Grantee Research Project Results

2008 Progress Report: Development and Evaluation of an Innovative System for the Concentration and Quantitative Detection of CCL Pathogens in Drinking Water

EPA Grant Number: R833003
Title: Development and Evaluation of an Innovative System for the Concentration and Quantitative Detection of CCL Pathogens in Drinking Water
Investigators: Tzipori, Saul , Walt, David , Zuckermann, Udi
Current Investigators: Tzipori, Saul , Zuckermann, Udi , Walt, David
Institution: Tufts University
EPA Project Officer: Aja, Hayley
Project Period: August 1, 2006 through August 1, 2009 (Extended to July 31, 2011)
Project Period Covered by this Report: August 1, 2007 through August 1,2008
Project Amount: $600,000
RFA: Development and Evaluation of Innovative Approaches for the Quantitative Assessment of Pathogens in Drinking Water (2005) RFA Text | Recipients Lists
Research Category: Drinking Water , Water

Objective:

In a previous EPASTAR award we had developed optimized and validated at Tufts University a novel method, the continuous flow centrifugation or CFC, for the concentration of three protozoa (Cryptosporidium Guardial, Microsporidia) from large volumes (1,000L) of water.

The specific objectives of this project are to: (1) optimize the parameters (flow rate, centrifugal force, various bowls, elusion buffers, etc.) in order to recover representative pathogens which include E. coli for bacteria, Microcystis aeruginosa for algae and MS2 bacteriophages for viruses. In this Objective, we will systematically optimize the concentration methodology for each representative pathogen with a view to generate reproducibly robust data; (2) integrate the concentration of protozoa (validated in the pervious award), bacteria, algae and viruses from water into a single concentration procedure. The PCFC will then be fine tuned for its ability to simultaneously concentrate representative pathogens from each group of the CCL list and; (3) focus on the detection and quantitative identification of CLL pathogens in water, using multiplex miniaturized fiber optic bead microarrays coupled with a compact confocal-type imaging system and comparing it with EPA approved methods.

Progress Summary:

Summary of progress: The automated system which simultaneously concentrates and recovers multiple types of microorganisms from large water volumes has been further validated under lab conditions including for protozoa, bacteria, and viruses present in volumes > 50L. In collaboration with Haemonetics, we have produced a third generation multipurpose device which is automated, with a disposable bowl modified to include a positive charged material. The overall automated process which was developed during year one, was validated using much the same procedures as described before but tested many times over to generate more robust information based on the following steps: a) Concentration by a peristaltic pump delivers a water sample through the bowls inlet port, the large particles such as protozoa, bacterial spores/cells and other suspended material is being sedimented due to the strong centrifugal forces and the much cleaner water sample is forced to flow through the designated passages in the bottom of the inner core where as the negatively charged viruses are captured by the strong electrostatic attraction, eliminating the clogging issues which are common in standard filtration components. b) Elution: protozoa/bacteria are dislodged by the addition of detergent and agitation, and the viruses which are trapped in the virus matrix are released using a beef extract/glycine/T80 buffer which inverts the negative charge. Both, virus and protozoa/bacteria concentrates are delivered to a sterile bag by a push of a button by reverse peristalsis. The bags could be processed in the field by using portable detection kits or delivered to the laboratory. Cryptosporidium oocysts, Bacillus anthracis spores and MS2 bacteriophages were simultaneously spiked in 10 -50L tap and turbid surface water samples, resulting in mean recoveries of 40%, 35%, and 50% respectively. To date, we have focused on the detection of DNA isolated from E. coli cells as a model system for detection through application of nucleic acid sensor technology developed by and is performed at Dr. David Walt’s lab. We have demonstrated the detection of PCR amplicons from three virulence genes using multiplexed bead-based microarrays. Current research is focused on expanding our protocol and microarray to include all bacteria and viruses listed as CCL3 candidates.

In the present study we describe the several stages that led to a successful development, application and validation of the following: a) a modified disposable bowl that simultaneously concentrate multiple pathogens from the same sample, b) an new phase fully automated, that can be operated from remote sites for continuous monitoring of presence of all classes of pathogens in the water supplies, using the portable continuous flow centrifuge that accommodates the modified bowl and a tubing harness and allows for sample concentration and elution from large water volumes. The new system was validated using the 10 - 50 liters volumes we had started with during year one of tap and source water samples spiked with low numbers of C. parvum oocysts, B. anthracis spores and MS2 bacteriophages and the recovered concentrate were assayed using standard methods.

