EPA/600/R-15/118 November 2015 I www2.epa.gov/research
United States
Environmental Protection
Agency
Literature Review on Processing and
Analytical Methods for Francisella
tularensis in Soil and Water
w
W
Office of Research and Development
National Homeland Security Research Center
-------
Disclaimer
The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed the research described here under Contract No. SP0700-00-D-3180,
Delivery Order 0729, Technical Area Task CB-11-0232 with the Defense Threat Reduction
Agency and the Department of Homeland Security under the Battelle/Chemical, Biological,
Radiological, and Nuclear Defense Information and Analysis Center. It has been subjected to the
Agency's review and has been approved for publication. Note that approval does not necessarily
signify that the contents reflect the views of the Agency. Mention of trade names, products, or
services does not convey official EPA approval, endorsement, or recommendation.
This report was generated using references (secondary data) that could not be evaluated for
accuracy, precision, representativeness, completeness, or comparability and therefore no
assurance can be made that the data extracted from these publications meet EPA's stringent
quality assurance requirements.
Questions concerning this document or its application should be addressed to:
Erin Silvestri
U.S. Environmental Protection Agency
National Homeland Security Research Center
26 W. Martin Luther King Drive, MS NG16
Cincinnati, OH 45268
513-569-7619
Silvestri.Erin@EPA.gov
-------
Executive Summary
Identifying pathogenic organisms within a soil sample can be a challenge, yet an understanding
of the environmental distribution of bacterial pathogens and their fate over time is needed for
multiple applications. Francisella tularensis, the etiological agent of tularemia in humans and
animals, can be acquired by handling infected carcasses, ingesting contaminated food or water,
from an infected arthropod bite, or inhaling infectious soil dust or aerosols1. F. tularensis is
listed as a Category A agent by the Centers for Disease Control and Prevention (CDC) due to its
extremely low infectious dose for humans2 so identifying F. tularensis within an environmental
soil or water matrix is a priority for protecting both human and animal lives. F. tularensis is
widely distributed in the environment and has been isolated from nearly 250 wildlife species,
ranging from mammals, invertebrates, birds, amphibians, and fish3. It is an environmentally
hardy organism and can survive for weeks at low temperatures in water, moist soil, hay, straw, or
decaying animal carcasses4.
The purpose of this report was to survey the open literature to determine the current state of the
science regarding the processing and analytical methods currently available for recovery of F.
tularensis from water and soil matrices, and to determine what gaps remain in the collective
knowledge concerning F. tularensis identification from environmental samples. Information for
this review came from unclassified reports, peer-reviewed journal articles, published books, and
government publications published in the last twenty years. The search was limited to articles
published in the English language, but no restrictions were placed on the geographic focus of the
documents.
The search identified three broad mechanisms of F. tularensis detection within environmental
samples: culture analysis, immunoassays, and genomic identification. Isolating environmental
cultures of F. tularensis is challenging as it is a slow-growing, nutritionally fastidious organism
requiring 24 to 72 hours for growth5 on supplemented media6. Even with antibiotic amended
media, colonies are often out-competed by background organisms present in environmental
samples. Antibiotic supplemented cysteine heart agar with blood (CHAB) was frequently cited in
the literature to culture F. tularensis from environmental samples. While CHAB, or modified
forms of CHAB, have been used to detect virulent F. tularensis from within environmental
samples, the process is long, labor intensive, and rarely yields positive isolates. There were
1 Fujita et al., 2006. Jpn J Infect Dis 59:46-51.
2 Cooper etal., 2011. Sensors 11:3004-19.
3 Broman et al., 2011. Int J Microbiol 2011.
4 Dennis et al., 2001. JAMA 285:2763-73.
5 Versage et al., 2003. J Clin Microbiol 41:5492-5499.
6 vanHoek, M. L. 2013. Virulence 4:833-46.
in
-------
multiple instances described in the literature where F. tularemis was identified by molecular
assays, but culture identification was limited in the laboratory setting.
Immunoassay detection of F. tularemis can be amenable to hand-held devices or multiagent
identification procedures; however, due to high limits of detection, the utility of these
immunoassay detection applications might only be seen in highly concentrated samples7. The
overall applicability of immunoassays are dependent upon the specificity of the selected
antigens. Some antigens can have cross-reactivity to other microorganism, thus impeding the
results.
Genomic identification of F. tularensis from environmental samples can rapidly yield detection
results. However, it must be noted that molecular identification of F. tularensis does not
necessarily indicate the presence of viable F. tularensis cultures. F. tularensis is a non-
sporulating Gram-negative organism; therefore, its deoxyribonucleic acid (DNA) can be
extracted for identification rather easily when compared to sporulated microorganisms. Yet,
inhibitory chemical constituents within environmental samples are often coextracted and lead to
confounding downstream polymerase chain reaction (PCR) responses. Therefore, special care
must be taken to efficiently clean environmental DNA extracts prior to downstream analysis. A
comparison of multiple commercial DNA recovery kits for isolating F. tularensis DNA from
within various soil types highlighted the efficiency of the UltraClean® Microbial DNA Isolation
kit and the PowerMax® Soil DNA Isolation kit, both products of MoBio Laboratories, Inc.,
Carlsbad, CA8. The UltraClean® Soil DNA Isolation kit (MoBio Laboratories, Inc., Carlsbad,
CA) was the recovery kit most commonly used within this literature search. One study
comparing two kits that used different amounts of the initial sample concluded that for samples
of unknown biological agents it is preferable to extract DNA from as much of the original
sample volume as possible9.
Direct genomic DNA extraction was not the only method for sample preparation found within
the literature. Sellek et al.10 developed a filtration method for processing soil samples that
allowed for both genomic analysis and immunologic analysis of the extracted sample with
limited efficiency. Trombley Hall et al.u focused on finding PCR reagents with inhibitor-
resistant capabilities. Use of inhibitor-resistant PCR reagents could eliminate the need for
sample-specific sample preparation and increase the sensitivity of downstream real-time PCR.
Multiple studies within this review demonstrated the capability of one assay to identify multiple
biothreat agents from a single sample. However, these studies also noted a trade-off between
achieving multiple organism detection and producing a minimized limit of detection (LOD).
7 Huelseweh et al., 2006. Proteomics 6:2972-81.
8 Whitehouse and Hottel. 2007. Mol Cell Probes 21:92-6.
9 Offermans and Zegers. 2007. Test Results 7th NATO-SIBCRA BW Round Robin Trial 2006. TD2007-0043.
10Sellek et al., 2008. J Environ Monit 10:362-9.
"Trombley Hall et al., 2013. PLoS One 8:e73845.
IV
-------
While each of the multiagent technologies is promising, efficacy data from environmental
assessments with complex matrices are lacking. Data from a broader range of complex
environments are needed to enable evaluation of the usefulness of the approach.
Two studies included in this review combined culture and genomic analysis to rapidly quantify
viable microorganisms. Using macrophage cell cultures to accelerate F. tularemis growth before
DNA extraction and amplification, Day and Whiting12 were able to detect viable F. tularensis in
contaminated foods at a LOD of 10 colony forming units (CPU)/ milliliter (mL). Rapid viability-
PCR (RV-PCR) is another promising technique that utilizes an enrichment step and the change in
cycle threshold time between two PCR reactions to determine the presence or absence of viable
cells13. While RV-PCR has not been optimized for F. tularensis detection, it has been shown to
be effective for Bacillus anthracis spore detection in environmental samples. Future work
incorporating a macrophage culture step with RV-PCR sample processes could significantly
improve viable F. tularensis detection capabilities in environmental soil and waters.
Other areas for future work could include a combined comparison of multiple soils with various
extraction kits and various inhibitor-resistant PCR reagents. Such an analysis would identify both
an optimum extraction kit and optimum PCR reagents to yield real-time PCR reactions with
increased sensitivity. Microarray detection technologies could be the future of high-throughput
environmental detection of multiple biothreat agents of interest. The introduction of whole
genome amplification prior to microarray detection might further improve sensitivity14. Future
work combining the use of internal controls for each analytical step, optimized DNA extraction,
whole genome amplification with inhibition-resistant polymerases, and multiagent microarray
detection could significantly expand the detection capabilities of F. tularensis in soil and water.
12Day and Whiting, 2009. J Food Protect 72:1156-1164.
13 Letant et al, 2011. Appl Environ Microbiol 77:6570-8.
14Brinkman et al., 2013. J Appl Microbiol 114:564-73.
-------
Table of Contents
Disclaimer ii
Executive Summary iii
Table of Contents vi
List of Acronyms and Abbreviations viii
Acknowledgements x
1 Introduction 1
1.1 Characteristics of F. tularensis 2
1.2 Persistence of F. tularensis in the environment 3
1.3 Purpose 4
1.4 Methods 5
2 Current State of the Science 5
2.1 Sample Processing 6
2.2 CulturingF. tularensis from the Environment 7
2.3 Immunoassay Detection of F. tularensis 13
2.4 Genomic Identification of F. tularensis 17
2.4.1 Extraction ofF. tularensisDNA 17
2.4.2 PCR amplification for genomic identification of F. tularensis 25
2.4.3 Methods for Environmental Sampling and Detection of Multiple Biothreat
Organisms 26
2.5 Combining Culture with PCR to detect liveF. tularensis 32
3 Conclusions and Identified Data Gaps 33
4 References 36
vi
-------
List of Tables
Table 1. Comparison of Francisella tularensis Culturing Studies 10
Table 2. Comparison of Francisella tularensis Immunoassay Studies 15
Table 3. Comparison of Francisella tularensis Genomic Studies 20
Table 4. Comparison of Developing Methods for Genomic Identification of Francisella tularensis Alone
and Simultaneously with Other Organisms 27
vn
-------
List of Acronyms and Abbreviations
°C
ABICAP
AFLP
BAL
CDC
CPU
CHAB
CHAB-A
ELISA
EPA
ESI-MS
fg
g
g
GE
kDa
L
LOD
LPS
LVS
Min
mL
mPCR-EHA
ng
PBS
PCR
PFGE
Pg
qPCR
R.A.P.I.D®
Rl-test
RPA
rRNA
RT-PCR-ESI-MS
RV-PCR
Degrees Celsius
Micrometer
Antibody immuno columns for analytical process
Amplified fragment length polymorphism
Bronchoalveolar lavage
Centers for Disease Control and Prevention
Colony forming units
Cysteine heart agar with blood
CHAB agar supplemented with colistin, amphotericin, lincomycin,
trimethoprim, and ampicillin
Enzyme-linked immunosorbent assay
U.S. Environmental Protection Agency
Electrospray ionization/time of flight mass spectrometry
Femtogram
Gravitational force
Grams
Genomic equivalents
Kilo Daltons
Liter
Limit of detection
Lipopolysaccharide
Live vaccine strain
Minutes
Milliliter
multiplex PCR enzyme hybridization assay
Nanogram
Phosphate buffer solution
Polymerase chain reaction
Pulsed-field gel electrophoresis
Picograms
Quantitative PCR
Ruggedized Advanced Pathogen Identification Device
Rapid immunochromatographic-test
Recombinase polymerase amplification
Ribosomal RNA
Reverse transcription-PCR coupled to electrospray ionization mass
spectrometry
Rapid viability-PCR
Vlll
-------
S Svedberg units
SETS Swab extraction tube system
TRF Time-resolved fluorescence
-------
Acknowledgements
The following individuals and organizations are acknowledged for their contributions to this
report:
U.S. Environmental Protection Agency
Office of Research and Development
National Homeland Security Research Center
Erin Silvestri
Frank W. Schaefer, III
Eugene Rice
Battelle, Contractor for the U.S. Environmental Protection Agency
-------
1 Introduction
Francisella tularensis is the etiological agent of tularemia, "rabbit fever," in humans and
numerous wild animals. Humans can acquire the disease by handling infected carcasses,
ingesting contaminated food or water, being infected by an infected arthropod bite, or inhaling
infectious soil dust or aerosols [1]. F. tularensis subspecies tularensis (type A) is virulent and
highly infectious, with an infective dose as little as 10-50 organisms for humans [2]. Therefore,
due to its ease of transmission, potential for substantial morbidity and mortality to large numbers
of people, and its capability to induce widespread panic, F. tularensis is listed as a Category A
select agent by the U.S. Centers for Disease Control and Prevention (CDC) [2]. Identifying F.
tularensis within a soil matrix and in surface, ground, and drinking water is a priority for
protecting both human and animal lives. There have been multiple studies dealing with clinical
samples. Due to the fastidious nature of the organism and the complexity of environmental
isolation there has been little work on identifying the organism from within soil samples. This
report is a compilation of soil and water sampling and processing information for microbial
detection acquired from research conducted within the last two decades, and describes research
gaps within the available literature.
Soil, in particular, is a complex matrix characterized by distinguishable layers, some of which
are capable of supporting rooted plants [3]. The overall properties of a soil fluctuate with time
due to changing weather patterns and plant growth cycles. For this reason, pH, soluble salts,
organic mass, flora, fauna, temperature, moisture, and the number and types of microorganisms
all change with the seasons and over extended periods of time [3]. Some naturally occurring
organisms can be pathogenic to animals and humans. Appropriate sampling methods for soil are
thus needed to help determine where, how, and to what extent soils might have been
contaminated following a tularemia event.
Water supplies are at risk of biological contamination through either natural or illicit means. A
tainted water source creates a significant disruption to society [4]. Drinking water might be
contaminated at the original source, during treatment, within distribution plumbing, or in
distribution containers [5]. Analytical methods for early detection of waterborne pathogens
within a variety of aqueous matrices (surface, ground, or drinking waters) are needed to help
maintain water security.
Identifying pathogenic organisms within an environmental soil or water sample can be a
challenging task. Direct culture of some bacteria can be difficult due to particular growth
requirements, extensive growing times, and potential risk to laboratory workers when an
organism is highly virulent. Identification can be impeded by chemical constituents within the
soil or water that can interfere with the chemistry involved in downstream molecular detection
methods [6-8]. An understanding of the environmental distribution of bacterial pathogens and
their fate over time in nature is needed for multiple applications, including determining risk to
-------
wildlife, livestock, and humans in a given area, and distinguishing between natural and
anthropogenic sources during an epidemic. However, due to the number of organisms and
impeding chemical constituents within soil and water, identifying a single virulent species within
an environmental sample can be a difficult task.
1.1 Characteristics of F. tularensis
F. tularensis is a gram-negative intracellular pathogen. It was first isolated from diseased
squirrels in Tulare county, California in 1912 [9], but was not officially named Francisella
tularensis until 1947 [10, 11]. There are now three commonly recognized subspecies: F.
tularensis subspecies tularensis (type A), F. tularensis subspecies holarctica (type B; previously
known as F. tularensis subspeciespalaearctica [12]), andF. tularensis subspecies mediasiatica.
F. tularensis type A and F. tularensis type B cause a majority of human tularemia infections
[13], while F. tularensis subspecies mediasiatica has only been isolated in Central Asia and
exhibits virulence in rabbits similar to type B organisms [10]. Each subspecies differs in
pathogenicity, prevalence, and geographic distribution [14]. F. novicida is a closely related
species that is sometimes considered a subspecies of F. tularensis, but is only very rarely
associated with human infections [9]. F. tularensis subspecies holarctica., F. novicida, and F.
philomiragia are all associated with environmental waters [15].
Tularemia within the United States is most commonly associated with hunting activities or tick
bites [16]. Tularemia incidents associated with the consumption of hunted animals typically
occur in the summer/ early autumn months, while waterborne tularemia often occurs during the
rainy season when swollen streams might extend onto contaminated animal carcasses in the
surrounding area [17]. F. tularensis type A is primarily found in North America, however
recently it was observed in Europe for the first time [13]. F. tularensis type A can be split further
into two distinct phylogenetic groups, Al and A2, based upon their geographic distribution and
primary vector species [10]. Type Al is found within the eastern United States and California
correlating to the tick vectors Dermacentor variabilis and Amblyomma americanum, with the
eastern cottontail rabbit as common tularemia host [10]. Type A2 is found at a significantly
higher elevation within the Rocky Mountain regions of western United States matching the
tularemia host mountain cottontail rabbit and the vectors D. andersoni (Rocky Mountain wood
tick) and Chrysops discalis (deer fly). Type Al can be further separated into two distinct clades,
Ala and Alb, based upon phylogenic analysis [18]. F. tularensis type Ala and Alb are found
primarily within the eastern United States, while type A2 strains are only found within the
western United States [18]. F. tularensis type Alb exhibit the highest mortality rate for human
mortality of all F. tularensis strains [18].
