Environ. Sci. Technol. 2008, 42, 1191–1200
Development and Application of Real-Time PCR Assays for Quantifying Total and Aerolysin Gene-Containing Aeromonas in Source, Intermediate, and Finished Drinking Water CHANG-PING YU,† SARA K. FARRELL,‡ BRUCE ROBINSON,§ AND K U N G - H U I C H U * ,† Zachry Department of Civil Engineering, Texas A&M University, College Station, TX 77843-3136, The Center for Environmental Biotechnology and Department of Civil and Environmental Engineering, University of Tennessee, Knoxville, Tennessee 37996
Received June 6, 2007. Revised manuscript received October 21, 2007. Accepted November 26, 2007.
Aeromonas spp., opportunistic pathogens, are listed as a microbiological contaminant on the Environmental Protection Agency’s (EPA) Drinking Water Contaminant Candidate List. Culture-based methods for identification and quantification of Aeromonas in drinking water are time-consuming and often fail to differentiate pathogenic species from nonpathogenic ones. This study reports successful development and applications of two real-time PCR assays, based on 16S rRNA gene sequences and a virulence gene (aerolysin gene), for rapid and effective quantification of total and aerolysin genecontaining Aeromonas spp. The assays successfully quantified total and aerolysin gene-containing Aeromonas in source, intermediate, and finished water samples collected from seven water works and one pilot plant. The effectiveness of Aeromonas removal by different drinking water treatment processes was examined by comparing the results obtained from the EPA culture-based method and developed real-time PCR assays. Regardless of the methods, our results indicated that conventional water treatment combination (prechlorination/ coagulation/sedimentation/rapid sand filtration) and membrane filtration alone could effectively remove Aeromonas. Slow sandfiltrationalonemightnotbeeffective.Theremovalefficiencies by different disinfection treatments were not determined, due to the lack of detectable Aeromonas. No Aeromonas was detected in samples with turbidity below 0.06 NTU.
Introduction The genus Aeromonas is a group of bacteria that are ubiquitous in aquatic environments (1–3) and may also be detected in fish, food, bottled water, and drinking water * Corresponding author phone: (979) 845-1403; fax: (979) 8621542; e-mail.:
[email protected]. † Texas A &M University. ‡ The Center for Environmental Biotechnology, University of Tennessee. § Department of Civil and Environmental Engineering, University of Tennessee. 10.1021/es071341g CCC: $40.75
Published on Web 01/17/2008
2008 American Chemical Society
(4–12). Since some Aeromonas spp. can produce a range of biotoxins (such as hemolysins, cytotoxins, aerolysins, and even enterotoxins), they are considered as opportunistic waterborne pathogens responsible for acute gastroenteritis and wound infections in humans (13) and animals (14). In 1996, Aeromonas was listed as a microbiological contaminant on the Environmental Protection Agency’s (EPA) Drinking Water Contaminant Candidate List. Monitoring for Aeromonas was conducted by 120 large and 180 small public water systems (PWS, list 2) in 2003 by using EPA approved culture-based method, Method 1605 (Federal Register, 40 CFR Part 141, March 7, 2002). Like other culture-based methods, this method is time-consuming. Furthermore, this method might not be effective for identifying pathogenic species from nonpathogenic ones. Quantitative culture-independent methods, like real-time PCR, might offer a better measure for the quantification and monitoring purposes of pathogens in drinking water. Recently, real-time PCR has been developed for quantifying Escherichia coli O157:H7 and hepatitis B viruses in environmental samples (15–17). Although studies have been reported using real-time PCR for monitoring Aeromonas spp. in stools (18) and fish tissue (19), no studies have been reported using real-time PCR for monitoring Aeromonas spp. in drinking water, and/or other environmental samples. Data on Aeromonas removal by water treatment facilities is essential for regulatory agencies to recommend any specific treatment process or to impose a practical regulation on Aeromonas for safe drinking water. To address the concern of Aeromonas in drinking water, we developed and applied quantitative and sensitive molecular assays to quantify total Aeromonas spp. and virulence gene-containing Aeromonas spp. in source, intermediate, and finished water of several selected water utilities. The results from the EPA method and the molecular assays were compared and used to determine the effectiveness of various water treatment processes.
