Sensitive Detection of E. coli in Artiﬁcial Seawater by Aptamer-Coated Magnetic Beads and Direct PCR

Featured Application: E. coli in seawater can be simply and e ﬃ ciently detected by direct PCR. The sensitivity of detection is signiﬁcantly improved using magnetic beads based bacterial pre-concentration. Abstract: Foodborne and waterborne E. coli remains a major economic burden worldwide. Assaying seawater for trace levels of E. coli is challenging since it applies time-consuming preparations, expensive instrumentation and complicated procedures. Therefore, there is a continued demand for new analytical technologies that can detect low bacterial concentrations in a more cost- and time-e ﬀ ective manner. In this study, combination of E. coli pre-concentration with a direct polymerase chain reaction (PCR) was shown to enable rapid bacterial detection without enrichment step or DNA extraction / puriﬁcation. The E1 aptamer that targets E. coli surface epitope grafted onto magnetic beads e ﬃ ciently concentrated E. coli from water samples containing high concentration of NaCl. When direct PCR was performed on bacteria attached to these aptamer-modiﬁed magnetic beads, a limit of 10 3 CFU / mL was obtained. The overall analysis was performed in less than 3 h. This approach may lead to a future PCR-based biosensor system for online monitoring of enteric bacteria in seawater.


Introduction
The 'One Health' approach recommended by WHO recognizes the inseparable link between human, animal and environmental health [1]. The most under investigated part of One Health is the environment. A seaside represents a unique ecological system that is highly influenced by industrial development and discharge of wastewater. Fecal contamination of seawater is characterized by an increase in concentration of Escherichia coli and intestinal enterococci. These bacteria are common members of gut microbiome in mammals and birds but are also opportunistic pathogens for human and animals [2]. E. coli and enterococci are used as standard for waterborne pathogens and as fecal indicator bacteria in water quality testing all over the world because these bacteria are easily culturable (EU BDW 2006/7/EC). The regular monitoring of E. coli content in water enables to avoid health risks, especially from exposure in the foreseeable context of short-term pollution on unusual circumstances. In addition, marine recreational waters were shown to be potential source of resistant bacteria because they may contribute to the dissemination of microorganisms carrying antimicrobial resistance genes [3].
France) was used as 1 g/L water solution. NaCl (99%) and Tween 20 were purchased from Merck (Fontenay sous Bois, France). Dynabeads M-280 streptavidin magnetic particles were purchased from Invitrogen (Saint Aubin, France). Promega Wizard Genomic DNA purification kit was used for extraction of bacterial genomic DNA. Colorant Midori green advance was purchased from Dutcher (Brumath, France), and 1 Kb Plus DNA ladder from New England Biolabs (Beverly, MA, USA). Phosphate-buffered saline (PBS) 10× was purchased from Lch Chime (Les Aires, France).

Bacterial Strain and Growth Condition.
Permanent stock of E. coli K-12 TG1strain was maintained at −80 • C in LB broth supplemented with 15% glycerol. Bacterial cells were routinely grown in LB broth or on LB agar (Difco) medium at 37 • C. An overnight liquid culture was washed and used to inoculate 25 mL various culture media in a 250 mL Erlenmeyer flask. Flasks were incubated with shaking (200 rpm) at 37 • C. The Wizard genomic DNA purification kit (Promega, Charbonnieres, France) was used for DNA extraction. The optical density of bacterial solutions was measured at 600 nm using Vis Spectrophotometer Genesys (Theromo Fisher).

Drop Plate Method for Enumerating Bacteria
Bacterial solutions were serially diluted by placing 100 µL of the suspension into a dilution tube containing 900 µL of BP. This tube was vortexed, and 100 µL was removed and placed into a second dilution tube containing 900 µL of BP. This process was repeated six times. Then, 10µL of each dilution was plated in triplicate on LB agar and incubated for 18 h at 37 • C. Colony counts from triplicate plates were converted to CFU/mL.

Aptamer
In this study, we used previously selected ssDNA E1aptamer that specifically binds to E. coli [6]. The aptamer was custom-synthesized by Eurogenotec (Angers, France) with biotin tag: 5 -biotin GCA  ATG GTA CGG TAC TTC CAC TTA GGT CGA GGT TAG TTT GTC TTG CTG GCG CAT CCA CTG  AGC GCA AAA GTG CAC GCT ACT TTG CTA A-3 . Before use, E1 aptamer was dissolved into a concentration of 20 µM in distilled water.

