We report an all-in-one platform – ScanDrop – for the rapid and specific capture, detection, and identification of bacteria in drinking water. The ScanDrop platform integrates droplet microfluidics, a portable imaging system, and cloud-based control software and data storage. The cloud-based control software and data storage enables robotic image acquisition, remote image processing, and rapid data sharing. These features form a “cloud” network for water quality monitoring. We have demonstrated the capability of ScanDrop to perform water quality monitoring via the detection of an indicator coliform bacterium, Escherichia coli, in drinking water contaminated with feces. Magnetic beads conjugated with antibodies to E. coli antigen were used to selectively capture and isolate specific bacteria from water samples. The bead-captured bacteria were co-encapsulated in pico-liter droplets with fluorescently-labeled anti-E. coli antibodies, and imaged with an automated custom designed fluorescence microscope. The entire water quality diagnostic process required 8 hours from sample collection to online-accessible results compared with 2–4 days for other currently available standard detection methods.
Citation: Golberg A, Linshiz G, Kravets I, Stawski N, Hillson NJ, et al. (2014) Cloud-Enabled Microscopy and Droplet Microfluidic Platform for Specific Detection of Escherichia coli in Water. PLoS ONE 9(1): e86341. doi:10.1371/journal.pone.0086341
Editor: James P. Brody, University of California, Irvine, United States of America
Received: November 3, 2013; Accepted: December 12, 2013; Published: January 27, 2014
Copyright: © 2014 Golberg et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: AG and MY acknowledge Shriners Foundation Grant #85120-BOS for the support of this work. The portion of this work conducted by the Joint BioEnergy Institute, and the U.S. Department of Energy Joint Genome Institute, was supported by the Office of Science, Office of Biological and Environmental Research, of the U.S. Department of Energy (Contract No. DE-AC02–05CH11231). This research is funded in part by the Singapore National Research Foundation and the publication is supported under the Campus for Research Excellence And Technological Enterprise (CREATE) program for the project ‘Nanomaterials for Energy and Water Management’. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Worldwide water-associated infectious diseases are a major cause of morbidity and mortality . It is estimated that 4.0% of global deaths and 5.7% of the global disease burden are caused by waterborne diseases –. Common waterborne diseases include diarrhea (bacterial, viral and parasitic), schistosomiasis, trachoma, ascariasis, and trichuriasis –. Low income countries are particularly vulnerable to waterborne diseases because of their under-developed infrastructure and poor water management –. Water and sewage distribution systems in high income societies also require pollutant and microorganism monitoring .
Escherichia coli, found in mammalian feces , has been a biological indicator for water quality since the 19th century . Testing for the presence of E. coli is obligatory for current water management systems –. Herein, we report a comprehensive system – ScanDrop – for the rapid and specific identification of E. coli in drinking water.
The identification of bacteria in a water sample includes two major steps: 1) the capture of target bacteria from the water sample, and 2) the identification of the captured bacteria. Traditional methods for E. coli detection include culture, fermentation, enzyme-linked immunosorbent (ELISA), and polymerase chain reaction (PCR) assays , . These traditional methods have disadvantages including long identification times (2–4 days), and/or high labor and reagent costs , . Despite high costs, rapid tests are necessary to enable quick responses to putative contamination threats. Recently, novel sensors and assays for rapid pathogen detection have been developed, including the capture of whole pathogen cells or molecular fragments for further amplification and identification –, with detection methods utilizing a variety of transducing technologies (optical, electrochemical, surface plasmon resonance and piezoelectric) –. Many of these newer methods remain expensive and/or require sophisticated instrumentation, and most have yet to reach the market place. Therefore, there remains a need for alternative platforms for the detection of bacteria in water samples.
It remains challenging to inexpensively perform water quality control testing at multiple locations along a distribution system, and to rapidly process and share the test results. To address these challenges, we have developed the ScanDrop platform. ScanDrop is a self-contained detection platform that enables the online control of water testing at multiple locations along the distribution system. ScanDrop integrates live-bacteria capturing and detection, droplet microfluidics, automated fluorescence microscopy, and cloud-based data management and sharing. Droplet microfluidics, applied in ScanDrop, is an emerging application of microelectromechanical systems (MEMS) technology, where assay reagents and biological sample are confined to the pico-liter reactors, composed of water in oil emulsion –. Small volumes, rapid reagent mixing and non-complex droplet control make droplet microfluidics an attractive choice for the next-generation of high-throughput assays – and herein detection of bacteria in water samples.
In this work, we demonstrate ScanDrop's capability to detect live E. coli in water samples. Magnetic beads, conjugated with specific antibodies, were used to quickly and effectively capture E. coli from contaminated water. The captured bacteria were then encapsulated into pico-liter droplets containing fluorescently labeled antibodies, for subsequent detection using a proprietary automated optical fluorescence signal registration system. Imaging system control was facilitated by leveraging a cloud-based laboratory automation system, coined Programing a Robot, PR-PR . We envision that multiple ScanDrop systems could be dispatched at multiple locations to form a cloud-enabled water quality assessment network. Each system could be managed in real-time from a remote control center. Such a network could potentially reduce the infrastructure, management, and labor costs required to perform multiple sample analysis and rapidly share results.
