Aptamer-nanobody based ELASA for detection of Vibrio cholerae O1

Background and Objectives: In recent years, the prevalence of diseases caused by Vibrio spp. is increasing in the world, and among them species, Vibrio cholerae is the most important Vibrio associated with pandemic and epidemic cholera outbreaks. Therefore, the development of a reliable method for early and accurate detection of V. cholerae for management of diseases is a real need. Aptamers with the ability to detect targets with high specificity and accuracy can be one of the candidates used for the whole cell and thereby V. cholerae detection. Materials and Methods: In this research high-affinity DNA aptamers against with two major serotypes of Inaba (ATCC 39315) and Ogawa (clinical sample) were selected from DNA aptamer library through 12 rounds of Systematic Evolution of Ligands by Exponential (SELEX) enrichment procedure using live cells as a target which monitored with flow cytometry. Results: The binding efficiency and dissociation constant of the isolated aptamers V.ch47 and V.ch27 were 56.4%, 53.3% and 15.404 ± 4.776 pM, 20.186 ± 3.655 pM, respectively. A sandwich Enzyme-linked aptamer sorbent assay (ELASA) was developed with the biotinylated V.ch47 aptamer and our previously developed nanobody anti-Lipopolysaccharides (LPS). We optimized this system with V. cholerae O1 and analyzed their cross reactivity with close physiological bacteria. The threshold of detection was obtained 104 CFU/ml in the sandwich ELASA process. Conclusion: Our results showed that the sandwich ELASA is sensitive enough for the rapid detection of V. cholerae from other bacteria.


INTRODUCTION
Vibrio cholerae is a Gram-negative, non-spore forming curved rod bacterium with polar flagella found mostly in marine environments. It is classified based on its major surface antigens into nearly 206 serogroups, of which O1 and O139 cause epidemic cholera. Cholera is an infectious disease characterized by intense vomiting and watery diarrhea which if left untreated, could lead to harsh dehydration and death (1). The ability of V. cholerae to survive in water supplies, aquatic ecosystems and inappropriate sewage disposal and short incubation period of the disease play a significant role in cholera epidemics. While only 2-5 % of V. cholerae infections will result in severe diarrhea; roughly 75% of them are asymptomatic. Cholera symptoms might start before 2 h or after 5 days infection and in severe disease could make to loss of up to 2 liters of water per hour, which may result in death within 24 h. Cholera outbreaks happen at regular intervals and it's still endemic in some countries. Furthermore, there have been seven pandemics since 1817 and it's still known as one of the significant health threats, especially in developing countries. Whereas V. cholerae O139 has been causing some limited number of outbreaks in the world recently, all the cholera pandemics initiated by O1 serotype (2,3). The gold standard for detection of V. cholerae from water is a culture method, but this requires laboratory infrastructure, besides it is time-consuming, especially in management a large number of water samples. There are some rapid detection methods, including nucleic acid-based methods such as PCR, immunomagnetic beads and DNA probe hybridization. The sensitivity of immunomagnetic bead is low and DNA probe hybridization and the PCR techniques require designing specific probes or primers both based on the detection of nucleic acids rather than live bacterial cells increasing the possibility of false positives results (4). Gubala et al. (2006) and Blackstone et al. (2007) developed a real-time PCR technique for detecting V. cholerae with high sensitivity. This technique is also based on nucleic acid detection and still requires isolation and culture of the sample after PCR confirmation (5,6). In 2006, surface plasmon resonance immunosensor was developed to target V. cholerae O1 using a monoclonal antibody. Their detection range was between 10 5 and 10 9 cells/ml (7). Chakraborty et al. introduced a rapid dipstick test based on the detection of LPS antigen directly from fecal samples. The test was shown to be sensitive >90% and roughly specific -70%; with limitations like not being able to detect Viable but nonculturable (VBNC) vibrios since the test relies on the ability of the bacteria to grow in the Alkaline peptone water (APW) culture medium or does not differentiate non-toxigenic from toxigenic strains; as the dipstick test detects the LPS of V. cholerae (8).
Aptamers are short (35-100 nucleotides) RNA or single-stranded DNA molecule, which can bind to specific targets due to their 3D conformations with high specificity. Aptamers are separated by in vitro screening and systematic evolution of ligands by exponential enrichment technology (SELEX). This is a high-throughput screening method that involves the selection of aptamers by repeated rounds of amplification and partitioning from a pool of random synthetic oligonucleic acid (9). Small size, cost-effectiveness, chemical stability and convenient and flexible structures of aptamers have led them to be used in highly selective and sensitive biosensors for the identification and detection of pathogens (10). Recently aptamers against several pathogens like Listeria monocytogenes, Streptococcus mutans, Haemophilus influenzae type b, Staphylococcus aureus and a mixture of GAS M-types by whole cell-SEL-EX technique have been characterized (11,12). The whole cell-SELEX is preferred for the detection of pathogens because in the live bacterial cells, the aptamers are selected for the surface molecules in their natural conformations condition. In another word, the conformation of the target remains the same from selection to detection (13).
In this research, we isolated a specific DNA aptamer with a high binding affinity toward V. cholerae through whole cell-SELEX. We have previously developed an Anti-Lipopolysaccharides nanobody (LPS) of V. cholerae O1 (14). Using this nanobody along with our selected aptamer, we developed a new detection system (an aptamer and nanobody) for recognition of V. cholerae O1 in an ELASA based system.