Magnetic Beads Functionalization
Prior to immobilization to magnetic beads, the E1 aptamer was denatured by heating at 95 • C for 5 min; then cooled at 4 • C for 15 min, and finally renatured at 25 • C for 5 min. Streptavidin magnetic beads were prepared according to the manufacturer's instructions, as previously described [7]. Initially, 200 µL of magnetic beads were gently washed with 0.5 mM ethylenediaminetetraacetic acid (EDTA), 1 M NaCl, 5 mM Tris-HCl buffer, pH 7, and concentrated using a magnetic stand DynaMag2, (Thermo Fisher Scientific, Lissieu, France). The washed beads were incubated with 2.5 µM E1 aptamer in 0.5 mM EDTA, 1 M NaCl, 5 mM Tris-HCl buffer, pH 7, in a bench-top rotator mixer at room temperature for 30 min. The aptamer-bead conjugates were concentrated using a magnetic stand and washed with 1 mL of 0.05% Tween20, PBS (PBST) for four times before storing at 4 • C.

Bacterial Capturing by Apta-Magnetic Beads
Varying concentrations of E. coli cells (from 10 2 CFU/mL to 10 8 CFU/mL, in BP + 0.4M NaCl, 1 mL volume) were each incubated with the aptamer-magnetic beads (10 µg/mL) under shaking for 30 min at room temperature. The complex of E. coli-aptamer-magnetic bead was washed three times in 1 mL PBST using a magnetic stand, to remove non-bound bacteria. In the final step, beads were concentrated using the magnetic stand, and dissolved in 50 µL PBS. A 20 µL of this bead suspension was used for direct PCR.

Optical Microscopy
Bacterial cells and magnetic beads were observed with an AxioObserver.Z1 Zeiss inverted optical microscope equipped with a Zeiss AxioCam MRm digital camera. Images were processed using the ZEN software package.

PCR Tests
All PCRs ware performed using specific primers targeting the yalO gene of E. coli as previously published [8].
Classical PCRs were performed on 50 µL solutions containing DreamTaq DNA polymerase (2× DreamTaq Green PCR Master Mix, Thermo Scientific), 0.02 mM forward YalO primer (5 -TGATTTCCGTGCGTCTGAATG-3 ) and 0.02 mM reverse YalO primer (5 -ATGCTGCCGTAGCGTGTTTC-3 ) and various concentration of template DNA (genomic DNA extracted from E. coli). Following template volumes were used for PCR reaction: 0.39 µL for DNA sourced from BP, 0.55 µL for DNA sourced from BP + 0.4 M NaCl, and 0.38 µL for DNA sourced from LB and LB + 0.4 M NaCl solutions. The concentration and quality of the DNA were determined by a spectrophotometer NanoDrop 2000c (Thermo Fisher Scientific, Lissieu, France) and Qubit Fluorometric quantification (Thermo Fisher Scientific, Lissieu, France). The DNA preparations were stored at −20 • C until use. The total volume was adjusted with RNase-free water.
Direct PCR was performed with functionalized magnetic beads carrying E. coli cells. Before detection, 7.5 µM biotinylated E1 aptamer conjugates with 0.5 mg of magnetic beads were mixed with bacterial culture and washed four times with PBST. Direct PCRs were performed on 50 µL total volume containing 20 µL of magnetic beads carrying bacteria cells, 25 µL of DreamTag Green PCR Master Mix 2×, 0.02 mM reversed and forward yalO primers (1 µL of each primer at 1 mM, and 3 µL distillated water). PCR was performed under the following conditions: initial denaturation at 95 • C for 3 min, followed by 35 cycles of denaturation at 95 • C for 30 s, primer annealing at 58 • C for 30 s, primer extension at 72 • C for 30 s, and a final extension at 72 • C for 10 min. In every assay, a buffer control, to which no DNA template was added, was used as a negative control.
Twenty µL of the PCR product were separated by electrophoresis on 1.5% agarose gel in 1 mM EDTA, 20 mM acetic acid, 40 mM Tris-base buffer, pH 8.0 (TAE buffer) and stained with midori green. Gels were visualized under UV light using a Fusion Fx gel documentation system (Vilber Lourmat, Marne-la-Vallee, France). The 1 Kb Plus DNA ladder was used as a molecular marker to indicate the size of the amplicons.