Results and Discussion
Bead-based E. coli capture and detection assay
Herein the isolation of bacteria and detection are conducted utilizing simple magnetic bead based immunoassay thus no bacteria agar plate cultivation step is necessary to identify a presumptive positive sample. This approach saves considerable time and resources. In our approach, magnetic beads conjugated with anti-E. coli antibodies are added to a water sample (Fig. 1). Within 10 min, the magnetic beads have captured the bacteria (if any) from the water sample. The beads are then concentrated with a simple magnet (Fig. 1), and a single immunoassay step labels the captured bacteria with a fluorescent antibody for subsequent detection (Fig. 1). Detection protocols are integrated into a droplet microfluidic device to reduce reagent volume and enhance reaction rates.
Figure 1. Bacteria capturing and detection assay.
Magnetic bead capture of E. coli from enriched water samples, and downstream chip encapsulation for fluorescent labeling and detection. 1L of water is passed through a 0.22 µm filter, which is then incubated for 6 hr in LB media. Dynabeads® MAX anti-E. coli O157 are added to the resulting cell culture (“sample”), incubated for 20 min, and concentrated via magnet. The beads (potentially conjugated with bacteria) are then co-encapsulated with secondary fluorescently labeled anti-E. coli antibodies in the chip and incubated up to 1 hour before imaging.doi:10.1371/journal.pone.0086341.g001
The ScanDrop sensor consists of two major components: 1) a droplet microfluidic device for bacteria labeling, and 2) a portable fluorescent optical system for signal detection and sharing.
Droplet microfluidic device.
To reduce reagent volumes and detection times, we designed a pico-liter droplet microfluidic chip. The design of poly(dimethylsiloxane) (PDMS) microfluidic device is shown in Fig. 2A, the generation of monodisperse droplets in a micro-channel through shearing flow at a flow-focusing zone in Fig. 2B, and the resulting droplet array in Fig. 2C. Three perpendicular inlet channels form a nozzle, (Fig. 2A rectangle), independent syringe pumps controlling flow rates for the oil, beads, and fluorescently labeled secondary antibodies streams. Each droplet in the array co-encapsulates fluorescently labeled anti-E. coli antibodies with captured bacteria (if any), to generate a localized fluorescent signal for subsequent detection. The chip enables the generation and incubation of 103 droplets with ~100 micron diameter (~520 pL). The advantages of this droplet-based array technique include the physical and chemical isolation of beads in droplets, and the rapid and efficient mixing of the reagents that occurs inside droplets providing fast reaction rates –. Importantly, this nano-liter microenvironment also enables gas exchange for bacterial viability if further studies are required –. Previous works in the field of droplet microfluidics showed that the chance to find a cell or a bead inside droplet follows Poisson distribution , . This puts certain theoretical limitations of the limit of detection of bacteria of droplet microfluidic system. Clausell-Tormos et al. showed that decreasing the number of cells in aqueous solution that is converted to droplets to less than 106 cell/mL reduced the probability to find droplets with encapsulated cells and increased the number of empty droplets . Therefore, at the end of incubation time we need to get at least 106 CFU/mL of bacteria. The relation between the bacteria in the analyzed sample and the bacteria subjected to encapsulation after enrichment is as follows:
Where No (CFU/mL) is the initial load of bacteria, n is the number of generations in enrichment phase, Nd (CFU/mL) is the concentration of bacteria subjected to encapsulation after enrichment.
Figure 2. Droplet microfluidic device for bacteria monitoring.
A) Schematic representation of ScanDrop droplet microfluidic chip and fluid control system. B) Droplet generator. C) Droplet incubation array (up to 103 droplets can be incubated simultaneously). D) Co-encapsulation of GFP- and RFP-expressing E. coli inside a single droplet (20× magnification, Zeiss microscopic imaging). Arrows indicate single bacteria cells.doi:10.1371/journal.pone.0086341.g002
Given the generation time of 20 min for E. coli in the optimum cultivation condition, to get 106 CFU/mL at the end of 6 hours of incubation (18 generations), proposed in our assay, the initial concentration should be at least 3.5 CFU/mL.
To demonstrate the feasibility of our ScanDrop system for multiplex analysis, we co-encapsulated red florescent protein (RFP) and green florescent protein (GFP) expressing E. coli in the same droplet (Fig. 2D). Capturing on a bead and latter encapsulation for detection of two different bacteria in the same microenvironment will enable multiplex future studies using several types of beads conjugated with different antibodies that bind different target bacteria and different fluorescent tags. The probability of capturing two different objects in a single droplets were analyzed in .