MATERIALS AND METHODS
Bacterial strains, culture conditions and preparation of cells. The V. cholerae O1 [Inaba (ATCC 39315) and Ogawa] used as targets were obtained from Shahed University. The other bacterial strains listed in Table 1 were used in cross-reactivity tests and counter-SELEX to raise the specificity of selection. All strains were verified with biochemical tests as well as amplification of 16S rRNAs (data not shown) and cultured under aerobic conditions in Luria Brentani (LB) broth at 37°C. For each round of SELEX, 10 8 CFU/ml of V. cholerae O1 (200 µl ogawa and 250 µl inaba (OD600 nm= 0.4)) in logarithmic phase were taken, washed with wash buffer (1× PBS plus 0.02% tween 20) at 7000 ×g for 10 min; then suspended in 200 μl of binding buffer (wash buffer containing 1% bovine serum albumin (BSA) was used.

Name
Aptamer library Forward -primer Reverse -primer FITC-forward primer Biotin-forward primer  Amplification of the aptamer library and preparation of ssDNA aptamers. 78-mer combinatorial dsDNA library procured from Metabion (Steinkirchen, Germany). Primers were synthesized at TAG Copenhagen A/S (Frederiksberg, Denmark). The sequences of DNA library, unlabelled and labelled primers are listed in Table 2. Aptamer library with an initial concentration of 10 μM was amplified in a 50 μl PCR reaction (Bio-Rad T100 Thermal Cycler, USA) by a three-step thermal protocol starting with an initial denaturation at 95°C for 5 min followed by 30 cycles of 95°C for 30 s, 62°C for 30 s, 72°C for 30 s, and a final extension at 72°C for 5 min. DNA library pools were melted to ssDNA by heating at 95°C for 10 min, then flash cooled in ice for 15 min. Converting dsDNA to ssDNA proved by size separation on 10% denaturing urea PAGE. The concentration of oligonucleotides was measured with Pico drop 200 (Pico Hinxton, United Kingdom).