Survival of E. coli in Highly Salted Media
In seawater, enteric bacteria encounter a variety of challenges to survive because high salt concentration and osmolarity induce a stress to bacterial cells [9]. Thus, we first compared the survival of E. coli cells in a rich bacterial medium, LB, and LB complemented with 0.4 M NaCl, and its survival in a poor bacterial medium, Peptone Bacto, and Peptone Bacto supplemented with 0.4 mM NaCl.
Because bacteria under the stress may become VBNC, two parameters were evaluated: the optical density of the solutions at 600 nm (OD600), which reflects total number of bacteria, and the colony forming ability, which is a viability parameter. Figure 1a shows that when solutions were inoculated to a cell density of approximately 10 8 CFU/mL, similar survival rates were obtained in both media supplemented or not with NaCl upon 30 h, although some differences were observed during the first 5 h post inoculation. In addition, no drastic morphological modification of E. coli cells was observed in solutions containing high NaCl concentration compared to the corresponding pure medium (Figure 1b). Recently, Zhang et al. [10] showed that there was no significant difference in E. coli cell growth when fresh water was replaced by 60% artificial seawater. Thus, it seems that viability of E. coli cells is no affected by high NaCl concentration (0.4 M) in solutions that normally support bacterial survival.

PCR Detection of E. coli Genomic DNA
Second, we wondered whether high NaCl concentration inhibits DNA polymerase and, in that way, decreases the sensitivity of PCR detection of bacteria. For this, genomic DNA were extracted from bacterial cells cultivated in four above media (LB, LB + NaCl; PB, and PB + NaCl) and tested for the efficiency of classical PCR. Previously validated primers that target consensus sequence of the orphan gene yaiO of E. coli yielding 115 bp PCR product [8] were used for PCR amplification. Figure  2 points out that the expected amplicon of 115 bp was detected with similar sensitivity using DNA from bacteria cultivated in pure bacterial media and in corresponding media supplemented with 0.4 M NaCl. This indicates that classical PCR technique can be successfully applied for E. coli detection in seawater.
However, although very robust, sensitive and specific classical PCR demands previously extracted and purified genomic DNA. Thus, we also tested whether direct PCR (without performing DNA isolation and purification steps) can be used to evidence E. coli in highly salted media.

PCR Detection of E. coli Genomic DNA
Second, we wondered whether high NaCl concentration inhibits DNA polymerase and, in that way, decreases the sensitivity of PCR detection of bacteria. For this, genomic DNA were extracted from bacterial cells cultivated in four above media (LB, LB + NaCl; PB, and PB + NaCl) and tested for the efficiency of classical PCR. Previously validated primers that target consensus sequence of the orphan gene yaiO of E. coli yielding 115 bp PCR product [8] were used for PCR amplification. Figure 2 points out that the expected amplicon of 115 bp was detected with similar sensitivity using DNA from bacteria cultivated in pure bacterial media and in corresponding media supplemented with 0.4 M NaCl. This indicates that classical PCR technique can be successfully applied for E. coli detection in seawater.

PCR Detection of E. coli Genomic DNA
Second, we wondered whether high NaCl concentration inhibits DNA polymerase and, in that way, decreases the sensitivity of PCR detection of bacteria. For this, genomic DNA were extracted from bacterial cells cultivated in four above media (LB, LB + NaCl; PB, and PB + NaCl) and tested for the efficiency of classical PCR. Previously validated primers that target consensus sequence of the orphan gene yaiO of E. coli yielding 115 bp PCR product [8] were used for PCR amplification. Figure  2 points out that the expected amplicon of 115 bp was detected with similar sensitivity using DNA from bacteria cultivated in pure bacterial media and in corresponding media supplemented with 0.4 M NaCl. This indicates that classical PCR technique can be successfully applied for E. coli detection in seawater.
However, although very robust, sensitive and specific classical PCR demands previously extracted and purified genomic DNA. Thus, we also tested whether direct PCR (without performing DNA isolation and purification steps) can be used to evidence E. coli in highly salted media.  However, although very robust, sensitive and specific classical PCR demands previously extracted and purified genomic DNA. Thus, we also tested whether direct PCR (without performing DNA isolation and purification steps) can be used to evidence E. coli in highly salted media.

Bacterial Pre-Concentration with Aptamer Decorated Magnetic Beads
The principle of the direct PCR method for E. coli detection in salted media is described in Figure 3. The first step was pre-concentration of bacterial cells with magnetic beads decorated with the specific E1 aptamer that recognizes an epitope on the surface of E. coli vegetative cells. The second step was a direct PCR detection of E. coli attached to beads.