Materials and Methods Drinking Water Treatment Plants and Sampling Locations. To examine the effectiveness of Aeromonas removal by various drinking water treatment processes, water samples were collected from different treatment units of seven drinking water treatment plants (plants 1–7) and one pilot plant (pilot plant). Table 1 lists information of these plants, including treatment capacity, process flow diagrams of each plant, and sampling locations. Two sampling events were conducted between April 2004 and August 2005. For each sampling event, seven source water, 12 intermediate, and eight finished samples were collected. Procedures for sample collection, preservation, and storage were followed as described in EPA Method 1605, an approved culture-based method for Aeromonas spp. monitoring (20). Water quality parameters, such as pH and turbidity, were measured on site during samples collection (for turbidity see Supporting Information Table S1). For plants 1–4, 4 L of water samples were collected from each sample location. For plants 5–7 and pilot plant, 2 L of water samples were collected from each sampling location. The collected samples were packed with ice and transported to the laboratory for analysis within 24 h. Strains. Both Aeromonas (22 strains) and non-Aeromonas (19 strains) bacteria were used to validate the quantitative molecular assays developed in this study (Table 3). The Aeromonas bacteria include four strains purchased from the American Type Culture Collection (ATCC) (ATCC 7966T, ATCC 15468, ATCC 35993, and ATCC 49657) and eighteen strains isolated from water treatment plants in this study. VOL. 42, NO. 4, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1191
TABLE 1. Unit Operation and Sampling Locations of Drinking Water Treatment Plants Surveyed in this Study
TABLE 2. Primer Sets Designed for Real-Time PCR Assays assay target aerolysin gene Aeromonas 16S rRNA gene
primer name, sequence (5′-3′)
references
AHCF1 5′-GAGAAGGTGACCACCAAGAACA-3′ AHCR1M 5′-ARCTGACATCGGCCTTGAACTC-3′ Aer66f 5′-GCGGCAGCGGGAAAGTAG-3′ Aer613r 5′-GCTTTCACATCTAACTTATCCAAC-3′
30 this study 25 this study
The non-Aeromonas bacteria include one E. coli. purchased from ATCC (ATCC 25922), 15 estrogen-degraders isolated from activated sludge (21, 22), and three strains (E. coli, Pseudomonas fluorescens, and Bacillus thuringiensis) obtained from the University of Tennessee, Knoxville (23). Enumeration and Isolation of Culturable Aeromonas. Culturable Aeromonas bacteria were enumerated by using ampicillin-dextrin agar with vancomycin (ADA-V) according to EPA Method 1605 (20). Different volumes of water samples were used for filtration. For source and intermediate water, sample volumes of 100, 10, and 1 mL were used. For finished water, 500 mL was filtered. Duplicate samples were used. Eighteen Aeromonas strains were isolated. These isolates were 1192
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 4, 2008
further identified according to their 16S rRNA gene sequences (∼500 bp) that were amplified with the primers Aer66f and Aer613r (Table 2). These isolates were also examined for the presence of aerolysin gene. DNA Extraction. Water samples were filtered through 0.2 µm pore-size polycarbonate membrane filters (Whatman, Clifton, NJ) to retain bacteria cells on the membranes for DNA extraction. In this study, DNA was extracted either by lysing cells eluted from the membrane or retained on the membrane. For samples collected during the first sampling event, the eluted cells were used. The retained bacterial cells on the filter membrane were eluted by immersing the membrane in 4 mL of phosphate-buffered saline, followed
TABLE 3. Validation of Real-Time PCR Assays by Using Both Aeromonas and Non-Aeromonas Strains conventional PCR strain model strains
Aeromonas isolates
non-Aeromonas strains
species
ATCC ATCC ATCC ATCC
7966T 15468 35993 49657
Aeromonas Aeromonas Aeromonas Aeromonas
plant plant plant plant plant plant plant plant plant plant plant plant plant plant plant plant plant plant
1, 2, 3, 4, 6, 2, 7, 1, 4, 1, 5, 6, 3, 3, 4, 5, 7, 7,
A. hydrophila (99%)
no. no. no. no. no. no. no. no. no. no. no. no. no. no. no. no. no. no.