The schematic for the portable optical system for fluorescent signal detection in the droplet microfluidic device is presented in Fig. 3A. The system enables remote microscope control as well as simultaneous top and inverted image registration (Fig. 3B). The top camera allows for whole chip bright field imaging, while the bottom camera allows for fluorescence imaging with 10× magnification. This combination allows for high-throughput droplet imaging. A robotic stage is used to scan the array of multiple droplets, with an XY microscope scanning range of 45 mm×45 mm and a resolution of 5 µm/step (10 mm/sec). Z-axis focus capabilities include 15 mm travel with 1 µm/step, at 2 mm/sec. The ScanDrop optical system is controlled by Python scripts which can be automatically generated by PR-PR.
Figure 3. Portable fluorescence microscope system.
A) Microscope design scheme. B) Robotic stage and two cameras (arrows) for chip observation and data acquisition. C) ScanDrop system set up, including 2 pressure pumps to create droplets, a microscope imaging system, internet access, and a monitor for local data viewing.doi:10.1371/journal.pone.0086341.g003
PR-PR cloud-based Laboratory Automation System, and Data Management and Sharing
In this work, we have further developed PR-PR, a biology-friendly high-level language for laboratory automation , to control ScanDrop's automated microscopy system and enable ScanDrop to be promptly and easily adjusted to changes in experimental protocol. In PR-PR, transfer of a material (e.g., a liquid) or system component (e.g., a robotic arm) is described by a Source, Destination, Quantity, and Method. For ScanDrop, the Source is the initial coordinates of the microscope stage (XY) and lens (Z), the Destination is the final target coordinates of interest, the Quantity is the number of pictures that should be taken, and the Method specifies imaging parameters: light, filters, and delay between image capture. PR-PR inputs a script for ScanDrop automated microscope control (such as that presented in Fig. S2) and outputs a Python script that can directly operate the ScanDrop automated microscope system. The PR-PR script protocol and the resulting data for each experiment are stored in a local folder within the ScanDrop sensor and can shared between users via Dropbox.
Detection of RFP-expressing E. coli in drinking water
As a positive control, we tested ScanDrop for the detection of 150 CFU/mL of RFP-expressing E. coli in drinking water (Fig. 4). The overall assay for E. coli detection is divided into three steps: enrichment, capture, and detection. For enrichment, 1L of contaminated water sample was filtered and the filter with captured bacteria was incubated for 6 hours in the LB medium. For capturing, the enriched solution was mixed with Dynabeads® MAX anti-E. coli O157 for 10 min and separated by magnet.
Figure 4. ScanDrop: E. coli detection.
A) Microfluidic chip operating inside the imaging system. B) Droplet array as viewed from the top camera. Arrows indicate tubing/chip connection locations as follows: o- oil (inlet); b- beads conjugated with E. coli (inlet); a- fluorescently labeled antibody (inlet); w- waste (outlets). C) Droplet image as seen from the top camera with white LED illumination. D) Antibody green fluorescence indicates the presence of RFP-expressing E. coli (positive control).doi:10.1371/journal.pone.0086341.g004
Fig. 4B shows the droplet-based microfluidic chip used to perform the immunoassay described in Fig. 1. For the detection step, beads conjugated to bacteria captured from contaminated water sample were co-encapsulated with secondary FITC fluorescently labeled anti-E. coli antibodies in the droplet array and incubated up to 1 hr in the chip array at room temperature. Fig. 4C shows a representative droplet, and Fig. 4D shows the green florescence signal detected in a single droplet containing E. coli capture on the bead and tagged by secondary FITC labeled antibodies. The presence of RFP expressing E. coli in water was confirmed by PCR (Fig. 5 C, D).
Figure 5. Detection of fecal E. coli in drinking water.
A) ScanDrop detection network. PR-PR generates a Python script to control the ScanDrop detector. The Python script is uploaded to Dropbox, and then run on the ScanDrop detector. The captured images are uploaded to Dropbox, and then distributed to various devices. B) Representative ScanDrop image demonstrating fecal E. coli detection in drinking water. C) Real-time PCR amplification plot for contaminated sample 3, and positive and negative controls. Red curves indicate amplification of primary 16S rRNA locus, cyan amplification of secondary 16S rRNA locus, green tuf locus, and blue uidA locus. Sample 3 and the positive control amplified similarly. The negative control did not amplify. D) Gel electrophoresis analysis of PCR reaction products for the four contaminated samples, along with positive and negative controls. For all samples, gel lanes correspond to the amplification of loci as follows: lane 1 - 16S rRNA primary locus, lane 2 - 16S rRNA secondary locus, lane 3 - tuf, lane 4 – uidA.doi:10.1371/journal.pone.0086341.g005
Detection of fecal E. coli in drinking water
Next, we tested the ScanDrop system for the detection of fecal E. coli in drinking water. We contaminated the water with rat feces and applied the droplet detection assay described above. Fig. 5A presents the procedure flow for the detection of fecal E. coli in water. Fig. 5B shows a representative resulting image, with fluorescence signal indicating E. coli contamination. We confirmed the ScanDrop detection with PCR (Fig. 5 C,D), which clearly showed that the water samples were contaminated with E. coli. The results from the ScanDrop tests were uploaded to Dropbox cloud data storage.
