Grantee Research Project Results
Final Report: A Hybrid Pathogen Detection System
EPA Contract Number: EPD05044Title: A Hybrid Pathogen Detection System
Investigators: Aguilar, Zoraida P.
Small Business: Vegrandis Inc.
EPA Contact: Richards, April
Phase: I
Project Period: March 1, 2005 through August 31, 2005
Project Amount: $70,000
RFA: Small Business Innovation Research (SBIR) - Phase I (2005) RFA Text | Recipients Lists
Research Category: Small Business Innovation Research (SBIR) , Watersheds , SBIR - Water and Wastewater
Description:
The primary objectives of Phase I were met and exceeded. The Phase I research project was designed to demonstrate an electrochemical antibody capture coupled with an mRNA hybridization assay approach for the detection of Cryptosporidium parvum oocysts that will provide improvements over U.S. Environmental Protection Agency Methods 1623 and 1622 in several respects. It also served as a feasibility study for the development of a fully automated system for Phase II.
Summary/Accomplishments (Outputs/Outcomes):
Vegrandis, LLC, intended to demonstrate a detection time of less than 8 hours for 10 L samples of C. parvum oocyst concentrations down to 50 oocysts/L. C. parvum oocysts were detected down to 5 oocysts/10 L, an order of magnitude lower than the Phase I target. The assay time from capture to signal generation was shortened to approximately 60 minutes using a microcavity with self-contained electrochemistry. The microcavity had a diameter of 50 μm, a depth of 8 μm, and was equipped with embedded microelectrodes along its wall and bottom—the bottom recessed microdisk electrode (RMD) was at 2 x 10 -5 cm 2 and the wall tubular nanoband electrode (TNB) was at 8 x 10 -8 cm 2. The electrochemical immunoassay of 5-250 oocysts/10 L successfully was demonstrated in an array of microcavities with self-contained electrochemistry following the manual transfer of reagents onto the top of each microcavity.
Vegrandis determined the optimal conditions of the immunoassay detection coupled with a DNA hybridization assay for the viability of C. parvum oocysts spiked in real water samples. For the immunoassay detection, optimal signals were harnessed with 100 μg/mL of 1 0 Ab. However, to maintain lower assay cost, 50 μg/mL of 1 0 Ab was used because the signal at this concentration is only about 27 percent less than that when using twice the amount. Vegrandis’ studies indicate that the best 2 0 Ab-AP concentration is 20 μ g/mL when using 50 μg/mL of 1 0 Ab. These concentrations of the 1 0 Ab and 2 0 Ab-AP were used to detect purified oocysts in buffer at concentrations of 6-6,000 oocysts/mL. Therefore, in all immunoassay with self-contained microelectrochemical detection studies, these 1 0 Ab and 2 0 Ab-AP concentrations were used.
The optimal conditions of the detection of hsp-mRNA in a DNA hyrbidization assay were determined using the self-contained microelectrochemistry before combining the assays in a hybrid chip. Results of these studies indicated that 50 μ g/mL 1 0 DNA with 50 μ g/mL 2 0 DNA gave optimal signals for the assay. These were used to detect target DNA at concentrations from 100 ng/mL to 50 μ g/mL.
The extent of interference from natural components of drinking water (e.g., water hardness as calcium and magnesium carbonates, residual chlorine, residual monochloramine, pH, and dissolved iron) was tested. Results indicated minimal decrease in the signals in the presence of chlorine and chloramines, which are within acceptable errors in biological assays. Studies on the effect of natural components of drinking water (i.e., calcium, magnesium, iron, chlorine, chloramine, and pH) were performed by soaking the oocysts in solutions prepared by dissolving sources of these materials in deionized water. The results indicate significant changes in the signals in the presence of chlorine and chloramines and no effect in the presence of the other components of water. The effects of chlorine and chloramines have to be further studied to establish the exact concentration at which these components affect the assay. Whether the oocysts or the antibodies are affected by the chlorine and the chloramines also must be determined.