2 1 3 2 2 2 3 3 3 1 2 1 1a 2b 1c 1b 1b 2d
ATCC 25922 CEB-1 CEB-2 CEB-3 ARI-1 KC1 KC2 KC3 KC4 KC5 KC6 KC7 KC8 KC9 KC10 KC11 KC12 KC13 KC14
hydrophila caviae sobria trota
A. encheleia (99%) A. media (99%) A. popoffii (99%) Aeromonas sp.
Escherichia coli Escherichia coli Pseudomonas fluorescens Bacillus thuringiensis Novosphingobium tardaugens Flavobacterium sp. Chitinophaga sp. Nocardioides simplex Rhodococcus ruber Microbacterium testaceum Aminobacter sp. Aminobacter sp. Sphingomonas sp. Sphingomonas sp. Sphingomonas sp. Sphingomonas sp. Brevundimonas vesicularis Escherichia coli Sphingomonas sp.
real-time PCR (values of Ct)
Aeromonas 16S rRNA gene
aerolysin
Aeromonas 16S rRNA genef
aerolysinf
+ + + +
+ + +e +e
10.4 10.7 13.2 13.1
16.7 24.6 32.2 28.6
+ + + + + + + + + + + + + + + + + +
+ + + + + + + +
12.3 12.7 12.0 11.6 10.1 12.9 13.4 13.4 13.3 10.6 13.5 10.1 13.0 12.6 12.9 10.8 10.5 11.4
15.9 15.3 21.3 18.7 16.0 17.0 18.4 16.0
-
-
-
-
-
-
a Partially sequenced 16S rRNA gene showed 96% similarity to both A. media and A. veronii. b Partially sequenced 16S rRNA gene showed 96–99% similarity to both A. hydrophila, A. salmonicida, and A. bestiarum. c Partially sequenced 16S rRNA gene showed 99% similarity to both A. veronii and A. culicicola. d Partially sequenced 16S rRNA gene showed 99% similarity to both A. hydrophila and A. veronii. e Weak band. f The Ct values are the means of duplicate determinations.
by vigorous agitation at the maximum velocity of a vortex mixer (Maxi MixII, Thermolyne) for 10 min. This elution step was repeated three times. The eluted cells were then transferred to centrifuge tubes and pelleted by centrifugation. The supernatant was discarded, and the eluted cells were then lysed as described by Kapley et al. (24). The derived DNA samples were concentrated 10-fold by ethanol precipitation. For samples collected during the second sampling event, DNA was extracted directly from the cells retained on the filter membrane. UltraClean Water DNA Kit (MoBio Laboratories, Inc.) was used for DNA extraction. The derived DNA samples were concentrated 20 times by ethanol precipitation. Both concentrated and unconcentrated samples were used as templates for SYBR Green real-time PCR assays. Real-Time PCR Assay for Quantifying Total Aeromonas spp. The region of 16S rRNA gene of Aeromonas spp. was used as the target region for designing real-time PCR assay for quantifying total Aeromonas spp. A forward primer Aer66f
and a reverse primer Aer613r (Table 2), were used for amplifying 16S rRNA gene of Aeromonas spp. The forward primer Aer66f was previously used as a fluorescence probe for in situ hybridization (25). The reverse primer Aer613r was modified from the primer Con607R (26) which has been used for sequencing a portion of the 16S rRNA gene of Aeromonas (27). By aligning 132 of 16S rRNA gene sequences of cultured Aeromonas (longer than 1000 bp that were deposited in GenBank database during the time the assays were designed), the forward primer (Aer66f) only has one mismatch with four sequences (X60416, X74682, AY538658, and AJ536821) and the reverse primer (Aer613r) only has one mismatch with three sequences (AY264937, AJ536820, and AJ536821). The designed primer sets were also examined for their uniqueness to the target gene by comparing the sequences in GenBank using the Basic Local Alignment Search Tool (BLAST) (28) and the Probe Match program of the Ribosomal Database Project (29). VOL. 42, NO. 4, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1193