The work presented here demonstrates the potential of automated microscale systems for water quality analysis. To detect E. coli in water samples, we developed and demonstrated a bead-based immuno-assay performed with a droplet microfluidic device to reduce reagent volume and enhance reaction rates. We integrated the microfluidic assay with a portable imaging system and remote control automation software. We demonstrated ScanDrop system capabilities through the detection of model coliform bacteria, E. coli, in feces–contaminated drinking water. Our successful multiplex detection assay results suggest that simultaneous multiple bacteria detection, using several types of beads conjugated with different antibodies that bind different target bacteria, will be possible with further development. The ScanDrop platform decreased reagent volumes, (the full chip uses 520 nL of reagents, while conventional assay require at least 10 µL of reagents) and allows for results within 8 hours from the time of water sampling. Our results demonstrate that a combination of droplet microfluidics with low cost optics and cloud network can provide a flexible and efficient alternative for pathogen detection in drinking water. The ScanDrop platform has the potential to significantly improve water diagnostics, particularly in low income countries where the infrastructure does not yet exist , .
We developed the ScanDrop platform for E. coli detection in water. The platform uses magnetic beads to capture bacteria, droplet microfluidics to encapsulate the captured bacteria with fluorescent antibodies, low cost portable optics for signal detection, PR-PR to facilitate microscopy control and data acquisition, and cloud-based storage for results sharing. The use of droplet microfluidics increases reaction kinetics and reduces reagent volumes (lowering the cost per test), the developed florescence microscopy system allows for data generation in multiple locations, and PR-PR facilitates ScanDrop control. A schematic illustration of an envisioned ScanDrop network for water quality analysis is shown in Fig. 6. The ScanDrop network would consist of 1) the PR-PR laboratory automation system and cloud-based data storage for remote control, image capturing, and result sharing; and 2) ScanDrop sensor stations deployed at multiple water distribution locations. The control station would perform image analysis for multiple sensors and shares the test results in the real time with multiple end users. This ScanDrop network could contribute to more rapid, cost-effective, and continuous water quality monitoring systems, with centralized facilities simultaneously monitoring multiple water sampling sites without complex imaging or data processing infrastructure.
Figure 6. Schematic overview of a ScanDrop cloud-based water quality assessment system.
The ScanDrop detector network is enabled by PR-PR and cloud-based data storage. Users send requests for water quality assessment at different locations in the distribution system. ScanDrop detectors perform the tests, the results are stored in the cloud, and the collected data is shared between users and applications.doi:10.1371/journal.pone.0086341.g006
Materials and Methods
ScanDrop optics system
A custom made, motorized, dual view, computerized portable microscopy system was designed for droplet microfluidic imaging (R&D Engineering Solutions, Netania, Israel). The dual view system was used for the simultaneous imaging of the whole chip (top view camera) and specific droplets (bottom view camera). The top view camera includes: 1280×768 resolution, color sensor, auto/computer-controlled focus, manually configurable [83×50 mm - 30×18 mm] field of view, 640×480 region of interest (ROI), and zoom functionality. The bottom view (microscope) camera includes: 752×582 resolution, monochrome 8.6 µm×8.3 µm pixels sensor, and a 10× objective. A single 3W 468 nm light emitting diode (LED) was used for florescence excitation. A 41017 - Endow GFP/EGFP bandpass fluorescence filter set (Chroma Inc., VT) was used for florescence detection. Top illumination was made by a single 30 mW white LED for chip observation and microscope camera positioning. An embedded ×86 dual core computer with HDMI display port outputs (CompuLab, Israel) was used for the local control of the system. An embedded computer runs custom software, which allows full control of the microscope, including XY position, focus, illumination, image acquisition and enhancement. A system can be controlled manually by the human operator via a standard PC console (keyboard, mouse and monitor). Alternatively, a system can be controlled programmatically via a program written in Python. We further improved the programmatic control aspect of our system by leveraging PR-PR , whereby a PR-PR microscope control script (Fig. S2) is translated into a Python script that can control the microscope system, as described above. Python script deployment and image retrieval across distributed microscope systems was performed with the Dropbox cloud-based storage service.
