December 2004
Environmental Technology
Verification Report
INVITROGEN CORPORATION
PATHALERT™ DETECTION KITS
FOR THE DETECTION OF
FRANCISELLA TULARENSIS,
YERSINIA PESTIS, AND
BACILLUS ANTHRACIS
Prepared by
Battelle
Baiteiie
The DuSiiieSs of 111 nnv.il ion
Under a cooperative agreement with
V>E!PA U.S. Environmental Protection Agency
-------
December 2004
Environmental Technology Verification
Report
ETV Advanced Monitoring Systems Center
Invitrogen Corporation
PathAlert™ Detection Kits
for the detection of
Francisella tularensis,
Yersinia pestis, and Bacillus anthracis
by
Stephanie Buehler
Amy Dindal
Zachary Willenberg
Karen Riggs
Battelle
Columbus, Ohio 43201
-------
Notice
The U.S. Environmental Protection Agency (EPA), through its Office of Research and
Development, has financially supported and collaborated in the extramural program described
here. This document has been peer reviewed by the Agency. Mention of trade names or
commercial products does not constitute endorsement or recommendation by the EPA for use.
11
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the
nation's air, water, and land resources. Under a mandate of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance between
human activities and the ability of natural systems to support and nurture life. To meet this
mandate, the EPA's Office of Research and Development provides data and science support that
can be used to solve environmental problems and to build the scientific knowledge base needed
to manage our ecological resources wisely, to understand how pollutants affect our health, and to
prevent or reduce environmental risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA to
verify the performance characteristics of innovative environmental technology across all media
and to report this objective information to permitters, buyers, and users of the technology, thus
substantially accelerating the entrance of new environmental technologies into the marketplace.
Verification organizations oversee and report verification activities based on testing and quality
assurance protocols developed with input from major stakeholders and customer groups
associated with the technology area. ETV consists of six verification centers. Information about
each of these centers can be found on the Internet at http://www.epa.gov/etv/.
Effective verifications of monitoring technologies are needed to assess environmental quality
and to supply cost and performance data to select the most appropriate technology for that
assessment. Under a cooperative agreement, Battelle has received EPA funding to plan,
coordinate, and conduct such verification tests for "Advanced Monitoring Systems for Air,
Water, and Soil" and report the results to the community at large. Information concerning this
specific environmental technology area can be found on the Internet at
http://www.epa.gov/etv/centers/centerl.html.
111
-------
Acknowledgments
The authors wish to acknowledge the support of all those who helped plan and conduct the
verification test, analyze the data, and prepare this report. We sincerely appreciate the
contribution of drinking water samples from the New York City Department of Environmental
Protection (Paul Bennett), the City of Orlando (Terri Slifko), and the Metropolitan Water
District of Southern California (Paul Rochelle). Also, thanks go to the Metropolitan Water
District of Southern California for concentrating each drinking water sample. We would also
like to thank Myriam Medina-Vera, U.S. Environmental Protection Agency National Exposure
Research Laboratory; Jorge Santo Domingo, National Risk Management Research Laboratory;
Kerri Alderisio, New York City Department of Environmental Protection; Ricardo DeLeon,
Metropolitan Water District of Southern California; and Stanley States, Pittsburgh Water and
Sewer Authority, for their careful review of the test/quality assurance plan and this verification
report.
IV
-------
Contents
Page
Notice ii
Foreword iii
Acknowledgments iv
List of Abbreviations viii
1 Background 1
2 Technology Description 2
3 Test Design and Procedures 3
3.1 Introduction 3
3.2 Test Samples 5
3.2.1 Performance Test Samples 5
3.2.2 Drinking Water Samples 6
3.2.3 Quality Control Samples 7
3.3 Reference Methods 8
3.3.1 Plate Enumeration 8
3.3.2 Drinking Water Analysis 8
3.4 Test Procedure 9
3.4.1 Sample Handling 9
3.4.2 Sample Preparation and Analysis 10
3.4.3 Drinking Water Characterization 11
4 Quality Assurance/Quality Control 13
4.1 Sample Chain-of-Custody Procedures 13
4.2 Equipment Calibration 13
4.3 Characterization of Contaminant Stock Solutions 13
4.4 Quality Control Samples 15
4.5 Audits 15
4.5.1 Technical Systems Audit 15
4.5.2 Audit of Data Quality 15
4.6 QA/QC Reporting 16
4.7 Data Review 16
5 Data Analysis 17
5.1 Accuracy 17
5.2 Specificity 17
5.3 False Positive/Negative Responses 17
5.4 Precision 18
-------
5.5 Interferences 18
5.6 Other Performance Factors 18
6 Test Results 19
6.1 Accuracy 21
6.1.1 F. tularensis 21
6.1.2 Y.pestis 21
6.1.3 B. anthracis 22
6.2 Specificity 23
6.2.1 F. tularensis 23
6.2.2 Y. pestis 24
6.2.3 B. anthracis 25
6.3 False Positive/Negative Responses 26
6.3.1 F. tularensis 26
6.3.2 Y. pestis 26
6.3.3 B. anthracis 26
6.4 Precision 30
6.5 Interferences 30
6.5.1 Interferent PT Samples 30
6.5.2 Drinking Water Samples 31
6.6 Other Performance Factors 32
7 Performance Summary 34
8 References 38
Figures
Figure 2-1. Invitrogen Corporation's PathAlert™ Detection Kit 2
Figure 6-1. Positive Electropherogram for B. anthracis 20
Tables
Table 3-1. Infective/Lethal Dose of Target Contaminants 5
Table 3-2. Performance Test Samples 5
Table 3-3. Drinking Water Samples 7
Table 3-4. ATEL Water Quality Characterization of Drinking Water Samples 12
Table 4-1. F. tularensis, Y. pestis, and B. anthracis Triplicate Plate Enumeration Data ... 14
Table 4-2. Data Recording Process 16
Table 6-la. F. tularensis Contaminant-Only PT Sample Results 21
Table 6-lb. Y. pestis Contaminant-Only PT Sample Results 22
Table 6-lc. B. anthracis Contaminant-Only PT Sample Results 23
Table 6-2a. F. tularensis Specificity Results 24
VI
-------
Table 6-2b. Y. pestis Specificity Results 25
Table 6-2c. B. anthracis Specificity Results 25
Table 6-3a. F. tularensis False Positive/Negative Results 27
Table 6-3b. Y. pestis False Positive/Negative Results 28
Table 6-3c. B. anthracis False Positive/Negative Results 29
Table 7-1. F. tularensis Summary Table 35
Table 7-2. Y. pestis Summary Table 36
Table 7-3. B. anthracis Summary Table 37
vn
-------
List of Abbreviations
AMS Advanced Monitoring Systems
ASTM American Society of Testing and Materials
ATEL AquaTech Environmental Laboratories, Inc.
ATCC American Type Culture Collection
BSL Biosafety Level
Ca calcium
cfu colony forming unit
cm centimeter
DI deionized water
DNA deoxyribonucleic acid
DW drinking water
EPA U.S. Environmental Protection Agency
EPC external positive control
ETV Environmental Technology Verification
ID identification
IPC internal positive control
L liter
LOD limit of detection
MB method blank
Mg magnesium
mg milligram
mL milliliter
MWD Metropolitan Water District
PBS phosphate buffered saline
PCR polymerase chain reaction
PT performance test
QA quality assurance
QC quality control
QMP Quality Management Plan
SOP standard operating procedure
TSA technical systems audit
Vlll
-------
Chapter 1
Background
The U.S. Environmental Protection Agency (EPA) supports the Environmental Technology
Verification (ETV) Program to facilitate the deployment of innovative environmental tech-
nologies through performance verification and dissemination of information. The goal of the
ETV Program is to further environmental protection by accelerating the acceptance and use of
improved and cost-effective technologies. ETV seeks to achieve this goal by providing high-
quality, peer-reviewed data on technology performance to those involved in the design,
distribution, financing, permitting, purchase, and use of environmental technologies.
ETV works in partnership with recognized testing organizations; with stakeholder groups
consisting of buyers, vendor organizations, and permitters; and with the full participation of
individual technology developers. The program evaluates the performance of innovative tech-
nologies by developing test plans that are responsive to the needs of stakeholders, conducting
field or laboratory tests (as appropriate), collecting and analyzing data, and preparing peer-
reviewed reports. All evaluations are conducted in accordance with rigorous quality assurance
(QA) protocols to ensure that data of known and adequate quality are generated and that the
results are defensible.
The EPA's National Exposure Research Laboratory and its verification organization partner,
Battelle, operate the Advanced Monitoring Systems (AMS) Center under ETV. The AMS Center
recently evaluated the performance of Invitrogen Corporation's Path Alert™ Detection Kits for
the detection of Francisella tularensis (F. tularensis), Yersinia pestis (Y. pestis), and Bacillus
anthracis (B. anthracis}.
-------
Chapter 2
Technology Description
The objective of the ETV AMS Center is to verify the performance characteristics of environ-
mental monitoring technologies for air, water, and soil. This verification report provides results
for the verification testing of the PathAlert™ Detection Kit. The following is a description of the
PathAlert™ Detection Kit based on information provided by the vendor. The information
provided below was not subjected to verification in this test.
The PathAlert™ Detection Kit is a multiplex polymerase chain reaction (PCR) reagent system
capable of detecting F. tularensis, Y. pestis, B. anthracis, or smallpox in individual assays. The
PathAlert™ Detection Kit comprises an optimized PCR SuperMix specific to the pathogen of
interest, as well as an external positive control (EPC) template for system validation. The kit
includes Taq polymerase, pre-complexed with antibodies to maintain "hot start" PCR (for
specificity and sensitivity); uracil deoxyribonucleic acid (DNA) glycosylase and deoxyuridane
triphosphate to eliminate post-PCR cross-contamination; and an internal positive control (IPC)
to identify potential PCR inhibition from sample contaminants or environmental samples.
Included in the kit is an EPC that has been engineered to produce different amplicon sizes than
either the IPC or the pathogen-specific loci. As a result, pathogen-specific results can be read
with minimal interference if contamination by the external control should occur.