Haemonetics Co., MA, constructed the automated machine, CFC 100A, modified from standard components of a blood separation system, as in previous prototypes. It houses a centrifuge (Magstar, MN) capable of running from 1 to 12,000 rpm (max 4,000 RCF), a peristaltic pump that runs from 0 to 1,000 mL/min and various pneumatic valves all controlled by a PLC computer system. It measures 12”x13”x16” and weighs approximately 30 lbs. The power source is 240/110 VAC and there is an optional transport and storage cart, which contains a rechargeable battery pack to run the machine with DC power. This machine enables the simultaneous concentration of protozoa, bacteria and virus from 10 L and 50 L samples of various matrices. The process of sample concentration and elution is completely automated and the resulting eluates are small in volume. Added to this is the ability to operate the centrifuge by remote control from another site which allows for continuous monitoring capabilities.

Water matrices and sample preparation: Tap water samples were obtained from the faucet of the Division of Infectious Diseases, Cummings School of Veterinary Medicine. Tap water is supplied by the North Grafton Water Utility and is initially collected from a local ground water source. The turbidity was measured at 0.002 NTU by a DRT-15CE Tubidimeter (HF Scientific, Inc., Ft. Myers, Florida). Surface water samples were collected from a near by pond, Pratt’s pond, North Grafton, MA, on the day of analysis. Samples were autoclaved and the mean turbidity post sterilization was measured at 3 NTU. 10 L of each water matrix was transferred to sterile, disposable plastic cubitainers. The cubitainer was placed on a stir plate and stirred while being seeded with an aliquot of each representative organism and during concentration. For 50 L samples 10 L of water from each matrix was transferred to one cubitainer, seeded with representative organisms and refilled with 9 L of water once 1 L of the original sample remained to be concentrated. The cubitainer was continually refilled in this manner until 50 L had been concentrated.

Cryptosporidium parvum oocysts: we continued to work with this pathogen as the representative protozoa. Water samples were spiked with one of two commercially available products. EasySeed, produced by Biotechnology Frontiers, LTD, NSW, Australia, consists of 100 flow-sorted; gamma irradiated C. parvum oocysts and Giardia lamblia cysts with a cell count standard deviation of less than 2.5 for each microorganism. Samples were spiked with EasySeed by decanting the contents of each vial as prescribed by the manufacturer. The second spiking suspension was produced by the Wisconsin State Laboratory of Hygiene, Madison, WI, and consists of 100 C. parvum oocysts in 10 mL of reagent grade water and 0.01% Tween 20 with a cell count standard deviation of less than 2.5. Pouring the contents of one 10 mL tube into the water sample and rinsing the tube with 2 mL of 0.01% Tween 20 PBST followed by two 2 mL rinses of reagent grade water seeded the water samples. Each rinse was shaken vigorously for 30 seconds before being decanted into the water sample. Enumeration of the protozoa eluate from spiked water samples was performed as described by Method 1622 (USEPA, 2001). Dynabeads anti-Cryptosporidium Kit (Invitrogen Dynal AS, Oslo, Norway) was employed for selective separation of oocysts from water sample concentrates using immunomagnetic separation according to the manufacturers instructions. Crypt-A-Glo fluorescent monoclonal antibodies (Waterborne, Inc., New Orleans, LA) were used against C. parvum oocysts for detection and enumeration with a fluorescent microscope. The recovered numbers were multiplied by a factor of 2.

Bacillus anthracis Sterne: a kanamycin resistant strain, was selected as representative bacteria. This strain of B. anthracis was constructed by the replacement of particular RNA coding sequences with an omega element, W-kan, conferring kanamycin resistance. The stock suspension of B. anthracis was kept at -80^oC. Each week during testing a spiking suspension was produced by diluting 100 mL of the stock suspension into 50 mL of maximum recovery diluent (Oxoid Ltd., Basingstoke, Hampshire, England) to an estimated concentration of 30 colony-forming units (cfu) per mL. Water samples were spiked by pipetting 1 mL of the spike suspension to the sample while stirring. The spike dose was quantitated by pipetting 1 mL of the spiking suspension to 50 mL of reagent grade water, vortexing and vacuum filtrating through a .45 mm, 47 mm membrane (Millipore, Billerica, MA). Post filtration the membranes were placed on LB agar (Acros Organics, NJ) plates containing kanamycin (Fisher Scientific, Fair Lawn, NJ), inverted and incubated overnight at 37^oC. After incubation, colony-forming units with the typical appearance of B. anthracis, opaque grey and white formations, were counted and recorded. To quantitate the recovery of B. anthracis from spiked water samples the bacteria eluate was vacuum filtrated and incubated in the same manner. The recovered numbers were multiplied by a factor of 2.