Bacterial strains and plasmids
Plasmids pFAB_SchPMK36GFP and pFAB_SchPMK36RFP (unpublished results, Vivek Mutalik, Drew Endy, and Adam Arkin; see Fig. S1), both carrying a kanamycin resistance marker, were transformed into E. coli BW25113. These bacterial strains and plasmids, along with their associated information (e.g., annotated Genbank-format DNA sequence files), have been deposited in the public instance of the JBEI Registry  (https://public-registry.jbei.org; corresponding Part IDs JPUB_001327-001329). For transformation, 1 µL pFAB_SchPMK36GFP or pFAB_SchPMK36RFP was mixed on ice with 40 µL chemically competent E. coli BW25113. The mixture was incubated on ice for 20 min, then placed at 42°C for 45 s (heat shock), and then returned to ice. 200 µL SOC media was then added to each tube of transformed cells and incubated with agitation at 37°C for 30 min. 100 µL of each transformation mixture was plated on solid LB media (Sigma-Aldrich, MO) supplemented with 30 µg/mL kanamycin (Sigma-Aldrich, MO) and then cultured at 37°C.
Microfluidic device for droplet generation
The droplet microfluidic flow focusing device mask was fabricated by soft lithography. Negative photo resist SU-8 2100 (MicroChem, Newton, MA) was deposited onto clean silicon wafers to a thickness of 150 µm, and patterned by exposure to UV light through a transparency photomask (CAD/Art Services, Bandon, OR). To manufacture consumable devices, Sylgard 184 poly(dimethylsiloxane) (PDMS) (Dow Corning, Midland, MI) was mixed with cross-linker (ratio 10:1) and poured onto the photoresist pattern, degassed thoroughly and cured for 12 hours at 75°C. After curing, the PDMS devices were peeled off the wafer and bonded to glass slides after oxygen-plasma activation of both surfaces. The microfluidic device was composed of two parts: 1) a droplet forming nozzle (channel cross section 6.25·10−8 m2) and 2) a 103 droplets storage array (channel cross section 3.13·10−7 m2). The bonded microfluidic channels were treated with Pico-Sur™ 2 (Dolomite Microfluidics, UK) by filling the channels with 10 µL of the solution as received and then flushing with air. This treatment was done to improve the wetting of the channels with mineral oil in the presence (1% w/w) of the surfactant (span80). 1 mL syringes were used to load the fluids into the devices through Tygon Micro Bore PVC Tubing 100f, 0.010″ ID, 0.030″ OD, 0.010″ Wall (Small Parts Inc, FL). Individual syringe pumps (Harvard Apparatus, USA) were used to control the flow rates of oil and other reagents. To form droplets, the flow-rate-ratio of water-to-oil was adjusted to Qw/Qo = 1.
Droplet microfluidics multiplex detection assay
E. coli expressing GFP or RFP were incubated for 12 hours at 37°C in LB media (Sigma-Aldrich, MO) to 106 CFU/mL and encapsulated into droplets. Fluorescence images were captured on a Zeiss 200 Axiovert microscope using an AxioCAM MRm digital camera and AxioVision 4.8 software at 20× magnification. Each experiment consisted of 4 repeats.
ScanDrop detection of E. coli in water
1 L drinking water was spiked with RFP-expressing E. coli to 150 CFU/mL. The spiked water was filtered through a 0.22 µm filter (Corning Inc., NY), and the filter was then inoculated in 10 mL LB media (Sigma-Aldrich, MO) and incubated for 6 hr at 37°C. 20 µL of Dynabeads® MAX anti-E. coli O157 beads (Life Technologies, CA) were added to 1.5 mL of the incubation media and further incubated for 20 min on a rotating stage at room temperature (RT). Beads with captured bacteria were separated by magnet and resuspended in 400 µL of Phosphate Buffered Saline (PBS). The resuspended solution was co-encapsulated 500:1 with green fluorescently labeled anti-E. coli antibody (FITC, ab30522, Abcam, MA) in droplet reactors inside the chip positioned on the ScanDrop robotic stage. After a further 1 hr of incubation at RT, images were taken from different locations on the chip. The objective position movements were controlled via PR-PR, and the generated images were automatically uploaded to Dropbox. Each experiment consisted of 4 repeats.