PathAlert™ Detection Kit is an endpoint assay; post-
amplification products may be analyzed using any
platform capable of distinguishing amplicon size, such
as the Agilent Bioanalyzer 2100, Agilent ALP (high
throughput), transgenomic WAVE high-performance
liquid chromatography, gel electrophoresis, and Caliper
AMS 90. The Agilent Bioanalyzer 2100 and the Agilent
ALP are recommended by Invitrogen Corporation for
use with the PathAlert™ Detection Kit because all field
testing to date has been performed with these systems,
and Agilent and Invitrogen Corporation have established
a co-marketing relationship for the complete system.
The cost of each PathAlert™ Detection Kit ranges from
Figure 2-1. Invitrogen Corporation's $ 12 to $ 16 per assay.
PathAlert™ Detection Kit
-------
Chapter 3
Test Design and Procedures
3.1 Introduction
The purpose of this verification test of rapid PCR technologies was to evaluate the ability of
these technologies to detect the presence of specific bacteria in water and to determine the
technologies' performance when specific interferents were added to pure water and when
interferents were inherently present in several drinking water matrices. The technologies for this
verification test operate based on the PCR process, which involves enzyme-mediated reactions
that allow for target DNA (from the bacteria of interest) replication and amplification through a
series of temperature cycles. Before the target DNA can be amplified, however, it must first be
extracted from the bacteria and then purified.
Because rapid PCR technologies are anticipated to serve mostly as screening tools in water
monitoring scenarios, providing rapid results as to whether or not a pathogen or biological agent
is present in the water, this verification test involved only qualitative results. This verification
test of the PathAlert™ Detection Kit was conducted according to procedures specified in the
Test/QA Plan for Verification of Rapid PCR Technologies^ The performance of the
PathAlert™ Detection Kit was verified in terms of the following parameters:
Accuracy
Specificity
False positive/negative responses
Precision
Interferences
Other performance factors.
The performance of the PathAlert™ Detection Kit was verified by challenging it with various
concentration levels of F. tularensis LVS [American Type Culture Collection (ATCC) #29684],
Y. pestis CO92, and B. anthracis Ames strain in American Society of Testing and Materials
(ASTM) Type n deionized (DI) water; ASTM Type n DI water spiked with various interferents;
and concentrated drinking water (DW) samples obtained from four water utilities from different
geographical locations in the United States. Each source of DW represented a unique water
treatment process. In addition, the interferent and DW samples were analyzed without adding
any contaminant to evaluate the potential for false positive results. The kit was only tested for
one bacteria at a time.
-------
Contaminant concentrations included the infective/lethal dose concentrations given in Table 3-1
for each contaminant and approximately 2, 5, 10, and 50 times the vendor-reported method limit
of detection (LOD) for each technology. The infective/lethal dose of each contaminant was
determined by calculating the concentration at which ingestion of 250 milliliters (mL) of water
is likely to cause the death of a 70-kilogram (approximately 154 pounds) person based on
human LD50 or JD50 data.(2) The results from quadruplicate analysis of the contaminant
performance test (PT) samples and comparison with the known sample compositions provided
information on the accuracy and precision of the PathAlert™ Detection Kit. The interferent PT
samples contained humic and fulvic acids at two concentrations, both spiked and unspiked with
contaminants. Each was analyzed in quadruplicate and provided information on potential matrix
interferences.
For the purposes of this test, IxlO4 colony forming units (cfu)/mL were used to calculate the
concentration levels of F. tularensis and B. anthracis spiked into the PT and DW samples;
100 cfu/mL were used to calculate levels of Y. pestis spiked in the PT and DW samples. These
vendor-provided concentration levels were anticipated to be the levels for the entire
experimental process at which quantifiably reproducible positive results could be obtained from
a raw water sample. These concentration levels are referred to as the "method LOD" for a
particular assay. The method LOD incorporates the sensitivities and uncertainties of not only the
PathAlert™ Detection Kit, but also the DNA purification step; and, as such, it is an
experimental detection limit rather than an instrument or reagent-specific detection limit. As
mentioned previously, the method LOD provided by the vendor was used specifically as a
guideline in calculating sample concentration ranges for use with the PathAlert™ Detection Kit
and all other components used in this verification test to analyze a sample, and it should be
noted that Invitrogen Corporation does not claim this to be the true LOD of the PathAlert™
Detection Kit alone. The vendor claims the absolute LOD (the least amount of target DNA that
would generate a positive result) for the PathAlert™ Detection Kit alone is as low as 1 to 10
copies of DNA, depending on the assay. This information was not verified in this test.
The verification test was conducted at Battelle's Medical Research and Evaluation Facility in
West Jefferson, Ohio, from June 9, 2004, through June 30, 2004. Aqua Tech Environmental
Laboratories, Inc. (ATEL) of Marion, Ohio, performed physicochemical characterization for
each DW sample, including turbidity, dissolved and total organic carbon, specific conductivity,
alkalinity, pH, magnesium (Mg), calcium (Ca), hardness, total organic halides, trihalomethanes,
and haloacetic acids. Battelle cultured the bacteria, provided the stock solutions of each bacteria
used in this test, and then confirmed the presence and quantity of F. tularensis, Y. pestis, and
B. anthracis bacteria in the stock solutions using plate enumeration. The stock solutions of
F. tularensis, and Y. pestis were stored frozen as 1 mL aliquots. The B. anthracis stock solutions
were refrigerated as 1 mL aliquots. A new 1 mL vial of stock solution was thawed and used for
each day of testing. All test samples were prepared from the stock solutions on the day of
analysis. All purified DNA was used the same day it was extracted and purified. Each set of
replicates for a sample came from the same batch of purified DNA.
-------
Table 3-1. Infective/Lethal Dose of Target Contaminants
Contaminant
F. tularensis
Y. pestis
B. anthracis
Disease
Caused by
Contaminant
Tularemia
Plague
Anthrax
Infective/Lethal Dose Concentration
(cfu/mL)
4xl05
0.28
200
3.2 Test Samples
Test samples used in this verification test included PT samples, DW samples, and quality control
(QC) samples. Each type of test sample, including QC samples, is described further below.
3.2.1 Performance Test Samples
Table 3-2 lists the PT samples analyzed in this verification test for each bacteria. The bacteria
were added individually to each spiked sample. PT samples were prepared in ASTM Type n DI
water. The first type of PT sample consisted of ASTM Type n DI water spiked at five
concentration levels of each individual contaminant. The contaminant PT sample concentrations
ranged from the infective/lethal dose concentration to 50 times the vendor-stated method LOD.
The infective/lethal dose concentration was analyzed to document the response of the
PathAlert™ Detection Kit at that important concentration level. Four concentration levels at 2,
5, 10, and 50 times the vendor-reported method LOD, in addition to the infective/lethal dose
concentration, were analyzed. Each concentration level for the PT samples was analyzed in
quadruplicate.
Table 3-2. Performance Test Samples
Type of PT
Sample
Contaminant-
only
Interferent
Sample Characteristics
F. tularensis
Y. pestis
B. anthracis
Contaminants in 0.5 milligram per
liter (mg/L) humic acid and
0.5 mg/L fulvic acid
Contaminants in 2.5 mg/L humic
acid and 2.5 mg/L fulvic acid
Approximate Concentrations
(cfu/mL)
2xl04to5xl05
0.28to5xl03
200to5xl05
F. tularensis — IxlO5
Y. pestis— IxlO3
B. anthracis — IxlO5
F. tularensis — IxlO5
Y. pestis— IxlO3
B. anthracis — IxlO5
-------
The second type of PT sample was potential interferent samples. Four replicates of each
interferent PT sample were analyzed to determine the performance of the PathAlert™ Detection
Kit in the presence of humic and fulvic acids. The interferent PT samples contained humic and
fulvic acids isolated from Elliot Soil near Joliett, IL, (obtained from the International Humic
Substances Society) spiked into ASTM Type n DI water. Each of these interferent mixtures was
prepared at two concentration levels. One concentration was near the upper limit of what would
be expected in DW (5 mg/L) and one was at a mid-low range of what would be expected (1
mg/L). The 1 mg/L interferent mixture was prepared as 0.5 mg/L humic acid and 0.5 mg/L
fulvic acid. Similarly, the 5 mg/L interferent solution was prepared as 2.5 mg/L humic acid and
2.5 mg/L fulvic acid. These interferent levels were confirmed through analysis of aliquots by
ATEL. Also, each bacteria was added separately to these samples, along with the potential
interferent, at a concentration of 10 times the method LOD and analyzed in quadruplicate.
In all cases, four replicates for each PT sample, DW sample, and QC sample were taken from the
extracted and purified product (unspiked) or DNA (spiked) of one sample solution. That is, only
one spiked or unspiked sample solution was prepared for each set of replicates and taken
through the DNA extraction and purification procedure. Four replicates were then taken from
the same purified product or DNA. In an effort to characterize the efficacy of the extraction and
purification procedure in the presence of inhibitory substances (humic and fulvic acids), four
solutions of humic and fulvic acids at 0.5 mg/L spiked with each contaminant at 10 times the
method LOD, were prepared in addition to the samples listed in Table 3-2. Each solution was
put through the DNA extraction and purification procedure, and then four replicates from each
of the four purified DNA solutions were analyzed using the PathAlert™ Detection Kit.
3.2.2 Drinking Water Samples
Table 3-3 lists the DW samples analyzed for each bacteria in this test. DW samples were
collected from four geographically distributed municipal sources (Ohio, California, Florida, and
New York) to evaluate the performance of the PathAlert™ Detection Kit with various sample
matrices. These samples varied in their source and treatment and disinfection process. All
samples had undergone either chlorination or chloramination prior to receipt. Samples were
collected from utility systems with the following treatment and source characteristics:
• Chlorinated filtered surface water
• Chloraminated filtered surface water
• Chlorinated filtered groundwater
• Chlorinated unfiltered surface water.