Tularemia manifests in a number of ways in humans depending upon the initial portal of
infection [19]. Ulceroglandular and glandular forms occur after handling contaminated carcasses
-------
or from an infected arthropod bite [20]. Oropharyngeal forms occur after ingesting contaminated
food or water. Oculoglandular and pneumonic forms occur after direct contamination of the eye
and inhalation of F. tularensis, respectively [20]. Grunow and Finke [21] established a list of
criteria for assessing tularemia events to determine if they began from natural or illicit
mechanisms. Two of the criteria included within the assessment are the natural geographic
distribution ofF. tularemis and the strain ofF. tularensis within the affected area [21].
1.2 Persistence ofF. tularensis in the environment
F. tularensis is widely distributed in the environment and has been isolated from nearly 250
wildlife species, ranging from mammals, invertebrates, birds, amphibians, and fish [10, 22, 23].
Whereas F. tularensis subspecies tularensis (type A) is found within wild mammals or blood-
feeding ticks and deerflies, F. tularensis subspecies holarctica (type B) is primarily found within
environmental surface waters [24].
F. tularensis is a hardy organism within the environment and can survive for weeks and
potentially years at low temperatures in water, moist soil, hay, straw, or decaying animal
carcasses [20, 25, 26]. Goethert and Telford [27] conducted a systematic analysis of dog ticks on
the island of Martha's Vineyard during a sustained tularemia outbreak. Their results point toward
dog ticks as a sustaining microfoci for F. tularensis for a minimum of four years. Davis-Hoover
et al. [28] spiked F. tularensis into microcosms filled with sterilized municipal solid waste
leachate. Replicate microcosms were stored either at 12 degrees Celsius (°C) or 37°C and
cultured at specified intervals [28]. Results show that F. tularensis was culturable for up to six
weeks within the microcosms, but were not culturable past six weeks at either incubation
temperature [28].
While F. tularensis is known to persist in the environment, and has been found in soils and
aerosols collected across the continental United States [29], the organism is extremely fastidious
within the laboratory setting. One study showed that F. tularensis live vaccine strain (LVS) and
F. tularensis NY98 were culturable in spiked tap water held at 8°C for 21 days [5]. However, if
the temperature was decreased to 5°C or increased to 25°C neither strain was culturable after 24-
hours. This exemplifies the specific nutrient conditions and high inoculum rates that are required
in addition to the strict safety precautions needed to prevent laboratory-acquired infections [29,
30].
The natural lifecycle ofF. tularensis within the environment is not fully understood. Many have
hypothesized that protozoa have a significant role in the Francisella sp. lifecycle, but the actual
activities have yet to be discerned [22, 31-33]. A parasitic interaction between F. tularensis and
Tetrahymenapyriformis, a ciliate protozoa commonly found within fresh water, was first
described by Kantardjiev and Velinov [34]. Their work indicates that F. tularensis can infect T.
pyriformis, replicate, and remain viable within the protozoan host for over 30 days; thereby
-------
providing a mode of transport for the bacterium within the environment [34]. More recently, Abd
et al. [31] found that F. tularemis LVS cultured in the presence of the soil amoeba
Acanthamoeba castellanii increased in concentration compared to F. tularemis LVS cultured
alone. F. tularensis bacteria go through multiple stages of infection within A. castellanii.
Bacterial growth was observed within intracellular vacuoles, released vesicles, and within
amoeba cysts [31]. It has been hypothesized that F. tularensis utilizes carbon dioxide produced
by live amoeba and the nutrients released from deceased amoeba to create an ideal setting for
proliferation over an extended period [31]. Work by Svensson et al. [35] corroborated these
findings by showing that identical F. tularensis genotypes overwinter at disease cluster sites;
this, in combination with the ability of amoebae to form cysts during periods of famine adds to a
potential F. tularensis-amooba relationship that helps F. tularensis survive long-term in the
environment [33, 35].
Another hypothesis is that long-term F. tularensis persistence is due to its survival within
biofilms [15, 36]. Biofilms are naturally formed communities of organisms held within an
extracellular polymeric matrix. Biofilm communities reduce the influence of shear stress from
flowing waters and increase nutrient capture, while simultaneously protecting the inner bacteria
from antibiotics and disinfecting chemicals [15]. A number of Francisella spp. have shown
biofilm formation capabilities, including F. novicida, F. tularensis subspecies holarctica LVS, F.
tularensis subspecies tularensis SchuS4, and F. philomiragia [15, 36]. While biofilm formation
has been noted in the laboratory and from within environmental samples, the precise role that
biofilm has in persistence is still uncertain. One hypothesis is that F. tularensis survive within
and among amoeba in biofilms [15]. Another hypothesis is that mosquito larvae aid in long-term
persistence of F. tularensis within the environment. Mahajan et al. [37] found that mosquito
larvae can ingest planktonic F. tularensis or F. tularensis within biofilms. Once inside, the larvae
could provide protection, nutrients, transportation, and a source of disease transmission for the F.
tularensis [37]. Thus, biofilm formation within environmental waters and moist soils might be an
additional mechanism by which F. tularensis could persist long-term in the environment [36].
1.3 Purpose
The purpose of this review was to survey the open literature on processing and analytical
methods currently available for detection of F. tularensis in soil and water (drinking, ground, and
surface), and to determine gaps in the collective knowledge concerning F. tularensis
identification from environmental samples. The information presented here could be used to
inform future development of standardizing methods used in detecting pathogens in
environmental matrices.
-------
1.4 Methods
Information about F. tularemis for this literature review was considered from unclassified
reports, peer-reviewed journal articles, published books, and government publications focusing
on the last twenty years. Published books were limited to the last ten years. The primary search
engines used were Science Direct and PubMed with Google Scholar and the Homeland Defense
and Security Information Analysis Center (U.S. Department of the Air Force) used secondarily.
Search terms included the agent name plus one or more of the following key words: soil, water,
environmental, methods, detection, extraction, recovery, and processing. The search was limited
to articles published in the English language, but there was no restriction on geographic focus.
This report was generated using references (secondary data) that could not be evaluated for
accuracy, precision, representativeness, completeness, or comparability and; therefore, no
assurance can be made that the data extracted from these publications meet the stringent quality
assurance requirements of the U.S. Environmental Protection Agency (EPA). However, the
sources of secondary data were limited to peer-reviewed documents wherever possible. In the
event that a pertinent study was found that had not been subject to review by fellow researchers,
the scientific and technical information from these non-peer reviewed sources were evaluated, as
outlined in the EPA General Assessment Factors for Evaluating the Quality of Scientific and
Technical Information (EPA/100/B-03/001) using the assessment factors: focus, verity, integrity,
rigor, soundness, applicability and utility, clarity and completeness, uncertainty and variability,
and evaluation and review.
2 Current State of the Science
Overall, there is not a great depth of knowledge regarding methods for F. tularensis
identification within a soil matrix. An initial PubMed search for "Francisella tularensis,"
"English language" and "soil" returns 11 references; expanding the search to "water" yielded
112 references. While this review did not limit the findings to these articles, it is an indication of
the limited breath of knowledge regarding F. tularensis in soil and water samples.
A review of these articles and others pointed to three broad mechanisms ofF. tularensis
detection within environmental samples: culture analysis, immunoassay, and genomic
identification. While some sampling methods targeted only F. tularensis spp., other methods
target multiple Category A and B agents within a single sample including, Bacillus anthracis,
Yersinia pestis, Brucella melitensis, Burkholderia mallei, and/or Burkholderia pseudomallei.
-------
2.1 Sample Processing
Environmental samples are often pre-processed before detection techniques are implemented to
eliminate inhibiting constituents. While some procedures simply suspend a soil aliquot in
buffered solution [38], others utilize filtration [16, 17, 39, 40], centrifugation [16, 17, 24, 28, 40-
42], or ultrafiltration [4, 43] to process environmental water and/or soil samples. Johansson et al.
[44] noted that the sampling method and transport medium (conditions) had a role in both the
culturability and genomic analysis of the samples.
Meric et al. [17] found a 0.45|im cellulose acetate filter to be more efficient for concentrating F.
tularensis from 1 liter (L) of reservoir water than centrifugation. The filters were washed with
sterile deionized water prior to deoxyribonucleic acid (DNA) extraction and real-time
polymerase chain reaction (PCR) analysis. In their analysis, only filtered water samples were
PCR positive, whereas centrifuge concentrated water samples were not PCR positive [17]. Sellek
et al. [39] developed a filtration method for processing soil samples that allowed for both
genomic analysis and immunologic analysis of the extracted sample. Their study assessed the
efficiency of two filters to capture F. tularensis and eliminate inhibiting constituents [39].
Briefly, 0.5 gram (g) of soil (sandy loam, silt loam, or clay) spiked withF. tularensis were mixed
with 1.5 milliliter (mL) of phosphate buffer solution (PBS) and sufficiently mixed. The
suspension was then collected into a sterile syringe and filtered through either an 8-micrometer
(|im) pore size glass fiber pre-filter or a 5-|im pore size polyvinylidene fluoride membrane filter.
F. tularensis cells within the flow-through were directly used for immunological analysis or the
flow-through was concentrated by centrifugation and heat lysed prior to PCR analysis. While
Sellek et al. [39] were able to show proof of concept for processing F. tularensis soils simply
through filtration, the results were not efficient. The glass fiber filters only recovered 6%-10% of
the F. tularensis cells while the polyvinylidene fluoride filters recovered approximately 20% of
the spiked bacteria [39]. Therefore, until more efficient filtration procedures are developed, other
soil sampling methods would appear to be more suitable for processing environmental soil
samples with potentially low F. tularensis concentrations.
Ultrafiltration techniques offer a more efficient method for concentrating contaminated water
samples. A study by the EPA [43] found ultrafiltration to be an effective sampling technique for
simultaneous recovery of diverse microbes from environmental waters. F. tularensis was the
most challenging microbe to recover during the experiments, yet the average recovery
efficiencies ranged from 13 to 62% depending upon the laboratory protocol used and the use of
ammonium chloride to treat ultrafiltration concentrates prior to culture. Francy et al. [4]
demonstrated the utility of ultrafiltration to concentrate 100 L samples of raw ground water or of
finished surface and ground waters into a 225 mL retentate for subsequent biological assessment.
F. tularensis was detectable in each of the 14 spiked water samples [4].
Swab sampling is a common interior surface sampling method employed during bioterrorism
investigations that yields solid particulates similar to some soils. Walker et al. [45] sought an
-------
optimized swab processing method for maximum recovery of F. tularemis cells, followed by
automated DNA extraction and real-time PCR detection. Four processing methods - heat,
sonication, vortexing, and the swab extraction tube system (SETS) - were tested against three
commonly used sampling swab materials: polyester, rayon, and foam. SETS is a disposable
centrifugal system composed of an inner and outer collection tube. The inner tube contains an
orifice to assist in separating collected bacteria from the swab tip. Rehydrated swab tips are
aseptically placed within the inner tube of the SETS system. After a brief centrifugation, the
rehydration fluids, along with recovered microorganisms, are collected in the outer SETS tube.
The sample suspension can then be cultured or processed further for genomic identification. A
careful statistical analysis determined that SETS was more efficient at recovering the spiked F.
tularensis cells from the various swab materials [45]. However, it must be noted that this work
utilized pure cultures ofF. tularensis at high concentrations (103 - 105 colony forming units
(CFU)/swab), and the correlation to field-collected environmental samples is still unknown [45].
2.2 Culturing F. tularensis from the Environment
While culturing an organism is considered the gold standard for identification, isolating
environmental cultures of F. tularensis is challenging as it is a slow-growing, nutritionally
fastidious organism that requires 24 to 72 hours for growth [46] on medium supplemented with
bio-available iron, cysteine, and up to 12 other nutrients [15]. Even with selective agars, F.
tularensis colonies are often out-competed by background organisms present in environmental
samples [46, 47]. In a study where tap water samples were spiked with F. tularensis and held at
various temperatures, the F. tularensis was not recovered after 24 hours when held at 5°C or
25°C, but was culturable for 21 days when held at 8°C [5]. Yet, when landfill leachates were
spiked with F. tularensis cultures the organism could be cultured for six weeks when held at
12°C or 37°C [28]. Thus, temperature seems to have a profound effect for some matrices, but not
all. Furthermore, due to the high risk of laboratory acquired infections, all F. tularensis culturing
must be conducted under biosafety level 3 conditions [2, 46]. Yet, F. tularensis culturing remains
a primary mechanism for confirming the presence of viable biothreat agents. Table 1 summarizes
the methods and findings of environmental culturing studies included herein.
Antibiotic-supplemented cysteine heart agar with blood (CHAB) has been frequently used to
culture F. tularensis from environmental samples [4, 16, 17, 24, 40-42, 46-49]. CHAB medium
has been used to attempt to locate the environmental origin of Francisella strains isolated within
the clinical setting. Two Utah hot springs were suspected as the original route of transmission to
a patient; therefore, Whitehouse et al. [24] collected soil, water, vegetation, sediment, and pond
scum samples from two suspect springs. Aliquots of the collected samples were centrifuged
before culturing on CHAB agar. Suspected Francisella spp. colonies were picked and further
processed to determine phytogeny and biochemical analysis [24]. While the study was unable to
-------
discern the origins of the human isolated strain, the authors were able to characterize multiple
presumptive Francisella isolates and identify them as either F. philomiragia or F. novicida [24].
CHAB agar has been modified with various antibiotics to aid environmental F. tularemis
isolation. Petersen et al. [49] developed a modified CHAB agar supplemented with colistin,
amphotericin, lincomycin, trimethoprim, and ampicillin (CHAB-A) to inhibit background
organisms when culturing environmental tissue samples in the field. CHAB has been further
modified for isolating Francisella spp. from environmental water and seaweed samples [42].
Utilizing CHAB containing polymyxin B, amphotericin B, cyclohexamide, cefepime and
vancomycin , Petersen et al. [42] were able to isolate three new Francisella spp. from seawater
and seaweed collected near Houston, Texas [42]. Their findings were significant, as BioWatch (a
federal bio-agent release detection technology program) filters stationed nearby had detected the
presence of Francisella spp. in the past [42, 50].
Following an outbreak of pneumonic tularemia on Martha's Vineyard, Massachusetts in 2000,
significant work was conducted to determine the natural foci for F. tularensis type A on the
island [16]. Water and soil samples collected from the island were initially screened for F.
tularensis by PCR detection of thefopA gene [46]. Samples PCR positive forfopA were cultured
on CHAB-A agar to investigate the culturability of the organism. F. philomiragia was cultured
from only one of fivefopA-positive water samples [16]. The isolate came from a brackish-water
sample and led Berrada and Telford [48] to hypothesize that brackish-water is a more suitable
environment for the persistence of F. tularensis Type A than freshwater. By culturing fresh and
brackish-water microcosms spiked with F. tularensis Type A, another study confirmed that
brackish-water is a superior environment for F. tularensis Type A persistence [48].
Other studies have attempted to culture F. tularensis from environmental waters with limited
success. §im§ek et al. [40] cultured multiple surface water samples on CHAB agar amended with
antibiotics in an effort to identify the source of a tularemia outbreak in Turkey. Of the 154 water
samples collected, four were culture positive for F. tularensis while 17 were PCR positive. Meric
et al. [17] attempted to identify F. tularensis by both PCR and culture techniques using filter
concentrated reservoir water samples; only PCR yielded positive results. Anda et al. [41] sought
the environmental source of a tularemia outbreak associated with crayfish fishing in a
contaminated freshwater stream in Spain. F. tularensis subspecies holarctica was identified as
the responsible agent; however, identification was accomplished through PCR analysis and DNA
sequencing and not culture, as no F. tularensis were isolated on the modified Thayer-Martin
chocolate agar utilized within the study. For each of these studies a maximum of 1.5 L of water
was concentrated before culture analysis.