Drinking water contaminated with rat feces
Fresh feces were collected from rat cages in the animal facility of Massachusetts General Hospital. 1.5 g feces was mechanicaly homogenized in 1 L of drinking water. The contaminated water was filtered twice through a 40 µm filter (BD Falcon™, BD Biosciences, CA). The detection of E.coli in permeate was done by ScanDrop assay (as described in the previous section) and by Real-Time PCR (as described in the following section). Each experiment consisted of 4 repeats.Detection of RFP expressing and fecal origin E. coli by PCR. We chose four primer sets (Table 1) to detect E. coli in the prepared drinking water contaminated with rat feces. The primers targeted specific sequences from different loci in the E. coli genome: two primer sets for 16S rRNA, one for tuf, and one for uidA –. For each primer set, we tested four contaminated samples and a positive and a negative control. Negative controls contained water only, and positive controls contained water supplemented with E. coli BW25113. After enrichment of the microbial population (described above), 5 µL of enriched culture was added to 45 µL H2O. All samples were incubated 15 min at 98°C and then diluted in additional 100 µL H2O. Each 30 µL PCR reaction contained 10 µL of the diluted cell lysate (as template), 10 µL of 3× qPCR master mix (H2O 3.3 µL, 5× Phusion HF 6 µL, dNTP 100 mM 0.25 µL, Phusion DNA Polymerase (NEB) 0.3 µL, SYBR® Green II 200× (Molecular Probes) 0.15 µL), and a pair of primers at 5 pmol each. PCR reactions were subjected to thermal cycling (3 min at 95°C, and then 30 cycles of 30 s at 95°C, 30 s at 58°C, and 30 s at 72°C, with a final hold step at 10°C) in a StepOnePlus™ Real-Time PCR System (Life Technologies, CA). We tracked the amplification curves and stopped the PCR amplifications after most reactions plateaued (Fig. 5c). We analyzed PCR fragments using electrophoresis by running the PCR products in 1% agarose gels (Fig. 5d). Each experiment consisted of 4 repeats.
Table 1. Primers and loci used for PCR detection of E. coli.doi:10.1371/journal.pone.0086341.t001
PR-PR software availability
PR-PR is open-source software under the BSD license and is freely available from GitHub (https://github.com/jbei/prpr), and is also available through its web interface on the public PR-PR webserver (http://prpr.jbei.org).
Plasmid maps for pFAB_SchPMK36GFP and pFAB_SchPMK36RFP.
Representative PR-PR script for ScanDrop. LOCATION declarations define microscope stage (XY) and lens (Z) locations. TRANSFER commands specify the starting and destination locations, the number of pictures to capture, and the capture parameters. In a single TRANSFER statement, multiple sequential destinations can be defined by location offset and number of repetitions.
The authors thank Vivek Mutalik for the gift of plasmids pFAB_SchPMK36GFP and pFAB_SchPMK36RFP.
Conceived and designed the experiments: AG GL TK. Performed the experiments: AG GL IK NS. Analyzed the data: AG GL TK. Contributed reagents/materials/analysis tools: AG GL IK NS NH TK MY. Wrote the paper: AG GL NH TK. Read the manuscript: MY RM Supervised the work: MY RM.
- 1. Yang K (2012) Global distribution of outbreaks of water-associated infectious diseases. PLoS Negl Trop Dis, 2012 6 (2) e1483. doi: 10.1371/journal.pntd.0001483
- 2. Lewin S (2007) Estimating the burden of disease attributable to unsafe water and lack of sanitation and hygiene in South Africa in 2000. S Afr Med J, 2007 97 (8 Pt 2) 755–62.
- 3. Kosek M, Bern , Guerrant RL (2000) The global burden of diarrhoeal disease, as estimated from studies published between 1992 and 2000. Bull World Health Organ, 2003 81 (3) 197–204.
- 4. Pruss A (2002) Estimating the burden of disease from water, sanitation, and hygiene at a global level. Environ Health Perspect, 2002 110 (5) 537–42. doi: 10.1289/ehp.02110537
- 5. Cann KF (2013) Extreme water-related weather events and waterborne disease. Epidemiol Infect 141 (4) 671–86. doi: 10.1017/s0950268812001653
- 6. Teschke K (2010) Water and sewage systems, socio-demographics, and duration of residence associated with endemic intestinal infectious diseases: a cohort study. BMC Public Health 10: 767. doi: 10.1186/1471-2458-10-767
- 7. Sarkar R (2013) Burden of childhood diseases and malnutrition in a semi-urban slum in southern India. BMC Public Health, 2013 13: 87. doi: 10.1186/1471-2458-13-87
- 8. Jafari N (2011) Prevention of communicable diseases after disaster: A review. J Res Med Sci, 2011 16 (7) 956–62.
- 9. Subbaraman R (2013) The social ecology of water in a Mumbai slum: failures in water quality, quantity, and reliability. BMC Public Health, 2013 13: 173. doi: 10.1186/1471-2458-13-173
- 10. Levy K (2012) Rethinking indicators of microbial drinking water quality for health studies in tropical developing countries: case study in northern coastal Ecuador. Am J Trop Med Hyg, 2012 86 (3) 499–507. doi: 10.4269/ajtmh.2012.11-0263
- 11. Stoler J (2012) When urban taps run dry: sachet water consumption and health effects in low income neighborhoods of Accra, Ghana. Health Place, 2012 18 (2) 250–62. doi: 10.1016/j.healthplace.2011.09.020
- 12. Gwimbi P (2011) The microbial quality of drinking water in Manonyane community: Maseru District (Lesotho). Afr Health Sci, 2011 11 (3) 474–80.