All samples were collected in pre-cleaned high density polyethylene containers. After sample
collection, to characterize the DW matrix, an aliquot of each DW sample was sent to ATEL to
determine the following water quality parameters: turbidity, organic carbon, conductivity,
alkalinity, pH, Ca, Mg, hardness, total organic halides, concentration of trihalomethanes, and
haloacetic acids. The DW samples were dechlorinated with sodium thiosulfate pentahydrate to
prevent the degradation of some of the contaminants by chlorine. Because real-world
applications of PCR technologies to screen water samples rely on pre-concentration of the water
-------
sample to be analyzed, approximately 100 L of each of the above sources of DW were
dechlorinated and then concentrated through ultrafiltration techniques to a final volume of
250 mL by the Metropolitan Water District (MWD) of Southern California. As shown in Table
3-3, each DW sample was analyzed without adding any contaminant (i.e., unspiked), as well as
after fortification with each individual contaminant at a single concentration level (10 times the
vendor-stated method LOD).
Table 3-3. Drinking Water Samples
Drinking Water Sample Description
Water Utility
Columbus, Ohio
(OH)
MWD of Southern
California (CA)
Orlando, Florida
(FL)
New York City,
New York (NY)
Water
Treatment
chlorinated
filtered
chloraminated
filtered
chlorinated
filtered
chlorinated
unfiltered
Source
Type
surface
surface
ground
surface
Approximate Contaminant Concentrations
(cfu/mL)
F. tularensis
unspiked and
IxlO5
unspiked and
IxlO5
unspiked and
IxlO5
unspiked and
IxlO5
Y. pestis
unspiked and
IxlO3
unspiked and
IxlO3
unspiked and
IxlO3
unspiked and
IxlO3
B. anthracis
unspiked and
IxlO5
unspiked and
IxlO5
unspiked and
IxlO5
unspiked and
IxlO5
3.2.3 Quality Control Samples
QC samples included method blank (MB) samples consisting of ASTM Type n DI water and
positive and negative controls, as provided by the vendor. All of the MB QC samples were
exposed to sample preparation and analysis procedures identical to the test samples. External
positive and negative controls were prepared and used according to the protocol provided by the
vendor. At least one EPC and negative (no-template) control was prepared with each batch of
samples placed on the thermal cycler. The MB samples were used to confirm negative responses
in the absence of any contaminant and to ensure that no sources of contamination were
introduced into handling and analysis procedures. At least 10% of the test samples (eight
replicates) for each bacteria were MB samples. The vendor-provided control samples indicated
to the technician whether the PathAlert™ Detection Kit was functioning properly. If the controls
failed for any reason, that batch of samples would be discarded and the extracts reanalyzed. To
the extent practicable, the test samples were analyzed blindly by having the technician label the
vials with only a sample number prior to the DNA purification step, so that the samples were
tracked through the purification, PCR, and detection steps by only a sample number. Due to
special facility use, the identity of the target bacteria was always known by the technician.
-------
3.3 Reference Methods
3.3.1 Plate Enumeration
For all contaminants, plate enumeration was used to quantify bacteria to confirm the
concentration of the stock solutions of these contaminants. The Battelle standard operating
procedure (SOP) followed was SOP No. MREF X-054, Standard Operating Procedure (SOP)
for the Enumeration ofBSL-2 and BSL-3 Bacteria Samples Via the Spread Plate Technique.
Prior to testing, the F. tularensis and Y. pestis were grown and then suspended in phosphate
buffered saline (PBS). Twenty-five or more individual 1 mL aliquots of stock solution were
prepared from each original PBS stock solution. Three 1 mL aliquots were randomly taken for
enumeration, while the others were frozen for later use in sample preparation. Each bacteria was
enumerated on each of the three selected 1 mL aliquots to confirm the determined concentration.
The B. anthracis came from a lot of spores prepared by Battelle and stored in a 1% stock
solution of phenol in water. Prior to testing, an aliquot of the B. anthracis solution described
above was centrifuged, the supernatant consisting of the phenol/water solution was decanted
from the spores, and the spores were reconstituted with DI water. This process was repeated two
times to ensure that the spores were suspended only in DI water. This DI water suspension of
spores was then aliquoted into 1-mL portions as with the F. tularensis and Y. pestis. Because of
the known stability of B. anthracis spores and based on general facility protocol, the aliquots
were refrigerated instead of frozen. An aliquot was enumerated in triplicate prior to testing to
confirm the concentration. Another aliquot was enumerated during the verification test to further
verify the concentration of B. anthracis in the stock solution vials.
3.3.2 Drinking Water Analysis
Because most of the contaminants tested can occur naturally in water, and because rapid PCR
technologies cannot distinguish between live and dead organisms, each unspiked concentrated
DW sample was plate enumerated to verify, to the extent practicable, the presence or absence of
the contaminant of interest. The samples were plated onto tryptic soy agar plates with 5% sheep
blood and incubated at 30 to 35°C. After 20 hours of incubation, the unspiked OH, CA, and NY
DW samples produced lawns of bacteria with a level of contamination estimated to be greater
than lx!03cfu/mL. The unspiked FL DW sample showed only 10 to 100 cfu/mL estimated
concentration levels after 20 hours. After further incubation, the FL DW sample produced
bacteria at a concentration estimated to be greater than lx!03cfu/mL. Each DW sample had at
least three distinct types of bacteria growing. Gram stains were performed on any distinct colony
types visible in each sample to gain further insight into the colony morphology. For OH and CA
DW, three Gram negative bacteria colonies were identified. For NY, four Gram negative
colonies were identified; and, for FL, both Gram negative and positive colonies were present.
The CA DW was further evaluated for the presence of F. tularensis based on the potential
positive results for unspiked CA DW samples during the verification test. An aliquot of the
water was plated onto cystine heart agar (F. tularensis selective media) and incubated at 30 to
8
-------
35°C. A single colony type (Gram negative rods) grew on the plates and was subjected to
biochemical tests (catalase, oxidase, (3-lactamase, and urease) for the presumptive identification
of F. tularensis. The biochemical test results came back oxidase positive, indicating that the
bacteria were not F. tularensis. Further identification tests were not conducted on other DW
samples because no confirmed positive responses were detected in the remaining unspiked DW
samples.
3.4 Test Procedure
3.4.1 Sample Handling
All testing for this verification test was conducted within Battelle laboratories staffed with
technicians trained to safely handle F. tularensis, Y. pestis, and B. anthracis bacteria. The
technician using the PathAlert™ Detection Kit had prior PCR experience. F. tularensis samples
were tested in a Biosafety Level 2 (BSL-2) laboratory, Y. pestis and B. anthracis samples were
tested in BSL-3 laboratories. Appropriate safety guidelines associated with each laboratory were
followed throughout the verification test. Each day, fresh samples were prepared from a thawed
vial of frozen or refrigerated stock solution in either DI water, an interferent matrix, or a DW
matrix. Concentration levels for spiked samples at various multiples of the method LOD for the
PathAlert™ Detection Kit and associated DNA purification (2, 5, 10, and 50 times the method
LOD for PT samples, and 10 times the method LOD for interferent and DW samples) were
calculated from the method LOD provided by Invitrogen Corporation. Sample solutions were
prepared to these concentrations based on the concentration of the bacteria stock solution, which
was determined through triplicate plate enumeration prior to testing. Each sample was prepared
in its own container and labeled only with a sample identification (ID) number that also was
recorded in a laboratory record book along with details of the sample preparation. Samples were
diluted to the appropriate concentration using volumetric pipettes and glassware. Each sample
was prepared in 1 mL quantities.
Despite rigorous sample preparation efforts, solutions consisting of low bacterial concentrations,
such as the Y. pestis infective/lethal dose, may have no DNA present in a given sample or
aliquot.(3>4) The rationale for this is based on the Poisson statistical distribution, where there is
some probability that a sample taken will contain no particles (i.e., bacteria or target DNA) and
thus yield a negative result.(3>4) As a practical example, assume that 1 mL contains exactly five
particles (i.e., bacteria or target DNA) of interest. If one takes ten 0.1 mL samples and analyzes
them, the maximum number of positives will be five out of the ten samples. From this it follows
that there will be at least five negatives. Random variation in the sampling will cause this ratio
to change. This verification test was not designed to differentiate between the stochastic nature
of the low concentration samples and the capabilities of the assays, but this phenomenon should
be noted.
-------
3.4.2 Sample Preparation and Analysis
For this verification test, the following components were used to analyze the samples: Roche
High Pure PCR Template Preparation Kit (for DNA extraction and purification); PathAlert™
Detection Kit for F. tularensis, Y. pestis, B. anthracis (each kit included bacteria-specific
PathAlert™ PCR SuperMix, PCR-grade water, and PathAlert™ EPC Plasmid); an MJ Research
DNA Engine® (PTC-200™) Peltier Thermal Cycler (for performing the PCR) with optical strip
tubes; and the Agilent 2100 Bioanalyzer (with a laptop computer, a miniature vortex, the 2100
Bioanalyzer, and a priming station) along with the 2100 Bioanalyzer DNA 500 chips and
reagent kit (for detecting the amplified PCR product). The 2100 Bioanalyzer and Bioanalyzer
DNA 500 chips are a microfluidics-based detection technology that provides amplicon sizing
information, as well as approximate product yields, using lab-on-a-chip technology to provide
rapid qualitative and quantitative information. The 2100 Bioanalyzer is approximately
6.5 inches x 16 inches x 11 inches.
Four steps were carried out to test a liquid sample for the presence of F. tularensis, Y. pestis, and
B. anthracis bacteria: (1) PCR SuperMix setup, (2) DNA purification, (3) PCR of the DNA, and
(4) 2100 Bioanalyzer loading and analysis. To perform these steps, the laboratory work area was
separated into three distinct areas: a "clean" area (DNA free), a "medium" area (a moderate
amount of DNA present), and a "dirty" area (high sample DNA concentrations). First, in the
"clean" area, the PCR SuperMix, part of the PathAlert™ Detection Kit, was added to the PCR
tubes. A volume of 12.5 microliters (|iL) was added for all controls, as well as the DI water PT
samples. For all DW and interferent samples (those samples with inhibitory substances present),
37.5 |iL SuperMix was added to the tubes. The negative controls were prepared at this time by
adding 12.5 |iL of sterile water to the appropriate PCR tubes.