Bacterial Pre-Concentration with Aptamer Decorated Magnetic Beads
The principle of the direct PCR method for E. coli detection in salted media is described in Figure  3. The first step was pre-concentration of bacterial cells with magnetic beads decorated with the specific E1 aptamer that recognizes an epitope on the surface of E. coli vegetative cells. The second step was a direct PCR detection of E. coli attached to beads.
The feasibility of E. coli detection by an integrated aptamer-magnetic direct PCR method was studied on bacterial cells cultured in 0.4 M NaCl, BP medium. This supplemented medium was considered as artificial seawater because most natural oligotrophic seawaters contain about 1 mg of dissolved organic carbon in addition to the high salt concentration [11]. Streptavidin-biotin conjugated chemistry was used to decorate magnetic beads with the E1 aptamers. Bright monodisperse metallic beads of a spherical shape were observed before bacterial immobilization (Figure 4a). The technique did not enable to visualize aptamer molecules on the beads surface because of nanometric sizes of biotinylated aptamer molecules. When admixed to a PB + 0.4 M NaCl culture containing E. coli, functionalized beads captured efficiently bacterial cells. Figure 4b shows that functionalized beads after 30 min incubation with bacteria were covered with bacterial cells, even after multiple washings.
The efficiency of capturing was calculated by enumerating bacteria in supernatant after magnetic concentration using a drop plate method remained. It was estimated to 90% for 10 4 CFU/mL of E. coli in BP + NaCl.  The feasibility of E. coli detection by an integrated aptamer-magnetic direct PCR method was studied on bacterial cells cultured in 0.4 M NaCl, BP medium. This supplemented medium was considered as artificial seawater because most natural oligotrophic seawaters contain about 1 mg of dissolved organic carbon in addition to the high salt concentration [11].
Streptavidin-biotin conjugated chemistry was used to decorate magnetic beads with the E1 aptamers. Bright monodisperse metallic beads of a spherical shape were observed before bacterial immobilization (Figure 4a). The technique did not enable to visualize aptamer molecules on the beads surface because of nanometric sizes of biotinylated aptamer molecules. When admixed to a PB + 0.4 M NaCl culture containing E. coli, functionalized beads captured efficiently bacterial cells. Figure 4b shows that functionalized beads after 30 min incubation with bacteria were covered with bacterial cells, even after multiple washings.

Bacterial Pre-Concentration with Aptamer Decorated Magnetic Beads
The principle of the direct PCR method for E. coli detection in salted media is described in Figure  3. The first step was pre-concentration of bacterial cells with magnetic beads decorated with the specific E1 aptamer that recognizes an epitope on the surface of E. coli vegetative cells. The second step was a direct PCR detection of E. coli attached to beads.
The feasibility of E. coli detection by an integrated aptamer-magnetic direct PCR method was studied on bacterial cells cultured in 0.4 M NaCl, BP medium. This supplemented medium was considered as artificial seawater because most natural oligotrophic seawaters contain about 1 mg of dissolved organic carbon in addition to the high salt concentration [11]. Streptavidin-biotin conjugated chemistry was used to decorate magnetic beads with the E1 aptamers. Bright monodisperse metallic beads of a spherical shape were observed before bacterial immobilization (Figure 4a). The technique did not enable to visualize aptamer molecules on the beads surface because of nanometric sizes of biotinylated aptamer molecules. When admixed to a PB + 0.4 M NaCl culture containing E. coli, functionalized beads captured efficiently bacterial cells. Figure 4b shows that functionalized beads after 30 min incubation with bacteria were covered with bacterial cells, even after multiple washings.
The efficiency of capturing was calculated by enumerating bacteria in supernatant after magnetic concentration using a drop plate method remained. It was estimated to 90% for 10 4 CFU/mL of E. coli in BP + NaCl.  The efficiency of capturing was calculated by enumerating bacteria in supernatant after magnetic concentration using a drop plate method remained. It was estimated to 90% for 10 4 CFU/mL of E. coli in BP + NaCl. Figure 5a shows that direct PCR obtained without the pre-PCR DNA extraction and purification steps can reveal the presence of E. coli in artificial seawater but only at high bacterial concentration of 10 8 CFU/mL. However, when bacterial cells were pre-concentrated with magnetic beads, the limit of detection was decreased to 10 3 CFU/mL (Figure 5b).   Figure 5a shows that direct PCR obtained without the pre-PCR DNA extraction and purification steps can reveal the presence of E. coli in artificial seawater but only at high bacterial concentration of 10 8 CFU/mL. However, when bacterial cells were pre-concentrated with magnetic beads, the limit of detection was decreased to 10 3 CFU/mL (Figure 5b).