The biggest achievement in this project was the application of Vegrandis’ self-contained microelectrochemical assay platforms in an array format for the detection of live oocysts that were spiked in 1 mL of the final 25 mL pellet from the filtration of 10 L water samples from Beaver Lake (Springdale, Arizona), White River (Elkins, Arizona), Lake Fayetteville (Fayetteville, Arizona) and Lake Springfield (Springfield, Missouri). Vegrandis staff gathered the samples from these enumerated sources on August 22, 2005 (see Figure 1). The water was filtered with Filta Maxx from IDEXX. The collected samples and the used filters were kept in properly labeled containers that were stored in a cooler filled with ice during the
entire sampling day. The filters and collected samples were processed at the laboratory. The filters were washed using a manual wash station also from IDEXX. After manual wash of the filters following published procedures, the pellets were spiked with oocysts to make 5, 25, 50, and 250 oocysts/10 L of raw water. These samples were used for the capture of the oocysts on Vegrandis’ 3 cm x 2.5 cm chips containing the 8 x 3 array of 50 mm diameter cavities with self-contained microelectrodes.
Figure 1. Onsite filtration of water from various locations. A) Water was drawn from a fishing dock at Lake Fayetteville. B) Vegrandis was able to bring its sampling setup down to Beaver Lake with enough cable connecting the pump to the generator. C) At the White River, Vegrandis did not get close enough to the riverbank, so water had to be hauled from the river using a carboy. D) At Lake Springfield, it was about to rain when Vegrandis staff arrived, so it was decided to carry out the filtration inside a van.
Vegrandis was able to detect the oocysts in the murky, green-colored, and gritty water pellets using its technology with insignificant interference effects and insignificant background signals that were overlapping with the signals from the enzyme substrate that was not exposed to any of the capture chips (data not shown). The timed response for the immunoassay capture and detection of 25 oocysts/10 L water gave the first distinguishable signal in 30 seconds after addition of the enzyme substrate (see Figure 2). This is very significant in minimizing response time required during an outbreak so that action to abate damage can be taken immediately.
Figure 2. Timed response during the immunoassay detection of 25 oocysts/10 L in pelleted sample taken from the White River where the water was greenish and gritty. The response curves show that there is a small signal at 0 seconds, which is to be expected because of the debris found in the water sample.
After the immunoassay capture and detection of the oocysts, the chip was placed inside a humid Petri dish. The cavities were covered with 20 μ L of hybridization buffer and then heated in a water bath that was placed on top of a hot plate. The temperature was monitored and kept between 42-45 ºC for no less than 20 minutes and then heated to approximately 60ºC for 5 minutes. After heating, the chip was washed thoroughly and exposed to the secondary DNA probe to complete the DNA or mRNA hybridization assay. Results showed no significant difference between the DNA assay in buffer and the DNA assay after capture and heating rupture of the oocysts. This indicates that the assay can be used for the detection of viable oocysts in real water samples.
The tabulated results of the assays, shown in Table 1, indicate that Vegrandis was able to detect down to 5 oocysts/10 L of real water samples that were pelleted down to 25-50 mL. The results of the hybrid assay for the detection of live oocysts in real water samples, also tabulated in Table 1, show that the technology is suitable for the capture and quantification followed by the test for viability of C. parvum oocysts. Viability was confirmed by subjecting the oocysts to 20 minutes of heating between 42-45 ºC followed by subsequent heating to approximately 60ºC for 5 minutes.
Table 1. Results of the hybrid assay for live oocysts that were spiked in pelleted sample resulting from the on-site IDEXX filtration of 10L water samples. (BL = Beaver Lake; LF = Lake Fayetteville; MO = Springfield Lake in Missouri; WR = White River).
Signals from Immunoassay/nA |
||||
Oocyst/10L |
BL |
LF |
MO |
WR |
250 |
12.8 |
13.3 |
10.4 |
15 |
50 |
7.9 |
6.2 |
6.5 |
9.2 |
25 |
3.6 |
3.4 |
3.5 |
4.9 |
5 |
1.8 |
1.99 |
2.6 |
2.3 |
|
Signals from mRNA hybridization/nA |
|||
250 |
15 |
11 |
8.7 |
NA |
50 |
6.6 |
8.3 |
7.2 |
NA |
25 |
2.8 |
2.7 |
2.6 |
3.6 |
5 |
2 |
1.5 |
1.9 |
2.2 |
The immunoassay capture and detection followed by testing for viability of the oocysts in the real water samples were performed on a microarray chip containing an 8 x 3 array of 50 mm diameter cavities with built-in microelectrodes embedded along the walls. This design proved to be the most reproducible and most efficient for the hybrid assay. Therefore, this design will be optimized in Phase II to increase the efficiency in fabrication, which at present, is about 75 percent. It is hoped to approach greater than 90 percent efficiency and reproducibility of chip fabrication in the future.