Ultrafiltration has been used to concentrate large volumes of water prior to biological
assessment. Using ultrafiltration techniques Francy et al. [4] concentrated 100 L samples of 14
spiked waters. They observed that F. tularensis was an extremely fastidious organism with a
maximum culture recovery rate of 40% (minimum 0.2% recovery) on CHAB agar with
-------
antibiotics. A study comparing two similar ultrafiltration techniques found F. tularemis to be the
most challenging organism to recover of those tested, which included viruses, bacteria, and
protozoa [43]. However, when ultrafiltration filtrates were exposed to 1% ammonium chloride
for two-hours prior to culturing on antibiotics amended CHAB the recovery rates dramatically
improved [43]. Humrighouse et al. [47] saw a similar effect when seeded water samples were
acid treated for 15 minutes before culture on antibiotic amended CHAB. Acid treatment reduced
the indigenous background organisms present in the environmental water samples, allowing
better F. tularensis recovery. Anda et al. [41] also demonstrated the use of acid shock to enhance
F. tularensis recovery on modified Thayer-Martin chocolate agar.
Johansson et al. [44] concluded that the successful culture of wound specimens (and therefore
other environmentally collected samples) was dependent upon the transport medium and
sampling techniques employed during collection. Consistent growth curves of F. tularensis
subspecies holarctica and tularensis Schu S4 have also been noted as difficult to achieve for
verification purposes in the laboratory setting [30]. Therefore, to circumvent the laboratory
challenges of directly culturing environmental isolates of F. tularensis, other methods of
identification, including immunoassays and genomic methods, have been developed to target F.
tularensis.
-------
Table 1. Comparison of Francisella tularensis Culturing Studies
Reference
Sample Matrix and Tested
Organism
Sample Preparation Method
Culture Media
Summary
Anda et al., River and sewage water naturally
2001 [41] contaminated with Francisella
tularensis subspecies holarctica
Berrada and Surface soil, sand, sediment, water
Telford, 2010 naturally contaminated with F.
[16] philomiragia
Berrada and
Telford, 2011
[48]
Microcosms of fresh or brackish
water spiked with F. tularensis Type
A, Type B live vaccine strain (LVS),
or_F. novicida
Davis-Hoover Landfill leachate microcosms spiked
et al., 2006 with pure cultures ofF. tularensis,
[28] Bacillus anthracis, Clostridium
botulinum, or Yersinia pestis
EPA, 2011 Drinking water spiked with F.
[43] tularensis LVS, Bacillus anthracis, B.
atrophaeus, Yersinia pestis,
Brevundimonas diminuta,
Clostridium perfringens,
EPA, 2012 Pure cultures ofF. tularensis Schu4
[30] and LVS
Francy et al., Raw water and drinking water spiked
2009 [4] withF. tularensis LVS, B. anthracis
Sterne, Salmonella typhi, Vibrio
cholerae, Cryptosporidium parvum
Gilbert and Autoclaved tap water spiked with F.
Rose, 2012 [5] tularensis LVS and NY98, Y. pestis,
Burkholderia pseudomallei, Brucella
melitensis, Bacillus suis
10 mL water samples were
concentrated by centrifugation, the
pellet was resuspended in 1 mL of
the original sample water and subject
to an acid shock to reduce
contaminants before plating
Large particles were removed from
100-300 mL samples by
centrifugation before filtering
through a 0.22 um cellulose nitrate
filter. Filters were washed and
resulting particles were collected for
culture or DNA extraction
10 uL from each microcosm was
directly cultured
5 mL microcosm samples were
centrifuged, and resuspended in
phosphate buffered saline (PBS)
before dilution and triplicate plating.
100L of drinking water was spiked
before concentration by
ultrafiltration. Ultrafiltration
concentrates were assayed
immediately for F. tularensis by
membrane filter plates.
None
100 L samples were concentrated to
225 mL by ultrafiltration before
culture.
1 mL aliquots of spiked water were
diluted in Butterfield's Buffer and
plated at each time point.
Modified Thayer-Martin
chocolate agar, supplemented
with Iso Vitalex™ and 1% L-
cysteine
Cysteine heart agar
supplemented with 9% sheep
blood (CHAB) and
antibiotics colistin,
lincomycin, trimethoprim,
and ampicillin
No F. tularensis isolates were detected in the
sewage or river water samples. One sewage
water sample was polymerase chain reaction
(PCR) positive for F. tularensis.
Only environmental samples that were PCR
positive forfopA were cultured. From these,
F. philomiragia was isolated from a single
brackish-water sample.
Cysteine heart agar
supplemented with 8% rabbit
blood and antibiotics as
supplied by Remel (Lenexa,
Kansas)
Cysteine heart agar,
chocolate agar, Thayer-
Martin agar, Buffered
Charcoal yeast extract
Cysteine heart agar with
chocolatized 9% sheep blood
and antibiotics colistin,
amphotericin, lincomycin,
trimethoprim, and ampicillin
Trypticase soy broth with
Isovitalex
Cysteine heart agar
supplemented with
hemoglobin, penicillin, and
polymyxin B and blood.
Chocolate II agar
F. tularensis Types A and B persist in
brackish water longer than freshwater.
F. tularensis was not viable after 7 weeks in
the landfill leachate held at 12°C or 37°C.
It was found that average recovery
efficiencies for F. tularensis were higher
when water sample ultrafiltration
concentrates were exposed to 1% ammonium
chloride for 2 hour prior to culturing.
Study attempted to determine recovery
methods from swabs, however F. tularensis
could not be reliably grown within broth
cultures and was thus cut from the study.
F. tularensis recoveries by culture were the
lowest of the six tested organisms. Whereas
F. tularensis was detectable within all 14
ultrafiltration water samples by quantitative
(or real time) PCR (qPCR).
F. tularensis LVS survived 8 days and NY98
survived 28 days in natural waters at 8°C.
10
-------
Reference
Sample Matrix and Tested
Organism
Sample Preparation Method
Culture Media
Summary
Humrighouse
et al., 2011
[47]
Creek water spiked with F. tularensis
subspecies holarctica and tularensis
Johansson et
al., 2000 [44]
Meric et al.,
2010 [17]
Clinical isolates collected from
infected wounds and F. tularensis
LVS
Reservoirs of spring water naturally
contaminated with F. tularensis
Petersen et Environmental tissues (prairie dog
al., 2004 [49] spleen and liver) naturally
contaminated with F. tularensis
subspecies holarctica
Petersen et Naturally contaminated seawater and
al., 2009 [42] seaweed from Houston, TX. F.
tularensis, F. novicida
and F. philomiragia
Simsek et al.,
2012 [40]
Versage et al.,
2003 [46]
Environmental water naturally
contaminated with F. tularensis
subspecies holarctica strain LVS
55 wild-type F. tularensis isolates
collected from naturally infected
tissues and laboratory infected
animals including F. tularensis
subspecies tularensis and holarctica,
F. novicida, and F. philomiragia
Spiked waters were subject to a 15-
min acid treatment (potassium
chloride- hydrogen chloride) before
neutralization (potassium hydroxide)
and streak plating on selective agar.
Wounds were directly cultured using
rayon-tipped applicators.
1 L water samples were filtered with
0.45 um cellulose acetate filters.
Filters were washed with distilled
water before filtrate was cultured
and DNA extracted.
Necropsied tissue samples were
plated and sealed on-site
100 uL was directly plated and an
additional 10 mL was centrifuged.
The resulting pellet was resuspended
in PBS and plated. Seaweed was
homogenized in 500 uL of PBS
before plating. DNA from selected
colonies were boil-lysed before
PCR.
0.3 -1.5 L water samples were
filtered through cellulose acetate
membranes (pore size 22 um). The
membranes were placed directly on
plates.
Tissue samples were directly
cultured while DNA from mouse and
prairie dog tissues were extracted.
Cysteine heart agar with
rabbit blood and antibiotics
as supplied by Remel
(Lenexa, Kansas)
Thayer-Martin agar
Glucose cysteine heart agar
with 2.5% blood
Cysteine heart agar with
chocolatized 9% sheep blood
supplemented with colistin,
amphotericin, rincomycin,
trimethoprim, and ampicillin
Cysteine heart agar with 9%
chocolatized sheep blood
supplemented with
antimicrobials
duced
;rved
ed.
The combination of acid treatment and
selective agar allowed the recovery ofF.
tularensis from water and effectively reduced
indigenous background organisms.
Differences in acid resistance were observed
among the 7 F. tularensis strains assessed
Study compared PCR to culture and found
PCR to more sensitive; however, sampling
methods can cause PCR difficulties.
Differentiating between a tularemic wound
and a non-infected wound can be difficult.
No cultures were recovered. PCR was
attempted following filter concentration,
however no samples were PCR positive.
Filtration is a better concentration method
than centrifugation. Sera, throat swabs,
lymph node aspirates, filter concentrated
reservoir waters were all culture negative.
Antibiotic supplementation of CHAD media
controlled the growth of contaminating
bacteria and significantly improved the
ability to recover F. tularensis and culture
sensitivity.
F. tularensis can be directly cultured from
environmental seawater and seaweed
samples. Presumptive F. tularensis colonies
were confirmed through PCR analysis.
Antibiotic (Oxoid SR147)-
added cysteine heart agar
base with blood
Cysteine heart agar with 9%
chocolatized blood
Real-time PCR was more sensitive as 17 of
154 samples were PCR positive and only 4
were culture positive. 16S rRNA sequencing
identified the cultured strains as F. tularensis
subspecies holarctica strain LVS.
Comparison of TaqMan PCR assays to
culturing determined that PCR was
significantly more sensitive than culturing.
11
-------
T> f Sample Matrix and Tested „ , _ ,.„ ,, , „ ,, ,.„ ,. „
Reference „ . Sample Preparation Method Culture Media Summary
Whitehouse et Water, vegetation, soil, sediment, and 50 mL samples were centrifuged at Cysteine heart blood Samples were centrifuged and the pellet was
al., 2012 [24] pond scum naturally contaminated 8,000 gravitational force (g), chocolate agar cultured. DNA from the isolates were further
with F. philomiragia and F. novicida supernatant was decanted and processed to identify F. philomiragia and F.
resuspended in sterile saline before novicida within the Utah natural warm
culturing. DNA from presumptive springs.
isolates was extracted.
12
-------
2.3 Immunoassay Detection of F. tularensis
Testing for F. tularensis antigens within environmental samples has been used as a means of
infection source tracking for some time due to the ability to incorporate immunoassays into
hand-held field-deployable systems. A summary of the immunoassay studies found within this
review are presented in Table 2. Care must be taken when developing assay antigens, as some
can have cross-reactivity to other microorganisms [51].
Berdal et al. [52] developed a rapid immunochromatographic-test (Rl-test) where upon direct
addition of environmental waters the presence of F. tularensis lipopolysaccharide (LPS) antigen
is indicated by a red line within the test window. A comparison of the Rl-test to enzyme-linked
immunosorbent assay (ELISA) and PCR analyses demonstrated that PCR performs best with
environmental water samples, while the Rl-test and ELISA were better suited for detecting F.
tularensis within tissue samples. However, no specifics were given regarding limit of detection
(LOD) for any of the three tested methods in the study [52].
Peruski et al. [53] demonstrated the effectiveness of time-resolved fluorescence (TRF), a
technology based on lanthanide chelate labels with unique fluorescence properties, within
various matrices including soil, serum, urine, and sewage water. The authors determined that
TRF improved assay sensitivity by 2000-fold when compared to standard capture ELISA.
Sewage water and urine did not impact the overall sensitivity, but soil and serum decreased the
capture efficiency. An overall lower LOD of the TRF assay was determined to be approximately
48 CFU/mL [53]. The authors concluded that the TRF assay is more sensitive, has a wider
dynamic range than standard ELISA, and could prove to be an invaluable tool for detecting low
levels of F. tularensis within environmental samples.
Grunow et al. [54] validated antibody immuno columns for analytical process (ABICAP) which
are an immunoaffinity chromatographic column test that includes ELISA detection chemistry
within a hand-held single use device. The ABICAP system uses small disposable plastic columns
within which all assay components are added in flow-through, and thus could allow larger
sample volumes to increase sensitivity. Bacterial LPS was directly extracted from environmental
waters (125 jiL) collected from a Swedish reservoir during a tularemia outbreak and from rabbit
and mouse fecal matter. The water samples contained various amounts of dissolved soil. The
concentration of mud within the initial water samples directly correlated with increasing
background signal during analysis and false positive test results [54]. Yet, the system offers a
field-deployable assay with a LOD comparable to capture ELISA [54]. Capture ELISA for
spiked silt loam samples processed through glass fiber filters have shown a LOD of 104 CFU/mL
[39].
Even with highly specific ELISA techniques, testing environmental samples can be challenging.
Capture ELISA tests were utilized to track F. tularensis contamination within water and fecal
samples in Kosovo during two tularemia outbreaks [55]. Unsanitary conditions were targeted as
the source of infection in Kosovo as rodent feces collected from food storage areas were found to
13
-------
contain F. tularemis LPS. Water sources were also tested as a probable contamination source,
however no water samples yielded positive capture ELISA results [55]. During a biodefense
training exercise Offermans and Zegers [38] determined that a 20% suspension of soil in PBS
was not suitable for their sandwich ELISA assay using monoclonal antibodies as neither the soil
samples nor the control positive sample yielded positive results.
New technologies are being developed that incorporate immunoassay detection chemistry.
Huelseweh et al. [56] have developed a protein chip for rapid detection of multiple biowarfare
agents. Their method was capable of simultaneously detecting two to five bioagents at similar
limits of detection as ELISA, but in less time. However, the overall quality of the immunoarray
is still dependent upon the individual affinities to the antibodies. Cooper et al. [2] recently
published details of their prototype biosensor for label-free, specific antibody and single-stranded
oligonucleotide detection of F. tularensis. Pohanka and Skladal [57] have developed a
piezoelectric immunosensor for direct detection of F. tularensis. While the detection limit is still
high, their method has been tested on drinking water and milk samples (LOD of 105 CFU/mL for
both). The utility of bidiffractive grating biosensor has been explored as a field deployable
biosensor for automated biodefense systems [58]. Sharma et al. [59] have developed a novel
competitive ELISA for clinical identification of F. tularensis, but there is potential that the
method could be useful for environmental samples in the future.
14
-------
Table 2. Comparison of Francisella tularensis Immunoassay Studies
Reference
Sample Matrix and Tested
Organism
Detection method (Sample
Amount)
Antibody
Summary
Berdal et al.,
2000 [52]
Environmental water
naturally containing F.
tularensis
Cooper et al., Pure cultures of F.
2011 [2]
Grunow et
al., 2008 [54]
Grunow et
al., 2012 [55]
Huelseweh et
al., 2006 [56]
O'Brien et
al., 2000 [58]
Offermans
and Zegers,
2007 [38]
Peruski et
al., 2002 [53]
tularensis subspecies
tularensis and holarcticc
Environmental water and
feces collected during a
tularemia outbreak or spiked
with F. tularensis live
vaccine strain (LVS).
