Development of Molecular Rapid Detection for Vibrio cholerae and Escherichia coli

Background: Vibrio cholerae and Escherichia coli were main causative agent foodborne diseases, especially in many developing countries, such as Indonesia. Thereby, rapid detection of these pathogenic bacteria is necessary to quickly detect infection that occurred so it can be treated immediately. In this case, multiplex PCR allows multiple genes amplication in one reaction thereby enable to perform rapid detection of these pathogenic bacteria. The objective of this study is to develop rapid molecular detection of V. Cholerae and E. coli and analyze the sensitivity and specicity of this assay. Result: In this study, we used various virulence genes in each pathogenic bacteria as marker to develop rapid molecular detection. Based on this research, optimum results of V. cholerae and E. coli rapid detection were obtained with a primer concentration of 16 µM for ctxA and ompU, 30 µM for ace, and 50 µM for zot, and toxR; 2 µM for elt and 5 µM for stx, respectively. Finally, based on the method standardization by ISO/TS 20836 these assays had 0% false positive, 0% false negative, 100% specicity, and 100% sensitivity; 0% false positive, 4% false negative, 100% specicity, and 96% sensitivity for V. cholerae and E. coli respectively. Conclusion: The optimized method was qualied to be used as a detection method for V. cholerae and E. coli detection according to ISO/TS 20836 (2017) and EHEC and ETEC contamination in drinking water samples.


Background
Contamination of food and beverage have been one of the main concerns in many developing countries, such as Indonesia and causing various kinds of diseases one of which is diarrheal. Some of the main causative bacteria which cause this contamination are V. cholerae and E. coli. These bacteria can spread through faecal and oral causing foodborne diseases, which potentially leading to high level of morbidity and mortality 1 . The pathogenicity of these bacteria comes from the expression of cluster of virulence genes. For example, expression of ctxA and ompU genes in V. choleraee leads to production of cholerae toxin and colonization of these bacteria in the small intestine, respectively 2 . Whereas in E.coli it is known that most of E. coli strain are harmless, but some serotypes are pathogen, such as EHEC and ETEC which produce shiga toxin and heat-labile enterotoxin, respectively 3 .
In this case, conventional method consists of the usage of selective media, microscopic examination, and biochemistry assay still were used to identify these pathogenic bacteria. However, this method is not reliable and quick enough to identify these bacteria in case of an outbreak happens. Advances in molecular techniques has led to a shift from conventional methods to molecular method, which are more sensitive, speci c, and more reproducible. PCR based detection test is fast and sensitive technique to identify pathogenic bacteria by detecting virulence genes which presence in bacteria. However, regular PCR is only capable of detecting single gene in one PCR reaction but multiplex PCR provides the