Aptamer selection process (SELEX).
To start the first round of SELEX, 2 nmol ssDNA library was mixed with 200 μl of 1× binding buffer and then added with 10 8 CFU/ml of target cells. The reaction mixture was incubated for 1 h at 4°C with gentle shaking. The cells were recovered by centrifugation at 7000 ×g at 4°C for 10 min. Cells were washed to eliminate unbounded/weakly bounded ssDNA aptamers and collected by centrifugation at 7000 ×g for 10 min. Cells were suspended by adding 70 μl of 1× PCR reaction buffer and heated at 70°C for 12 min. After centrifugation the supernatant containing bounded aptamers was used as a template and amplified with PCR to be used for the next rounds of SEL-EX. Some modifications were applied in PCR protocol for amplification of cell-bound aptamers after each SELEX. To strengthen the binding condition, the washing frequency was increased from SELEX 2 onward with ascending concentration of tween 20 (0.02% to 0.07%). Simultaneously the incubation time was gradually decreased from 60 min in the first to 15 min in the last round of SELEX. A total of twelve rounds of SELEX and three rounds of counter-SELEX (3 rd , 7 th and 8 th rounds) were performed to select high-binding aptamers toward V. cholerae O1. In counter-SELEX we also used 10 8 CFU/ml of bacterial cocktail. The SELEX procedure was similar to that of positive SELEXs. The supernatant containing unbounded aptamers was used as a template to amplify and use in the next round of SELEX. Bacteria used for the counter-SELEX were: E. coli O157:H7, Shigella dysenteriae, Salmonella enteritidis, Salmonella Typhimurium, Yersinia and Shigella flexneri.
Flow-cytometric analysis of aptamer binding affinity. The binding efficiency of ssDNA aptamer pool toward target cells, SELEX steps were monitored with flow cytometry using FITC-labeled primers. The outcome of 2 nd , 5th, 10 th , 11 th , 12 th SELEXs and initial ssDNA library were amplified and fluorescence intensity was assessed using Attune NxT (Thermo Fisher, USA). In brief, incubation of 50 pM of FITC labeled aptamer with 10 8 CFU/ml of target cells in binding buffer, followed by incubating at 4°C for 15 min. Cells were washed with washing buffer, harvested with centrifugation and resuspended in 0.5 ml binding buffer. The binding affinity was measured by the flow cytometer.
Cloning procedure. Aptamers obtained from the last round of SELEX were amplified by PCR and were cloned into a pTG19-T vector (Vivantis, Selangor, Malaysia). The ligation reaction was transformed into E. coli DH5α competent cells (Thermo Fisher Scientific, Waltham, MA USA) and plated on LB agar containing 50 µg.ml -1 ampicillin. Colonies were analyzed with colony PCR. Each positive clone was amplified with FITC-labeled primers and the fluorescence intensity was estimated by flow cytometer. Then aptamer candidates with the highest binding affinities were selected for further analysis.
Affinity measurement of the bounded aptamer. To study the cross-reactivity of selected aptamers, individual samples of Enterobacter, Klebsiella and Proteus vulgaris were tested for cross-reactivity with the same procedure described above. In brief, 10 8  Sequencing and structural analysis of selected aptamers. Two aptamers with the highest affinity toward V. cholerae and minimum affinity to other bacteria were sequenced (Tehran University, Iran). The secondary structure of the selected aptamer sequences was predicted using the online software Oligo Analyzer3.1 (https://www.idtdna.com/calc/analyzer). The Prediction of structural folding was carried out under the circumstances set up of 144 mM Na + at 21°C.
ELASA based detection of V. cholerae. ELASA assay was carried out for detection of V. cholerae O1 (Inaba and Ogawa) using anti-LPS single-domain antibody fragments (VHHs) for V. cholerae O1 recognizing Inaba and Ogawa serotypes as capture and biotinylated aptamer V.ch47 as a detector. Salmonella typhimurium, Salmonella enteritidis, Shigella flexneri, Shigella dysenteriae, E. coli O157:H7, Yersinia, Enterobacter, Klebsiella and Proteus vulgaris were used for assessment of any possible cross-reactivity. All the reactions were repeated 3 times.
The expression and purification of VHH were done as described previously (14). 100 µl of VHH (20 µg.ml -1 ) in carbonate-bicarbonate buffer (pH 9.5) was coated on the microplate and incubated over-night at 4°C. Wells were washed three times with PBS-T (PBS + 0.05% Tween 20) and unbound locations were blocked with 100 µl skim milk at 37°C for 1 h and washed three times with PBS-T. The bacterial cocktail having 10 8 CFU.ml -1 was added to the wells and incubated at 37°C for 2 h. After washing with PBS, 100 µl of biotinylated aptamer V.ch47 was added to the wells and incubated at RT for 2 h. Wells were washed and 100 µl of 1:1000 dilution of horseradish peroxidase (HRP) conjugated with streptavidin (Biolegend, San Diego, USA) was added to the wells and incubated at 37°C for 90 min, then washed with PBS. Finally, 100 µl of tetramethylbenzidine (DNAbiotech Co., Iran) as a substrate was added. The plate was incubated at room temperature (RT) for 10 min in a dark condition. The reaction was stopped with 100 µl H 2 SO 4 3N and the OD 450 was measured using Infinite F50 microplate reader (TE-CAN, Mannedorf, Switzerland). Cross-reacting Bacteria was also used for negative control. The same experiment was carried out with different enumeration of bacteria (10 4 -10 8 CFU/ml) and different concentrations of aptamer (25, 50 and 100 pM), to obtain the limit of detection (LOD).

