater samples collected from Plant A and Plant B are shown in Figure 2. Arcobacter was detected in all influent samples from both Plant A and Plant B and in ten (83%) and nine (75%) of twelve effluent samples at each plant. Average concentrations of Arcobacter 16S rRNA gene were 7.9 ± 0.7 and 8.5 ± 1.2 log copies L −1 in influent samples of Plant A and Plant B, respectively. Those in effluent samples were 6.4 ± 0.6 and 6.1 ± 0.9 log copies L −1 in Plant A and Plant B, respectively.
In our previous study, concentrations of total bacterial 16S rRNA gene obtained using qPCR and abundances of Arcobacter obtained using NGS analysis were utilized to compare results obtained from the two methodologies [31]. Here, concentrations of Arcobacter 16S rRNA gene were calculated similarly for both Plant A and Plant B. The correlation coefficient of Arcobacter 16S rRNA gene concentrations calculated by the two different methods were −0.95 and −0.99 for Plant A and Plant B, respectively.
The ciaB gene was detected in eleven (92%) and twelve (100%) of twelve influent samples collected from both Plant A and Plant B, respectively, while the pldA gene was detected in eight (67%) and six (50%) influent samples from Plant A and Plant B, respectively. Average concentrations of ciaB and pldA genes in influent samples were 7.3 ± 0.7 and 5.9 ± 0.2 log copies L −1 in Plant A, respectively, whereas those in influent samples were 7.8 ± 1.2 and 6.8 ± 0.8 log copies L −1 in Plant B, respectively ( Table 2).

Occurrence of Total Bacteria, Arcobacter and Associated Virulence Genes in Wastewater Samples
The concentrations of total bacterial 16S rRNA gene in influent samples from both Plant A and Plant B were 9.7 ± 0.3 and 9.7 ± 0.4 log copies L −1 , respectively, and in effluent samples, the concentrations of total bacterial 16S rRNA gene were 8.0 ± 0.4 and 8.3 ± 0.2 log copies L −1 in Plant A and Plant B, respectively. The concentrations of Arcobacter 16S rRNA gene in wastewater samples collected from Plant A and Plant B are shown in Figure 2. Arcobacter was detected in all influent samples from both Plant A and Plant B and in ten (83%) and nine (75%) of twelve effluent samples at each plant. Average concentrations of Arcobacter 16S rRNA gene were 7.9 ± 0.7 and 8.5 ± 1.2 log copies L −1 in influent samples of Plant A and Plant B, respectively. Those in effluent samples were 6.4 ± 0.6 and 6.1 ± 0.9 log copies L −1 in Plant A and Plant B, respectively.
In our previous study, concentrations of total bacterial 16S rRNA gene obtained using qPCR and abundances of Arcobacter obtained using NGS analysis were utilized to compare results obtained from the two methodologies [31]. Here, concentrations of Arcobacter 16S rRNA gene were calculated similarly for both Plant A and Plant B. The correlation coefficient of Arcobacter 16S rRNA gene concentrations calculated by the two different methods were −0.95 and −0.99 for Plant A and Plant B, respectively.
The ciaB gene was detected in eleven (92%) and twelve (100%) of twelve influent samples collected from both Plant A and Plant B, respectively, while the pldA gene was detected in eight (67%) and six (50%) influent samples from Plant A and Plant B, respectively. Average concentrations of ciaB and pldA genes in influent samples were 7.3 ± 0.7 and 5.9 ± 0.2 log copies L −1 in Plant A, respectively, whereas those in influent samples were 7.8 ± 1.2 and 6.8 ± 0.8 log copies L −1 in Plant B, respectively ( Table 2).

Reduction Ratios of Arcobacter and Virulence Genes During Wastewater Treatment
The log reduction ratios of 16S rRNA genes of total bacteria and Arcobacter in Plant A, where a conventional activated sludge process is utilized, were 1.6 ± 0.4 (n = 12) and 1.7 ± 0.9 (n = 10), respectively. Those of total bacteria and Arcobacter at Plant B, which utilizes a biological trickling filter process, were 1.5 ± 0.4 (n = 12) and 2.3 ± 1.5 (n = 9), respectively. For Arcobacter, the highest reduction ratios in Plant A and Plant B were obtained in March (3.1 log) and June (4.7 log), respectively. Even though the methodologies for the treatment of wastewater were different between locations, there was no significant difference in log reductions of total bacteria or Arcobacter between Plant A and Plant B. The log reduction ratio of the ciaB gene was 1.7 ± 1.0 (n = 8) and 2.1 ± 1.8 (n = 7) in Plant A and Plant B, respectively ( Table 3). The log reduction ratio of the pldA gene was 1.1 ± 0.1 (n = 2) in Plant A.