Environmental water and
feces collected during a
tularemia outbreak
Pure cultures ofF.
tularensis WIS 140,
Yersinia pestis,
Burkholderia pseudomallei,
B. mallei, Brucella
melitensis, Escherichia coli
Pure cultures ofF.
tularensis LVS
Soils spiked with unknown
quantities ofF. tularensis,
B. anthracis, Brucella
pseudomallei or Vaccinia
Serum, urine, dirt, sewage
water spiked with F.
tularensis LVS
Rapid immunochromatographic-
test (RI- test) (200 uL)
Enzyme-linked immunosorbent
assay (ELISA) (50 uL)
Antibody and DNA photonic
biosensors
•
Antibody immuno columns for
analytical process (ABICAP) (125
uL)
Capture ELISA
Protein chip
Bidiffractive grating biosensors
ELISA (2g)
Time-resolved fluorescence (TRF)
and ELISA (100 uL or lOmg
dirt/mL phosphate buffered saline
(PBS))
Lyophilized mouse monoclonal
IgG antibody specific for F.
tularensis lipopolysaccharide
(LPS) antigen
Mouse monoclonal F. tularensis
LPS antibody Ft-27 and antibody
Ft-11.
IgG LPS directed antibody
F. tularensis LPS monoclonal
IgGlantibodyFF/11/6
F. tularensis LPS-specific
monoclonal IgGlantibody FF/11/6
as capture antibody bound to the
solid phase
LPS-specific capture and detector
monoclonal antibody:
FT140/11/1/06
Goat capture antibody
Capture antibody: Monoclonal
antibody FF27/1/7 anti F.
tularensis; Detector antibody:
Monoclonal antibody FF1 1/1/6-
biotin anti F. tularensis
Biotinylated capture antibodies:
monoclonal antibody Ft-03 or
polyclonal antibodies to F.
tularensis and detected by TRF
Europium-labeled antibodies
Three wells were tested. One gave a weakly
positive signal for 2 of 4 collected samples. Liver,
spleen and kidney supernatants from a lemming
carcass tested strongly positive.
At least 105 colony forming units (CPU) ofF.
tularensis were detectable to present a proof-of-
principle.
The ABICAP system is useful for identifying
reservoirs ofF. tularensis from within
environmental waters and feces. Limit of detection
(LOD) 103 bacteria.
Kosovo outbreak due to poor conditions followin
war activity. No water samples yielded positive
capture ELISA results.
Microarray detection limits were comparable to
ELISA, but require less time even when detecting
multiple bioagents. LOD 106 CFU/mL.
ing
'
Proof of principle to assess if bidiffractive g
biosensors could be used for biosurveillance
monitoring. LOD 105 CFU/mL.
grating
6
No F. tularensis was detected with ELISA
techniques in either the spiked samples or the
positive control.
Capture biotinylated antibodies to F. tularensis
were used to compare ELISA to TRF. TRF was
2000 times more sensitive for F. tularensis than
standard ELISA. Additionally when tested within
sera, urine, sewage water, and dirt (10 mg/mL PBS)
the sensitivity decreased for sera and dirt, but not
for sewage water or urine.
15
-------
„ ,. Sample Matrix and Tested Detection method (Sample . ^., ,
Reference Amount) Antlbody
Pohanka and Pure cultures ofF. Piezoelectric immunosensor (500 Mouse polyclonal antibody The novel device was capable of detecting F.
Skladal, tular ensis LVS uL) tularensis within tap water and milk at a LOD of
2007 [57] 105 CFU/mL.
Sellek et al., Sandy loam, silt loam, or Capture ELISA, quantitative- Anti-F. tularensis LVS Comparisons showed that qPCR had a lower LOD
2008 [39] clay soil spiked with FMA- polymerase chain reaction (qPCR) monoclonal antibody T14 than capture ELISA (102 and 104 CFU/mL,
inactivated F. tularensis respectively)
ubspecies holarctica LVS
Sharma et Human and animal serum Competitive ELISA LPS monoclonal antibody M14B 1 Currently this method is limited to clinical
al., 2013 [59] and/7. tularensis subspecies identification of tularemia.
holarctica
/er LOD
16
-------
2.4 Genomic Identification of F. tularensis
While culture is considered the gold standard for identifying viable pathogenic microorganisms
from environmental or clinical samples, multiple reports have shown that PCR identification is
faster and more sensitive than culture or immunoassay [17, 39]. However, these assays also have
limitations. Versage et al. [46] compared the capabilities of culture versus real-time PCR to
identify F. tularensis from contaminated animal tissues. Their analysis determined that real-time
PCR assays were significantly more sensitive than culture. Anda et al. [41] and Meric et al. [17]
each attempted to identify F. tularensis within environmental water samples. Neither study was
able to recover F. tularensis isolates, but both detected F. tularensis through genomic analysis.
Work at Martha's Vineyard following a pneumonic tularemia outbreak screened water and soil
samples for F. tularensis [16]. Of the 156 samples assessed, 23 were PCR positive for F.
tularensis genes, yet only one sample yielded a F. philomiragia culture. §im§ek et al. [40] also
attempted to identify F. tularensis by culture and real-time PCR following a tularemia outbreak
in Turkey. Here again, culture was less sensitive than PCR as only four water samples were
culture positive, but 17 of the 154 samples were PCR positive for lSFtu2. This review identified
a number of studies that utilized genomic analysis to identify F. tularensis within environmental
samples. Table 3 outlines the studies discussed in sections 2.4.1 and 2.4.2, their processing
methods, and a brief summary of their conclusions.
2.4.1 Extraction of F. tularensis DNA
F. tularensis is a non-sporulating Gram-negative organism; therefore, its DNA can be extracted
for identification rather easily when compared to sporulated microorganisms. However, humic
acids and other inhibitory compounds within environmental soil and water samples are often
coextracted and lead to confounding downstream PCR responses [7]. Therefore, special care
must be taken to efficiently clean environmental DNA extracts prior to analysis. An efficient
method of DNA extraction ought to produce an unbiased yield of quality DNA suitable for
downstream analysis, meaning that a high concentration of long-DNA segments from diverse
species present within a single sample is needed [60].
Whitehouse and Hottel [61] conducted a comparison of five commercial DNA recovery kits for
isolating F. tularensis DNA from three types of soil: silt loam, clay, and commercial potting soil.
They determined that the UltraClean® Microbial DNA Isolation kit (MoBio laboratories, Inc.,
Carlsbad, CA) and the PowerMax® Soil DNA Isolation kit (MoBio laboratories, Inc.) yielded the
most consistent and lowest limits of detection (LOD) of the tested kits. The UltraClean®
Microbial DNA Isolation kit yielded an LOD of 20 colony forming units (CFU)/g soil, and the
PowerMax® Soil DNA Isolation kit yielded an LOD of 100 CFU/g soil. These limits of detection
were similar to the positive control LOD of 10 CFU/mL achieved for pure culture F. tularensis
extraction and real-time PCR [61]. Klerks et al. [62] conducted a similar study comparing five
commercial DNA recovery kits for isolating Salmonella enterica DNA, another non-sporulating
Gram-negative organism, from within various environmental matrices. They determined that the
UltraClean® Soil DNA Isolation kit (MoBio laboratories, Inc.), BiolOl extraction kit (Q-Biogene
17
-------
Inc. Carlsbad, CA), and the UltraClean® Fecal DNA Isolation kits (MoBio laboratories, Inc.),
yielded superior quality and quantity of DNA from the tested soil, manure, and compost samples
[62]. The Soilmaster™ DNA extraction kit (Epicentere, Madison, WI) and QIAGEN plant
DNeasy™ DNA (QIAGEN, Westburg, The Netherlands) extraction kit were not found to be
optimal for S. enterica from soil samples [62]. Yet, Broman et al. [22] successfully utilized the
Soilmaster™ DNA extraction kit to identify the presence ofF. tularensis in 20% of the sediment
samples and 32% of the surface water samples collected within two regions of reoccurring
tularemia outbreaks in Sweden.
Berrada and Telford [16] utilized the UltraClean® Soil DNA Isolation kit in their analysis of soil,
mud, and sediment samples collected on Martha's Vineyard. They followed the manufacturer's
protocol with one exception; in an effort to reduce DNA shearing, the bead-beating time was
reduced from ten minutes (min) to five min [16, 63]. Utilizing the modified UltraClean® Soil
DNA Isolation kit protocol, the study identified four brackish-water soil/sediment samples that
were PCR positive for specific primers (i.e., Francisella spp. 16 svedberg units [S] ribosomal
ribonucleic acid [rRNA] primers [16S rRNA]) and three samples positive for F. tularensis
specific sequences [16]. Barns et al. [50] also utilized the UltraClean® Soil DNA Isolation kit to
broadly survey the Houston, Texas area for Francisella species and relatives. Following aF.
tularensis positive sample by the BioWatch aerosol monitors in October 2003, 364 soil and
water samples were collected around the Houston area. The 16S rRNA sequencing results from
one water sample showed the presence ofF. philomiragia while the 16S rRNA sequencing
results from seven soils pointed to the presence of new subspecies ofF. tularensis with unknown
pathogenicity [50].
Defense Research and Development Canada included the UltraClean® Soil DNA Isolation kit as
part of two biothreat response readiness exercises [64, 65]. During the 2001 exercise [64], F.
tularensis was spiked into a single liquid sample. No processing was conducted for liquid
samples before PCR analysis, but DNA was extracted from the unknown soil samples with
UltraClean® Soil DNA Isolation kits prior to PCR analysis. All 13 laboratories involved in the
exercise accurately identified F. tularensis within the aqueous sample [64]. In 2002, a single soil
sample spiked with Brucella suis and a chemical nerve agent stimulant was assessed as an
unknown sample by Defense Research and Development Canada, Suffield. Through the course
of the exercise Brucella spp. was accurately identified through SYBR® Green PCR, culture, and
BIOLOG™ within the unknown soil sample [65]. Another biodefense training exercise
conducted in 2006 sought to identify the presence of biological warfare agents within supplied
unknown soil samples through either real-time PCR or immunochemical assays [38]. Fourteen
soils were screened for the presence of B. anthracis, F. tularensis, B. pseudomallei, or vaccinia
individually and as a mixture. Prior to real-time PCR analysis, soil samples were extracted with
either the UltraClean® Soil DNA isolation kit (200 |iL) or the PowerMax® DNA isolation kit (6.0
mL). The significant difference in loading size of the two extraction kits was clearly seen in the
real-time PCR analyses results. Only soil samples spiked with the high concentrations of
18
-------
biological agents were detectable within the UltraClean® Soil extracts. Therefore, it was
concluded that for samples of unknown biological agents it is preferable to extract DNA from as
much of the original sample volume as possible [38].
Trombley Hall et al. [66] recognized the need for purified nucleic acids from environmental
samples; however, rather than seeking an optimum extraction kit that removes inhibiting
constituents, they sought inhibitor-resistant PCR reagents. Use of inhibitor-resistant PCR
reagents eliminates the need for sample-specific preparation and increases the sensitivity of real-
time PCR [66]. Among the five PCR chemistries tested, KAPA Blood PCR Kit (KAPA
Biosystems, Wilmington, MA) yielded the most consistent estimated LOD results across the
range of tested matrices (buffer, whole blood, sputum, stool, soil, sand, and swab) [66]. When
looking at soil results alone, the KAPA Blood PCR Kit, Ampdirect® buffer (Rockland
Immunochemicals, Gilbertsville, PA) with Phire® Hot Start DNA Polymerase (Finnzymes/New
England Biolabs, Ipswich, MA), and STRboost™ buffers (Clontech Laboratories Inc., Mountain
View, CA) with Phire® Hot Start DNA Polymerase all yielded the same LOD, 0.2 picograms
(pg)F. tularensis DNA, when the PCR reaction was composed of 0.05% soil [66].
19
-------
Table 3. Comparison of Francisella tularensis Genomic Studies.
Reference
Sample Matrix and
Tested Organism(s)
Sample Preparation Method
Detection method
Summary
Ahlinder et
al., 2012 [67]
Anda et al.,
2001 [41]
Francisella tularensis
subspecies holarctica,
mediasiatica,
tularensis, F. novicida,
F. hispaniensis, F.
philomiragia
River and sewage
water naturally
contaminated with
Francisella tularensis
subspecies holarctica
None given
Bader et al.,
2003 [64]
Phosphate buffered
saline (PBS) spiked
with F. tularensis Schu
S4
ImL of water was centrifuged and the pellet
was resuspended in 100 uL of sample water.
After a low speed centrifugation to eliminate
solids, cells within the supernatant were
chemically lysed and DNA precipitated.
1 mL liquid was directly processed
Bader et al.,
2004 [65]
Barns et al.,
2005 [50]
Berdal et al.,
2000 [52]
Berrada and
Telford,
2010 [16]
Soil spiked with
Brucella suis and
simulant nerve agent
Houston, TX surface
soil, grab water
naturally containing F.
tularensis
Well water naturally
containing/7, tularensis
Environmental surface
soil, sand, sediment,
water naturally found
with F. philomiragia
0.25 g of soil was processed in a MoBio
UltraClean Soil DNA isolation kit
0.25 g soil directly processed while cells
within the 50 mL water samples were
pelleted by centrifugation before DNA
extraction with a MoBio UltraClean Soil
DNA isolation kit
100 ul incubated with lysis buffer before
using 2 ul directly in PCR reactions.
DNA within 0.25 to 0.5 g of sediments, mud,
or soil were directly extracted using a MoBio
UltraClean Soil DNA isolation kit. Large
particulates from 100-300 mL of water were
removed by centrifugation before filtering
through a 0.22 um cellulose nitrate filter.
Filter wash collected for culture or DNA
extraction with MoBio UltraClean Soil DNA
isolation kits.
in silico polymerase chain
reaction (PCR) analyses
were conducted for a large
dataset of primers and
Francisella genomes.
PCR targeting the F.
tularensis specific 16S
rRNA
No single marker topology of the entire genus is
currently available to classify all Francisella spp. to
their proper subspecies. This indicates that several
markers utilized for detection are unspecific resulting
in false positives. No environmental samples were
assessed.
No F. tularensis isolates were detected in the sewage
or river water samples- one sewage water sample was
PCR positive for F. tularensis.
PCR targeting tul4 gene
PCR targeting tul4 and
fopA genes
PCR detecting F.
tularensis specific 16S
rRNA, lSFtu2, tul4,fopA,
23kDa
PCR followed by
restriction analysis with
endonuclease Oral
PCR: sdhA, ISFtu2, tul4,
fopA
Dwns
'
A higher number of false positive and false negative
identifications were reported for soil sample unknowns
than for liquid sample unknowns. F. tularensis was
properly identified within a phosphate buffered
solution (PBS) by 10 of 13 reporting laboratories.
Soil sample was correctly identified to not be spiked
with F. tularensis but rather with B. suis and a G nerve
agent simulant.
DNA from soil and water samples collected from
Houston, TX showed the presence of new F. tularensis
subspecies.
Three wells were tested. One gave a positive PCR
signal in 4 of 4 collected samples. Liver, spleen and
kidney supernatants from a tested lemming carcass
were all PCR negative, but ELISA and Rl-tests were
positive.
All samples collected near the freshwater pond and th
marsh were PCR negative for F. tularensis 16S rRNA,
but only one brackish water sample was culture
positive for F. philomiragia.
20
-------
Reference
Sample Matrix and
Tested Organism(s)
Sample Preparation Method
Detection method
Summary
Broman et
al., 2011 [22]
Buzard et
al., 2012 [68]
Duncan et
al., 2013 [14]
Escudero et
al., 2008 [69]
Forsman et
al., 1995 [70]
Francy et al.,
2009 [4]
Environmental soil and
water naturally
contaminated with F.
tularensis subspecies
holarctica
Pure cultures of
F. tularensis LVS,
Bacillus anthracis
Ames, Brucella
melitensis, B. mallei
F. tularensis
subspecies tularensis,
holarctica,
mediasiatica, F.
novicida
Clinical and
environmental tissues
contaminated with F.
tularensis subspecies
tularensis, holarctica,
orF. novicida
Environmental water
spiked with F.
tularensis LVS
Raw water and
drinking water spiked
with F. tularensis LVS,
B. anthracis Sterne,
Salmonella typhi,
Vibrio cholerae,
Cryptosporidium
parvum
2.0 mL of soil or water centrifuged. Cell
pellet processed through Soil Master™ DNA
Extraction (Epicenter Biotechnologies) to
yield sample DNA.