Recovery of B. anthracis from 10 L tap water
samples using a standard HS Core Bowl and a
manually operated centrifuge.
Vol. analyzed	Spike dose	Percent Recovery
(L) (# replicates)	cfu	cfu
	(mean +/- SD)	(mean +/- SD)
10 (4)	98.5 +/- 0.7	30.9 +/- 8.9
10 (10)	58.4 +/- 9.6	30.9 +/- 10.8
10 (10)	10.3 +/- 0.4	44.8 +/- 11.9

Escherichia coli ATCC 15597 and 700891: E. coli (American Type Culture Collection, Manassas, VA) was used as host bacteria for MS2 bacteriophage. Freeze dried samples were purchased from ATCC, re-hydrated with .150 mL of sterile glycerol and .850 mL of LB broth (Acros Organics, NJ) for E. coli 15597 or Tryptic Soy broth (Becton, Dickinson and Company, MD) prepared with ampicillin (Roche, CH) and streptomycin sulfate (Acros Organics, NJ) according to EPA Method 1602 for E. coli 700891. The re-hydrated pellet was then aliquoted to a volume of 50 mL and stored at -80^oC. To propagate the bacteria one 50 mL aliquot was diluted in 10 mL of LB broth for E. coli 15597 or 10 mL of Tryptic Soy broth prepared with ampicillin and streptomycin sulfate for E. coli 700891 and incubated overnight at 37^oC and 200 rpm on a shaking incubator. After incubation 1.5 mL of sterile glycerol was added to the host bacteria broth, which was subsequently aliquoted to a volume of 200 mL and stored at -80^oC.

Two methods were used to quantitate the recovery of MS2 bacteriophage from spiked water samples and to enumerate the spike dose of phage. For high spike doses, approximately 10⁸plaque-forming units (pfu) per sample, LB agar plates were prepared with a field of actively growing E. coli 15597 to spot the phage on. Prior to preparation of the plates 800 mL of LB broth was added to one 200 mL aliquot of overnight grown E. coli and incubated for two hours at 37^oC. After incubation the host bacteria broth was spread evenly over an LB agar plate and the excess aspirated off. Serial dilutions of phage were than spotted onto the plate immediately after the E. coli had air-dried. For low spike doses, approximately 200 pfu per sample, Tryptic Soy agar (Becton, Dickinson and Company, MD) containing ampicillin and streptomycin was prepared with actively growing E. coli 700891 according to the Single Agar Layer (SAL) procedure outlined in EPA Method 1602.

MS2 Escherichia coli bacteriophage ATCC 15597-B1: MS2 bacteriophage (American Type Culture Collection, Manassas, VA) was purchased from ATCC, re-hydrated with 1 mL of either LB broth or Tryptic Soy broth depending on method of quantization, aliquoted to a volume of 50 mL with an estimated concentration of 10⁸ pfu per mL and stored at -80^oC. Water samples were either spiked with a low or high concentration of phage. Samples spiked with a high concentration of phage, approximately 10⁸ pfu, received one 50 mL aliquot of phage prepared in LB broth while stirring. The spike dose was quantitated by pipetting one 50 mL aliquot to 50 mL of reagent grade water, vortexing and vacuum filtrating through a ViroCap 1 mm, 47 mm membrane (Scientific Methods, Granger, IN) positively charged with aluminum hydroxide, Al(OH)₃. After filtration the membrane was soaked in 5 mL of virus elution buffer to reverse the positive charge and vacuum filtrated again to extract the phage as prescribed by the manufacturer. The resulting phage extract was serially diluted in reagent grade water. 5 mL of each dilution was spotted on the surface of an LB Agar plate prepared with an actively growing field of E. coli 15597, which was subsequently inverted and incubated overnight at 37^oC. The virus eluate from water samples spiked with a high concentration of phage was serially diluted and spotted on LB Agar plates in the same manner to quantitate the recovery of phage. Samples spiked with a low concentration of phage received a 200 mL aliquot of phage diluted to approximately 200 pfu in Tryptic Soy broth while stirring. The spike dose was quantitated by pipetting a comparable 200 mL aliquot to 20 mL of reagent grade water, which was subsequently added to 20 mL of 2X Tryptic Soy agar containing ampicillin and streptomycin sulfate with 2 mL of actively growing E. coli 700891 according to the SAL procedure outlined in EPA Method 1602. The virus eluate from water samples spike with a low concentration of phage was added to 20 mL of 2X Tryptic Soy agar containing ampicillin and streptomycin sulfate with 2 mL of actively growing E. coli 700891 according to the SAL procedure outlined in EPA Method 1602. After drying, the plates were inverted and incubated overnight at 37^oC. Post incubation, plaque-forming units, hazy clearings of the E. coli field with large halos, was counted and the concentration was calculated.