Untreated water samples from various sources were collected and brought to the laboratory. After establishing the concentrations of the possible interferents in the real water samples, these were spiked with oocysts to make a 50:50 mixture of the 0.1 M phosphate buffer saline (PBS) and the real water samples. The oocysts were incubated in the 50:50 mixture of 0.1 M PBS and real water samples for 30 minutes before they were captured on the chips. The presence of the interferents that were detected using the Hach DR 850 still demonstrated the performance of the assay. The data for the different oocyst concentrations shown on Table 3 indicate that the assay works well with a slight decrease in the signal that also is exhibited by a decrease in the slope of the concentration versus the signal curve when the assay was performed on oocysts that were incubated in the real untreated water samples. This is very promising for the technology, which Vegrandis plans to use in the detection of Cryptosporidium oocysts in real water samples.
Table 3 . Studies on the effect of possible interferents that are commonly found in drinking water.
Water Samples |
Location |
Current (nA):Oocyst/mL |
|||
5 |
25 |
50 |
250 |
||
Lake Fayetteville |
Veterans Park, Fayetteville, AR |
0.25 |
0.61 |
0.92 |
1.61 |
Beaver Lake |
Highway 412, Springdale, AR |
0.24 |
0.56 |
0.82 |
1.24 |
White River |
Highway 72, Elkins, AR |
0.12 |
0.48 |
1.02 |
1.36 |
Buffer (0.1 M PBS) |
Laboratory |
0.33 |
0.69 |
0.97 |
1.91 |
There was a decrease in the slope of the signal when the oocysts were incubated for 30 minutes in water from Beaver Lake, Lake Fayetteville, and the White River as shown in Figure 3. The decrease in the slope of the signal is significantly higher in the studies conducted on Beaver Lake water that contained 1.97 mg/mL calcium reported as CaCO 3. Additional studies will be performed using these real water samples to establish how signals at even lower oocyst concentrations are affected.
Figure 3. Plot of signal-versus-oocyst concentration spiked in real water samples. There was a slight decrease in the slope of the signal when the oocysts were incubated for 30 minutes in water from Beaver Lake and Lake Fayetteville. Diamonds = Buffer; Circles = Lake Fayetteville; Triangles = White River; Squares = Beaver Lake.
�It also was Vegrandis’ goal to show capture of the oocysts on the chip using scanning electron microscopy (SEM). The results indicate that the batch of the live oocysts purchased from Waterborne had smaller oocysts than average. Most of the oocysts on the capture surface were less than 5 μm in diameter. If possible, Vegrandis will find another source of oocysts to compare the sizes with those received from its present supplier.
The self-contained microelectrochemical detection of C. parvum oocysts also was successfully partially automated using a digital syringe pump for the delivery of reagents onto the chip through a PEEK capillary tubing. The signals generated using the TNB electrode of a 50 μm diameter cavity in an array of 1 mm distance between cavities were used in the detection of 5 oocysts/10 L to 250 oocysts/10 L. Detection of 250 oocysts/L was performed in a disposable cartridge consisting of a chip with an array of 50 μm diameter microcavities with a capture surface equipped with a 1 mm Kapton channel and a glass slide cover (see Figure 4).
Figure 4. a) Preliminary design of the hybrid chip. The oval cavities for the mRNA assay are 150 μm x 200 μm while the dimensions of the gold strip for the oocyte capture are 1.4 mm long and 100 μm wide. b) The captured oocysts (Waterborne, LA) were small as seen from the SEM after oocysts were captured on the hybrid chip.