Then, in the "dirty" area, the DNA was isolated and purified from the sample. The entire 1 mL
sample was taken through this isolation procedure. The vendor-provided instructions were
followed, which were the Roche High Pure Prep Kit instructions for the isolation of bacteria or
yeast with the following modifications to accommodate the 1 mL sample: 25 |iL of lysozyme
was used; the sample was split into two 500 |iL aliquots, with each aliquot going through the
second incubation steps; 500 |iL of Binding Buffer and 100 |iL of Protinease K were used prior
to the second incubation step; 250 |iL of isopropanol were used; and additional initial
centrifuging was necessary to pass both 500 |iL aliquots through the first filter tube. According
to the Roche High Pure Prep Kit instructions, 200 |iL of elution buffer were used in the final
step.
In the "medium" area, 12.5 |iL of purified DNA were added to the appropriate PCR tubes,
which already contained the SuperMix. Not all of the purified DNA obtained using the Roche
High Pure Prep Kit was used during the verification test. Any unused purified DNA was frozen
for possible later use. The EPCs were also prepared at this time. The capped PCR tubes were
then loaded onto the thermal cycler, which was pre-programmed by the vendor at the time of
training. After the thermal cycler had completed its PCR program run, the amplified product was
loaded onto the 2100 Bioanalyzer DNA 500 chip according to the directions provided with the
PathAlert™ Detection Kit. Briefly, the chip was inserted into a priming station, and gel-dye mix
10
-------
was added to the priming well and then pushed throughout the sample wells on the chip using
the priming station. Then, the appropriate reagents were added to the remaining wells, and 1 |iL
amplified DNA was added to each sample well. Each DNA 500 chip held up to 12 samples. In
general, at least two positive and two negative control samples were placed on the first
2100 Bioanalyzer chip to be run on a given day to verify the efficacy of the PCR process and the
reagents. Before each day's 2100 Bioanalyzer use, a cleaning chip was used. After the chip
containing the samples was loaded onto the 2100 Bioanalyzer, the DNA 500 assay was loaded in
the 2100 Bioanalyzer software, and the chip run was started. The resulting electropherograms for
each sample were analyzed to determine the results for each sample. The bacteria were
considered present in the sample if the 2100 Bioanalyzer's electropherogram for a given sample
showed three peaks of appropriate amplicon (base pair) size: two for the bacteria being
monitored and one for the IPC. The bacteria were considered not present if only the single IPC
peak was present. The negative control was considered successful if only the IPC peak was
present, and the EPC samples were considered successful if three peaks of the appropriate
amplicon size were present. The EPC peaks differ from the sample peaks to help distinguish
these samples. If the ladder for a given chip was unsuccessful, the chip was reloaded and run
again. The ladder is an external standard for the 2100 Bioanalyzer DNA 500 chip that ensures
the proper performance of the chip assay. The technician recorded the sample ID number on a
sample data sheet along with the qualitative results (positive or negative) for each sample.
3.4.3 Drinking Water Characterization
An aliquot of each DW sample, collected as described in Section 3.2.2, was sent to ATEL prior
to concentration to determine the following water quality parameters: turbidity; concentration of
dissolved and total organic carbon; conductivity; alkalinity; pH; concentration of Ca and Mg;
hardness; and concentration of total organic halides, trihalomethanes, and haloacetic acids.
Table 3-4 lists the methods used to characterize the DW samples, as well as the characterization
data from the four water samples used in this verification test. Water samples were collected and
water quality parameters were measured by ATEL in January 2004. Some of the water quality
parameters may have changed slightly prior to verification testing.
11
-------
Table 3-4. ATEL Water Quality Characterization of Drinking Water Samples
Sources of Drinking Water Samples
Parameter
Turbidity
Dissolved organic
carbon
Total organic carbon
Specific conductivity
Alkalinity
pH
Ca
Mg
Hardness
Total organic halides
Trihalomethanes
Haloacetic acids
Unit
NTU
mg/L
mg/L
microSiemens
mg/L
mg/L
mg/L
mg/L
Hg/L
|j,g/L/analyte
|j,g/L/analyte
Method
EPA 180.1(5)
SM5310(6)
SM5310(6)
SM2510(6)
SM 2320(6)
EPA 150.1(7)
EPA 200.8(8)
EPA 200.8(8)
EPA 130.2(7)
SM 5320(6)
EPA 524.2(9)
EPA 552.2(10)
Columbus,
Ohio
(OH DW)
0.2
1.9
1.6
357
55
7.33
42
5.9
125
360
26.9
23.2
MWD,
California
(CA DW)
0.1
2.3
2.1
740
90
7.91
35
1.5
161
370
79.7
17.6
Orlando,
Florida
(FL DW)
0.5
1.7
1.8
325
124
7.93
41
8.4
137
370
80.9
41.1
New York City,
New York
(NY DW)
1.3
1.5
2.1
85
4
6.80
5.7
19
28
310
38.4
40.3
NTU = nephelometric turbidity unit
|j,g = microgram
12
-------
Chapter 4
Quality Assurance/Quality Control
Quality assurance/quality control procedures were performed in accordance with the quality
management plan (QMP) for the AMS Center(11) and the test/QA plan for this verification test.(1)
4.1 Sample Chain-of Custody Procedures
Sample custody was documented throughout collection, shipping, and analysis of the samples.
Sample chain-of-custody procedures were generally those provided in the guidelines in
ASAT.n-007, Standard Operating Procedure for Chain of Custody for Dioxin/Furan Analysis.
The chain-of-custody forms summarized the samples collected and analyses requested and were
signed by the person relinquishing samples once that person had verified that the custody forms
were accurate. The original sample custody forms accompanied the samples; the shipper kept a
copy. Upon receipt at the sample destination, sample custody forms were signed by the person
receiving the samples once that person had verified that all samples identified on the custody
forms were present in the shipping container.
4.2 Equipment Calibration
The PathAlert™ Detection Kit, Agilent 2100 Bioanalyzer, and all associated reagents and
supplies specific for the detection of F. tularensis, Y. pestis, and B. anthracis were provided to
Battelle by the vendor. This system required no calibration. The performance of the system was
monitored through ladders, EPC, IPC, and negative controls. For DW characterization and
confirmation of the possible interferent, analytical equipment was calibrated by ATEL according
to the procedures specified in the appropriate standard methods. Pipettes used during the
verification test were calibrated according to Battelle SOP VI-025, Operation, Calibration, and
Maintaining Fixed and Adjustable Volume Pipettes.
4.3 Characterization of Contaminant Stock Solutions
F. tularensis, Y. pestis, and B. anthracis were grown and prepared by Battelle. All bacteria were
plate enumerated in triplicate for confirmation of the concentration of the 1 mL aliquot stock
solutions. Prior to enumeration, the B. anthracis, originally stored as a 1% stock solution of
13
-------
phenol in water, was aliquoted and washed twice with DI water and resuspended in only DI
water for analysis.
The lot of B. anthracis spores used for this verification test was previously characterized in
September 2003 by Battelle and the Centers for Disease Control and Prevention. This
characterization involved 11 criteria, including the percent of vegetative cells present, the viable
spore count, the guinea pig 10-day LD50, as well as DNA fingerprinting and gene sequencing.
This lot of spores met all 11 acceptance criteria, proving that they were viable and of the
specified strain (Ames). The vegetative cell count indicated that the stock solution of spores was
99.94% pure spores, with only 0.06% of the solution containing vegetative cells.
The Battelle SOP No. MREF X-054, Standard Operating Procedure (SOP) for the Enumeration
ofBSL-2 and BSL-3 Bacteria Samples Via the Spread Plate Technique, was followed for the
plate enumeration of F. tularensis, Y. pestis, and B. anthracis. The results of the plate enumera-
tions for each bacteria are presented in Table 4-1. For all bacteria, the plate enumeration was
conducted prior to testing. Because the B. anthracis stock solution aliquots were stored at 2 to
8°C, another 1 mL aliquot stock solution vial was enumerated during testing activities to further
confirm the concentration of the aliquots. The average of triplicate enumerations for each
bacteria was used to calculate and prepare all spiked sample solutions. The percent difference
between the concentration of the initial preparation of B. anthracis spores and the second
analysis of these spores during testing was 23%. Because this difference falls within the bounds
of expected plate enumeration errors and is close to the standard deviations found for the plate
enumerations of other bacteria used in this verification test, the concentration determined from
the initial set of plate enumerations on the B. anthracis spores was used in calculating solution
concentrations.
Table 4-1. F. tularensis, Y. pestis, and B. anthracis Triplicate Plate Enumeration Data
Bacteria
F. tularensis
Y. pestis
B. anthracis
(initial prep)
B. anthracis
(second
analysis)
Plate 1
Concentration
(cfu/mL)
l.OxlO9
5.8xl07
8.7xl07
5.7xl07
Plate 2
Concentration
(cfu/mL)
l.lxlO9
6.5xl07
S.lxlO7
5.7xl07
Plate 3
Concentration
(cfu/mL)
1.2xl09
5.0xl07
7.8xl07
7.6xl07
Average
(cfu/mL)
l.lxlO9
5.8xl07
8.2xl07
6.3xl07
Relative
Standard
Deviation
9%
13%
6%
17%
14
-------
4.4 Quality Control Samples
MB samples consisting of ASTM Type n DI water, and EPC and negative control samples, as
provided in the PathAlert™ Detection Kit, were analyzed to help identify potential cross-
contamination issues as well as verify that the PCR process was functioning properly. IPCs were
part of each sample that was analyzed and provided further checks on the performance of the
PathAlert™ Detection Kit, especially in identifying the presence of potential inhibitory
substances. EPC and negative control samples were run with each set of samples placed on the
thermal cycler. Eight MB replicates were analyzed over the course of the verification test for
each bacteria. Ladder samples, an external standard for the 2100 Bioanalyzer DNA 500 chip,
were analyzed with each chip to ensure the proper performance of the 2100 Bioanalyzer.