Discussion
The optimization of the multiplex PCR method was done by optimizing all of the primers concentration until the optimum concentrations were acquired, which was 16 µM for ctxA and ompU, 30 µM for ace, 50 µM for zot, and toxR in V. cholerae and 2 µM for elt, and 5 µM for stx in ETEC and EHEC, respectively. Below the optimal concentration, all of the primers could not produce a clear band, consistently. Conversely, more than the optimal concentration all of the primers produced too strong band. It might happen due to the difference between the ampli cation product size of all the primers. The possibility of DNA sequence to be ampli ed was higher if the sequence product was smaller, conversely the possibility was lower if the product was larger.
The sensitivity of the V. cholerae optimized multiplex assay proved to be high, as little as 0.25 ng/µL of DNA was su cient to produce clear bands. In addition, we also found that the sensitivity of ETEC and EHEC optimized multiplex assay proved to be high, as little as 1 ng/µL and 2 ng/µL, respectively. Previous study conducted by Mehrabdi et al. (2012) 10 which used three sets of primers ctxA, tcpA, and ompW stated that it was possible to detect even at lower numbers, down to between 8.5-85 pg of genomic DNA.
Signi cant sensitivity difference might was resulted from complex formation and competition between primers, therefore the more set of primer used, the more primer competition will happens 11 .
To con rm the speci city of the multiplex assay, we performed speci city evaluation on V. choleraee. No ampli ed product was seen with other non-V. choleraee bacterial strains using this multiplex PCR, this result indicates the high speci city of selected primers only speci c to V. choleraee. Mehrabdiet al (2012) have tested ctxA, tcpA, and ompW gene to Shigella dysenteriae, Aeromonas hydrophila, no ampli cation product was detected 10 . However the other study reported that V.mimicus might present ompU and toxR genes. This might result a cross reaction when both V. choleraee and V. mimicus present in the sample, although V. mimicus lack the core of the cholerae toxin element ctxA 7 .
Arti cal contamination was performed to analyze the capability of the assay to detect the contamination of pathogenic bacteria tested directly from the sample. The result showed that there was no signi cant difference between direct and indirect method. Pathogenic bacteria detection method in general normally used bacterial cultivation for selection and enrichment before going into the detection step. However, in outbreak cases that caused by pathogenic bacteria, immediate detection is needed to give the rapid treatment to the patients. The direct test was meant to see if the optimized method can be used straight to the contaminated water without growing the bacteria in advanced. Therefore, this assay was considered to be important to produce rapid diagnosis, where time is an important factor 12 . In addition, based on the result we also able to detect all of the pathogenic bacteria tested in all given temperature condition (room, refrigerator, and freezer), which is most common food and water storage placement. In this case, several bacteria such as V. cholerae could enter into a viable but non-culturable state in response to unfavorable temperature conditions. In this case, cultural identi cation method cannot detect V. choleraee contamination in sample 13 . Since the PCR does not distinguish among viable and dead bacterial cells, this method can be used to even detect all V. choleraee contamination in mineral water sample.
Finally we tested this assay using ISO/TS 20836:2017, Polymerase chain reaction for the detection of food-borne pathogens. We found that V. cholerae optimized methods had 0% false positive, 0% false negative, 100% speci city, and 100% sensitivity. While ETEC and EHEC optimized method had 0% false positive, 4% false negative, 100% speci city, and 96% sensitivity. Therefore, all of the optimized methods has met the requirement for PCR detection of food-borne pathogen according to ISO/TS 20836 limits, where acceptance limits for speci city and sensitivity are ≥ 70%, and ≤ 5% for false positive and false negative.

Limitation
This study only optimizes V. cholerae, EHEC, and ETEC rapid detection. Therefore, further studies need to be perform to optimize rapid detection of other food borne pathogenic bacteria. In addition, there is a possibility that a non-speci c band might appear during this detection.

Conclusion
In this research, several virulence genes primers were used to detect all pathogenic bacteria tested, using optimized multiplex PCR. This assay is able to detect V. cholerae which has ctxA, ompU, zot, toxR, and ace genes up to 0.25 ng genomic DNA. In addition, this assay is also able to detect ETEC and EHEC which has elt and stx genes up to 1 ng 2 ng, respectively. Based on the method standardization by ISO/TS 20836 these optimized methods are considered acceptable to detect food-borne pathogen tested.

Pathogenic bacteria cultivation
In this research, we used several of pathogenic bacteria V. cholerae C43 and E. coli ATCC 25922 which provided by BPOM; EHEC and ETEC which acquired from US Namru 2 Then, the bacteria were streaked onto LA, except for V. cholerae which were streaked onto LA + 2% (w/v) NaCl. In order to con rm the bacteria then each bacteria were grown in their selective media. For example, V. cholerae and E. coli were grown in TCBS and EMB, respectively. Subsequently, bacteria which showed the right morphology then streaked onto their growth media for further assay.

Genomic DNA extraction
The isolates were cultured on LB for overnight at 37 °C. The extraction of genomic DNA was performed by using boiling method 5 . Firstly, 1 mL of broth culture was centrifuged at 12000 x g for 5 min. Then the pellet was resuspended in 1 mL of NaCl (0.85% w/v), boiled for 5 min, and centrifuged again. The supernatant was stored at -20 °C for further use. Quantity, quality, and concentration of the extracted DNA were analyzed using Nanodrop instrument and gel electrophoresis.