Discussion
Microbiological quality of seawater is a public health problem that can have serious repercussions in the economy and tourism. The prevention of waterborne diseases relies on microbiological detection methods. Recently, several new analytic methods have been proposed to replace standard procedures that take a few days to provide results. For instance, optical biosensors that use filters and camera enable to detect bacteria in water by fluorescence or hemiluminescence [12][13][14][15]. In the literature, electrochemical sensors are proposed that detect pathogens thanks to the current or potential variations that occur upon pathogen or its biomarker interaction with an electrode [16][17][18][19][20][21]. Among various detection methods, the most robust are those based on ELISA immunological tests and on PCR-based molecular tests [22]. Both methods may provide results within 24 h. However, their sensitivity is not sufficient for on-field applications, as an enrichment step is needed prior to detection. The enrichment step is particularly essential for detection of waterborne bacteria that are typically highly diluted. Moreover, PCR-based methods (like qPCR, digital PCR, isothermal methods) are performed on extracted and purified bacterial DNA. Purification of DNA enables to avoid possible inhibitors of the downstream amplification. Multisteps that include bacterial concentration by filtration, pathogen recovery/enrichment, and DNA extraction/purification significantly increase the cost and time of water analysis.
Detection of E. coli is of great importance to the quality control of food and water. The combination of pathogen magnetic pre-concentration and direct PCR strategy proposed in this study enables detection of E. coli without long multi-step DNA extraction. In addition, amplification of DNA sequences by PCR allows detection of VBNC cells that pose problem in conventional culture methods.
In our procedure, the enrichment step (that takes from few hours to a few days) was replaced by the magnetic pre-concentration that was performed in less than 1 h. The pre-concentration

Discussion
Microbiological quality of seawater is a public health problem that can have serious repercussions in the economy and tourism. The prevention of waterborne diseases relies on microbiological detection methods. Recently, several new analytic methods have been proposed to replace standard procedures that take a few days to provide results. For instance, optical biosensors that use filters and camera enable to detect bacteria in water by fluorescence or hemiluminescence [12][13][14][15]. In the literature, electrochemical sensors are proposed that detect pathogens thanks to the current or potential variations that occur upon pathogen or its biomarker interaction with an electrode [16][17][18][19][20][21]. Among various detection methods, the most robust are those based on ELISA immunological tests and on PCR-based molecular tests [22]. Both methods may provide results within 24 h. However, their sensitivity is not sufficient for on-field applications, as an enrichment step is needed prior to detection. The enrichment step is particularly essential for detection of waterborne bacteria that are typically highly diluted. Moreover, PCR-based methods (like qPCR, digital PCR, isothermal methods) are performed on extracted and purified bacterial DNA. Purification of DNA enables to avoid possible inhibitors of the downstream amplification. Multi-steps that include bacterial concentration by filtration, pathogen recovery/enrichment, and DNA extraction/purification significantly increase the cost and time of water analysis.
Detection of E. coli is of great importance to the quality control of food and water. The combination of pathogen magnetic pre-concentration and direct PCR strategy proposed in this study enables detection of E. coli without long multi-step DNA extraction. In addition, amplification of DNA sequences by PCR allows detection of VBNC cells that pose problem in conventional culture methods.
In our procedure, the enrichment step (that takes from few hours to a few days) was replaced by the magnetic pre-concentration that was performed in less than 1 h. The pre-concentration increased the limit of detection by five orders of magnitude (from 10 8 CFU/mL to 10 3 CFU/mL of E. coli). Regarding the low concentration of contaminating bacteria in potable, recreational, and bathing waters, we believe that this procedure based on aptamer-coated magnetic beads may successfully replace membrane filtration to concentrate bacteria. It can be performed on-site, as no electricity source is needed nor a specific equipment for sample preparation. A simple magnet enables to collect beads carrying bacterial cells for analysis.
As the present work was a proof of principle, a lab PCR was used to verify the feasibility of the system. Portable PCR have already been presented in [14,18,[23][24][25], and the aim of future work is to miniaturize the system for direct PCR detection in order to obtain a transportable system for in-field measurements.