Each set of eight MB sample replicates for each bacteria returned negative results. IPC peaks
were present in all contaminant-only PT samples and interferent and DW samples using 37.5 |iL
of SuperMix for the PCR. In those interferent and DW samples containing 12.5 |iL of SuperMix,
the IPC peak was often suppressed, indicating the presence of inhibitors in the sample (see
section 6.5.1 for further details). For all three bacteria tested, no EPC or negative controls failed.
In two separate analysis events, the ladder on the DNA 500 chip failed because of improper
loading onto the chip. DNA 500 chips had to be reloaded and rerun on the 2100 Bioanalyzer
before results could be obtained for the samples on the failed chips.
4.5 Audits
4.5.1 Technical Systems Audit
The Battelle Quality Manager conducted a technical systems audit (TSA) on June 11, 2004, to
ensure that the verification test was performed in accordance with the test/QA plan(1) and the
AMS Center QMP.(11) As part of the audit, the Battelle Quality Manager reviewed the standards
and methods used, compared actual test procedures to those specified in the test/QA plan, and
reviewed data acquisition and handling procedures. Observations and findings from this audit
were documented and submitted to the Verification Test Coordinator for response. No findings
were documented that required any significant action. The records concerning the TSA are
stored for at least seven years with the Battelle Quality Manager.
4.5.2 Audit of Data Quality
At least 10% of the data acquired during the verification test was audited. Battelle's Quality
Manager traced the data from the initial acquisition, through reduction and statistical analysis,
to final reporting, to ensure the integrity of the reported results. All calculations performed on
the data undergoing the audit were checked.
15
-------
4.6 QA/QC Reporting
Each assessment and audit was documented in accordance with Sections 3.3.4 and 3.3.5 of the
QMP for the ETV AMS Center.(11) Once the assessment report was prepared, the Verification
Test Coordinator ensured that a response was provided for each adverse finding or potential
problem and implemented any necessary follow-up corrective action. The Battelle Quality
Manager ensured that follow-up corrective action was taken. The results of the TS A were sent to
the EPA.
4.7 Data Review
Records generated in the verification test were reviewed before these records were used to
calculate, evaluate, or report verification results. Table 4-2 summarizes the types of data
recorded. The review was performed by a Battelle technical staff member involved in the
verification test, but not the staff member that originally generated the record. The person
performing the review added his/her initials and the date to a hard copy of the record being
reviewed.
Table 4-2. Data Recording Process
Data to Be
Recorded
Where Recorded How Often Recorded
Disposition of Data(a)
Dates and times of
test events
ETV data sheets
Start/end of test and at
each change of a test
parameter
Used to organize/check test
results; manually incorporated
in data spreadsheets as
necessary
Sample collection
and preparation
information,
including chain-of-
custody
ETV data sheets
and chain-of-
custody forms
At time of sample
collection and
preparation
Used to organize/check test
results; manually incorporated
in data spreadsheets as
necessary
PathAlert™
Detection Kit
procedures and
sample results
Enumeration data
ETV data sheets
and data
acquisition system
Enumeration data
forms and ETV
data sheets
Throughout test
duration
With every
enumeration
Manually incorporated in data
spreadsheets
Used to organize/check test
results
Reference method
procedures and
sample results
Data acquisition
system, as
appropriate
Throughout sample
analysis process
Transferred to spreadsheets
) All activities subsequent to data recording were carried out by Battelle, except for the reference method analyses
(DW characterization), which were carried out by ATEL.
16
-------
Chapter 5
Data Analysis
The PathAlert™ Detection Kit was evaluated for qualitative results (i.e., positive/negative
responses to samples) based on the expected application of rapid PCR technologies as rapid
screening tools. All data analyses were based on these qualitative results. QC and MB samples
were not included in any of the analyses.
5.1 Accuracy
Accuracy was assessed by evaluating how often the PathAlert™ Detection Kit results were
positive in the presence of a concentration of contaminant above the method LOD.
Contaminant-only PT samples were used for this analysis. An overall percent agreement was
determined by dividing the number of positive responses by the overall number of analyses of
contaminant-only PT samples above the method LOD.
5.2 Specificity
The ability of the PathAlert™ Detection Kit to provide a negative response when the
contaminant was absent was assessed. The specificity rate was determined by dividing the
number of negative responses by the total number of unspiked samples.
5.3 False Positive/Negative Responses
A false positive response was defined as a detectable or positive PathAlert™ Detection Kit
response when the ASTM Type n DI water (including interferent samples) or DW samples were
not spiked. A false positive rate was reported as the frequency of false positive results out of the
total number of unspiked samples.
A false negative response was defined as a non-detectable response or negative response when
the sample was spiked with a contaminant at a concentration greater than the method LOD.
Spiked PT (contaminant and interferent) samples and spiked DW samples were included in the
analysis. A false negative rate was evaluated as the frequency of false negative results out of the
total number of spiked samples for a particular contaminant.
17
-------
5.4 Precision
The precision of the four replicates of each sample set were assessed. Responses were
considered consistent if all four replicates gave the same result. The precision of the PathAlert™
Detection Kit was assessed by calculating the overall number of consistent responses for all the
sample sets.
5.5 Interferences
The potential effect of the DW matrix on the PathAlert™ Detection Kit performance was
evaluated qualitatively by comparing the results for the spiked and unspiked DW samples to
those for the PT samples. Similarly, the potential effect of interferent PT samples containing
fulvic and humic acids at two levels, both spiked and not spiked with bacteria, were evaluated.
5.6 Other Performance Factors
Aspects of the PathAlert™ Detection Kit performance such as ease of use and sample through-
put are discussed in Section 6. Also addressed are qualitative observations of the verification
staff pertaining to the performance of the PathAlert™ Detection Kit.
18
-------
Chapter 6
Test Results
The results for the PathAlert™ Detection Kit were evaluated based on the responses provided by
the 2100 Bioanalyzer electropherogram output. An example positive electropherogram for
B. anthracis is presented in Figure 6-1. The electropherogram displays the fluorescence intensity
(fluorescence units) versus migration time (seconds) for each sample component, which is
displayed as a peak in the electropherogram. Only qualitative (positive/negative) responses were
recorded for each sample. To determine the results of each sample, peak sizes were monitored in
the electropherogram display. In Figure 6-1, peak sizes are shown above each peak.
Approximate amplicon sizes were listed on each bacteria-specific PathAlert™ Detection Kit
(consisting of reagents necessary for the PCR of each sample) for each expected peak in a
positive sample. These were used as guidelines in identifying positive responses. The threshold
for the peak height was set to 10 fluorescence units, and the peak filter width was set to 1
second, at the direction of the vendor. Peaks that met these criteria, as well as other criteria left
at the default settings, were automatically picked and integrated by the 2100 Bioanalyzer
software. The bacteria were considered present in the sample if the electropherogram for a given
sample showed three peaks of appropriate amplicon (base pair) size that appeared at
approximately the same time throughout the samples: two for the bacteria being monitored and
one for the IPC. The bacteria were considered not present, and thus a negative response was
recorded, if the only peak present was for the IPC. The results for a sample were considered
inconclusive if the IPC and only one bacteria peak were present in the electropherogram results.
In a real-world scenario, samples with inconclusive results would likely be rerun, as well as
subjected to a battery of other tests to confirm the presence or absence of the bacteria of interest.
Negative controls and EPCs were monitored with each day's sample set. No controls failed
throughout the testing process. The ladder well results for a given 2100 Bioanalyzer DNA 500
chip were also monitored to ensure the integrity of the chip analysis. The ladder is an external
standard for the 2100 Bioanalyzer DNA 500 chip that ensures that each chip is properly
working. If the ladder failed or was unsuccessful, no results were recorded for that chip, and the
samples were loaded onto another chip and run again.
19
-------
Sample 5,1 A
B. anthracis Sample Peak*
30 40 50 60 70
Figure 6-1. Positive Electropherogram for B. anthracis
90
100 110
120 [s]
-------
6.1 Accuracy
The results for the PathAlert™ Detection Kit using the contaminant-only PT samples containing
F. tularensis, Y. pestis, and B. anthracis are discussed in this section. The infective/lethal dose
samples for each bacteria were included in the contaminant-only PT samples. In the case of
Y. pestis and B. anthracis, the infective doses (see Table 3-1) were below the vendor-stated
method LOD. The results for each bacteria at the infective/lethal dose are presented in the
following tables, but those for Y. pestis, and B. anthracis were not included in the overall
accuracy calculations for those bacteria.
6.1.1 F. tularensis
The results obtained for the PT samples containing F. tularensis are given in Table 6-la. All
concentration levels analyzed generated 4 out of 4 positive responses. An overall percent
agreement was determined by dividing the number of positive responses by the overall number
of analyses of contaminant-only PT samples. This resulted in 100% agreement for the overall
accuracy of the PathAlert™ Detection Kit in detecting F. tularensis.
Table 6-la. F. tularensis Contaminant-Only PT Sample Results
Concentration^ Positive Results Out of
Sample Type (cfu/mL) Total Replicates
4xl05(b) 4/4
2xl04 4/4
PT samples 5xl04 4/4
IxlO5 4/4
5xl05 4/4
Overall accuracy 100% (20/20)
-------
samples above the method LOD. This resulted in 100% agreement for the overall accuracy of the
PathAlert™ Detection Kit in detecting Y. pestis above the method LOD.
Table 6-lb. Y. pestis Contaminant-Only PT Sample Results
Concentration^ Positive Results Out of
Sample Type (cfu/mL) Total Replicates
0.28(b) 0/4
2xl02 4/4
PT samples 5x102 4/4
IxlO3 4/4
5xl03 4/4
Overall accuracy 100% (16/16)(c)
(a) Sample solutions were prepared at 2, 5, 10, and 50 times the vendor-stated method LOD from stock solutions
based on the enumeration data (see Table 4-1).
(b) Infective/lethal dose—below the method LOD for Y. pestis.
(c) Excludes infective/lethal dose concentration, which was below the method LOD.