SELEX optimization and flow cytometry.
Twelve rounds of SELEX including nine positive-SELEX and three counter-SELEX were carried out to reach the saturation state. The fluorescent intensity of the ssDNA aptamer pool bound to the target was increased from 14.7% in the second round to 50.3% and 50.9% in the 11 th and 12 th rounds of SELEX, respectively (Fig. 1). The binding efficiency during the last two SELEX rounds (11 th and 12 th ) didn't show any impressive growth; therefore, the SELEX process was terminated at round 12 th .
Our data revealed descending binding affinity toward counter bacteria and the results of SELEX 12 th showed approximately 4-fold improvement in binding efficiency compared to the 2 nd round. The ssDNA aptamer pool of the latest round was cloned and transferred into the E. coli DH5α cells. Thoroughly 97 clones were achieved, of which 67 aptamer transformants were confirmed as positive clones with colony PCR. The binding affinity of 31 positive clones to target bacteria was analyzed (Table 3). In the next step flow cytometer Results obtained from flow cytometry shows increased binding affinity of 11 th (a) and 12 th (b) rounds of SELEX compared to that of 2 nd round. The binding efficiency increased from 14.7% in the 2 nd round to 50.3% and 50.9% in the 11 th and 12 th rounds respectively.  determined the binding efficiency to cross bacteria (Table 4).

Aptamer name
Finally, three aptamers with the highest affinity to V. cholerae O1 and the lowest affinity to other bacteria used for cross-reactivity were selected for fur-ther analysis. The affinity of 3 aptamers for the target were estimated as 56.4%, 53.3% and 52.5% and here onward named as V.ch47, V.ch27 and V.ch50, respectively (Fig. 2).

Dissociation constants.
The percent binding efficiency between the aptamer and target cells was estimated by considering a constant amount of 10 8 CFU/ml -1 of V. cholerae O1 cells against varied aptamer concentrations from 0 to 250 pM. The dissociation constant values (Kd) of V.ch47 and V.ch27 were calculated as 15.404 ± 4.776 and 20.186 ± 3.655 pM, respectively. The related binding saturation curve is displayed in Fig. 3.