- 13. Mandeville KL (2009) Gastroenterology in developing countries: issues and advances. World J Gastroenterol, 2009 15 (23) 2839–54. doi: 10.3748/wjg.15.2839
- 14. Al-Bayatti KK, Al-Arajy KH, Al-Nuaemy SH (2012) Bacteriological and physicochemical studies on Tigris River near the water purification stations within Baghdad Province. J Environ Public Health, 2012 2012: 695253. doi: 10.1155/2012/695253
- 15. Ashbolt N, Grabow W, Snozzi M(2001) Indicators of Microbial Water Quality, in Water Quality: Guidelines, Standards and Health, LFewtrell and JBartman, Editors. IWA Publishing: London,UK. p. 289–316.
- 16. Edberg SC (2000) Escherichia coli: the best biological drinking water indicator for public health protection. Symp Ser Soc Appl Microbiol (29) 106S–116S. doi: 10.1111/j.1365-2672.2000.tb05338.x
- 17. (2006) EPA, Total Coliform Rule Requirements. Available: http://water.epa.gov/lawsregs/rulesregs/sdwa/tcr/regulation.cfm. Accessed Nov 25 2013.
- 18. WHO (2006), Guidlines for drinking-water Quality. Third Edition.
- 19. Leclerc, H., Relationships between common water bacteria and pathogens in drinking-water, in Heterotrophic Plate Counts and Drinking-water Safety, JBartman, et al.., Editors. 2003, IWA Publishing: London, UK. p. 80–118.
- 20. (2005) StandardMethods, For the examination of water and wastewater. 9020 Quality Assurance/Quality control. Available: http://www.standardmethods.org/store/ProductView.cfm?ProductID=538. Accessed Nov 25 2013
- 21. Mairhofer J, Roppert K, Ertl P (2009) Microfluidic systems for pathogen sensing: a review. Sensors (Basel), 2009 9 (6) 4804–23. doi: 10.3390/s90604804
- 22. Kao MC, Durst RA (2010) Detection of Escherichia coli Using Nucleic Acid Sequence-Based Amplification and Oligonucleotide Probes for 16S Ribosomal RNA. Analytical Letters 43 (10–11) 1756–1769. doi: 10.1080/00032711003654005
- 23. Xiao XL (2013) Detection of viable but nonculturable Escherichia coli O157:H7 using propidium monoazide treatments and qPCR. Can J Microbiol 59 (3) 157–63. doi: 10.1139/cjm-2012-0577
- 24. Langer V, Niessner R, Seidel M (2013) Stopped-flow microarray immunoassay for detection of viable E. coli by use of chemiluminescence flow-through microarrays. Anal Bioanal Chem, 2013 399 (3) 1041–50. doi: 10.1007/s00216-010-4414-0
- 25. Rompre A (2002) Detection and enumeration of coliforms in drinking water: current methods and emerging approaches. J Microbiol Methods, 2002 49 (1) 31–54. doi: 10.1016/s0167-7012(01)00351-7
- 26. Liu Y (2008) Detection of viable but nonculturable Escherichia coli O157:H7 bacteria in drinking water and river water. Appl Environ Microbiol 74 (5) 1502–7. doi: 10.1128/aem.02125-07
- 27. Noble RT, Weisberg SB (2005) A review of technologies for rapid detection of bacteria in recreational waters. J Water Health 3 (4) 381–92.
- 28. Setterington EB, Alocilja EC (2011) Electrochemical Biosensor for Rapid and Sensitive Detection of Magnetically Extracted Bacterial Pathogens. Biosensors 2 (1) 15–31. doi: 10.3390/bios2010015
- 29. Bogomolny E, Swift S, Vanholsbeeck F (2013) Total viable bacterial count using a real time all-fibre spectroscopic system. Analyst 138 (14) 4112–9. doi: 10.1039/c3an00254c
- 30. Allevi RP (2013) Quantitative analysis of microbial contamination in private drinking water supply systems. J Water Health 11 (2) 244–55. doi: 10.2166/wh.2013.152
- 31. Johnson PE (2006) Fountain Flow cytometry, a new technique for the rapid detection and enumeration of microorganisms in aqueous samples. Cytometry A 69 (12) 1212–21. doi: 10.1002/cyto.a.20351
- 32. Sun H (2009) Nano-silver-modified PQC/DNA biosensor for detecting E. coli in environmental water. Biosens Bioelectron 24 (5) 1405–10. doi: 10.1016/j.bios.2008.08.008
- 33. Abu-Rabeah K (2009) Highly sensitive amperometric immunosensor for the detection of Escherichia coli. Biosensors and Bioelectronics 24: 3461–3466. doi: 10.1016/j.bios.2009.04.042
- 34. Liebes Y (2009) Immobilization strategies of Brucella particles on optical fibers for use in chemiluminescence immunosensors. Talanta 80: 338–345. doi: 10.1016/j.talanta.2009.06.070
- 35. Liebes Y, Marks R, Banai M (2009) Chemiluminescent optical fiber immunosensor detection of Brucella cells presenting smooth A antigen. Sensors and Actuators B 140: 568–576. doi: 10.1016/j.snb.2009.04.050
- 36. Abdulhalim I (2009) Surface enhanced fluorescence from metal sculptured thin films with application to biosensing in water. App Phys Lett 94: 063206. doi: 10.1063/1.3081031
- 37. Abdulhalim I, Zourob M, Lakhtakia A Surface plasmon resonance sensors-a mini review. J Electromagnetism 2008 28 (3) 213–242. doi: 10.1080/02726340801921650
- 38. Karabchevsky A (2012) Microspot biosensing based on surface enhanced fluorescence from Nano-STFs. J NanoPhotonics, 2012 6 (6) 061508–1. doi: 10.1117/1.jnp.6.061508
- 39. Shalabney A, Abdulhalim I (2011) Sensitivity enhancement methods for surface plasmon sensors. Lasers and Photonics Reviews 5: 571–606. doi: 10.1002/lpor.201000009
- 40. Zourob M, Elwary S, Turner A (2008) Principles of Bacterial Detection - Biosensors, Recognition Receptors and Microsystems. NY: Springer.