Reduction Ratios of Arcobacter and Virulence Genes During Wastewater Treatment
The log reduction ratios of 16S rRNA genes of total bacteria and Arcobacter in Plant A, where a conventional activated sludge process is utilized, were 1.6 ± 0.4 (n = 12) and 1.7 ± 0.9 (n = 10), respectively. Those of total bacteria and Arcobacter at Plant B, which utilizes a biological trickling filter process, were 1.5 ± 0.4 (n = 12) and 2.3 ± 1.5 (n = 9), respectively. For Arcobacter, the highest reduction ratios in Plant A and Plant B were obtained in March (3.1 log) and June (4.7 log), respectively. Even though the methodologies for the treatment of wastewater were different between locations, there was no significant difference in log reductions of total bacteria or Arcobacter between Plant A and Plant B. The log reduction ratio of the ciaB gene was 1.7 ± 1.0 (n = 8) and 2.1 ± 1.8 (n = 7) in Plant A and Plant B, respectively ( Table 3). The log reduction ratio of the pldA gene was 1.1 ± 0.1 (n = 2) in Plant A. Table 3. Reduction of total bacteria and Arcobacter at WWTPs.

Discussion
The overall diversity and abundance of bacterial genera were identified in influent and effluent of wastewater samples using NGS. The phylum Proteobacteria was the most abundant phylum, followed by Firmicutes, Bacteroidetes, and Actinobacteria in both WWTPs (Figure 1). These data are in agreement with results of previous studies that tested untreated sewage [27,28] and WWTP samples [29] in the United States. This is also in agreement with the study by Zhang et al. (2012), in which Proteobacteria were the most abundant phylum at 14 WWTPs in samples collected from Asia (China, Hong Kong, and Singapore) and North America (Canada and the United States) [32]. Average abundance of Bacillus in both plant locations was found to be highest in the effluent (among potentially pathogenic bacteria) ( Table 1). Lee et al. (2008) found that concentrations of Bacillus cereus were highest in final effluents (disinfected by chlorination and UV radiation) in two WWTPs in Canada [33]. The genus Pseudomonas was also abundant and has commonly been detected in wastewater samples collected in the United States [29]. Pseudomonas is ubiquitous in the environment, and pathogenic species of Pseudomonas can cause infections in hospital patients and/or those with weakened immune systems, such as pneumonia and blood infections [34]. The abundance of Mycobacterium significantly increased from influent to effluent samples in Plant A. This taxon has also been found in effluent and activated sludge of a WWTP in Hong Kong via metagenomic analysis [35,36].
The occurrence of potentially pathogenic bacteria in influent and effluent samples emphasizes their ability to persist and be discharged in the environment. WWTPs have been considered a potential hub for evolution and dissemination of antibiotic resistance and virulence genes [37]. Focusing on construction and maintenance of treatment plants, treatment methodologies, and disinfection processes before releasing treated wastewater into the environment will help reduce the spread of potentially pathogenic bacteria into the water environment.
Prevalence of Arcobacter has been found to be higher in wastewater compared to other aquatic environments, such as lakes, river, recreational beaches, groundwater, seawater, and drinking water [38]. This is also supported by metagenomic analysis of wastewater samples from various locations where Arcobacter has been detected as one of the most abundant genera [19,27,28]. Arcobacter is ubiquitous in the environment, and this taxon has been associated with both humans and animals and can cause gastroenteritis, septicemia, mastitis, reproductive disorders, and abortion in livestock [22,39]. Arcobacter was detected in all influent samples tested in the current study, and detection in the effluent of both plant locations indicates a high tolerance capability of Arcobacter, which can ultimately lead to persistence and spread of the pathogenic bacteria.
Wastewater samples tested in this study have been previously tested for viruses and protozoa. Most of the viruses and protozoa tested did not show any significant differences in log reductions between Plant A and Plant B [40,41]. Similarly, there was no significant difference in the reduction of Arcobacter in either Plant A or Plant B. The efficiencies of either plant, operating the activated sludge process or the biological trickling filter process, were not effective in removing pathogens; therefore, they must be improved. When less effectively treated water is used for agricultural or recreational uses, there is a high risk of contamination with Arcobacter, as one of the routes of transmission of this taxon is water. Seasonal variations, especially during extreme rainfall, can affect the transportation of Arcobacter from WWTPs to groundwater [38]. The concentration of Arcobacter was lowest in February but increased in March in Plant A and Plant B. In this study, sample collection for a year may not capture the seasonal variation in Arcobacter in WWTPs, and a longer survey period could help gain a better understanding of Arcobacter dynamics.
In this study, the ciaB gene was detected more frequently than the pldA gene in wastewater samples of both plant locations ( Table 2). This is likely due to a higher detection frequency for the ciaB gene in all Arcobacter strains compared to the pldA gene [42]. These virulence genes have been detected in Arcobacter strains obtained from a variety of species, including humans, chickens, pigs, cattle, sheep, horses, dogs, clams, mussels, and in milk [42][43][44]. These genes have also been identified from isolates of Arcobacter recovered from fecal samples originating from humans and from animals [45]. It is evident from these reports that these virulence genes are ubiquitous in the environment. The detection of ciaB and pldA genes in effluent samples may influence nearby environmental microbial communities, and the dispersal and fate of these genes in water environments suggests negative impacts associated with contaminated wastewater effluent on the gene content of water bodies [46]. The presence of potentially pathogenic organisms and their virulence genes in water demonstrates the importance of monitoring effluent water samples before release into the water bodies and the need for disinfection of the effluent before discharge.
In summary, the characterization of the bacterial community in two WWTPs via NGS detected 113 (13%) of 847 genera as potential pathogenic bacteria. In all influent samples of both plant locations, the Arcobacter 16S rRNA gene was quantified and detected in ten and nine of twelve effluent samples at Plant A and Plant B, respectively. There was no significant difference in reduction ratios of the Arcobacter 16S rRNA gene between the two plants. Virulence genes, ciaB and pldA, were also detected in both influent and effluent samples from Plant A and Plant B. The presence of potential pathogenic bacteria and quantification of Arcobacter and its virulence genes in effluent samples of WWTPs demonstrate the need for disinfection before discharge into the environment.