Sequence analysis of selected aptamers.
The sequence of the variable regions of the selected aptamers is given in Table 5. The Gibb's free energy of each sequence was calculated by the UNAFold online program. The secondary structures were predicted at 21°C with a minimum free energy of 16.39 kcal.mol -1 for aptamer V.ch27 and -8.62 kcal.mol -1 for aptamer V.ch47 are displayed in Fig. 4.      Table 6. ELASA results of V. cholerae O1 and bacteria used for cross-reactivity. Results are displayed as an OD 450 mean ± standard deviation (n=3) in Fig. 5. The value of reaction containing 10 4 CFU. ml -1 V. cholerae O1 and 25 pM ssDNA aptamer became nearly 0.77, was selected as an optimum readout for further analysis. Four different bacteria are selected to investigate the cross-reactions and the results are exhibited in Table 6. There was a significant difference between the optical density gained from V. cholerae O1 and other bacteria used as a cross. The OD 450 obtained for V. cholerae O1 was 0.77, while the highest OD 450 related to other bacteria was 0.3 belonging to S. flexneri ( Table 6). The optical density of the reactions in the absence of aptamer/bacteria was always less than 0.2.

DISCUSSION
V. cholerae which causes an epidemic infectious disease of cholera, generally spread in marine environment. This bacterium is mostly present in water supply environments of cholera patient's existence (1). WHO has estimated, having 1.3 million to 4.0 million cases of cholera, and 21000 to 143000 deaths in the world annually. Moreover, outbreaks continued to impress numerical countries. Cholera often affects developing countries with no proper contact with suitable water and sanitation resources and thereby remains a major public health problem yet. Although the O1 and O139 serogroups are related to epidemic disease, V. cholerae O1 was the main cause of all recent outbreaks. Therefore rapid detection of V. cholerae O1 in polluted waters or clinical samples like stool; in early stages is critical, mainly for the prevention of the outbreaks (17).
In recent years, aptamers and recombinant antibodies have been used in different diagnostic systems such as ELASA, blotting assay, flow cytometry, immunocytochemistry, immunofluorescence microscopy and microarray (18,19). In the present work, we have developed an ELASA system for detection of V. cholerae O1, where nanobody and aptamer, both developed specifically against V. cholerae O1, were used as capture and detector respectively.
Aptamers are widely used for the detection of bacteria with SELEX procedure where they show high specificity and binding affinity toward their targets (13,16). In this research, a whole-cell SELEX method was used to isolate ssDNA aptamers for V. cholerae O1. Theoretically, the whole-cell SELEX allows candidate aptamers to recognize multiple cell surface targets for binding in their native conformation. The cell-SELEX approach was analyzed by flow cytometry for both binding affinity and selection. In the traditional protein SELEX, on the other hand, one cannot firmly determine the aptamer binding site in a physiological condition, and this may create problem with post-selection where the selected aptamer/ aptamers may not bind to their targets on the cell surface (20). V.ch47 and V.ch27 aptamers showed affinity of 15 ± 4 and 20 ± 1 pM toward V. cholerae O1 respectively. The data are in agreement with our and other previous reports, which indicate the efficiency of our selection process (11,13,21). The best Kd obtained was 15.404 ± 4.776 pM, characterized after 12 rounds of SELEX. Compared to our previous reports, for Haemophilus influenzae type b, Streptococcus pyogenes type M3 and Acinetobacter baumannii; the Kd for V. cholerae O1 was somewhat higher (18,22). The reason could be due to having two serotypes as a target instead of having a single one in the SELEX procedure. But in comparison to other mention works, our selected aptamers showed better affinity (11,13,23). The higher affinity obtained here can be attributed to suitable selection stringency and using bacterial strains with similar physiological properties to those used in counter-SELEX, which considerably eliminates a non-specific, poor binding and non-target cells. Furthermore, amplification of the whole DNA pool after each SELEX