Recovery of C. parvum oocysts and B. anthracis from 10 L tap water samples using
a standard HS Core Bowl and an automated centrifuge.
Vol. analyzed	Spike dose	Percent Recovery	Spike dose	Percent recovery
(L) (# replicates)	oocysts	oocysts	cfu	cfu
	(mean +/- SD)	(mean +/- SD)	(mean +/- SD)	(mean +/- SD)
10 (5)	100 +/- 2.5	36.0 +/- 15.2	14.8 +/- 3.8	90.2 +/- 9.0

Detection methods using Nucleic acid sensor technology developed at David Walt’s lab: To date, we have concentrated on the detection of DNA isolated from E. coli cells as a model system. We have demonstrated the detection of PCR amplicons from three virulence genes using multiplexed bead-based microarrays. Current research is focused on expanding our protocol and microarray to include all bacteria and viruses listed as CCL3 candidates in the table below:

Microbial Contaminant Name	Information
Caliciviruses	Virus (includes Norovirus) causing mild self-limiting gastrointestinal illness
Campylobacter jejuni	Bacterium causing mild self-limiting gastrointestinal illness
Entamoeba histolytica	Protozoan parasite which can cause short as well as long-lasting gastrointestinal illness
Escherichia coli (0157)	Toxin-producing bacterium causing gastrointestinal illness and kidney failure
Helicobacter pylori	Bacterium sometimes found in the environment capable of colonizing human gut that can cause ulcers and cancer
Hepatitis A virus	Virus that causes a liver disease and jaundice
Legionella pneumophila	Bacterium found in the environment including hot water systems causing lung diseases when inhaled
Naegleria fowleri	Protozoan parasite found in shallow, warm surface and ground water causing primary amebic meningoencephalitis
Salmonella enterica	Bacterium causing mild self-limiting gastrointestinal illness
Shigella sonnei	Bacterium causing mild self-limiting gastrointestinal illness and bloody diarrhea
Vibrio cholerae	Bacterium found in the environment causing gastrointestinal illness

Pathogens highlighted in blue were included in the last iteration of the microarray. We have designed additional capture probes and PCR primers for the remaining bacteria and viruses and are scheduled to validate the performance of these sequences at the end of the year. Each microarray probe has been matched with multiple potential PCR primer pairs that will be tested with cultures prior to microarray hybridization experiments. The microarrays have been ordered from Illumina, Inc. and are expected to be delivered by 12/31/08.

We have designed and validated a detection protocol for the amplification of cells concentrated from spiked drinking water samples via continuous flow centrifugation. Cells are first lysed by heating the samples (~1 mL) in a water bath at 90^oC for 10 min. Solids are then removed from the lysate by filtering with Millex-HV (Millipore, Bedford, MA) cartridges and 10 mL syringes. Each 200 uL tube contains 25 uL PCR master mix (Promega, Inc., Madison, WI), 5 uL of each primer (1 uM final concentration), 5 uL of cell lysate, and 10 uL of DNAse-free water water. PCR amplification of DNA has been demonstrated using the following temperature cycling program:

T (C)	time
95	5 min
95	30 s
50	30 s	35 cycles
72	30 s
72	7 min

Reactions are analyzed by gel electrophoresis using E-gel cartridges with 4% agarose (Invitrogen, Inc., Carlsbad, CA) prior to array hybridization. Multiple reactions are pooled for each sample, with approximately 15 uL total volume for each microarray analysis. Reaction mixtures are heated to 90^oC for 5 min and immediately chilled on ice for 1 min. Hybridization at room temperature for 30 min has proved to be sufficient for the detection of ~6 CFU/mL in the case of E. coli.

Future Activities:

We have designed all PCR primers to have similar lengths and annealing temperatures and expect the amplification temperature cycling protocol to be used with only minor modifications. Assymetric PCR has shown initial promise to increase the hybridization signal intensities and will be studied further. Using a higher antisense primer concentration tends to bias the reaction towards the labeled strand, increasing the probability that it will hybridize to the capture probe instead of the complementary product strand in solution. Multiple primers have been designed and purchased (Integrated DNA Technologies, Coralville, IA) with the strategy that those sequences with highest specificity and sensitivity will be used. At least two capture probes per species of interest have been designed to insure redundancy in the event that any sequences exhibit cross reactivity. Before the microarrays arrive, we will focus on validating the primers for the other CCL3 species of interest. Microarray hybridization conditions and stringency washes will then be refined to maximize the specificity of the capture probes.

Supplemental Keywords:

RFA, Scientific Discipline, Water, Environmental Chemistry, Drinking Water, Environmental Engineering, Environmental Monitoring, E. Coli, contaminant candidate list, analytical methods, contaminant removal, pathogens, cyanobacteria, drinking water contaminants, drinking water treatment, drinking water monitoring, Giardia, continuous flow centrifugation, CCL, cryptosporidium

Progress and Final Reports:

Original Abstract

The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.