Uniplex PCR
All primer pairs (Table 3) were tested in uniplex PCR at the estimated optimal annealing temperature to con rm correct ampli cation of the desired genes. Each primer pair was tested on uniplex PCR assay to ensure primer ampli cation ability and also con rming primer melting temperatures 6 . Mixture of the reaction and PCR condition was shown at Table 4 and Table 5, respectively. After PCR reaction, the ampli cation products were separated in 2.5% (w/v) agarose gel electrophoresis at 75 V for 95 minutes and visualized using GelDoc with EtBr dye. Multiplex PCR PCR ampli cation of the target DNA was carried out in a thermal cycler. The bacterial cell lysate was used for the template DNA to multiplex PCR using virulence and regulatory genes as their primers ( Table 1).
The mixture of PCR and PCR condition was shown at Table 6 and Table 7. Subsequently, the ampli cation products were separated in 2.5% agarose gel electrophoresis at 75 V for 90 minutes and visualized with GelDoc using EtBr dye. Sensitivity and speci city evaluation The sensitivity of the primes was tested with serial of dilution of the V. cholerae and E. coli (EHEC and ETEC) genomics, which was serially diluted from 10 ng; 5 ng; 1 ng; 0.5 ng; 0.25 ng and 10 ng; 5 ng; 2 ng; 1 ng; 0.5 ng; 0.1 ng, respectively. Limit of detection was determined with the lowest DNA concentration that gives clear bands 9 . Speci cally for V. cholerae speci city evaluation was performed in order to con rm whether primer pairs amplify only with the targeted bacteria and do not cross-react with DNA from other species. This test was performed by testing the primers with genomic DNA samples from V. choleraee, E. coli, S. typhi, and V. vulni cus.
Speci cty, sensitivity, false positive, and false negative assay DNA from pathogenic bacteria were tested and ampli ed using procedure according to ISO/TS 20836, the acceptance limit for speci city and sensitivity is ≥ 70%, and ≤ 5% for false positive and false negative were listed in Table 8 and Table 9. In this study we used two kinds of approach, which were growing the bacteria and isolate the genomic DNA (indirect method), then DNA from the samples was directly extracted (direct method). After arti cial contamination, mineral water sample was streaked onto TCBS or EMB agar and incubated overnight at 37 °C. The positive colonies were streaked into LA (37 °C, overnight) and then cultured in LB medium at 37 °C, 120 rpm overnight. Afterwards, one milliliter of the suspension was centrifuged at 7513 x g for two minutes. Following centrifugation, supernatant was discarded and DNA was extracted from the pellet using Wizard® Genomic DNA Puri cation Kit (Promega) based on manufacturer's guidelines for indirect method. At the same time, without growing the bacteria, mineral water samples that were arti cially contaminated were resumed for genomic DNA extraction using Wizard® Genomic DNA Puri cation Kit (Promega) based on manufacturer's guidelines for direct method.
The extracted genomic DNA from bacterial colonies as well as genomic DNA extracted directly from the samples were continued for multiplex PCR detection. PCR was employed with the same primers concentration and PCR condition as the previous step. The PCR condition is the same as the uniplex and multiplex method. The ampli cation products were separated in 2.5% agarose gel electrophoresis at 75 V for 90 minutes and visualized with GelDoc using EtBr dye. Availability of data and materials. The data of this study is available with the corresponding author up on request.
Competing interests. The authors declare that they have no competing interests.
Funding. This study was funded by Catholic University of Atma Jaya. The funder has no contribution in this study.
Author's contribution. DEW involved in research design and advisory. JP and RK gathering data and analysis. GR contribute in data analysis and manuscript preparation. All authors read and approved the nal manuscript.  Multiplex PCR performed on ETEC (1, 3, and 5) and EHEC (2, 4, and 6) with three repetitions.