6.1.3 B. anthracis
The results obtained for the PT samples containing B. anthracis are given in Table 6-lc. All
samples with concentration levels above the vendor-stated method LOD generated 4 out of 4
positive responses. The infective/lethal dose of B. anthracis was below the method LOD for this
bacteria, but produced one positive response in four replicates. For the remaining three replicates
of the infective dose sample, the IPC and one bacteria peak were present in the
electropherogram. This indicated inconclusive results for those replicates (i.e., the sample could
be declared neither positive nor negative). In a screening scenario, inconclusive results would
lead to further testing of the sample, but this was beyond the scope of this test. The overall
accuracy of the PathAlert™ Detection Kit in detecting B. anthracis above the method LOD was
100%.
22
-------
Table 6-lc. B. anthracis Contaminant-Only PT Sample Results
Concentration^ Positive Results Out of
Sample Type (cfu/mL) Total Replicates
200(b)
2xl04
PT samples 5xl04
IxlO5
5xl05
1/4(0
4/4
4/4
4/4
4/4
Overall accuracy 100% (16/16)(d)
(a) Sample solutions were prepared at 2, 5, 10, and 50 times the vendor-stated method LOD from stock solutions
based on the enumeration data (see Table 4-1).
*• ' Infective/lethal dose—below the method LOD for B. anthracis.
(c) Three replicates had an IPC and one B. anthracis peak in the electropherogram. This indicated an inconclusive
result (neither positive or negative) for each replicate.
*• ' Excludes infective/lethal dose concentration, which was below the method LOD.
6.2 Specificity
Specificity assesses the PathAlert™ Detection Kit's ability to provide a negative response when
the contaminant was absent. The results from all unspiked interferent PT samples and unspiked
DW samples are presented in this section. Negative results out of total replicates are presented in
each table.
6.2.1 F. tularensis
The results obtained for F. tularensis for the unspiked interferent and DW samples are given in
Table 6-2a. All unspiked interferent PT samples showed negative responses. All OH, FL, and
NY unspiked DW samples showed negative responses also, indicating that the bacteria were not
present in these samples, as would be expected. For the unspiked CA DW samples, the IPC and
one bacteria peak were present in the electropherogram (with a baseline correction) for one
replicate. This indicated inconclusive results for this replicate (i.e., the sample could be declared
neither positive nor negative). The CA DW was further analyzed to determine the presence (or
absence) of F. tularensis naturally in the water (see Section 3.3.2). F. tularensis could not be
identified in the sample.
An overall specificity rate was determined by dividing the number of negative responses by the
overall number of analyses of unspiked samples. This resulted in 96% agreement for the overall
specificity of the PathAlert™ Detection Kit for F. tularensis.
23
-------
Table 6-2a. F. tularensis Specificity Results
Negative Results Out of
Sample Type Sample Total Replicates
0.5 mg/L humic acid and ...
0.5 mg/L fulvic acid, unspiked
Interferent PT samples
2.5 mg/L humic acid and ...
2.5 mg/L fulvic acid, unspiked
DW samples
Overall specificity
OH DW, unspiked
CA DW, unspiked
FL DW, unspiked
NY DW, unspiked
4/4
3/4(a)
4/4
4/4
96%(23/24)(a)
(a) One sample had an IPC and one bacteria peak in the electropherogram. This indicated an inconclusive result.
6.2.2 Y.pestis
The results obtained for unspiked interferent PT and DW samples using the Y. pestis
PathAlert™ Detection Kit are given in Table 6-2b. All unspiked interferent PT samples showed
negative responses. All OH, CA, and NY unspiked DW samples showed negative responses
also, indicating that the bacteria were not present in these samples, as would be expected. For
the unspiked FL DW samples, the IPC and one bacteria peak were present in the
electropherogram (with a baseline correction) for one replicate. This indicated inconclusive
results for this replicate (i.e., the sample could be declared neither positive nor negative).
An overall specificity rate was determined by dividing the number of negative responses by the
overall number of analyses of unspiked samples. This resulted in 96% agreement for the overall
specificity of the PathAlert™ Detection Kit for Y. pestis.
24
-------
Table 6-2b. Y. pestis Specificity Results
Negative Results Out of
Sample Type Sample Total Replicates
0.5 mg/L humic acid and ...
0.5 mg/L fulvic acid, unspiked
Interferent PT samples
2.5 mg/L humic acid and ...
2.5 mg/L fulvic acid, unspiked
DW samples
Overall specificity
OH DW, unspiked
CA DW, unspiked
FL DW, unspiked
NY DW, unspiked
4/4
4/4
3/4(a)
4/4
96% (23/24)(a)
(a) One sample had an IPC and one bacteria peak in the electropherogram. This indicated an inconclusive result.
6.2.3 B. anthracis
The results obtained using B. anthracis reagents for the analysis of unspiked interferent and DW
samples are given in Table 6-2c. All unspiked interferent PT samples and unspiked DW samples
showed negative responses for all of the replicates. The overall specificity rate of the
PathAlert™ Detection Kit for B. anthracis was 100%.
Table 6-2c. B. anthracis Specificity Results
Negative Results Out of
Sample Type Sample Total Replicates
Interferent PT samples
DW samples
Overall specificity
0.5 mg/L humic acid and
0.5 mg/L fulvic acid, unspiked
2.5 mg/L humic acid and
2.5 mg/L fulvic acid, unspiked
OH DW, unspiked
CA DW, unspiked
FL DW, unspiked
NY DW, unspiked
3/3(a)
3/3(a)
4/4
4/4
4/4
4/4
100% (22/22)
a) These samples had to be rerun because of suspected sample preparation problems. Samples were run in triplicate
because of limited supplies. Only the rerun results are presented.
25
-------
6.3 False Positive/Negative Responses
Contaminant-only PT samples, interferent PT samples, and DW samples were evaluated to
determine false positive and false negative results for the PathAlert™ Detection Kit. Included in
the calculations were the 16 additional interferent samples (0.5 mg/L each humic and fulvic
acids) tested to determine the effects of the DNA extraction and isolation on the results. A false
positive response was defined as a positive result when bacteria were not spiked into the sample.
A false negative response was defined as a negative result when the sample was spiked with a
contaminant at a concentration greater than the method LOD for that bacteria.
It should be noted that false positive responses cannot be absolutely confirmed as false because
there is a possibility of cross-contamination. All appropriate steps were taken throughout the
verification test to avoid this issue by using three work areas ("clean," "medium," and "dirty"),
by following daily cleanup procedures, and by loading only one set of samples into the PCR
tubes at a time (thus only the optical tubes for one set of replicates were uncapped at a time).
However, cross-contamination is always a possibility in any PCR process.(12) No appropriate
reference method was available to cross-check the amplified PCR product to confirm the
PathAlert™ Detection Kit responses. When sample preparation error was suspected (e.g., a
sample appeared to be unspiked when it should have been spiked or spiked when it should have
been blank), the sample was reevaluated. If sample preparation errors or cross-contamination
were suspected after reanalysis, only the results of the reruns were presented.
6.3.1 F. tularensis
Table 6-3a presents the false positive/negative results for F. tularensis. The number of positive
samples out of the total replicates analyzed is presented in the table. No false positive or false
negative samples were found in any of the sample matrices. One replicate for unspiked CA DW
did show one bacteria peak along with the IPC peak in the electropherogram results. Because
neither F. tularensis peaks was apparent in the electropherogram, the result was determined to
be inconclusive.
6.3.2 Y.pestis
Table 6-3b presents the false positive/negative results for Y. pestis. The number of positive
samples out of the total replicates analyzed is presented in the table. As with F. tularensis, no
false positive or false negative samples were found in any of the sample matrices. One replicate
for unspiked FL DW did show one of the two Y. pestis peaks along with the IPC peak in the
electropherogram results. Because neither Y. pestis peak was apparent in the electropherogram,
the result was determined to be inconclusive.
6.3.3 B. anthracis
Table 6-3c presents the false positive/negative results for B. anthracis. The number of positive
samples out of the total replicates analyzed is presented in the table. No false positive or false
negative samples were found in any of the sample matrices.
26
-------
Table 6-3a. F. tularensis False Positive/Negative Results
Sample Type
Contaminant-only
PT samples
Interferent PT
samples
DW samples
False positive rate
False negative rate
Sample
DI water
DI water
DI water
DI water
DI water
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
OHDW
OHDW
CADW
CADW
FLOW
FLOW
NYDW
NYDW
Concentration*3'
(cfu/mL)
4xl05(b)
2xl04
5xl04
IxlO5
5xl05
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Positive Results Out of
Total Replicates
4/4
4/4
4/4
4/4
4/4
0/4
20/20
0/4
4/4
0/4
4/4
0/4(c)
4/4
0/4
4/4
0/4
4/4
0/24(c)
0/60
(a) Sample solutions were prepared at 2, 5, 10, and 50 times the vendor-stated method LOD from stock solutions
based on the enumeration data (see Table 4-1).
(b) Infective/lethal dose.
(c) One unspiked CA DW replicate had an IPC and one F. tularensis peak in the electropherogram. This indicated an
inconclusive result.
27
-------
Table 6-3b. Y. pestis False Positive/Negative Results
Sample Type
Contaminant-only
PT samples
Interferent PT
samples
DW samples
False positive rate
False negative rate
Sample
DI water
DI water
DI water
DI water
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
OHDW
OHDW
CADW
CADW
FLOW
FLDW
NYDW
NYDW
Concentration'3'
(cfu/mL)
2xl02
5xl02
IxlO3
5xl03
Blank
IxlO3
Blank
IxlO3
Blank
IxlO3
Blank
IxlO3
Blank
IxlO3
Blank
IxlO3
Positive Results Out of
Total Replicates
4/4
4/4
4/4
4/4
0/4
20/20
0/4
4/4
0/4
4/4
0/4
4/4
0/400
4/4
0/4
4/4
0/24(b)
0/56(c)
-------
Table 6-3c. B. anthracis False Positive/Negative Results
Sample Type
Contaminant-only
PT samples
Interferent PT
samples
DW samples
False positive rate
False negative rate
Sample
DI water
DI water
DI water
DI water
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
0.5 mg/L humic acid and
0.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
2.5 mg/L humic acid and
2.5 mg/L fulvic acid
OHDW
OHDW
CADW
CADW
FLOW
FLDW
NYDW
NYDW
Concentration'3'
(cfu/mL)
2xl04
5xl04
IxlO5
5xl05
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Blank
IxlO5
Positive Results Out of
Total Replicates
4/4
4/4
4/4
4/4
0/3°°
20/20
0/3°°
4/4
0/4
4/4
0/4
4/4
0/4
4/4
0/4
4/4
0/22
0/56(c)
(a) Sample solutions were prepared at 2, 5, 10, and 50 times the vendor-stated method LOD from stock solutions
based on the enumeration data (see Table 4-1).