DNA extracts from pure cultures were
obtained for this study.
Pure DNA was procured for the study.
QIAamp DNA blood extraction kit (Qiagen)
used to extract DNA from human and tick
tissue samples.
A) 1 mL filtered and freeze thaw DNA lysis;
B) 1 mL centrifuged, pellet treated with a
commercial ion exchange suspension to
purify DNA; C) Treated by alkaline method
to prep DNA; D) Chromosomal DNA from
water samples were prepared; E) 1 ml
filtered, filtered bacteria chemically lysed,
DNA purification by phenol chloroform
isoamyl alcohol, DNA filter purified by
microspin column
Ultrafiltration retentate filtered through 0.4
um polycarbonate filters. DNA from the
organisms collected on the filters was
(S)
extracted using MoBio PowerSoil DNA
extraction kit (filters directly placed into the
extraction tubes.)
Real-time PCR detecting
IpnA and FtM19 internal
deletion region
real-time PCR for tul4
hierarchical PCR analysis
using electrospray
ionization/time of flight
mass spectrometry (ESI-
MS) detection
PCR detecting IpnA
followed by hybridization
to various probes for
subspecies differentiation
PCR with genus specific
F. tularensis primers
Real-time PCR targeting
fopA and tul4
Clinically relevant subspecies F. tularensis subspecies
holarctica found in water and sediment samples
during three consecutive years.
Ten commercial PCR master mixes and three real-time
PCR instruments were compared: all ten yielded
positive results for F. tularensis on the 7500 Fast Dx
and Smart Cycler instruments, but only seven were
positive on the Light Cycler instrument.
This method can differentiate between pathogenic and
nonpathogenic F. tularensis strains for
epidemiological or investigation studies.
*
.
Method able to differentiate pathogenic F. tularensis
from non-pathogenic subspecies within tissue samples.
Limit of detection (LOD) 1 plasmid copy OR 10
genomic equivalents (GE).
An early study that looked at various methods for
processing environmental water samples for F.
tularensis detection. LOD 10 bacteria/mL.
Determined qPCR of ultrafiltration retentate is an
effective method to sample large-scale drinking water
samples.
21
-------
Reference
Fujita et aL,
2006 [1]
Garcia Del
Blanco et aL,
2002 [71]
Sample Matrix and
Tested Organism(s)
Pure cultures ofF.
tularensis subspecies
tularensis, holarctica,
philomiragia, F.
novicida
Clinical and
environmental isolates
of -F. tularensis
subspecies tularensis,
holarctica, andF.
novicida
Sample Preparation Method
Genomic DNA from pure cultures was
manually extracted or with the SepaGene
DNA Extraction Kit (Sanko Junyaku Co.,
Tokyo, Japan)
DNA from pure cultures grown on Thayer-
Martin agar was obtained through manual
extraction.
Detection method
Real-time PCR detecting
fopA gene
Pulsed-field gel
electrophoresis (PFGE),
amplified fragment length
polymorphism (AFLP),
and 16S rRNA
amplification for
Summary
Development of real-time PCR primers for identifying
F. tularensis in Ihour. LOD 1.2 colony forming units
(CPU) or 10 copies oftiaefopA gene.
PFGE and AFLP can discriminate between
Fmncisella species, but 16S rRNA amplification
cannot.
Johansson et
aL, 2000 [44]
Klerks et aL,
2006 [62]
Kuske et aL,
2006 [29]
Clinical isolates
collected from wounds
andF. tularensis
subspecies holarctica
LVS
Soil, manure, and
compost spiked with
Salmonella enterica
USA soil and aerosol
samples targeting F.
tularensis subspecies
holarctica, B.
anthracis, Y. pestis,
Clostridium
perfringens
Heat-killed whole cell extraction
DNA from lOOmg of spiked sample was
extracted by 1 of 5 commercial kits:
Ultraclean soil DNA isolation kit (MoBio);
Ultraclean fecal DNA kit (MoBio); BiolOl
extraction kit (Q-Biogene); Soilmaster DNA
extraction kit (Epicenter); plant DNeasy
DNA extraction kit (QIAGEN); or a
combination of the microbial DNA extraction
kit (MoBio) with bacterial isolation using
Optiprep™
DNA from 0.5 g of soil was manually
extracted with bead beating, ethanol DNA
precipitation and spin Sephadex® G-200
column cleanup. Aerosol filters washed in
PBS before DNA extracted with same
process.
subspecies identification
of F. tularensis.
Cotton-tipped applicator
used to collect material
fromF. tularensis
suspected ulcers. Cotton
applicators transported in
guanidine isothiocyanate
buffer before 450uL
manually extracted for
DNA and assessed for tul4
gene presence.
Real-time PCR using S.
enterica-specific detection
probe
Study compared PCR to culture and found PCR to
more sensitive; however, sampling methods can cause
PCR difficulties. Differentiating between tularemic
wounds and a non-infected wounds can be difficult.
MoBio soil, BiolOl, and MoBio fecal were found to
be most efficient for DNA extraction from,
respectively, soil (eight different substrates), manure
(six substrates), and compost (two substrates).
PCR targeting F.
tularensis specific 16S
rRNA and tul4 gene
F. tularensis 16S rRNA found in aerosol samples from
two US cites: Denver and San Diego. No tul4 genes
were detected. No soil samples were positive. LOD
0. Ipg or 17-46 GE
22
-------
Reference
Sample Matrix and
Tested Organism(s)
Sample Preparation Method
Detection method
Summary
Matero et
al., 2011 [72]
Meric et al.,
2010 [17]
O'Connell et
al., 2004 [73]
Offermans
and Zegers,
2007 [38]
Sellek et al.,
2008 [39]
Sim$ek et al.,
2012 [40]
Svensson et
al., 2009 [35]
B. thuringiensis, F.
tularensis, B.
anthracis, Yersinia
pestis, Bmcella spp.
Reservoir spring water
naturally contaminated
with/7, tularensis
Creamer, cornstarch,
baking powder, flour
spiked with F.
tularensis subspecies
holarctica LVS
Soils spiked with
unknown quantities of
F. tularensis, B.
anthracis, Brucella
pseudomallei or
Vaccinia
Sandy loam, silt loam,
or clay soil spiked with
FMA-inactivated F.
tularensis subspecies
holarctica LVS
Environmental water
naturally contaminated
with/7, tularensis
subspecies holarctica
LVS
62 Francisella isolates
of diverse genetic and
geographical origins
DNA from pure cultures were extracted using
MagNA Pure Nucleic Acid Isolation Kit I
PCR targeting 23kDa gene
1 L water samples filtered with 0.45 um
cellulose acetate filters. Filters washed with
sterile distilled water before filtrate was
cultured or DNA extracted with QIAamp
DNA mini kits.
DNA from pure cultures of/7, tularensis
extracted with QIAGEN DNeasy mini spin
columns.
UltraClean Soil DNA Isolation Kit (200 uL);
PowerMax Soil DNA Isolation Kit (6 mL)
0.5 g soil in PBS processed through Glass
fiber pre-filter (pore size 8 um) to separate
cells followed by heat lysis OR Millex®-SV
filter unit (pore size 5.0 um) followed by heat
lysis
0.3 -1.5 L water samples filtered through
cellulose acetate membranes (pore size 22
um). Membranes place directly on cysteine
heart agar with blood (CHAD) plates. For
PCR detection, filters were washed with
sterile water before DNA extracted using a
QIAamp DNA Mini Kit.
DNA from pure cultures was manually
extracted for this study.
Culture and real-time PCR
targeting: ISFtu2 element,
23 kDa gene, and the tul4
gene.
Direct PCR in BioSeeq8
handheld system
Real-time PCR targeting
tul4 gene and ELISA
qPCR with SYBR Green I
targeting tul4 gene
Culture and real-time PCR
targeting ISFtu2 gene
68 real-time PCRs for
hierarchical identification
of/7, tularensis
Study compared RAZOR to ABI instrumentation and
assessed a B. thuringiensis protocol with
environmental samples to show proof-of-principle for
F. tularensis.
No cultures were recovered. PCR was also attempted
following centrifugation concentration, however no
samples were PCR positive. Filtration is a better
concentration method than centrifugation. Sera, throat
swabs, lymph node aspirates, centrifuge concentrated
reservoir water, and filter concentrated waters were all
culture negative.
Bio-Seeq® technology is a novel system for use in
areas with high concentrations of bacteria. LOD of/7.
tularensis determined to be 103 cells/reaction or less
when the consumable sampling assembly is utilized
with household powders.
PowerMax DNA isolation kit extracts yielded much
stronger reactions than the UltraClean extracts. No F.
tularensis was detected by ELISA techniques.
Millex filter was more efficient for filtering soils
samples; however, it allowed more PCR inhibiting
compounds through to the final sample than the glass
fiber filter. Comparisons between qPCR and capture
ELISA show that qPCR has a lower LOD (102 and 104
CFU/mL, respectively)
Real-time PCR was more sensitive than culture as 17
of 154 samples were PCR positive and only 4 were
culture positive. 16S rRNA sequencing identified the
cultured strains as F. tularensis subspecies holarctica
strain LVS.
Study established a 68-well assay for differentiating
between/7, tularensis strains.
23
-------
Reference
Sample Matrix and
Tested Organism(s)
Sample Preparation Method
Detection method
Summary
Trombley
Hall et al.,
2013 [66]
Versage et
al., 2003 [46]
Whitehouse
et al., 2007
[61]
Pure DNA from F.
tularensis SCHU S4
spiked into sand and
soil
Pure cultures isolated
from tissues of
laboratory infected
animals including F.
tularensis subspecies
tularensis, holarctica,
philomiragia, F.
novicida
Silt loam, clay, and
potting soil spiked with
F. tularensis Shu-4
No extraction; direct PCR with 5 ul sample
slurry and added pure DNA
Tissue samples were directly cultured while
DNA from mouse and prairie dog tissues
were extracted with MasterPure™
Purification Kit (epiCenter).
0.1 to 10 g of soil processed through
Puregene® DNA purification Kit OR QIAmp
DNA Stool Mini Kit OR Epicentre
SoilMaster DNA Extraction Kit OR MoBio
UltraClean Soil DNA Isolation Kit OR
MoBio PowerMax soil DNA isolation Kit
PCR targeting tul4 gene
Multitarget PCR targeting
tul4,fopA, lSFtu2, 23kDa
compared to culture
Study assessed various PCR chemistries for inhibitor
resistant capacity for use with environmental samples.
Phire Hot Start DNA polymerase with SRT Boost
reagents was the best combination found for detecting
spiked DNA in soil samples.
Comparison of TaqMan PCR assays to culturing
determined that PCR was significantly more :
than culturing.
PCR targetingfopA gene
iring
; sensitive
UltraClean and PowerMax soil DNA isolation kits
were the most consistent and sensitive methods for
extracting F. tularensis from soil.
24
-------
2.4.2 PCR amplification for genomic identification of F. tularensis
PCR identification has progressed significantly in recent years. F. tularensis identification within
environmental waters by PCR amplification was initially conducted by manual DNA extraction
followed by genus specific Francisella PCR amplification [70] or restriction enzyme analysis
[52] and visual gel electrophoresis detection. Now, commercial sample extraction kits [61] and
rapid real-time PCR analysis allow for sensitive detection at low concentrations [1]. Genes
commonly targeted in genomic identification studies were tul4,fopA, lSFtu2, and 23kDa genes.
The tul4 andfopA genes are outer membrane proteins [46] encoding for a!7-kiloDalton (kDa)
protein [4] and a 43-kDa protein [16], respectively. !SFtu2 targets an insertion element-like
sequence in F. tularensis [50]. The 23kDA gene encodes a protein that is expressed during
macrophage infection [46].
PCR analysis was used to determine the natural presence of F. tularensis among soil and aerosol
samples collected across the United States. In total, 89 soils from across the US and over 15,000
aerosol samples from 15 major US cities were evaluated [29]. Utilizing 16S rRNA primers, the
study found that F. tularensis or its near relatives are naturally present in urban aerosols;
however, no 16S rRNA sequences for F. tularensis were found within the studied soils [29].
Following a natural tularemia outbreak at Martha's Vineyard, Berrada and Telford [16] were
able to identify diverse Francisella spp. within the environment through PCR analysis. Of the
156 samples assessed, 23 were positive for F. tularensis 16S rRNA, 15 were positive forfopA,
19 positive for lSFTu2, and 14 were positive for tul4. Of the PCR positive samples, only one
fopA PCR positive sample yielded a culture of F. philomiragia. Meric et al. [17] linked a
tularemia outbreak in Turkey to consuming reservoired spring water by targeting lSFtu2, 23kDa,
and tul4 genes in their PCR analyses. Targeting fopA, Fujita et al. [1] established a specific and
sensitive real-time PCR assay for rapid detection of F. tularensis within a prepared DNA sample.
This method can achieve detection equivalent to 1.2 CFU of bacterial cells/reaction.
Molecular methods have been developed to discriminate between F. tularensis and Francisella-
like organisms. Differentiation between pathogenic and nonpathogenic strains of F. tularensis is
critical to epidemiological and outbreak investigation studies. Recognition of a 36 base pair
deletion in IpnA sequences within F. tularensis subspecies allowed Escudero et al. [39] to
develop a genomic method for differentiating between F. tularensis and Francisella-like
organisms. One study compared three molecular methods for separating F. tularensis strains:
pulsed-field gel electrophoresis (PFGE), amplified fragment length polymorphism (AFLP), and
16S rRNA gene sequencing [71]. PFGE and AFLP were able to distinguish^, tularensis
subspecies that could be useful for epidemiological tracking during a tularemia event. Duncan et
al. [14] and Svensson et al. [35] used PCR assays for hierarchical identification of Francisella
isolates. Duncan et al. [14] utilized 24 multilocus PCR reactions followed by electrospray
ionizati on/time of flight mass spectrometry (ESI-MS) detection to differentiated, tularensis
subspecies. Svensson et al. [35] utilized specific deletions and insertions within the F. tularensis
genome to generate a hierarchical identification system using 68 individual real-time PCR
25
-------
reactions. While both of these studies have the ability to differentiate between F. tularemis
subspecies, their utility would be most useful for tracking analysis in a tularemia outbreak
situation.
Results of a study focusing on published PCR primers and their specificity among whole-
genome sequences now available showed that many primers previously developed for F.
tularensis contain extremely low specificity, and therefore yield false positives [67].
Identification of specific species or subspecies can be challenging. Real-time PCR assays
incorrectly identified F. tularensis andF. novicida during an outbreak [74]. This finding points
to the need for thorough characterization of isolates that share close sequence identities. To
mitigate false positive PCR results, primer sequences need to be continually evaluated and
redesigned using up-to-date genomic databases. Furthermore, as no single-marker was capable of
distinguishing all the Francisella strains within the Ahlinder et al. [67] study, an optimized
combination of markers could be used to improve Francisella strain resolution.
It has been recognized that PCR master mixes and PCR thermocycler instruments do not all
function equally. In a comparison of the ABI 7300/7500 (Applied Biosystems, Foster City, CA)
to the RAZOR (Idaho Technology Inc., Salt Lake City, UT) real-time PCR thermocyclers the
LOD for F. tularensis 23kDa gene was found to be the same at 10 fentagram (fg) genomic DNA
per reaction [72]. However when Buzard et al. [68] compared ten commercially available PCR
master mixes and three real-time PCR instruments, all ten master mixes tested yielded positive
results for F. tularensis on the 7500 Fast Dx (Applied Biosystems) and SmartCycler (Cepheid,
Sunnyvale, CA) instruments, but only seven were positive on the LightCycler (Roche,
Indianapolis, IN) instrument.
2.4.3 Methods for Environmental Sampling and Detection of Multiple Biothreat
Organisms
New technologies utilizing genomic techniques to detect pathogenic organisms alone or in
concert with other organisms are constantly being developed. Table 4 gives details on developing
assays for F. tularensis identification and multiagent identification methods discussed in this
review.