Real-Time PCR Assay for Quantifying Aerolysin GeneContaining Aeromonas spp. In this study, aerolysin gene was used as a biomarker of pathogenic Aeromonas spp. A real-time PCR assay using primers AHCF1 and AHCR1M (Table 2) was developed for quantifying aerolysin-genecontaining Aeromonas spp. These primers (AHCF1 and AHCR1M) were modified from a primer set (AHCF1 and AHCR1) designed by Kingombe et al. (30). AHCF1 and AHCR1 have been used for detecting aerolysin gene in many studies (13, 18, 26, 31, 32). Compared to AHCR1 primer, the reverse primer AHCR1M designed in this study is more specific to five aerolysin genes of A. hydrophila that were deposited in GenBank (accession numbers M84709 (33), M16495 (34), X65043, X65044, X65045 (35)). Real-Time PCR Analysis. The SYBR Green real-time PCR assays for quantifying aerolysin gene and 16S rRNA gene of Aeromonas were performed similarly, except using different annealing temperatures. Each reaction was performed in a total volume of 25 µL, with QuantiTect SYBR Green PCR Master Mix (QIAGEN Inc., Valencia, CA), 600 nM forward and reverse primers, and 5 µL of DNA templates. The PCR thermal cycle was 95 °C for 10 min, followed by 35 cycles of 95 °C for 30 s, 61 °C (for 16S rRNA genes of Aeromonas) or 57 °C (for aerolysin genes of Aeromonas) for 45 s, and 72 °C for 30 s. PCR amplification and detection were performed by using a DNA Engine Opticon continuous fluorescence detection system (MJ Research, Waltham, MA). The cycle threshold (Ct) value was determined automatically with the computer software (Opticon Monitor, version 1.4, MJ Research). All concentrated and unconcentrated samples were measured twice by each assay. A subset of each sample was spiked with 1 µL of 105 copies of standard DNA (described below) as an internal control during PCR amplification. This was done as a means to determine whether inhibition of PCR had occurred. The negative controls containing only HPLC water were also included in each PCR run. The melting temperatures of amplification products were determined by melting curves. The melting curves were obtained by operating PCR reactions as follows: heating to 95 °C for 1 min, cooling to 55 °C, and then ramping to 95 °C. Melting curves were checked routinely to confirm the quantification of the desired products. Plasmid DNA was used as a template for constructing standard curves. By using the designed primer sets, a 232 bp-fragment of partial aerolysin gene and a 548 bp-fragment of partial 16S rRNA gene were amplified from A. hydrophila (ATCC 7966). These products were cloned into the vector pCR4-TOPO (TA cloning; Invitrogen, Carlsbad, CA.). The inserts were confirmed by sequencing using M13 primers that flank the cloning region. The sequences of inserts were confirmed by an Applied Biosystems 3100 DNA sequencer (Perkin-Elmer, Foster City, CA). Selected clones were grown overnight in 5 mL of LB broth with kanamycin, and the plasmids were purified using Wizard Plus SV Minipreps (Promega, Madison, WI). The plasmid DNA concentration was determined using a Hoefer DyNa Quant 200 Fluorometer (Hoefer, Pharmacia Biotech, San Francisco, CA). The copy numbers in samples were determined as described by Yu et al. (23). Statistical Analysis. The experimental data were log10 transformed as described by Dionisi et al. (36). The correlation coefficients among the results obtained from EPA method 1605 and from real-time PCR assays were calculated by using bivariate two-tailed Pearson correlation in SPSS version 14.0.