(b)
(c)
These samples had to be rerun because of suspected cross-contamination problems. Samples were run in triplicate
because of limited supplies. Only the rerun results are presented.
The infective/lethal dose for B. anthracis was below the method LOD and thus was not included in this
calculation.
29
-------
6.4 Precision
The performance of the PathAlert™ Detection Kit F. tularensis assay within sample sets of four
replicates was consistent. Only one set of replicates, that for unspiked C A DW, was inconsistent,
with one of the replicates showing inconclusive results, while the other samples were negative.
All other samples showed the same results within a set of replicates. Thus, for F. tularensis, one
of the 21 sets of replicates that was analyzed was determined to be inconsistent, indicating that
95% of the sample sets showed consistent results among the replicates.
The results for Y. pestis were similar to those for F. tularensis, with only one set of replicate
sample sets showing inconsistent results. One of the four quadruplicate samples for unspiked FL
DW showed inconclusive results (only one of the two Y. pestis peaks was apparent on the
electropherogram, along with the IPC), while the other three samples were all negative. All of
the remaining 20 sets of replicate samples analyzed for Y. pestis by the PathAlert™ Detection
Kit showed the same results within the sample set. Thus, as with F. tularensis, 95% of the
sample sets showed consistent results among the replicates.
As with F. tularensis and Y. pestis, 95% of the sample sets for B. anthracis (20/21) showed
consistent results among the replicates. For this bacteria, the one sample set with inconsistent
results was the infective dose PT sample. In this set of replicates, three of the four samples had
inconclusive results, while the fourth sample was positive for B. anthracis. The infective dose of
B. anthracis was below the method LOD for this bacteria, so the discrepancy between replicate
samples likely has more to do with this fact than the actual precision of the PathAlert™
Detection Kit.
6.5 Interferences
6.5.1 Interferent PT Samples
In the 0.5 mg/L and 2.5 mg/L fulvic and humic acids solution, both spiked with the bacteria of
interest and unspiked, the PathAlert™ Detection Kit provided expected results for F. tularensis,
Y. pestis, and B. anthracis. In the absence of the bacteria, the samples tested negative; in the
presence of the bacteria, the samples tested positive. For the interferent PT samples for
F. tularensis, Y. pestis, and B. anthracis, 37.5 |iL of the PathAlert™ Detection Kit SuperMix
were used for the PCR process. In the case of F. tularensis, the first bacteria tested, the spiked
interferent PT samples were also analyzed using 12.5 |iL of SuperMix in addition to the
interferent PT samples analyzed using 37.5 |iL of SuperMix. This was done to verify that fulvic
and humic acids were acting as potential inhibitors in the PCR process. The presence of an
inhibitory substance is signified by the suppression of the IPC peak in the electropherogram
results for the sample. For the 0.5 mg/L each humic and fulvic acids solution spiked with IxlO5
cfu/mL of F. tularensis, the IPC peak was suppressed in two of the replicate samples, with one
of those replicates showing only one sample peak in the electropherogram. The remaining two
replicates indicated positive results, with the IPC clearly present. For the 2.5 mg/L each humic
and fulvic acids solution spiked with IxlO5 cfu/mL of F. tularensis, the IPC was suppressed in
30
-------
three of the four replicate samples, though both sample peaks were apparent in those three
samples. The remaining replicate sample was positive, with the IPC and two peaks indicative of
F. tularensis present in the electropherogram. The suppression of the IPC peaks indicated that
the humic and fulvic acids were acting as inhibitory substances in the spiked PT interferent
samples. After discussions with Invitrogen Corporation, and in the interest of time, all remaining
interferent PT samples for Y. pestis and B. anthracis were analyzed using only 37.5 |iL of
SuperMix to overcome the inhibitory actions of the humic and fulvic acids.
As discussed in Section 3.2.1, four solutions of fulvic and humic acids at 0.5 mg/L each, spiked
with each contaminant at IxlO3 and IxlO5 cfu/mL (depending on the bacteria), were prepared in
addition to the initial 0.5 mg/L and 2.5 mg/L fulvic and humic acid solutions. Each solution was
put through the DNA extraction and isolation procedure, and then four replicates from each of
the four purified DNA solutions were analyzed using the PathAlert™ Detection Kit. These
samples were included in the verification test in an effort to evaluate the efficacy of the DNA
extraction and isolation procedure in the presence of inhibitory substances. These samples also
contribute to the precision evaluations of the PathAlert™ Detection Kit. For F. tularensis,
Y. pestis, and B. anthracis, all of the samples tested resulted in positive responses. Thus, 20 out
of the 20 spiked 0.5 mg/L humic and fulvic acid samples tested resulted in positive responses.
6.5.2 Drinking Water Samples
The PathAlert™ Detection Kit DW sample results for F. tularensis, Y. pestis, and B. anthracis
are presented in Tables 6-3a, 6-3b, and 6-3c, respectively. In general, the PathAlert™ Detection
Kit showed positive results for each set of replicates for the spiked DW samples and negative
results for each set of replicates for the unspiked DW samples, with two exceptions. For the
detection of F. tularensis in unspiked CA DW, one of the four replicates had an inconclusive
result, where one of the two F. tularensis peaks was apparent on the electropherogram, along
with the IPC. Similarly, one of the replicates for the detection of Y. pestis in unspiked FL DW
had an inconclusive result. Analysis of the DW samples did not indicate the presence of
F. tularensis in the CA DW and could not confirm the presence of Y. pestis in the FL DW. The
possibility of cross-contamination causing the inconclusive results for these DW samples cannot
be ruled out.
As with the interferent PT samples, DW samples spiked with F. tularensis were also analyzed
using both 12.5 |iL and 37.5 |iL of SuperMix. Because of time constraints and the amount of
SuperMix readily available, this was only done for spiked C A, FL, and NY DW. For two of the
spiked FL DW replicates, there was a baseline shift and few identifiable peaks in the
electropherogram. After a baseline correction was performed, all F. tularensis and IPC peaks
were present in the electropherogram. It is not clear whether the baseline rise/shift is attributable
to the amount of SuperMix in the sample. The remaining two replicates were positive, with all
appropriate peaks present in the electropherogram. For spiked NY DW analyzed using 12.5 |iL
of SuperMix, one replicate required baseline correction to best view the electropherogram. Upon
correction, the sample response was positive. The other three replicates all had suppressed IPC
peaks, indicative of the presence of inhibitory substances. Two of the three samples were
inconclusive, showing only one F. tularensis sample peak, while both F. tularensis bacteria
31
-------
peaks were present in the third replicates, with only the IPC not found. For spiked CA DW
samples, all replicate results were positive. In the case of one replicate, all peak sizes were
slightly smaller than the other three samples. As with the interferent PT samples, all remaining
DW samples for Y. pestis and B. anthracis were analyzed using only 37.5 |iL of SuperMix to
overcome the potential inhibitory actions of the DW matrix.
The contaminant-only PT samples spiked at IxlO5 cfu/mL with F. tularensis and B. anthracis
and IxlO3 cfu/mL with Y. pestis, the level at which the DW samples were spiked, showed
consistent positive responses across all bacteria. The interferent PT samples at both 0.5 mg/L
and 2.5 mg/L humic and fulvic acids also spiked at IxlO5 and IxlO3 cfu/mL showed consistent
positive responses for all replicates across all bacteria using 37.5 |iL of SuperMix. The IPC peak
was present in all of the aforementioned interferent PT samples. The consistency of responses in
these PT samples, as well as other contaminant-only PT samples above the method LOD, would
seem to indicate that the DW matrices used in this test do not have inhibitory effects on the PCR
process for the Path Alert™ Detection Kit using 37.5 |iL of SuperMix.
6.6 Other Performance Factors
The Path Alert™ Detection Kit was operated by the same Battelle technician throughout the
verification test. This technician had prior PCR experience and was trained by Invitrogen and
Agilent in operating the PathAlert™ Detection Kit and 2100 Bioanalyzer, respectively, before
testing began. This training included the use of the Roche High Pure Prep Kit, the PathAlert™
Detection Kit, the operation of the thermal cycler used for testing, and the use of the 2100
Bioanalyzer. The Battelle technician was familiar with general DNA extraction and isolation
techniques, PCR plating techniques, and general thermal cycler operation, as well as general
PCR theory prior to training. The overall operation of the PathAlert™ Detection Kit was
straightforward, and the experienced technician found the kit easy to use and had no major
difficulties using the reagents. The need to use only the SuperMix for the PCR setup added to
the ease of operation of the kit, since all of the necessary components for the PCR process were
contained in one solution, instead of two or more. Though the DNA extraction procedure was
straightforward, many steps were involved in the process. The operation of the 2100 Bioanalyzer
was straightforward, though some degree of laboratory skill was required to properly load the
chips without bubbles. Training on the 2100 Bioanalyzer software was very helpful in
understanding how to interpret the results because an understanding of the software and the
expected peak sizes was necessary for the data interpretations.
All testing was performed in a laboratory setting because the PathAlert™ Detection Kit is not
field portable. Three distinct and separate testing areas were required in each laboratory to
operate the PathAlert™ Detection Kit: "clean" or DNA-free, "medium" or moderate amount of
DNA, and "dirty" or high DNA concentration. The PathAlert™ Detection Kit PCR reagents had
to be stored at -20°C and thawed before use. The SuperMix was aliquoted into smaller portions
to avoid thawing and refreezing the entire allotment of SuperMix each day. The PCR reactions
had to be assembled on ice, and the plated SuperMix had to be incubated on ice or placed in the
refrigerator until the sample DNA could be added to it.