26
-------
Table 4. Comparison of Developing Methods for Genomic Identification of Francisella tularensis Alone and Simultaneously with Other Organisms.
Reference
Organism(s)
Sample Matrix
Sample Preparation Method
Detection Method
Summary
Brinkman et
al., 2013 [75]
Cooper et al.,
2011 [2]
Euler et al.,
2012 [13]
Euler et al.,
2013 [76]
He et al., 2009
[77]
Janse et al.,
2010 [78]
Francisella tularensis
LVS, Bacillus
anthracis,
Crypto sporidium.
parvum, C. hominis,
Enterococcus faecium
Tap water
F. tularensis
subspecies tularensis,
holarctica
F. tularensis
subspecies tularensis,
holarctica, F.
hispaniensis, F.
novicida F.
philomiragia
F. tularensis (Ft 12),
B. anthracis, Yersinia
pestis, variola virus
F. tularensis, B.
anthracis, Y. pests,
variola major
F. tularensis
subspecies holarctica
and tularensis, and F.
novicida, B.
anthracis, Y. pestis
Phosphate
buffered saline
(PBS)
Rabbit tissue
Spiked human
plasma
Spiked clinical
samples
Spiked milk
powder, soy
powder, silica,
and maize
powder
DNA extracted from pure cultures with
Centra PureGene Genomic Prep kit (Qiagen)
added to a background of concentrated tap
water. 1000 L of tap water was repeatedly
filtered before ultracentrifugation and solvent
extraction to remove PCR inhibitors.
Pure cultures were boiled to lyse cells before
ethanol precipitation to concentrate DNA.
DNA extracted from pure cultures with
QIAamp DNA blood extraction kit (Qiagen)
small pieces of tissue was homogenized
before DNA extracted by same kit.
DNA from spiked plasma was extracted with
an innuPREP MP basic kit A (Jena Analytik)
Manual DNA extractior
NucliSens Magnetic Extraction Reagents
(bioMerieux) were used to extract DNA from
pure cultures. DNA was added to interfering
agents before analysis.
Microarray
Antibody and DNA
photonic biosensors
targeting yhh W gene for
type A strains, IpnA gene
for both type A and B
strains
Real-time recombinase
polymerase amplification
(RPA) assay on an
isothermal amplification
methods ESEQuant tube
scanner device detecting
tul4
Real-time RPA assay on
an isothermal
amplification methods
ESEQuant tube scanner
device detecting tul4
lultiplex PCR-enzyme
hybridization assay
targeting tul4
Multiplex quantitative
PCR targeting/op^,
ISFtu2, andpdpD
Designed to identify F. tularensis
and other human pathogens under
periods of high concentration from
within tap water samples.
rom
Photonic biosensor only requires
nanogram quantities of target DNA
to differentiate F. tularensis
subspecies without polymerase
chain reaction (PCR) amplification.
RPA is comparable to real-time
PCR with —10 min run times and
limit of detection (LOD) of 10-100
molecules.
RPA performed equally as PCR
and showed not cross-detection
among targets. RPA run time is
-10 min with a LOD of 10
molecules.
Only spiked clinical samples were
assessed for method development
of multiplex PCR enzyme
hybridization assay (mPCR-EHA).
LOD established at 10 copies/mL.
This multiplex reaction
incorporates an internal positive
control (B. thuringiensis spores) for
both nucleic acid extraction and
amplification. It allows rapid
detection of three pathogen-specific
targets simultaneously without
compromising sensitivity. F.
tularensis LOD 0.6-11.8 fg
DNA/reaction.
27
-------
Reference
Organism(s)
Sample Matrix
Sample Preparation Method
Detection Method
Summary
Janse et al.,
2012 [79]
Jeng et al.,
2013 [80]
F. tularensis
subspecies holarctica,
tularensis, and
novicida, B.
anthracis, Y. pestis,
Coxiella burnetii
F. tularensis, B.
anthracis, Y. pestis,
Brucella spp.,
Burkholderia spp,
Rickettsia prowazekii
et al., r. tularensis Lv&
O'Connell et
al., 2004 [82]
F. tularensis LVS
Spiked blood,
water, surface
swab
Clinical
bronchoalveolar
lavage (BAL)
fluids spiked
with purified
DNA
Purified DNA
Spiked creamer,
cornstarch,
baking powder,
flour
Rachwal et al., F. tularensis Schu4, Purified DNA
2012 [83]
B. anthracis, B.
mallei, B.
pseudomallei, Y.
pestis
Schweighardt
et al., 2014
[84]
F. tularensis
subspecies tularensis,
B. anthracis, Y. pestis,
C. botulinum
Purified DNA
200 L surface water was filter concentrated.
DNA in filtrate extracted with NucliSens
Magnetic Extraction Reagents (bioMerieux).
Cotton swab samples collected at a goat farm
were added to 10 mL of NucliSens lysis
buffer and vortexed before DNA extraction.
B. thuringiensis spore suspension (105
spores) added as an internal control to each
sample before DNA extraction.
DNA from 300 uL of BAL fluid was
extracted with the automated Roche Magna
Pure LC robot with DNA isolation kit III
protocol
DNA from pure cultures was extracted with
MagNA Pure Nucleic Acid Isolation Kits.
DNA from pure cultures ofF. tularensis
extracted with QIAGEN Dneasy mini spin
columns.
None- purified DNA was acquired for the
study
None- purified DNA was acquired for the
study
qPCR detected through
direct hybridization to
microarray probes OR
target-specific primer
extension followed by
universal hybridization
targeting:/?^, wbtK,
ISFtu2, pdpD
Reverse transcription-
PCR-electrospray
ionization mass
spectrometry
R.A.P.I.D® platform for
real-time PCR in the field
Direct PCR in BioSeeq8
handheld system
TaqMan Array Cards
developed for multiple
biothreat organisms
Bead-based liquid
hybridization assay,
Luminex ® 100TM,
targeted ribosomal rrl and
F. tularensis toxicity
target
The microarrays were capable of
detecting multiple signature
sequences with an internal control,
making it possible to identify
targeted pathogens and assess
virulence potential.
High-throughput reverse
transcription-PCR coupled to
electrospray ionization mass
spectrometry analysis (RT-PCR-
ESI-MS) can be used to detect
biothreat agents in clinical samples.
Study established a real-time PCR
protocol for sputum and blood
samples in the field using the
R.A.P.I.D. system. LOD of 10 fg of
DNA or 5 genomic equivalents
(GE).
Eg of
Bio-Seeq technology is a novel
system for use in areas with high
concentrations of bacteria. LOD of
F. tularensis determined to be 103
cells/reaction or less when the
consumable sampling assembly is
utilized with household powders.
TaqMan® Array Card was capable
of detecting all five organisms,
with a LOD one order of
magnitude greater than the
singleplex reactions (10 vs 100
fg/reaction).
Proof of principle for laboratory
samples for simultaneously
identifying multiple pathogenic
microorganisms. Achieved LODs
of 0.1 to lOngDNA.
28
-------
Reference
Organism(s)
Sample Matrix
Sample Preparation Method
Detection Method
Summary
Seiner et al.,
2013 [85]
Turingan et
al., 2013 [12]
Yang et al.,
2012 [86]
F. tularensis
subspecies tularensis
and holarctica, B.
anthracis, Y. pestis
Purified DNA Pure genomic DNA purchased for the study.
F. tularensis, B.
anthracis, Y. pestis
F. tularensis
(410101), B.
anthracis, Y. pestis,
Brucella spp, B.
pseudomallei
Biowatch filters Air filters were washed in sterile water, cells
were lysed by sonication before DNA was
purified by a Qiagen spin column
Pure cultures DNA from pure cultures of F. tularensis was
extracted from cells that were lysed by
boiling.
Multiplexed PCR-based
assay for 17 pathogens
and toxins
Microfluidic multiplexed
PCR and sequencing
assays
Multiplex PCR targeting
fopA
Proof of principle study for
FilmArray platform as complete
sample-to-answer system,
combining sample preparation,
PCR and data analysis. LOD at 250
GE.
Study demonstrated a proof-of-
principle for F. tularensis, B.
anthracis, and Y. pestis detection
and subspecies differentiation
within environmental aerosol
(Biowatch) samples using B.
subtilis.
Results suggest that the liquid array
method would be capable of
detecting bioagents of interest from
environmental samples. (LOD 0.95
pg DN A/reaction).
29
-------
Field-deployable detection systems are needed for first responders. Ideally, field-deployable
systems would be rugged, sensitive, specific, and easily manipulated within protective gear. Bio-
Seeq®, Ruggedized Advanced Pathogen Identification Device (R. A.P.I.D.®), and FilmArray®
systems are three technologies available for first responders that were discussed in the literature.
The Bio-Seeq® instrument is a self-contained, portable, handheld, real-time PCR system that
includes a consumable sampling and reaction tube assembly. The consumable assembly includes
a sampling swab, buffer, and assay reagents. The operator simply uses the swab to sample a
surface, and then inserts the swab into the system and twists to release the prepared buffer.
Manual shaking completes the sample processing before inserting the unit into the Bio-Seeq®
instrument for PCR analysis [82]. The Bio-Seeq® technology is a novel system for use in high
concentration areas as the LOD of F. tularensis was determined to be 103 cells per reaction when
the consumable sampling assembly is used. Furthermore, as the sample DNA is not purified,
there could be significant inhibition when used for environmental soils or waters. F. tularensis
was detectable when spiked into wheat flour, cornstarch, baking soda, and coffee creamer;
however, inhibition was noted [82]. R. A.P.I.D. is a field deployable real-time PCR platform for
which F. tularensis specific primers have been established [81]. A newly developed FilmArray®
system utilizes a "Lab-in-a-Pouch" approach for conducting sample-to-answer detection of 17
biothreat agents [85]. A liquid sample is placed within the system pouch, which contains all the
reagents required for sample preparation, cell lysis, PCR, and end-point detection. Thus far, the
system has only been assessed with B. anthracis cells and spores, Y. pestis cells, and F.
tularensis genomic DNA to demonstrate its proof-of-principle [85]. Therefore, future work
evaluating field-deployable detecting systems for environmental liquids and soils is needed.
Multiplex real-time PCR detection methods could save both time and valuable resources during a
crisis event. Janse et al. [78] developed a multiplex qPCR for simultaneously detecting three
genes ofF. tularensis (fopA, ISFtu2,pdpD) while also incorporating an internal positive control
(B. thuringiensis spores) for both nucleic acid extraction and amplification. The multi-target PCR
was initially developed to reduce false positive and false negative results from environmental
samples. While this method has not been verified specifically with soils, the authors stated that
the method has been utilized for hundreds of solid and liquid samples [78]. More recently, the
same research group developed a multiplex asymmetric PCR protocol that amplifies 16 DNA
signatures for simultaneous detection of four biothreat agents. Four gene signatures are targeted
from F. tularensis, Y. pestis, and Coxiella burnetii; three signatures are targeted from B.
anthracis, and a single signature is dedicated to the internal positive control, B. thuringiensis
[79]. Due to the number of amplified signatures, standard multiplex platforms are unable to
differentiate the PCR products. Therefore, two labeling chemistries for microarray detection
were compared [79]. Direct hybridization uses in-house labeled primers in the multiplex PCR to
generate labeled PCR products, while target-specific primer extension followed by universal
hybridization incorporates a unique capture tag sequence during strand extension by DNA
polymerase. Both microarray formats allowed multiple pathogens to be simultaneously detected
with high specificity and sensitivity [79]. The LOD for F. tularensis through either microarray
30
-------
detection technology was determined to be 12 copies/reaction (target amplicon of 4.1) when
targeting the internal spacer region, ISFtu2 [79].
Rachwal et al. [83] noted the trade-off between achieving multiple organism detection and
minimizing LOD. They developed a TaqMan® Array Card that incorporated ten PCR reactions
targeting five biothreat agents: B. anthracis, B. mallei, B. pseudomallei, Y. pestis, and F.
tularensis. A comparison of PCR performance of the TaqMan® Array Card and singleplex real-
time PCR using pure genomic DNA showed that while the TaqMan® Array Card was capable of
detecting all five organisms, its LOD was one order of magnitude greater than the singleplex
reactions [83]. In an attempt to minimize LODs and still achieve multiple pathogen
identification, Brinkman et al. [75] developed a microarray-based method for simultaneously
detecting Cryptosporidium parvum, C. hominis, Enterococcus faecium, B. anthracis andF.
tularensis in concentrated aqueous samples. DNA microarrays can identify thousands of loci
within a single sample, and their microarray assay was capable of detecting F. tularensis
genomic DNA at 20 genomic copies without PCR preamplification. While this method has not
been tested with soil samples, the concentrated tap water sample used within the study was
equivalent to 33 L of tap water [75]. It is therefore conjectured, that after adequate optimization,
soil sample suspensions might be suitable for analysis using this technology.
Other groups have also sought to detect multiple biothreat agents within a single assay. Turingan
et al. [12] utilized a microfluidic biochip to develop a multiplexed PCR and sequencing assay for
simultaneous detection of three pathogens, 10 loci per pathogen. Schweighardt et al. [84]
developed a protocol using a Luminex® system to detect B. anthracis, Clostridium botulinum, Y.
pestis, and F. tularensis. The Luminex® liquid array platform system uses genetically marked
beads to simultaneously identify multiple pathogenic microorganisms, and can achieve LODs of
0.1 to 10 nanograms (ng) DNA [84]. Yang et al. [86] assessed a multi-targeted liquid array
method for simultaneously detecting B. anthracis, Y. pestis, B. pseudomallei, Brucella spp., and
F. tularensis within a simulated white-power sample. Universal 16S rRNA primers were used for
amplification before identification using pathogen-specific hybridization probes. The Bio-Plex
assay was then assessed using B. anthracis and Y. pestis spiked into various household white
powders (milk powder, corn starch, wheat flour, instant drink mix). Results suggest that the
liquid array method would be capable of detecting bioagents of interest from environmental
samples [86]. A multiplex PCR enzyme hybridization assay (mPCR-EHA) has also been
developed by He et al. [77] to simultaneously detect variola major, B. anthracis, Y. pestis,
varicella zoster virus, and F. tularensis from within clinical samples. Jeng et al. [80] assessed the
utility of high-throughput reverse transcription-PCR coupled to electrospray ionization mass
spectrometry analysis (RT-PCR-ESI-MS) for detecting biothreat agents in clinical
bronchoalveolar lavage (BAL) fluid specimens. Their analysis determined that RT-PCR-ESI-MS
could provide accurate detection of multiple biothreat organisms from within polymicrobial
clinical matrices [80]. Development of a qualitative real-time isothermal recombinase
polymerase amplification (RPA) assay for F. tularensis alone [13] or in combination with Y.
31
-------
pestis, B. anthracis, and variola virus [76] shows potential as a field deployable method for quick
results (-10 min).
Cooper et al. [2] developed an assay for detecting F. tularemis from aqueous samples, but
without PCR amplification. The prototype photonic biosensor utilizes label-free single-stranded
oligonucleotides to consistently detect F. tularemis at low concentrations (minimum
concentration tested, 1.7 ng) without PCR amplification. While the method needs to be
optimized for field use, the initial studies demonstrate that the method could be a promising tool
to rapidly detect F. tularensis in the field or with limited laboratory facilities [2]. While each of
these technologies are promising, environmental assessments with complex environmental
matrices are lacking and will need to be conducted to assess efficacy.
2.5 Combining Culture with PCR to detect live F. tularensis
The downfall of PCR techniques are their inability to discriminate between viable and non-viable
target microorganisms. This review found two methods for rapid detection of viable pathogenic
cells from various matrices by combining culture with PCR. Day and Whiting [87] utilized
mammalian macrophage cell cultures to detect F. tularensis from contaminated foods. The
macrophage cell cultures were exposed to contaminated foods (liquid baby formula, liquid egg
whites, and iceberg lettuce mixed 1:1 with PBS) for two-hours to allow cell contact and
engulfment of F. tularensis. After this initial incubation with the contaminated matrix, the
macrophage monolayers are then washed with PBS to remove food particles and reconstituted
with macrophage growth medium before an additional five to 18 hours of incubation. The
additional incubation allows for proliferation of the engulfed F. tularensis within the
macrophages. Finally, the macrophage monolayers are scraped from the plates, cleaned, and
boiled to lyse the cells. The resulting supernatant is then used directly for real-time PCR analysis
[87]. Using this method Day and Whiting [87] were able to detect viable F. tularensis from food
matrices at a LOD of 10 CFU/mL formula or egg whites and 10 CFU/g lettuce within 22 hours.