In preparation for the full automation for the hybrid assay, the extent of nonspecific adsorption (NSA) on the components of the cartridge was studied. This involved examining NSA on different capillary tubing materials and channel materials. Studies on the automation of the electrochemical immunoassay involved testing the materials that included three types of capillary tubings (tygon, PEEK, and silicone), three types of Kapton, and polydimethylsiloxane (PDMS) that were used as the channel material (see Figure 5). These materials were tested to establish compatibility with the proteins in the immunoassay. The results of these tests indicate that NSA on these materials could be eliminated by incubating the blocking buffer in the capillary tubing or the channel material for at least 5 minutes before use. The NSA on PEEK tubing and either Kapton 1 or PDMS was easily eliminated by washing and blocking buffers making these useful in the development of the automated delivery and detection of live C. parvum oocysts using Vegrandis’ approach.
Figure 5. Flow cell setup for the semi-automated delivery of reagents and samples in the immunoassay detection of captured oocysts. A 20 μL solution of 20Ab-AP in 0.1 M PBS was paced on: A - Kapton 500 FN 131, B - Kapton 500 HN, C - Kapton 500 HPP-St, and D - polydimethylsiloxane. E - The semi-automated setup using the flow cell.
Before the flow cell was assembled, 20 μ L of the alkaline phosphatase labeled anti-C. parvum IgG (2 0Ab-AP)was incubated on different possible channel materials for the flow cell for 5 minutes. The materials tested were Kapton 500 FN131, Kapton 500 HN, Kapton 500 HPP-St, and PDMS. As shown in Figure 5, the results indicate that Kapton 500 FN and PDMS are the most hydrophobic, with a contact angle at approximately 40 º . The contact angle was measured by drawing a line tangent to the bottom of the droplet, drawing a line from the tangential point that touches the side of the droplet and then measuring the angle with a protractor. Kapton 500 HN and Kapton 500 HPP-St have a contact angle of approximately 12 º, indicating that the materials are very hydrophilic. A 1 mm channel was created on Kapton 500 FN that was slotted on top of the hybrid chip. The hybrid chip was sandwiched between two pieces of glass slide and screwed tight with a plastic clamp as shown in Figure 5. The inlet and outlet reservoirs were plugged with pipette tip (0.1-1 μ L capacity) cut at the thin end to fit the 1 mm opening to prevent leakage. A PEEK tubing inserted into a luer (Upchurch Scientific) that was slotted snuggly unto the pipette tip was used for the sample/reagent delivery from the syringe pump. A second PEEK tubing inserted into a luer was inserted to the other pipette tip for the waste disposal.
Conclusions:
Overall, the goals for Phase I were achieved and mostly exceeded. Vegrandis was able to successfully show the immunoassay and DNA hybridization assay detection of C. parvum oocysts in an array format that is very significant in developing the system into an automated array assay. An array of microcavities can be configured for the simultaneous assay of a single water sample for the detection of pathogens of high risk to humans.
Supplemental Keywords:
pathogen detection system, hybridization, Cryptosporidium parvum , C. parvum , oocysts, microcavity, electrochemistry, microelectrochemistry, recessed microdisk electrode, RMD, tubular nanoband electrode, TNB, immunoassay, assay, DNA, mRNA, drinking water, detection, scanning electron microscopy, SEM, phosphate buffer saline, PBS, polydimethylsiloxane, PDMS, contact angle, EPA, small business, SBIR,, RFA, Scientific Discipline, PHYSICAL ASPECTS, Water, Ecosystem Protection/Environmental Exposure & Risk, Environmental Chemistry, Health Risk Assessment, Monitoring/Modeling, Biochemistry, Physical Processes, Drinking Water, cryptosporidium parvum oocysts, microbial contamination, Safe Drinking Water, pathogens, monitoring, microbial risk assessment, assays, water quality parameters, waterborne disease, exposure and effects, fecal contamination, mRNA hybridization assay, exposure, other - risk assessment, drinking water distribution system, cryptosporidium , public health, water quality, drinking water contaminants, immunofluorescent assay, drinking water systemSBIR Phase II:
A Hybrid Pathogen Detection System | Final ReportThe 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.