Results Aeromonas Strains Isolated from Water Utilities Surveyed. Eighteen Aeromonas strains were isolated from this study. Based on their partial 16S rRNA sequences (∼500 bp) (Table 3), they are A. hydorphila (five isolates), A. media (two 1194
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 4, 2008
isolates), A. encheleia (two isolates), A. popoffii (three isolates), and Aeromonas sp. (six isolates). No A. sobria, A. caviae, or A. trota were isolated. Our observation was consistent with a previous study conducted by the U.S. EPA in 2001–2002 (26). In the EPA’s study, no A. sobria or A. trota was identified from 205 Aeromonas strains that were isolated from 16 water treatment utilities. Only six isolates were identified as A. caviae (26). Validation of Real-Time PCR Assays. The developed realtime PCR assays for quantifying total and aerolysin genecontaining Aeromonas were vigorously validated. First, the specificity of primers for Aeromonas 16S rRNA genes was determined by using the Probe Match Program of the Ribosomal Database Project (http://rdp.cme.msu.edu/ probematch/, as assessed on April 8, 2007). Only sequences that had been verified using Pintail software (37) and have near-full-length sequences (>1200 bases) were selected for the specificity analysis. When no base pair mismatches are allowed, the forward primer (Aer 66f) will anneal to 858 out of 967 Aeromonas genus 16S rRNA genes (including uncultured clones) and the reverse primers (Aer613r) will anneal to 927 out of the 967 Aeromonas genus 16S rRNA genes. For nontarget 16S rRNA gene sequences analysis, the forward primer (allowed no mismatches) will anneal to only six sequences (3.3 log reductions for plant 1, from >0.3 to >3.1 log reductions for plant 2, and from >0.6 to 4 log reductions for plant 3 (Supporting Information Table S2). Plant 1 and plant 2 showed similar removal effectiveness of Aeromonas. River water was used as source water for plants 1 and 3. Lake water was used as source water for plant 2. The source water used by plant 3 had the highest turbidity (NTUs were 6.5–7) compared to other source water (NTUs 0.04 to >3.8 log reductions, are shown in Supporting Information Table S2. Aeromonas concentrations were measured by different methods: 7.0 × 103 to 4.9 × 103 CFU/100 mL of culturable Aeromonas, 7.2 × 103 to 1.0 × 104 copies/100 mL of aerolysin gene, and 2.7 × 105 to 4.3 × 105 copies/100 mL of Aeromonas 16S rRNA gene. While no Aeromonas colony was detected in water collected after C/F/S, RSF, and UV+Cl2, Aeromonas 16S rRNA genes were detected in water samples collected after C/F/S and RSF. During the second sampling event, aerolysin gene was detected in water collected after C/F/S (200 aerolysin gene copies/100 mL, Table 4). Similarly, no Aeromonas colonies or genes were detected in finished water like those collected from the other drinking water treatment plants. (iii) Conventional Water Treatment Processes with Slow Sand Filtration (SSF). (SWfSSFfCl2 or Chloramine): Cases of plant 6–7. Measured turbidities of water samples are available in the (Supporting Information Table S1). Removals of Aeromonas by different treatment process units, ranging from >0.6 log to >1.8 log as log reductions, are shown in Supporting Information Table S2. Case of Plant 6 (using Cl2). The source water for plant 6 is a brook, where the turbidity was below 0.5 NTU. However, detectable Aeromonas concentrations were still observed by different methods: 56.5 CFU/100 mL of culturable Aeromonas, 3.2 × 102 copies/100 mL of aerolysin gene, and 5.4 × 103 copies/100 mL of Aeromonas 16S rRNA gene (Table 4). All the measured concentrations of Aeromonas in source water for plant 6 were 1 to 2 orders of magnitude lower than those used by plants 1–5. The finished water was free of Aeromonas based on results from all the methods. Case of Plant 7 (Using Chloramine). Removals of Aeromonas by different treatment process units, ranging from >0.5 to >1.0 log as log reductions (Supporting Information Table S2). Lake water was used as source water for plant 7. The turbidity of the pond water was below 0.5 NTU (Supporting Information Table S1). Low concentrations of Aeromonas in the source water were detected: 1–10 CFU/100 mL culturable Aeromonas, and 1.7 × 103 copies/100 mL of Aeromonas 16S rRNA gene (Table 4). Based on turbidities and concentrations of Aeromonas measured by different methods, the quality of source water used by plant 7 was comparable to that used by plant 6. As shown in Supporting Information Table S2, slow sand filtration alone had 0.5 log reduction of the Aeromonas based on the real-time PCR assay. This result is close to the removal efficiency of plant 6 in Vermont. Although the source water quality appeared to be better than those VOL. 42, NO. 4, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1195
1196
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 4, 2008
plant 7 EPA method 1605 (CFU/100 mL)
plant 6 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
plant 5 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
plant 4 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
plant 3 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
plant 2 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
plant 1 EPA method 1605 (CFU/100 mL) aerolysin gene (copies/ 100 mL) 16S rRNA gene (copies/100 mL)
method
C7