helps us to reduce the possibility of losing any sequences (possibly high binder). Finally, the V.ch47 aptamer showed the highest specificity toward V. cholerae O1 (Inaba and Ogawa) with no affinity toward other related bacteria used in the cross and counter-SELEX exhibited in Table 1. RNA aptamers have also been selected against whole cell bacteria such as E. coli O157:H7(24); however RNA aptamers are harder to produce, less stable compared to DNA aptamers, and become cost-effective if modified in order to make them RNase-resistant. Yu et al. accomplished a total of 25 rounds of selection against live E. coli O157:H7. Their selected DNA aptamer exhibited a Kd of 10.30 nM (25)., whereas we obtained better a kd after 12 rounds of SELEX, which can be due to proper selection performance and the binding buffer used in procedures.
The affinity of the VHH against V. cholerae O1 LPS (lipopolysaccharide) used in ELASA was 5 × 10 -8 M, described previously (14). Generally in the sandwich ELASA, either aptamer or antibody is fixed as a capture (26). The nanobody was immobilized on the surface of the wells because aptamer has smaller size compared to antibodies and fixing aptamer as a capture substance may modify the structural conformation of the aptamer which subsequently affects its binding to the target on the cell surface. Moreover, aptamer labeling is easier and does not affect their conformation, which makes them more suitable detectors compared to the antibodies (27). Our method was evaluated in numerical steps to verify the specificity and efficiency of the technique with different concentrations of target cells and aptamer. We could identify 10 4 CFU.ml -1 (OD 450 = 0.762) with 25 pM of biotinylated aptamer and only 20 µg.ml -1 of VHH without any cross-reactivity. Wang et al. gen-erated an aptamer against prion protein and developed a sandwich ELISA with 0.5 ng.ml -1 aptamer as a capture and 2 ng.ml -1 prion, using antibodies in a common sandwich ELISA technique (28). In 2009, Chen et al. developed an ELASA system for the detection of Hepatitis C virus. They used aptamer as a detector and a 250 nM polyclonal antibody as a capturing agent (29). We previously designed an ELA-SA system for the detection of A. baumannii (ATCC 19606) by a biotinylated aptamer and a nanobody against biofilm-associated protein (Bap) in 2016. The detection threshold in our sandwich ELASA for A. baumannii was 10 3 CFU/ml (18). Better cut off value in the ELASA system for A. baumannii could be due to the development of higher affinity aptamers (7.547 ± 1.353 pM) and nanobody (3.8 × 10 -8 M/L). Whereas in the present work, the Kd for aptamer and VHH were 15.404 ± 4.776 pM and 5 × 10 -8 M, respectively (30). For detection of OmpW and CtxB antigens of V. cholerae, Tuteja et al. developed a dip-stick sandwich ELISA where the analytical sensitivity of the designed assay was 60 pg/ml for CtxB and 10 4 CFU•ml -1 for OmpW (31). In another research, based on anti-lipopolysaccharide antibodies, Nato et al. designed an immunochromatographic method which able to detect 10 7 CFU/ml -1 of O1 strain and 10 6 CFU/ ml -1 of O139 (32). Also, Obihiro university designed a immunochromatographic test strip by Yamasaki et al. the reported detection limit was 10 ng/ml of cholera toxin (33). In the current research, Bayat et al. used IgY based on ELISA for the detection of toxigenic V. cholerae for the first time. They reported the limit of detection of OmpW is 10 3 CFU/ml and that of CtxB is nearby 33 pg/ml (34) University of East Anglia developed a colorimetric bioassay method for Cholera toxin (CTB) with a limit detection of 54 nM (35). Lee et al. also used the whole-cell SELEX strategy to generate aptamers with high affinity against live Listeria monocytogenes. Their aptamer-based sandwich assay (ABSA) exhibited a linear response over an extensive concentration range of L. monocytogenes from 20 to 2 × 10 6 CFU/ml (18).

CONCLUSION
In conclusion, we have introduced a specific and easy with high-affinity detection ELASA system based on a combination of V.ch47 aptamer and anti LPS nanobody; for detection of V. cholerae O1 with-out needing for robotic equipment or costly preparations. As we have mentioned previously, our ELISA system is flexible enough to be used in the detection of a wide range of pathogens.