In a similar manner, rapid viability (RV)-PCR utilizes an enrichment step and the change in
cycle threshold time between two PCR reactions to determine the presence or absence of viable
cells [88, 89]. RV-PCR has been used to detect viable B. anthracis spores from within dust,
water, and dirty air filters [89]. For B. anthracis spore samples, spores within the environmental
samples are separated from other particles and suspended in a growth medium. Prior to
incubation, an aliquot of the sample is collected and processed for genomic identification by
real-time PCR. After a minimum of nine hours of incubation an additional aliquot is collected
and processed for real-time PCR. Comparing PCR cycle threshold numbers before and after
incubation allows the discrimination between viable and non-viable B. anthracis spores [89].
While the literature review conducted herein did not find a study where environmental soil or
waters were detected by either macrophage cell cultures or RV-PCR, culturing prior to PCR
shows promise as a means to detect viable F. tularensis at low concentrations. Future work
32
-------
expanding one or both of these methodologies might provide increased detection capabilities for
environmental samples.
3 Conclusions and Identified Data Gaps
Limited work regarding F. tularensis detection in soil has been conducted. More information
regarding F. tularensis detection in environmental waters is available. However, questions
remain regarding the complete lifecycle of F. tularensis within the environment. The role
protozoa and biofilms have in F. tularensis persistence needs to be elucidated [15, 31].
Additional information regarding how F. tularensis persists in the environment will be helpful in
guiding research in the development of appropriate detection technologies targeting F. tularensis
in microenvironments. Further, once the ecology of F. tularensis is understood, proper
disinfection technologies for combating sustained F. tularensis outbreaks can be developed.
Culturing F. tularensis from environmental samples is challenging, yet isolating viable F.
tularensis cultures from samples is required to evaluate factors such as pathogenicity and
antibiotic sensitivity of environmental isolates. It is also the current approach for evaluating the
efficacy of decontamination procedures. F. tularensis is slow-growing, nutritionally fastidious
organism that requires 24 to 72 hours for growth [46] on supplemented medium [15]. Even with
selective agars, F. tularensis colonies are often out-competed by background organisms present
in environmental samples [46]. The review herein found 14 studies that utilized culture analyses
with varying success. Future studies focused on the integration of culture and genomic
identification could be the future for rapid viable detection [87, 89].
Immunoassay detection of F. tularensis can be amenable to hand-held devices, however due to
high limits of detection their utility might only be seen in highly concentrated samples [52].
Protein chip immunoarrays can rapidly identify multiple bioagents within a single sample; yet
again high limits of detection (~106 CFU/mL) limit its utility for screening potential low
concentrations in environmental matrices [56]. The overall quality of immunoassays, whether in
a single reaction or as part of an immunoarray chip, is dependent upon the specificity of the
selected antigens. Some antigens can have cross-reactivity to other microorganism, thus
impeding the results [51, 56]. Other immunosensor assays are on the horizon and could offer
environmentally applicable methods after further development and optimization [59, 90].
Genomic identification of F. tularensis was the most common mode of identification seen in this
review. Four genes were repeatedly used to identify F. tularensis: tul4, fopA, ISFtu2, and 23kDa.
Sampling methods [44], sample purification methods [61], and the PCR primers used within a
study can impact the overall findings [67]. A study focused on the specificity of various primers
for F. tularensis noted that many published primers are not very specific, and therefore evaluate
their primers against up-to-date genomic databases before their use is needed during future
investigations [67]. It was also noted that no single F. tularensis marker was capable
33
-------
distinguishing all Francisella strains; therefore, to maximize resolution multiple markers could
be targeted [67].
As mentioned previously in this literature review, extracting DNA from F. tularemis is relatively
easy when compared to sporulated microorganisms. However, constituents within soil and
environmental waters must be removed from DNA samples to increase processing efficiency.
UltraClean® DNA extraction kits were widely used for extracting DNA from various
environmental sample matrices, and have demonstrated their ability to produce DNA of
sufficient quantity and quality for downstream genomic analyses [16, 38, 50, 61, 62, 65].
However, while Whitehouse and Hottel [61] conducted a systematic comparison of DNA
extraction kits for isolating F. tularensis DNA, laboratory inoculated soils are not equivalent to
environmentally contaminated soils. Cells within real environmental samples might be
aggregated with other constituents making DNA extraction more complicated [62]. In light of the
complexity of soils and the potential for unknown inhibiting compounds in environmental
samples, each analytical step should have internal controls. Janse et al. [79] has suggested B.
thuringiensis spores as a possible agent for both extraction and amplification internal controls.
Spores are added prior to sample extraction to ensure that even the most recalcitrant cells within
the soil aliquot are lysed, while PCR inhibition is identified by using well studied PCR primers.
Primers for the internal control could ideally have the same melting and annealing temperatures.
A relatively new mechanism to prevent PCR inhibition is using inhibitor-resistant PCR reagents
[66]. While one study herein optimized the use of inhibitor-resistant PCR reagents for detecting
F. tularensis in soil samples, a detailed comparison of multiple soils with various extraction kits
and various inhibitor-resistant PCR reagents might be needed to make generalizations about its
applicability. Such an analysis could identify an optimum extraction kit in conjunction with
optimum PCR reagents to yield real-time PCR reactions with increased sensitivity.
Microarray detection technologies offer the potential for high-throughput environmental
detection. Several groups have utilized microarray technology to simultaneously detect multiple
biothreat agents of interest [12, 75, 80, 84, 86] while few have assessed the technology with
environmental samples and their associated complexities. Brinkman et al. [75] and Francy et al.
[4] have demonstrated the utility of detecting F. tularensis genomic DNA from within highly
concentrated tap water samples, and thus offers insight into the potential use with other
environmental matrices. The introduction of whole genome amplification prior to microarray
detection could further improve sensitivity [75]. Future work combining optimized DNA
extraction, whole genome amplification with inhibition-resistant polymerases, and multiagent
microarray detection could significantly expand biothreat detection capabilities.
Two groups identified by this review have utilized a combination of culture and genomic
analysis to rapidly, quantify viable microorganisms. Using macrophage cell cultures to accelerate
F. tularensis growth before DNA extraction and amplification, Day and Whiting [87] were able
to detect viable F. tularensis within contaminated foods at a LOD of 10 CFU/mL. RV-PCR is
another promising technique that utilizes an enrichment step between two PCR reactions to
34
-------
quantify the concentration change of a targeted microorganism [89]. While RV-PCR has not
been optimized for F. tularemis detection, it has been shown to be effective for B. anthracis
spore detection from within environmental samples. Future work incorporating a macrophage
culture step with RV-PCR sample processes could significantly improve viable F. tularensis
detection capabilities from within environmental soil and waters.
35
-------
4 References
1. Fujita, O., M. Tatsumi, K. Tanabayashi, and A. Yamada. (2006). Development of a real-
time PCR assay for detection and quantification of Francisella tularensis. JpnJInfect
Dis, 59(1): 46-51.
2. Cooper, K.L., A.B. Bandara, Y. Wang, A. Wang, and T.J. Inzana. (2011). Photonic
biosensor assays to detect and distinguish subspecies of Francisella tularensis. Sensors,
11(3): 3004-19.
3. USD A, Soil Survey Staff. (1999). Soil taxonomy: A basic system of soil classification for
making and interpreting soil surveys. 2nd ed.: Natural Resources Conservation Service.
U.S. Department of Agriculture Handbook.
4. Francy, D.S., R.N. Bushon, A.M. Brady, E.E. Bertke, C.M. Kephart, C.A. Likirdopulos,
B.E. Mailot, F.W. Schaefer, 3rd, and H.D. Lindquist. (2009). Comparison of traditional
and molecular analytical methods for detecting biological agents in raw and drinking
water following ultrafiltration. J ApplMicrobiol, 107(5): 1479-91.
5. Gilbert, S.E. and LJ. Rose. (2012). Survival and persistence of nonspore-forming
biothreat agents in water. Lett Appl Microbiol, 55(3): 189-94.
6. Balestrazzi, A., M. Bonadei, C. Calvio, A. Galizzi, and D. Carbonera. (2009). DNA
extraction from soil: comparison of different methods using spore-forming bacteria and
the swrAA gene as indicators. Ann Microbiol, 59(4): 827-832.
7. Robe, P., R. Nalin, C. Capellano, T.M. Vogel, and P. Simonet. (2003). Extraction of
DNA from soil. Euro Journal SoilBiol, 39(4): 183-190.
8. Zhou, J., M.A. Bruns, and J.M. Tiedje. (1996). DNA recovery from soils of diverse
composition. Appl Environ Microbiol, 62(2): 316-22.
9. Kingry, L.C. and J.M. Petersen. (2014). Comparative review of Francisella tularensis
and Francisella novicida. Front Cell Infect Microbiol, 4: 35.
10. Keim, P., A. Johansson, and D.M. Wagner. (2007). Molecular epidemiology, evolution,
and ecology of Francisella. Ann NY AcadSci, 1105: 30-66.
11. McCoy, G.W. and C.W. Chapin. (1912). Bacterium tularense, the cause of a plaguelike
disease of rodents Public Health Bull 53:17-23.
12. Turingan, R.S., H.U. Thomann, A. Zolotova, E. Tan, and R.F. Selden. (2013). Rapid
focused sequencing: a multiplexed assay for simultaneous detection and strain typing of
Bacillus anthracis, Francisella tularensis, and Yersiniapestis. PLoS One, 8(2): e56093.
13. Euler, M., Y. Wang, P. Otto, H. Tomaso, R. Escudero, P. Anda, F.T. Hufert, and M.
Weidmann. (2012). Recombinase polymerase amplification assay for rapid detection of
Francisella tularensis. J Clin Microbiol, 50(7): 2234-8.
14. Duncan, D.D., A.J. Vogler, M.J. Wolcott, F. Li, D.S. Sarovich, D.N. Birdsell, L.M.
Watson, T.A. Hall, R. Sampath, R. Housley, L.B. Blyn, S.A. Hofstadler, D.J. Ecker, P.
Keim, D.M. Wagner, and M.W. Eshoo. (2013). Identification and typing of Francisella
tularensis with a highly automated genotyping assay. Lett Appl Microbiol, 56(2): 128-34.
15. van Hoek, M.L. (2013). Biofilms: an advancement in our understanding of Francisella
species. Virulence, 4(8): 833-46.
16. Berrada, Z.L. and S.R. Telford, 3rd. (2010). Diversity of Francisella species in
environmental samples from Martha's Vineyard, Massachusetts. Microb Ecol, 59(2): 277-
83.
36
-------
17. Meric, M., M. Sayan, D. Dundar, and A. Willke. (2010). Tularaemia outbreaks in
Sakarya, Turkey: Case-control and environmental studies. SingaporeMed J, 51(8): 655-
9.
18. Nakazawa, Y., R.A. Williams, A.T. Peterson, P.S. Mead, KJ. Kugeler, and J.M.
Petersen. (2010). Ecological niche modeling of Francisella tularemis subspecies and
clades in the United States. Am JTropMedHyg, 82(5): 912-8.
19. Reintjes, R., I. Dedushaj, A. Gjini, T.R. Jorgensen, B. Cotter, A. Lieftucht, F. D'Ancona,
D.T. Dennis, M.A. Kosoy, G. Mulliqi-Osmani, R. Grunow, A. Kalaveshi, L. Gashi, and I.
Humolli. (2002). Tularemia outbreak investigation in Kosovo: Case control and
environmental studies. EmergInfect Dis, 8(1): 69-73.
20. Dennis, D.T., T.V. Inglesby, D.A. Henderson, J.G. Bartlett, M.S. Ascher, E. Eitzen, A.D.
Fine, A.M. Friedlander, J. Hauer, M. Layton, S.R. Lillibridge, I.E. McDade, M.T.
Osterholm, T. O'Toole, G. Parker, T.M. Perl, P.K. Russell, and K. Tonat. (2001).
Tularemia as a biological weapon: medical and public health management. JAMA,
285(21): 2763-73.
21. Grunow, R. and EJ. Finke. (2002). A procedure for differentiating between the
intentional release of biological warfare agents and natural outbreaks of disease: its use in
analyzing the tularemia outbreak in Kosovo in 1999 and 2000. din Microbiol Infect,
8(8): 510-21.
22. Broman, T., J. Thelaus, A.C. Andersson, S. Backman, P. Wikstrom, E. Larsson, M.
Granberg, L. Karlsson, E. Back, H. Eliasson, R. Mattsson, A. Sjostedt, and M. Forsman.
(2011). Molecular detection of persistent Francisella tularensis subspecies holarctica in
natural waters. Int JMicrobiol, 2011.
23. Johansson, A., A. Ibrahim, I. Goransson, U. Eriksson, D. Gurycova, I.E. Clarridge, 3rd,
and A. Sjostedt. (2000). Evaluation of PCR-based methods for discrimination of
Francisella species and subspecies and development of a specific PCR that distinguishes
the two major subspecies of Francisella tularensis. J Clin Microbiol, 38(11): 4180-5.
24. Whitehouse, C.A., K.E. Kesterson, D.D. Duncan, M.W. Eshoo, and M. Wolcott. (2012).
Identification and characterization of Francisella species from natural warm springs in
Utah, USA. Lett ApplMicrobiol, 54(4): 313-24.
25. Forsman, M., E.W. Henningson, E. Larsson, T. Johansson, and G. Sandstrom. (2000).
Francisella tularensis does not manifest virulence in viable but non-culturable state.
FEMSMicrobiolEcol, 31(3): 217-224.
26. Mitscherlich, E. and E.H. Marth. (1984). Microbial Survival in the Environment. Berlin:
Springer-Verlag. 741-742.
27. Goethert, H.K. and S.R. Telford, 3rd. (2009). Nonrandom distribution of vector ticks
(Dermacentor variabilis) infected by Francisella tularensis. PLoS Pathog, 5(2):
e!000319.
28. Davis-Hoover, W., W.G. Wade, Y. Li, T.D. Biggs, and P.G. Koga. (2006). Persistence of
Bacillus anthracis spores and Clostridium botulinum and destruction of Francisella
tularensis and Yersiniapestis in municipal solid waste landfill leachates, in Fourth
Intercontinental Landfill Research Symposium: Gallivare (Lapland), SWEDEN, 2006.
29. Kuske, C.R., S.M. Barns, C.C. Grow, L. Merrill, and J. Dunbar. (2006). Environmental
survey for four pathogenic bacteria and closely related species using phylogenetic and
functional genes. JForensic Sci, 51(3): 548-58.
37
-------
30. EPA and CDC. (2012). Method development for optimum recovery of Yersiniapestis
from transport media and swabs. Environmental Protection Agency and Centers for
Disease Control and Prevention. EPA/600/R-12/620.
31. Abd, H., T. Johansson, I. Golovliov, G. Sandstrom, andM. Forsman. (2003). Survival
and growth of Francisella tularensis in Acanthamoeba castellanii. Appl Environ
Microbiol, 69(1): 600-6.
32. Sjostedt, A. (2006). Intracellular survival mechanisms of Francisella tularensis, a stealth
pathogen. Microbes Infect, 8(2): 561-7.
33. Visvesvara, G.S. (2010). Free-Living amebae as opportunistic agents of human disease. J
Neuroparasitology, l:(Article ID N100802) (13 pages).
34. Kantardjiev, T. and T. Velinov. (1995). Interaction between protozoa and
mircoorganisms of the genus Francisella. Problems Infect Dis, 22: 34-35.