Collection of WWTP Samples
As described previously [40], during a 12-month period between August 2011 and July 2012, monthly sampling of influent and effluent wastewater was conducted at two WWTPs in southern Arizona. Wastewater samples from Plants A (conventional activated sludge process) and B (biological trickling filter process) were collected as grab samples. Each sampling was conducted at~10:00.

Bacterial DNA Extraction
Water samples (100 mL of influent and 1000 mL of effluent) were used to measure viral concentrations using a mixed cellulose ester membrane (pore size: 0.45 µm, diameter: 90 mm; Merck Millipore, Cat. No. HAWP-090-00, Billerica, MA, USA) as described previously [40]. The membrane filter was used for extraction of bacterial DNA after virus elution. In brief, the membrane filter was cut in half, and one piece was mixed with 10 mL of surfactant-based elution buffer in a 50-mL tube. The tube was vortexed vigorously for~5 min, and the eluate was transferred to a new tube. This step was repeated by adding 5 mL of the elution buffer to the original tube, resulting in~15 mL of eluate. The tube was centrifuged at 2000× g for 10 min at 4 • C, and the supernatant was removed. A volume of 200 µL phosphate buffered saline was added to the tube containing the pellet, mixed, and transferred into a new 2-mL microtube. This step was repeated until the final volume of the bacterial concentrate reached 1 mL. Bacterial DNA (200 µL) was extracted from 200 µL of the concentrate using a QIAamp DNA mini kit (QIAGEN, Hilden, Germany).

NGS for Characterization of Bacterial Communities
The bacterial DNA extracts of wastewater samples were used for metagenomic sequencing via a MiSeq gene sequencer (Illumina, San Diego, CA, USA) as described previously [47]. Operational taxonomic units obtained were analyzed based on the bacterial domain, phylum, family, and genus. A genus was considered as a potentially pathogenic bacterium if any one species of the genus was categorized as biosafety level 2 or 3 by the American Biological Safety Association (https: //my.absa.org/tiki-index.php?page=Riskgroups) as described previously [48]. The raw sequences obtained were registered in the NCBI Sequence Read Archive under the accession number PRJNA525124.

qPCR of Total Bacteria and Arcobacter
For total bacterial 16S rRNA gene, qPCR was performed using 515F and U806R primers [49,50] with the thermal conditions, qPCR mixture components, and qPCR reaction conditions as previously described [48]. Arcobacter 16S rRNA gene was quantified via qPCR using 2 µL of template DNA, 12.5 µL of a MightyAmp for Real Time (SYBR Plus) (Takara Bio, Kusatsu, Japan), 0.1 µL each of 1 µM of Arco-F and Arco-R-rev primers [31], and 10.3 µL of ultrapure water. For the virulence genes ciaB and pldA, 2 µL of template DNA, 12.5 µL of a SYBR Premix Ex Taq II (Tli RNase Plus) (Takara Bio), 0.1 µL each of 1 µM of ciaB-F and ciaB-R primers (for the ciaB gene) [42] or pldA-F and pldA-R primers (for the pldA gene) [42], and 10.3 µL of ultrapure water were used. qPCR was performed with a Thermal Cycler Dice Real Time System Single TP850 (Takara Bio) under the following thermal conditions: for Arcobacter 16S rRNA gene, 98 • C for 2 min, followed by 35 cycles at 98 • C for 10 s, 55 • C for 30 s, and 68 • C for 40 s; for the ciaB gene, 95 • C for 30 s, followed by 35 cycles at 94 • C for 15 s, 55 • C for 30 s, and 72 • C for 20 s; and for the pldA gene, 95 • C for 30s, followed by 35 cycles at 94 • C for 15 s, 56 • C for 45 s, and 72 • C for 20 s. A melting curve analysis was performed to confirm the generation of specific qPCR products.

Statistical Analysis
Student's t-tests were performed with Microsoft Excel 2018 (Microsoft Corporation, Redmond, WA, USA) to determine the difference in abundance of bacteria in influent and effluent wastewater samples. The test was also used to determine whether log reductions of total bacteria and Arcobacter at Plants A and B were statistically different. Differences were considered statistically significant if the resulting p value was <0.05. The log reduction of Arcobacter was calculated from the samples, which were qPCR-positive for both influent and effluent samples.