32
-------
F. tularensis samples were tested in a BSL-2 laboratory, while Y. pestis and B. anthracis were
tested in a BSL-3 laboratory. Because live bacteria were being handled, special safety require-
ments and protocols had to be implemented in both the BSL-2 and BSL-3 laboratories. Some of
these requirements impacted the analysis time for the PathAlert™ Detection Kit and are
inherently present in any throughput estimations for this verification test. Thus, performance
factors mentioned also incorporate the safety and facility requirements necessary for this test.
The PathAlert™ Detection Kit was used to test 92 or more sample replicates (including MBs)
for each bacteria. Dispensing the SuperMix into the PCR tubes took approximately 15 minutes
for each set of samples analyzed on a given day. Loading the sample DNA into the PCR tubes
after purification took between 15 and 30 minutes, depending on the number of samples being
analyzed. A maximum 36 replicates (nine sample solutions) plus controls were analyzed on a
given day. Most sample sets averaged between five and eight sample solutions. On average, the
DNA extraction and isolation step for between five and nine solutions took approximately
2.5 hours. The completion of the thermal cycle program to amplify the sample DNA took
approximately 1.5 hours. Loading and analyzing each 2100 Bioanalyzer chip took
approximately 45 minutes. The verification staff analyzed on average three DNA 500 chips a
day, in some instances up to 4 chips a day. This equates to approximately 36 sample replicates a
day for three chips and 48 sample replicates a day for four chips, including controls. The
PathAlert™ Detection Kits can perform up to 320 assays per kit using 12.5 |oL of SuperMix per
reaction.
33
-------
Chapter 7
Performance Summary
The PathAlert™ Detection Kit results for this verification test for samples containing
F. tularensis, Y. pestis, and B. anthracis are presented in Tables 7-1 through 7-3. The results for
each bacteria assay are presented in a separate table. Qualitative responses for each set of sample
replicates as well as accuracy, specificity, false positives and negatives, and precision are
presented in each table. A summary of the other performance factors associated with the
PathAlert™ Detection Kit is presented at the end of this chapter. These performance factors
apply to each kit across all bacteria.
34
-------
Table 7-1. F. tularensis Summary Table
Parameter
Qualitative
results
Contaminant-
only PT
samples
Interferent
PT samples00
DW samples*'
Accuracy
Specificity
False positives
False negatives
Precision
Number Detected/
Sample Information Concentration Number of Samples
4xl05 cfu/mL(a) 4/4
2xl04cfu/mL 4/4
DI water 5xl04cfu/mL 4/4
lx!05cfu/mL 4/4
5xl05cfu/mL 4/4
Humic and . In5 - , T i/i/o/i
,, , . ., IxlO^cfu/mL 24/24
fulvic acids
Concentrated DW lxl05cfu/mL 16/16
100% (20 out of 20) of the contaminant-only PT samples above the
method LOD were positive.
96% (23 out of 24) of the unspiked interferent and DW samples were
negative. One unspiked CA DW replicate returned an inconclusive
result. (c)
No false positives resulted from the analysis of the unspiked
interferent or DW samples. One inconclusive result was obtained for
CA DW(C)
No false negative results were obtained from the analysis of the
interferent and DW samples spiked with F. tularensis above the
method LOD.
95% (20 out of 21) of the sample sets showed consistent results
among the individual replicates within that set.(c)
(a)
(c)
Infective/lethal dose.
tnterterent Pi and D\v sample results retlect the use ot 3 /.5 j^-L ot ouperJVLix. oome samples were analyzed using
12.5 u,L of SuperMix, and many of the interferent PT and DW samples showed suppressed IPC peaks, indicating
the presence of inhibitory substances.
One unspiked CA DW replicate had an IPC and one bacteria peak in the electropherogram. This indicated an
inconclusive result and would require reanalysis in a real-world scenario. The remaining three replicates were
negative.
35
-------
Table 7-2. Y. pestis Summary Table
Parameter
Qualitative
results
Contaminant-
only PT
samples
Interferent
PT samples00
DW samples00
Accuracy
Specificity
False positives
False negatives
Precision
Number Detected/
Sample Information Concentration Number of Samples
0.28 cfu/mL(a) 0/4(a)
2xl02 cfu/mL 4/4
DI water 5xl02cfu/mL 4/4
IxlO3 cfu/mL 4/4
5x1 03 cfu/mL 4/4
Humic and 1 1A3 , o/i/o/i
,. , . ., 1x10 cfu/mL 24/24
fulvic acids
Concentrated DW lxl03cfu/mL 16/16
100% (16 out of 16) of the contaminant-only PT samples above the
method LOD were positive.
96% (23 out of 24) of the unspiked interferent and DW samples were
negative. One unspiked FL DW replicate returned an inconclusive
result. (c)
No false positives resulted from the analysis of the unspiked
interferent or DW samples. One inconclusive result was obtained for
FL DW(C)
No false negative results were obtained from the analysis of the
interferent and DW samples spiked with Y. pestis above the method
LOD.
95% (20 out of 21) of the sample sets showed consistent results
among the individual replicates within that set.(c)
Infective/lethal dose—below the method LOD for Y. pestis.
00 Interferent PT and DW sample results reflect the use of 37.5 uL of SuperMix.
(c) One unspiked FL DW replicate had an IPC and one bacteria peak in the electropherogram. This indicated an
inconclusive result and would require reanalysis in a real-world scenario. The remaining three replicates were
negative.
36
-------
Table 7-3. B. anthracis Summary Table
Parameter
Qualitative
results
Contaminant-
only PT
samples
Interferent
PT samples (c)
DW samples (c)
Accuracy
Specificity
False positives
False negatives
Precision
Number Detected/
Sample Information Concentration Number of Samples
200 cfu/mL(a) l/4<®
2xl04cfu/mL 4/4
DI water 5xl04cfu/mL 4/4
lxl05cfu/mL 4/4
5xl05cfu/mL 4/4
Humic and 1 1A5 , o/i/o/i
,. , . ., 1x10 cfu/mL 24/24
rulvic acids
Concentrated DW lxl05cfu/mL 16/16
100% (16 out of 16) of the contaminant-only PT samples above the
method LOD were positive.
100% (22 out of 22) of the unspiked interferent and DW samples
were negative/*
No false positives resulted from the analysis of the unspiked
interferent or DW samples.
No false negative results were obtained from the analysis of the
interferent and DW samples spiked with B. anthracis above the
method LOD.
95% (20 out of 21) of the sample sets showed consistent results
among the individual replicates within that set.*'
(a) Infective/lethal dose—below the method LOD for B. anthracis.
(b) Three samples in the infective/lethal dose PT sample replicates had an IPC and one B. anthracis peak in the
electropherogram. This indicated an inconclusive result for each sample.
(c) Interferent PT and DW sample results reflect the use of 37.5 uL of SuperMix.
(d) Three replicates were run for both unspiked fulvic and humic acid samples because of limited supplies for rerun
analysis.
Other performance factors: A technician with prior PCR experience operated the PathAlert™
Detection Kit at all times. The kit was straightforward and easy to use. All components
necessary for the PCR process (excluding the sample DNA) were contained in one solution,
which had to be stored at -20°C. Three separate work areas were needed for testing: a "clean"
area free of DNA, a "medium" area with moderate DNA presence, and a "dirty" area, with high
DNA presence. SuperMix preparation took approximately 15 minutes, and loading the sample
DNA into the PCR tubes took between 15 and 30 minutes for each set of samples. For this
verification test, sample throughput (from DNA purification to amplified product detection) was
36 to 48 samples per day. The PathAlert™ Detection Kit cost is around $15 per assay
(approximately $15 per sample using 12.5 |iL of SuperMix). PathAlert™ Detection Kits can
perform up to 320 assays per kit.
37
-------
Chapter 8
References
1. Test/QA Plan for Verification of Rapid PCR Technologies, Battelle, Columbus, Ohio,
May 2004.
2. Burrows, W. D.; Renner, S. E. "Biological Warfare Agents as Threats to Potable Water,"
Environmental Health Perspectives, 107, 975-984, 1999.
3. Stenman, J.; Orpana, A. "Accuracy in amplification," Nature Biotechnology, 19, 1011-
1012,2001.
4. Hughes, J.; Totten, P. "Estimating the accuracy of polymerase chain reaction-based tests
using endpoint dilution," University of Washington Biostatistics Working Paper Series,
Working Paper 196, 2003.
5. U.S. EPA Method 180.1, "Turbidity (Nephelometric)" in Methods for the Determination of
Inorganic Substances in Environmental Samples, EPA/600/R-93-100, August 1993.
6. American Public Health Association, et al. Standard Methods for Examination of Water
and Waste-water. 19th Edition. 1997. Washington D.C.
7. U.S. EPA, Methods for Chemical Analysis of Water and Wastes, EPA/600/4-79-020,
March 1983.
8. U.S. EPA Method 200.8, "Determination of Trace Elements in Waters and Wastes by
Inductively-Coupled Plasma Mass Spectrometry" in Methods for the Determination of
Organic Compounds in Drinking Water, Supplement I, EPA/600/R-94/111, October 1994.
9. U.S. EPA Method 524.2, "Permeable Organic Compounds by Capillary Column GC/Mass
Spectrometry" in Methods for the Determination of Organic Compounds in Drinking
Water, Supplement m, EPA/600/R-95-131. August 1995.
10. U.S. EPA Method 552.2, "Haloacetic Acids and Dalapon by Liquid-Liquid Extraction,
Derivatization and GC with Electron Capture Detector" in Methods for the Determination
of Organic Compounds in Drinking Water, Supplement ffl, EPA/600/R-95-131, August
1995.
38
-------
11. Quality Management Plan (QMP)for the ETV Advanced Monitoring Systems Center,
Version 5.0. EPA Environmental Technology Verification Program, prepared by Battelle,
Columbus, Ohio, March, 2004.
12. Shapiro, D. S. "Quality Control in Nucleic Acid Amplification Methods: Use of
Elementary Probability Theory," Journal of Clinical Microbiology, 37, 848-851, 1999.
39
------- |