- 41. Niu X, deMello AJ (2012) Building droplet-based microfluidic systems for biological analysis. Biochem Soc Trans 40 (4) 615–23. doi: 10.1042/bst20120005
- 42. Choi K (2012) Digital microfluidics. Annu Rev Anal Chem (Palo Alto Calif) 5: 413–40.
- 43. Guo MT (2012) Droplet microfluidics for high-throughput biological assays. Lab Chip 12 (12) 2146–55. doi: 10.1039/c2lc21147e
- 44. Linshiz G (2012) PaR-PaR Laboratory Automation platform. ACS Synthetic Biology 2 (5) 216–222. doi: 10.1021/sb300075t
- 45. Baroud CN, Gallaire F, Dangla R (2010) Dynamics of microfluidic droplets. Lab on Chip 10: 2032–2045. doi: 10.1039/c001191f
- 46. Bringer MR (2004) Microfluidic systems for chemical kinetics that rely on chaotic mixing in droplets. Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences 362 (1818) 1087–1104. doi: 10.1098/rsta.2003.1364
- 47. Handique K, Burns MA (2001) Mathematical modeling of drop mixing in a slit-type microchannel. Journal of Micromechanics and Microengineering 11: 548–554. doi: 10.1088/0960-1317/11/5/316
- 48. Sarrazin F (2007) Mixing characterization inside microdroplets engineered on a microcoalescer. Chemical Engineering Science 62 (4) 1042–1048. doi: 10.1016/j.ces.2006.10.013
- 49. Mazutis L (2013) Single-cell analysis and sorting using droplet-based microfluidics. Nature Protocols 8: 870–891. doi: 10.1038/nprot.2013.046
- 50. Clausell-Tormos J (2008) Droplet-based microfluidic platforms for the encapsulation and screening of Mammalian cells and multicellular organisms. Chem Biol, 2008 15 (5) 427–37. doi: 10.1016/j.chembiol.2008.04.004
- 51. Reyes DR (2002) Micro total analysis systems. 1. Introduction, theory, and technology. Anal Chem 74 (12) 2623–36. doi: 10.1021/ac0202435
- 52. Buffi N (2011) Development of a microfluidics biosensor for agarose-bead immobilized Escherichia coli bioreporter cells for arsenite detection in aqueous samples. Lab Chip, 2011 11 (14) 2369–77. doi: 10.1039/c1lc20274j
- 53. Ham TS (2012) Design, implementation and practice of JBEI-ICE: an open source biological part registry platform and tools. Nucleic Acids Res, 2012 40 (18) e141. doi: 10.1093/nar/gks531
- 54. Maheux AF (2009) Analytical comparison of nine PCR primer sets designed to detect the presence of Escherichia coli/Shigella in water samples. Water Res, 2009 43 (12) 3019–28. doi: 10.1016/j.watres.2009.04.017
- 55. Tsen HY, Lin CK, Chi WR (1998) Development and use of 16S rRNA gene targeted PCR primers for the identification of Escherichia coli cells in water. J Appl Microbiol 85 (3) 554–60. doi: 10.1046/j.1365-2672.1998.853535.x
- 56. Tantawiwat S (2005) Development of multiplex PCR for the detection of total coliform bacteria for Escherichia coli and Clostridium perfringens in drinking water. Southeast Asian J Trop Med Public Health 36 (1) 162–9.
- 57. Sabat G (2000) Selective and sensitive method for PCR amplification of Escherichia coli 16S rRNA genes in soil. Appl Environ Microbiol, 2000 66 (2) 844–9. doi: 10.1128/aem.66.2.844-849.2000