35. Svensson, K., M. Granberg, L. Karlsson, V. Neubauerova, M. Forsman, and A.
Johansson. (2009). A real-time PCR array for hierarchical identification of Francisella
isolates. PLoS One, 4(12): e8360.
36. Durham-Colleran, M.W., A.B. Verhoeven, and M.L. van Hoek. (2010). Francisella
novicida forms in vitro biofilms mediated by an orphan response regulator. Microb Ecol,
59(3): 457-65.
37. Mahajan, U.V., J. Gravgaard, M. Turnbull, D.B. Jacobs, and T.L. McNealy. (2011).
Larval exposure to Francisella tularensis LVS affects fitness of the mosquito Culex
quinquefasciatiis. FEMSMicrobiol Ecol, 78(3): 520-30.
38. Offermans, M.T.C. and N.D. Zegers. (2007). Test Results 7th NATO-SIBCRA BW
Round Robin Trial 2006. The Netherlands. TNO Defence, Security, and Safety, Delft,
Netherlands. TD2007-0043.
39. Sellek, R., O. Jimenez, C. Aizpurua, B. Fernandez-Frutos, P. De Leon, M. Camacho, D.
Fernandez-Moreira, C. Ybarra, and J. Carlos Cabria. (2008). Recovery of Francisella
tularensis from soil samples by filtration and detection by real-time PCR and cELISA J
Environ Monit, 10(3): 362-9.
40. §im§ek, H., M. Taner, A. Karadenizli, M. Ertek, and H. Vahaboglu. (2012). Identification
of Francisella tularensis by both culture and real-time TaqMan PCR methods from
environmental water specimens in outbreak areas where tularemia cases were not
previously reported. Euro J Clin Microbiol & Infect Dis, 31(9): 2353-2357.
41. Anda, P., J. Segura del Pozo, J.M. Diaz Garcia, R. Escudero, F.J. Garcia Pena, M.C.
Lopez Velasco, R.E. Sellek, M.R. Jimenez Chillaron, L.P. Sanchez Serrano, and J.F.
Martinez Navarro. (2001). Waterborne outbreak of tularemia associated with crayfish
fishing. Emerg Infect Dis, 7(3 Suppl): 575-82.
42. Petersen, J.M., J. Carlson, B. Yockey, S. Pillai, C. Kuske, G. Garbalena, S. Pottumarthy,
and L. Chalcraft. (2009). Direct isolation of Francisella spp. from environmental
samples. Lett Appl Microbiol, 48(6): 663-7.
43. EPA and CDC. (2011). Comparison of ultrafiltration techniques for recovering biothreat
agents in water. Environmental Protection Agency and the Centers for Disease Control
and Prevention. EPA 600/R-l 1/103.
44. Johansson, A., L. Berglund, U. Eriksson, I. Goransson, R. Wollin, M. Forsman, A.
Tarnvik, and A. Sjostedt. (2000). Comparative analysis of PCR versus culture for
diagnosis of ulceroglandular tularemia. J Clin Microbiol, 38(1): 22-6.
38
-------
45. Walker, R.E., J.M. Petersen, K.W. Stephens, and L.A. Dauphin. (2010). Optimal swab
processing recovery method for detection of bioterrorism-related Francisella tularensis
by real-time PCR JMicrobiolMethods, 83(1): 42-7.
46. Versage, J.L., D.D.M. Severin, M.C. Chu, and J.M. Petersen. (2003). Development of a
multitarget real-time TaqMan PCR assay for enhanced detection of Francisella tularensis
in complex specimens. J Clin Microbiol, 41(12): 5492-5499.
47. Humrighouse, B.W., N.J. Adcock, and E.W. Rice. (2011). Use of acid treatment and a
selective medium to enhance the recovery of Francisella tularensis from water. Appl
Environ Microbiol, 77(18): 6729-32.
48. Berrada, Z.L. and S.R. Telford. (2011). Survival of Francisella tularensis Type A in
brackish-water. Arch Microbiol, 193(3): 223-6.
49. Petersen, J.M., M.E. Schriefer, K.L. Gage, J.A. Montenieri, L.G. Carter, M. Stanley, and
M.C. Chu. (2004). Methods for enhanced culture recovery of Francisella tularensis. Appl
Environ Microbiol, 70(6): 3 73 3 -3 73 5.
50. Barns, S.M., C.C. Grow, R.T. Okinaka, P. Keim, and C.R. Kuske. (2005). Detection of
diverse new Francisella-\ike bacteria in environmental samples. Appl Environ Microbiol,
71(9): 5494-500.
51. Pohanka, M. and P. Skladal. (2009). Bacillus anthracis, Francisella tularensis and
Yersinia pestis. The most important bacterial warfare agents - review. Folia Microbiol,
54(4): 263-72.
52. Berdal, B.P., R. Mehl, H. Haaheim, M. Loksa, R. Grunow, J. Burans, C. Morgan, and H.
Meyer. (2000). Field detection of Francisella tularensis. Scand J Infect Dis, 32(3): 287-
91.
53. Peruski, A.H., L.H. Johnson, 3rd, and L.F. Peruski, Jr. (2002). Rapid and sensitive
detection of biological warfare agents using time-resolved fluorescence assays. J
ImmunolMethods, 263(1-2): 35-41.
54. Grunow, R., P. Miethe, W. Conlan, E.-J. Finke, S. Friedewald, and M. Porsch-
Ozcurumez. (2008). Rapid detection of Francisella tularensis by the immunoaffmity
assay ABICAP in environmental and human samples. JRapidMethods Autom Microbiol,
16(1): 30-54.
55. Grunow, R., A. Kalaveshi, A. Kuhn, G. Mulliqi-Osmani, and N. Ramadani. (2012).
Surveillance of tularaemia in Kosovo, 2001 to 2010. Euro Surveill, 17(28).
56. Huelseweh, B., R. Ehricht, and H.J. Marschall. (2006). A simple and rapid protein array
based method for the simultaneous detection of biowarfare agents. Proteomics, 6(10):
2972-81.
57. Pohanka, M. and P. Skladal. (2007). Piezoelectric immunosensor for the direct and rapid
detection of Francisella tularensis. Folia Microbiol, 52(4): 325-330.
58. O'Brien, T., L.H. Johnson 3rd, J.L. Aldrich, S.G Allen, L.-T. Liang, A.L. Plummer, S.J.
Krak, and A.A. Boiarski. (2000). The development of immunoassays to four biological
threat agents in a bidiffractive grating biosensor. Biosens Bioelectron, 14(10-11): 815-
828.
59. Sharma, N., A. Hotta, Y. Yamamoto, O. Fujita, A. Uda, S. Morikawa, A. Yamada, and K.
Tanabayashi. (2013). Detection of Francisella tularensis-specific antibodies in patients
with tularemia by a novel competitive enzyme-linked immunosorbent assay. Clin
Vaccine Immunol, 20(1): 9-16.
39
-------
60. Gillings, M.R. (2014). Rapid extraction of PCR-competent DNA from recalcitrant
environmental samples. Methods Mol Biol, 1096: 17-23.
61. Whitehouse, C.A. and H.E. Hottel. (2007). Comparison of five commercial DNA
extraction kits for the recovery of Francisella tularensis DNA from spiked soil samples.
Mol Cell Probes, 21(2): 92-6.
62. Klerks, M.M., A.H. van Bruggen, C. Zijlstra, and M. Donnikov. (2006). Comparison of
methods of extracting Salmonella enterica serovar Enteritidis DNA from environmental
substrates and quantification of organisms by using a general internal procedural control.
ApplEnviron Microbiol, 72(6): 3879-86.
63. MoBio Laboratories. (2010). UltraClean® Soil DNA Isolation Kit. MoBio Laboratories,
Inc. p. 21.
64. Bader, D.E., G.R. Fisher, and LJ. McLaws. (2003). Molecular genetic analysis of killed
biological agents in sample unknowns: NATO SIBCA Exercise III. TR-2003-043.
65. Bader, D.E., G.R. Fisher, and LJ. McLaws. (2004). Analysis of a simulated
biological/chemical agent mixture in soil. TM 2004-164.
66. Trombley Hall, A., A. McKay Zovanyi, D.R. Christensen, J.W. Koehler, and T. Devins
Minogue. (2013). Evaluation of inhibitor-resistant real-time PCR methods for diagnostics
in clinical and environmental samples. PLoS One, 8(9): e73845.
67. Ahlinder, J., C. Ohrman, K. Svensson, P. Lindgren, A. Johansson, M. Forsman, P.
Larsson, and A. Sjodin. (2012). Increased knowledge of Francisella genus diversity
highlights the benefits of optimised DNA-based assays. BMC Microbiol, 12: 220.
68. Buzard, G.S., D. Baker, M.J. Wolcott, D.A. Norwood, and L.A. Dauphin. (2012). Multi-
platform comparison often commercial master mixes for probe-based real-time
polymerase chain reaction detection of bioterrorism threat agents for surge preparedness.
Forensic Sci Int, 223(1-3): 292-7.
69. Escudero, R., A. Toledo, H. Gil, K. Kovacsova, M. Rodriguez-Vargas, I. Jado, C. Garcia-
Amil, B. Lobo, M. Bhide, and P. Anda. (2008). Molecular method for discrimination
between Francisella tularensis and Francisella-\ike endosymbionts. J Clin Microbiol,
46(9): 3139-43.
70. Forsman, M., A. Nyren, A. Sjostedt, L. Sjokvist, and G. Sandstrom. (1995). Identification
of Francisella tularensis in natural water samples by PCR FEMS Microbiol Ecol, 16(1):
83-92.
71. Garcia Del Blanco, N., M.E. Dobson, A.I. Vela, V.A. De La Puente, C.B. Gutierrez, T.L.
Hadfield, P. Kuhnert, J. Frey, L. Dominguez, and E.F. Rodriguez Ferri. (2002).
Genotyping of Francisella tularensis strains by pulsed-field gel electrophoresis,
amplified fragment length polymorphism fingerprinting, and 16S rRNA gene sequencing.
J Clin Microbiol, 40(8): 2964-72.
72. Matero, P., H. Hemmila, H. Tomaso, H. Piiparinen, K. Rantakokko-Jalava, L. Nuotio,
and S. Nikkari. (2011). Rapid field detection assays for Bacillus anthracis, Brucella spp.,
Francisella tularensis and Yersinia pestis. Clin Microbiol Infect, 17(1): 34-43.
73. Oyston, P.C., A. Sjostedt, and R.W. Titball. (2004). Tularaemia: bioterrorism defence
renews interest in Francisella tularensis. Nat Rev Microbiol, 2(12): 967-78.
74. Brett, M.E., L.B. Respicio-Kingry, S. Yendell, R. Ratard, J. Hand, G. Balsamo, C. Scott-
Waldron, C. O'Neal, D. Kidwell, B. Yockey, P. Singh, J. Carpenter, V. Hill, J.M.
Petersen, and P. Mead. (2014). Outbreak of Francisella novicida bacteremia among
inmates at a Louisiana correctional facility. Clin Infect Dis, 59(6): 826-33.
40
-------
75. Brinkman, N.E., R. Francisco, T.L. Nichols, D. Robinson, F.W. Schaefer, 3rd, R.P.
Schaudies, and E.N. Villegas. (2013). Detection of multiple waterborne pathogens using
microsequencing arrays. JApplMicrobiol, 114(2): 564-73.
76. Euler, M., Y. Wang, D. Heidenreich, P. Patel, O. Strohmeier, S. Hakenberg, M. Niedrig,
F.T. Hufert, and M. Weidmann. (2013). Development of a panel of recombinase
polymerase amplification assays for detection of biothreat agents. J Clin Microbiol,
51(4): 1110-7.
77. He, J., AJ. Kraft, J. Fan, M. Van Dyke, L. Wang, M.E. Bose, M. Khanna, J.A. Metallo,
and KJ. Henrickson. (2009). Simultaneous detection of CDC category "A" DNA and
RNA bioterrorism agents by use of multiplex PCR & RT-PCR enzyme hybridization
assays. Viruses, 1(3): 441-459.
78. Janse, I, R.A. Hamidjaja, J.M. Bok, and BJ. van Rotterdam. (2010). Reliable detection
of Bacillus anthracis, Francisella tularemis and Yersinia pestis by using multiplex qPCR
including internal controls for nucleic acid extraction and amplification. BMC Microbiol,
10: 314.
79. Janse, I, J.M. Bok, R.A. Hamidjaja, H.M. Hodemaekers, and B.J. van Rotterdam. (2012).
Development and comparison of two assay formats for parallel detection of four biothreat
pathogens by using suspension microarrays. PLoS One, 7(2): e31958.
80. Jeng, K., J. Hardick, R. Rothman, S. Yang, H. Won, S. Peterson, Y.H. Hsieh, B.J. Masek,
K.C. Carroll, and C.A. Gaydos. (2013). Reverse transcription-PCR-electrospray
ionization mass spectrometry for rapid detection of biothreat and common respiratory
pathogens. J Clin Microbiol, 51(10): 3300-7.
81. McAvin, J.C., M.M. Morton, R.M. Roudabush, D.H. Atchley, and J.R. Hickman. (2004).
Identification of Francisella tularensis using real-time fluorescence polymerase chain
reaction. Mil Me d, 169(4): 330-3.
82. O'Connell, K.P., P.E. Anderson, J.J. Valdes, and J.R. Bucher. (2004). Testing of the Bio-
SEEQ® (Smiths detection handheld PCR instrument): sensitivity, secificity, and effect of
interferents on Francisella tularensis assay performance. Aberdeen Proving Ground,
MD: U.S. Army Research, Development and Engineering Command. ECBC-TR-380.
83. Rachwal, P.A., H.L. Rose, V. Cox, R.A. Lukaszewski, A.L. Murch, and S.A. Weller.
(2012). The potential of TaqMan Array Cards for detection of multiple biological agents
by real-time PCR PLoS One, 7(4): e35971.
84. Schweighardt, A.J., A. Battaglia, and M.M. Wallace. (2014). Detection of anthrax and
other pathogens using a unique liquid array technology. J Forensic Sci, 59(1): 15-33.
85. Seiner, D.R., H.A. Colburn, C. Baird, R.A. Bartholomew, T. Straub, K. Victry, J.R.
Hutchison, N. Valentine, and C.J. Bruckner-Lea. (2013). Evaluation of the FilmArray®
system for detection of Bacillus anthracis, Francisella tularensis and Yersinia pestis. J
Appl Microbiol, 114(4): 992-1000.
86. Yang, Y., J. Wang, H. Wen, and H. Liu. (2012). Comparison of two suspension arrays for
simultaneous detection of five biothreat bacterial in powder samples. J Biomed
Biotechnol, 2012: 831052.
87. Day, J.B. and R.C. Whiting. (2009). Development of a Macrophage Cell Culture Method
To Isolate and Enrich Francisella tularensis from Food Matrices for Subsequent
Detection by Real-Time PCR. J Food Protect, 72(6): 1156-1164.
88. Kane, S.R., S.E. Letant, G.A. Murphy, T.M. Alfaro, P.W. Krauter, R. Mahnke, T.C.
Legler, and E. Raber. (2009). Rapid, high-throughput, culture-based PCR methods to
41
-------
analyze samples for viable spores of Bacillus anthracis and its surrogates. JMicrobiol
Methods, 76(3): 278-84.
89. Letant, S.E., G.A. Murphy, T.M. Alfaro, J.R. Avila, S.R. Kane, E. Raber, T.M. Bunt, and
S.R. Shah. (2011). Rapid-viability PCR method for detection of live, virulent Bacillus
anthracis in environmental samples. ApplEnviron Microbiol, 77(18): 6570-8.
90. Rydzewski, K., T. Schulz, E. Brzuszkiewicz, G. Holland, C. Luck, J. Fleischer, R.
Grunow, and K. Heuner. (2014). Genome sequence and phenotypic analysis of a first
German Francisella sp. isolate (W12-1067) not belonging to the species Francisella
tularensis. BMC Microbiol, 14(1): 169.
42
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |