Electric DNA arrays for determination of pathogenic ...

Electric DNA arrays for determination of pathogenic Bacillus cereus

Yanling Liu

School of Biotechnology

Royal Institute of Technology

Stockholm 2007

© Yanling Liu Stockholm 2007 Department of Bioprocess Technology School of Biotechnology Royal Institute of Technology 106 91 Stockholm Sweden ISBN: 978-91-7178-655-5 TRITA-BIO-Report 2007:4 ISSN 1654-2312

Electric DNA arrays for determination of pathogenic Bacillus cereus – Yanling Liu

Yanling Liu (2007): Electric DNA arrays for determination of pathogenic Bacillus cereus.

School of Biotechnology, Royal Institute of Technology (KTH), Stockholm, Sweden

Abstract

Silicon-based electric chip arrays were developed for characterization of Bacillus

cereus with respect to the capacity to produce toxins involved in food poisoning and

foodborne infections. Bacteria of the B. cereus group contain different sets of four

toxins encoded by eight genes. The purpose of this work was to develop a fast method

for determination of the presence of these genes in colonies from primary enrichment

cultures. The specific DNA detection was based on immobilization of DNA capture

probes, which hybridize to specific sites on the target genes. Biotin-labeled detection

probes were designed to hybridize with the target DNA adjacent to the capture

probes. An extravidin - alkaline phosphatase complex was subsequently bound to the

hybridized detection probes. Finally, p-aminophenyl phosphate was added as

substrate for the enzyme, and the product p-aminophenol was brought in contact with

the interdigitated gold electrode on the silicon chips surface. The p-aminophenol was

oxidized at the anode to quinoneimine, which was then reduced back to p-

aminophenol at the cathode. This redox recycling generates a current that was used as

the DNA-chip response to the target DNA. Two versions of the assay were used. In

the first version the capture probes were immobilized on magnetic beads and all

chemical reactions until and including the enzymatic reaction took place in an

eppendorf tube while the redox recycling was used to measure the amount of p-

aminophenol produced after transfer from the tube to the silicon chip surface. In the

second version a silicon chip array was used with 16 parallel electrode positions, each

activated by immobilization of one type of capture probes on the gold electrodes.

With this system all chemical reactions took place at the chip surface. The kinetics of

cell disruption and DNA fragmentation from B. cereus by ultrasonication was

determined. Maximum cell disruption was achieved within 5 min and the chip

response increased in proportion to the ultrasonic time. Further ultrasonication up to

10 min resulted in further increasing current although no further cell disruption was

observed. If the sonication time was extended above 10 min the signal declined.

Based on analysis of the DNA size distribution by early end-point PCR and gel


electrophoresis, it is suggested that the first 5 min ultrasonication increased the signal

by increasing the release of target DNA molecules. Thereafter the signal was

increased by fragmentation of target DNA which increases the diffusion rate and also

the accessibility of the hybridization site. Finally, the DNA fragment sizes approached

that of the hybridization site (51-bp) which may reduce the signal because of cleavage

of the target DNA in the hybridization region. These studies were performed with the

bead-based hybridization assay. The assay was highly specific to the target gene

(hblC) of both B. cereus and B. thuringiensis with no response from negative control

cells of B. subtilis. The 16 positions of the silicon chip array were activated by

immobilization of all known toxin-coding genes of B. cereus and also included both a

positive control and a negative control electrode positions. When these chips were

exposed to ultrasonicated B. cereus, the gold electrodes were fouled by some

component in DNA cell lysates. To circumvent this, the released large DNA was first

extracted and then ultrasonicated again, since the extract mainly contains large

molecular weight DNA. This DNA extract was applied to characterize one “diarrheal”

and one “emetic” strain of B. cereus with the DNA chip arrays. The results agreed

with PCR control analysis which means that these electric DNA chip arrays can be

used to characterize bacterial colonies with respect to the genes coding of all known

toxins of B. cereus: haemolysin (hblA, hblC, hblD), non-haemolytic enterotoxin

(nheA, nheB, nheC), cytotoxin K-2 (cytK-2), and cereulide (ces). The chip assay

required about 30 min after application of DNA samples. Due to the generic

properties of the chips, this technique should also be applicable for characterization of

the pathogenicity potential of many other organisms.

Keywords: Bacillus cereus, haemolysin, non-haemolytic enterotoxin, cytotoxin K-2,

cereulide, toxin-coding genes, bacterial colony, electric DNA chip, ultrasonication,

DNA fragmentation.


List of publications

This thesis is based on the following papers, which in the text will be referred to by their

Roman numerals:

I. Magdalena Gabig-Ciminska, Yanling Liu, Sven-Olof Enfors. Gene-based

identification of bacterial colonies with an electric chip. Analytical

Biochemistry 345 (2005) 270 – 276.

II. Yanling Liu, Bruno Elsholz, Sven-Olof Enfors and Magdalena Gabig-

Ciminska. Confirmative electric DNA array-based test for food poisoning

Bacillus cereus. Journal of Microbiological Methods. Accepted.


Table of contents AIM OF THE STUDY 1 INTRODUCTION 2 Bacillus cereus group and classification 2 B. cereus pathogenicity 3 Toxin detection methods 4 DNA-based detection methods 5 The concept of DNA chip 6 DNA chip classification and characteristics 7 Electrochemical DNA chip 8 PRESENT INVESTIGATION 10 Gene-based identification of bacterial colonies with an electric chip (paper I) 10 The detection principle 10 DNA oligonucleotide-functionalized beads 11 BBSH and electrical signal generation 11 Results 13 Conclusions 15 Confirmative electric DNA array-based test for food poisoning Bacillus cereus (paper II) 17 The DNA chip array and its activation by eight toxin sensing probes 17 The instrument and assay program 17 Results 19 Conclusions 23 CONCLUDING REMARKS 25 ABBREVIATIONS 27 ACKNOWLEDGEMENTS 28 REFERENCES 29


1

Aim of the study

In order to develop a method for fast characterization of pathogenic microorganisms,

an electric DNA chip array-based detection method is investigated. Firstly, an

efficient cell lysate sample preparation from bacteria must be established. This

method must release the genomic DNA from cells and also fragment the DNA to

produce molecules that can diffuse and hybridize at the chip surface. Secondly, since

several genes usually are involved in the pathogenicity of bacteria, a chip array with

the capacity to determine several genes simultaneously should be developed. Bacillus

cereus, which is known to produce up to four different toxins, was chosen as a model

for the study. The regulatory demands on the protocols for analysis of many

foodborne pathogens require detection of a single cell in 25 grams of food samples.

This means that a primary enrichment cultivation must be included before a

characterization. In most analytical protocols this enrichment results in a bacterial

colony. For this reason a Bacillus cereus colony was chosen as the target for the DNA

chip and therefore also the number of cells in a colony should be determined to get a

measure of the demand on sensitivity of the DNA chip.


2

Introduction

Bacillus cereus group and classification

Bacillus cereus are ubiquitous in the environment and also in food products. They are

gram-positive, aerobic spore forming bacteria. Vegetative cells are normally large and

rod-like (Fig. 1A). They can survive in different adverse conditions, e.g. during food

processing, because of their ability to form resistant endospores (Fig. 1B).

Figure 1. A microscopic image of vegetative cells (A) and endospores (B) of B. cereus ATCC14579.

B. cereus has been reported to be an important pathogen responsible for food

poisoning and food spoilage (Pruss, Dietrich et al. 1999; Ghelardi, Celandroni et al.

2002; Thaenthanee, Wong et al. 2005). Six different species have been classified and

characterized in the Bacillus cereus group. Bacillus anthracis, a primary virulent

pathogen for animals, possesses two specific virulence-encoding plasmids coding for

a capsule and the anthrax toxin; Bacillus thuringiensis, a pathogen for some insects,

produces parasporal crystal proteins encoded by its plasmid and has been

commercialized for the use as an insecticide; the rhizoid colony forming Bacillus

mycoides (Nakamura 1998) has been classified from other bacteria of B. cereus group

due to their different fatty acid composition; Bacillus pseudomycoides (Nakamura

1998) recently was separated from Bacillus mycoides due to their distinct genotypes,

but they are not distinguishable by their phenotypes; Bacillus weihenstephanensis

(Lechner, Mayr et al. 1998) is classified by its psychrotolerant properties, and grows

at 4 - 7 °C, not at 43 °C; Bacillus cereus, a common soil inhabitant and often

identified from a variety of foods, including grains, dairy products and meat, is a

foodborne pathogen and its toxins are encoded by both the chromosome and the

endospore

B. A.


3

Foodborne pathogenic B. cereus species

hblA hblC hblD

haemolytic toxin HBL

cytK

cytotoxin K

nheA nheB nheC

nonhaemolytic toxin NHE

Enterotoxins

Ces cluster

cereulide peptide

Emetic toxin

plasmids. This organism grows in a broad temperature range of 8 - 55 °C, and

optimally around 28 - 35 °C. B. cereus and B. thuringiensis are very similar according

to their genomics (Priest, Barker et al. 2004). They can only be distinguished by

production of an intracellular crystal protein from the B. thuringiensis plasmid.

B. cereus pathogenicity

B. cereus can secrete two main types of toxins, enterotoxins (Hansen and Hendriksen

2001) and emetic toxin (Paananen, Mikkola et al. 2002), which cause foodborne

human illness (Fig. 2) (Lund and Granum 1996; Ghelardi, Celandroni et al. 2002;

Schoeni and Wong 2005).

Figure 2. Schematic display of different food-poisoning toxins secreted by B. cereus and their responsible genes, respectively.

The enterotoxins consist of the haemolytic toxin BL (HBL), the nonhaemolytic toxin

(NHE) and cytotoxin K (cytK). Both HBL and NHE are enterotoxin complexes

composed of three proteins (Beecher and Macmillan 1991; Lund and Granum 1996):

HBL is encoded cooperatively by the genes hblA, hblD and hblC in one operon; NHE

is encoded cooperatively by the genes nheA, nheB and nheC. HBL and NHE are

responsible for human diarrhoeal illness. This type of illness is normally mild and can

last less than 24 h. Cytotoxin K is a cytotoxic pore-forming protein encoded only by

one gene (Fagerlund, Ween et al. 2004). It is necrotic, haemolytic and very toxic to

human intestinal epithelial cells. A severe outbreak involved by cytotoxin K in 1998

led to the death of three people. Recently two different forms, encoded by cytK-1 and

cytK-2 genes, have been described (Fagerlund, Ween et al. 2004). CytK-1 shows 89%

protein sequence homology with that of cytK-2, but carries much higher toxicity.

These toxins have been evaluated and studied well by different research groups until

now. BceT and entFM, which before were believed to express enterotoxins, are now

not considered to responsible for such diseases (Choma 2002; Hansen, Hoiby et al.


4

2003). Cereulide, a cyclic dodecadepsipeptide (Andersson 1998), can cause acute

human emesis after 1 to 5 h of its ingestion. This peptide is a mitochondrion-toxin and

synthesized by a non-ribosomal peptide synthetase (NRPS) from B. cereus (Horwood,

Burgess et al. 2004). This small peptide is very thermo-stable and inert, can tolerate

heat, protease, acid, alkali, even autoclaving. So once it’s secreted by the food

contaminating B. cereus, cereulide can’t be inactivated by the food processing and

brings a risk to consumers.

Around 60% of B. cereus carry HBL genes and most of them carry NHE genes, while

cereulide - producing B. cereus are relatively rare. B. cereus bacteria are commonly

present in raw and processed foods. They can survive different food treatments by

forming spores, and pre-formed cereulide can’t be degraded during food processing.

These facts pose a potential food safety problem. It’s normally believed that food

containing > 103 B. cereus/g is not safe enough for consumption (Rosenquist, Smidt et

al. 2005). Furthermore, we need to know accurately what types of toxins can be

expressed by the contaminating B. cereus bacteria of different foods. An additional

food safety problem is the large similarities between the species in the B. cereus

group. The B. cereus enterotoxins are produced by several other Bacillus species, so it

is very important to identify the pathogenic potential of the bacteria of the B. cereus

group when they are found in food, rather than the phenotypic properties that are

determined by the traditional cultivation protocols.

Toxin detection methods

To date some methods have been developed to detect different toxins secreted by B.

cereus. For diarrhoeal toxin detection, a gel diffusion assay for HBL is based on a

distinct serological entity of ring shaped haemolysis (Rhodehamel, Harmon et al.

2001); biology-based methods for the detection of diarrhoeal toxins, include vascular

permeability reaction assay, different cell lines cytotoxicity assay. Based on

immunochemical reactions, two diarrhoeal enterotoxin immunoassay kits were

commercialized, OXOID Reverse Passive Latex Agglutination (BCET-RPLA) for the

L2 component (encoded by the hblC gene) detection of the HBL complex, and

TECRA Visual Immunoassay for the 41kDa protein component (encoded by the nheA

gene) detection of the NHE complex. For emetic toxin detection, the test of motility


5

loss of boar spermatozoa (Andersson 1998), Hep-2 cell culture-based MTT

conversion assay (Finlay, Logan et al. 1999), a chemical assay for cereulide based on

HPLC-MS (Haggblom, Apetroaie et al. 2002), and a rat liver mitochondrial

respiratory uncoupling activity assay (Kawamura-Sato, Hirama et al. 2005) have been

developed. These methods have some internal limitations: time-consuming and

laborious procedure, detection not specific for each individual function proteins.

Recently, antibodies against each of the three components of the B. cereus HBL and

NHE complexes (Dietrich 1999; Dietrich, Moravek et al. 2005) have been produced

and characterized, so the detection for each individual toxic protein can be realized.

But no antibodies against cytK or cereulide have been reported until now.

DNA-based detection methods

Due to the dramatic development of genome-wide sequencing techniques, a variety of

organism’ genomes or genes have been sequenced. Genes (hblA, hblC, hblD)

responsible for HBL toxin, genes (nheA, nheB, nheC) responsible for NHE toxin,

genes cytK-1, cytK-2 and the gene ces responsible for emetic toxin cereulide, all have

been fully sequenced and can be downloaded from several major database websites

freely. The genetic sequence-based methods have impelled the identification of

pathogenic microorganisms and types of the secreted toxins from the phenotypic level

to the genotypic molecular level.

Southern blotting, but especially PCR-based methods for all eight genes, i.e. hblA,

hblC, hblD, nheA, nheB, nheC, cytK, ces, have been intensively explored (Mantynen

and Lindstrom 1998; Hansen and Hendriksen 2001; Stenfors, Mayr et al. 2002;

Ehling-Schulz, Fricker et al. 2004; Ehling-Schulz, Guinebretiere et al. 2006;

Guinebretiere, Fagerlund et al. 2006). These methods are much faster and highly

sensitive when compared with conventional ones. They can differentiate particular

genes coding for each type of toxin specifically, even small differences between genes

representing different forms or variants of the same type, e.g. cytK-1 and cytK-2

(Guinebretiere, Fagerlund et al. 2006), which can’t be fulfilled by any other methods.

More recently, different DNA chips or sensors have been derived and applied for the

detection of foodborne or clinical pathogens (Gabig-Ciminska, Liu et al. 2005;

Elsholz, Worl et al. 2006). These DNA chip-based methodologies provide potentially


6

fast, simple and highly automated detection. Since these DNA-based methods detect

the toxin related genes carried by the food contaminating bacteria, the presence of

these genes does not indicate the expression and high producing level of toxins by the

bacteria (Ivanova, Sorokin et al. 2003; Slamti, Perchat et al. 2004). B. cereus toxin

expression is also regulated by its PlcR (Agaisse, Gominet et al. 1999). Thus DNA-

based analysis only shows that the cells have the capacity to produce the toxins and

therefore are potential food-poisoning pathogens. Moreover, it is reported that the

enterotoxins, except emetic toxin, produced by B. cereus are subject to inactivation in

the human gut, so the gastrointestinal diseases caused are not directly related to the

presence and ingestion of enterotoxins in food; but rather to toxin production by the

proliferation of vegetative cells germinated from B. cereus spores in the small

intestine (Jaaskelainen, Haggblom et al. 2004). On the other hand, enterotoxins and

emetic toxin are not produced by B. cereus present in food at any circumstance. For

example, the production of the emetic toxin cereulide is induced at the end of the

exponential growth phase and enhanced by agitation (Haggblom, Apetroaie et al.

2002), enough oxygen (Jaaskelainen, Haggblom et al. 2004) and mild temperatures

(15 - 30 °C). Its production is low or undetectable at temperatures below 8 °C or

above 37 °C. Also the production of HBL and NHE toxins is strongly influenced by

the bacterial growth rate and carbohydrate (Ouassila, Thierry et al. 2006). The

expression of food-poisoning toxins is regulated by the bacterial metabolism and

stimulated at certain conditions (Fermanian, Lapeyre et al. 1997). This limits the

application of the phenotype-based detection methods, but do not influence that of the

genomic sequence-based detection. Thereby the DNA chip-based detection method

has become a promising tool to detect all existing disease-causing toxin encoding

genes and their carriers, and assess the risk level of implicated food statistically.

When combined with other assays, it can be used for the differential toxigenicity

study among all such gene carrying bacteria.

The concept of DNA chip

DNA chips are developed from the integration of multidisciplinary fields, such as

molecular biology, chemistry, bioinformatics, microelectronics, microfabrication and

automation technologies, and other fields from the early 90s (Fodor, Read et al. 1991;

Homs 2002). A DNA chip is generally produced by the immobilization of single-


7

stranded DNA (ss-DNA) probes or mimics on the solid-phase with the coordination of

a transducer. Its detection scheme is: attached sensing probes can recognize and

hybridize with the target DNA or RNA in a sample solution specifically according to

the nucleotide base pairing, then this binding event can be translated to a detectable

signal by the transducer.

DNA chip classification and characteristics

DNA chips can be divided into high-density DNA microarrays and low-density DNA

arrays/chips/sensors approximately. High-density DNA microarrays are specialized

for high-throughput screening analysis (Schena, Shalon et al. 1995; Chee, Yang et al.

1996; Panda, Sato et al. 2003), mainly applied in the genome-wide DNA sequencing,

gene expression profiling and genetic disease related areas. Low-density DNA chips

offer simple, fast procedures and reusability for the limited number of certain

sequence analysis. Up to date, various types of DNA chips have been established,

which cater applications in different circumstances, respectively. Otherwise stated,

DNA chips mentioned later refer to the low-density format. Detecting events can be

measured by the change of optical, gravimetric, electrical and other detectable signals

according to different transducers, before and after the application of an analyte onto

the sensing surface. Present optical chips are mainly based on fluorescence,

chemiluminescence, colorimetry and surface plasmon resonance (SPR) detection

techniques (Cheek, Steel et al. 2001; Broude 2002; Wirtz 2003; Zezza, Pascale et al.

2006). Optical chips are very sensitive (Hoang A. Ho 2005), and suitable for parallel

DNA sequence analysis in the same sample, e.g. DNA microarrays (Cheung VG,

Morley M et al. 1999). But they normally need sophisticated and expensive

instrumentations and are mainly used in a laboratory. On the other hand, DNA

analytes need to be extracted purely from crude samples or non-nucleic acid

contaminants before applied to an optical chip, so no obvious photo-induced

background from samples can interfere in the final signal. Typical gravity-based DNA

sensing instrument includes quartz crystal microbalance (QCM) and microcantilever

(Hansen, Ji et al. 2001; Minunni, Tombelli et al. 2005). This type of chips needn’t

target labeling step for final signal generation and can be re-used multiple times.

Meanwhile, they are sensitive and can carry out the real-time monitoring of target

DNA binding reactions. Their primary limitations are: delicate and expensive


8

Capacitive signal Conductive signal Amperometric or potentiometric signal

instruments are needed; mainly suitable for the laboratory research and careful

operation is necessary. DNA chips, which finally generate the electrical readout, are

called electrochemical chips. Up to now, different nanometer-scaled particles have

been introduced to the established DNA sensing systems (So-Jung Park 2002),

different nanowires and nanotubes have also emerged as new types of DNA sensors

(Bunimovich, Shin et al. 2006), all these innovations are in order to amplify final

signals and improve the detecting sensitivity or specificity.

Electrochemical DNA chip

Recently, increasing interest has been drawn to electrical signal-generating DNA

chips (Fig. 3), many related review articles have been written by worldwide research

groups (Wang 2000; Albers, Grunwald et al. 2003; Drummond, Hill et al. 2003;

Kagan Kerman 2004; Wang 2006).

target DNA-binding

events being sensed readout Transducers

Figure 3. The general working principle of an electrochemical chip-based DNA detection. Also a number of electrochemistry-based signaling strategies and instruments have

been rapidly developed with distinct characteristics. Overall, these DNA chips have

been reported to exhibit the following promising features: low-cost platform and

simple operation, no need for expensive and complicated instrumentation, e.g. silicon-

based sensors (Gabig-Ciminska, Liu et al. 2005; Elsholz, Worl et al. 2006); simple

analyte treatment from crude or multi-component samples, resistant to non-nucleic

acid contaminants (Lee and Hsing 2002; Zwirglmaier, Ludwig et al. 2004; Liao,

Mastali et al. 2006; Lubin, Lai et al. 2006); high signal detection sensitivity and

specificity (Zhang, Pothukuchy et al. 2004; Liao, Mastali et al. 2006); instrument

integration and automation, the whole process from sample introduction to final signal

generation can be realized automatically without manual aid (Liu, Yang et al. 2004;

target DNA

capture probe


9

Elsholz, Worl et al. 2006); instrument miniaturization and portability, insensitivity to

the surrounding environment are suitable for point-of-care analysis (Wang 2002; Liu,

Yang et al. 2004; Wang 2006). All these offer the electrochemical DNA chip the

potential candidate for realistic samples tests from environmental, clinical, food and

other sources, and for genetic or infectious disease diagnosis, genetically modified

organisms detection, foodborne pathogenicity detection, DNA – other molecules

interaction detection and other utilities (Boon EM 2002; Drummond, Hill et al. 2003;

Wang 2006). Label-free and label-based detections are two main design directions in

this field. For label-free detections, different strategies based on DNA electrical

properties and reduction/oxidation of purine bases have been used to signal the target

capturing (Boon EM 2002; Ozkan, Erdem et al. 2002). For label-based detections,

many sensitive methods have been developed: DNA specific redox-active label

detection (Boon, Ceres et al. 2000; Kara, Kerman et al. 2002), sensing probe – target

sequence – enzyme labeled detection probe sandwich hybridization based techniques

(Umek, Lin et al. 2001; Gabig-Ciminska, Holmgren et al. 2004), metal nanoparticle

tagged detection (Peng, Soeller et al. 2006). Compared with label-free techniques,

much higher detection sensitivity and sequence selectivity can be achieved by label-

based techniques (Boon, Ceres et al. 2000; Zhang, Pothukuchy et al. 2004).

Amperometric transducer is one of the most commonly used in combination with

electrochemical chips. In the present investigation, an amperometric DNA chip

coupled with the enzyme-labeled sandwich detection method has been employed.


10

Present investigation

Gene-based identification of bacterial colonies with an electric chip (paper I)

In order to develop a fast DNA sensor-based method for the confirmative analysis of

pathogenic bacteria from enrichment cultures, i.e. bacterial colonies or liquid cultures,

without additional target DNA amplification step, different solutions were explored: a

simple and efficient crude sample treatment for the detecting sequence site of target

DNA to be more accessible by sensing probes; a functionalized solid surface design

for immobilized sensing probes to recognize and capture target DNA kinetically and

therefore enhance hybridization efficiency; a sensitive transduction strategy for final

signal amplification.

The detection principle

In this paper, a micro bead-based DNA sensor exhibits a large contact surface and

high-density of capture probes for sensing events to take place. The gene hblC was

chosen as target for the analysis. Signal amplification was realized by the p-amino

phenyl phosphate (pAPP) hydrolytic reaction of the captured enzyme in a tube and

following electric redox recycling by transferring the generated p-aminophenol (pAP)

to the silicon chip surface. The whole analysis procedure is illustrated in figure 4.

Figure 4. The schematic illustration of the BBSH-based identification procedure for bacterial colonies with an electric chip.

Addition of detection probes B. cereus DNA analyte preparation by ultrasonication and centrifugation Denaturation Addition of functionalized beads of ds–DNA analytes by 95 °C water bath BBSH reaction in an eppendorf tube Addition of the extravidin-ALP Magnetic collection and washing of reacted beads Enzyme conjugate binding to target Washing of the beads DNA-probes hybridization complexes Addition of substrate pAPP and enzymatic reaction in an eppendorf tube Transfer of resulting pAP solution to silicon chip electrodes and electric signal generation by redox recycling


11

This work focused on the sample treatment and its effect on the signal generation

level. An ultrasound technique can be applied for cell disruption, release and

fragmentation of genomic or plasmid DNA. Ultrasonic disruption efficiency is

influenced by different parameters, such as output power, ionic strength, ultrasound

duration, amount and volume of bacteria samples, and also bacterial age (Fykse,

Olsen et al. 2003; Mann and Krull 2004). In order to obtain target DNA samples with

expected characteristics, applied ultrasonic time should be adjusted according to each

specific operation parameters and sample situations. It is believed that the longer an

ultrasonic disruption time is applied for cells, the more DNA is released out and

shorter sizes DNA is fragmented. Gel electrophoresis picture with the smeared

staining DNA band was used to estimate overall resulted DNA amount and average

lengths. One emerging problem is to define the correlation of fragmentation to the

hybridization of particular DNA sequences carrying the detecting sites. Furthermore,

the influence on disintegration of detecting sequence sites and different parameters’

contribution to the signal generation level during ultrasonication need to be clarified.

DNA oligonucleotide-functionalized beads

The paramagnetic beads with a diameter of 2,8 µm, which were coated by carboxylic

acid group, were activated by immobilization with ss-hblC capture probe via the

covalent binding by the amino group labeled at the 5’ end of the capture probe to the

bead surface. This functionalized magnetic bead provides a large sensing surface for

coupled hblC capture probes to recognize and hybridize with the haemolysin L2

component encoding gene hblC present in cell lysate DNA samples.

BBSH and electrical signal generation

The prepared supernatant of ultrasonicated B. cereus was mixed well with the 3’

biotin-labeled ss–hblC detection probe and followed by denaturation of the ds-target

DNA at 95 °C water bath. After that, the denatured solution was immediately

transferred to the magnetic beads with immobilized capture probes for bead-based

sandwich hybridization (BBSH) (Fig. 5A), which means simultaneous hybridization

of target DNA with the capture and detection probes. The beads were then washed

and extravidin – alkaline phosphatase conjugate (extravidin-ALP) was applied to bind

to the detection probe complex via the strong avidin-biotin binding affinity. After

another washing of the beads, the substrate pAPP was added to be hydrolyzed to the


12

NH2

O

PO32-

NH2

OH

NH

O

ALP !2H+, !2e

!PO32-

+2H+, +2e

pAPP pAP quinoneimine

redox recycling on the electrodes

(-50 mV) (+250 mV)

redox active product pAP (Fig. 5B). Then this resulting solution was transported to an

electrical silicon chip and the pAP – quinoneimine redox recycling (Fig. 5B) was

triggered on the chip electrode surface under applied potentials to generate an

electrical signal (Fig. 5C).

Figure 5.

A. The schematic illustration of the BBSH-based target DNA detection principle.

B. The principles of the

substrate pAPP enzymatic reaction and the pAP redox

recycling on the chip electrode surface.

C. The schematic illustration of the BBSH single-electrode chip and the electric signal generation.

pAPP

pAP

biotin - labeled detection probe

extravidin-ALP

target

capture probe

electric signal generation

1-2 µm

+ 250 mV

- 50 mV

1000 µm

0,8 µm

pAP -

+

single-IDA (interdigitated arrays) electrode chip

Time (sec)

Current (nA)

Background Target DNA signal

signal summation

anode signal

cathode signal


13

Results

Ultrasonication technique was employed to disrupt B. cereus cells from one colony,

release and fragment its genomic DNA for the detection of hblC gene by using BBSH

and an electric chip. Different electric signals were generated when cell lysates treated

with different ultrasonic time were applied (Fig. 6). Electric signals increased

continuously during extended ultrasonication time up to 10 min. When even longer

ultrasonication, 13 min, was applied, the electric signal started to decline. The

maximal signal after 10 min sonication and the fact that the protein release reached a

maximum after 5 min (Fig. 7) raised the hypothesis that part of the signal increase

was due to improved hybridization of the target DNA with increased sonication time.

Figure 6. Effects of different ultrasonic time on the electric signal from analysis of the hblC gene of one B. cereus colony.

In order to study the effect of cell ultrasonication on hybridization dynamics, we need

to understand the situation of targeting-site containing DNA sequences after different

ultrasonic time. Three methods, i.e. protein release analysis (Fig. 7), agar gel

electrophoresis (Fig. 8), semiquantitative early endpoint PCR analysis (Fig. 9), were

used in this work. From protein release analysis, the maximal cell disintegration was

Figure 7. Analysis of the protein amount released from single B. cereus colony by the ultrasonic disruption as the function of time applied.

achieved within 5 min. DNA pattern (Fig. 8) showed that the average size of the

released genomic DNA gradually declined with increasing ultrasonication time. After


14

2,5 min ultrasonication, very little large-molecule DNA of more than 2000-bp can be

observed, instead DNA fragmentation took place dominantly; more than 5 min

treatment shortened most DNA to sizes in the range of 200 – 600 bp.

Figure 8. The fragment distribution image of B. cereus genomic DNA treated by different ultrasonic time in the agar gel.

An alternative method to analyze the DNA fragment sizes is the semiquantitative

early endpoint PCR analysis. The traditional PCR is completed at the plateau phase.

Analysis taken at this saturated phase, i.e. endpoint analysis, will not reflect the initial

amount of target DNA proportionally. Analysis taken at the earlier phase (exponential

phase), i.e. early endpoint PCR analysis, can represent the initial target DNA number

more accurately. Since only ordinary PCR was carried out, analysis was not so precise

as that of the real-time PCR and then is called semiquantitative analysis. In this

experiment, different PCR primer pairs were designed (Fig. 10).

Figure 10. Illustration of the six primer pairs targeting sites along the target hblC strands and the resulting sizes of amplicons used in the early endpoint PCR analysis.

The targeting site of all upper primers was equivalent to that of the capture probe.

sonication time

1’ 30’’ 2.5’ 5’ 10’ 13’ nt

8000 2000 1000 750 500 400 300 200 100


15

Targeting sites of lower primers were arranged downstream in a distance from upper

primers according to the different sizes of resulting PCR amplicons, which encompass

the hybridization sites (51-bp) of the capture probe (25-bp) and the adjacent detection

probe (25-bp). Any cut in the hybridization or PCR-amplified regions causes the

reduction of the amplicon production of different sizes. The kinetics of both cellular

DNA release and DNA fragmentation regarding to the detecting sequence site during

the ultrasonic processing can be characterized by using this early endpoint PCR

method (Fig. 9).

Figure 9. The relative amount and sizes distribution of targeting - site carrying DNA strands according to early endpoint PCR analysis after treatment with different ultrasonic time.

Initially, all target DNA fragments with sizes in the range of 75 – 911 bp increased in

number. Then, this number declined with the prolonged ultrasonication time.

However, the smaller the DNA size, the slower and less pronounced was the size

reduction. The largest DNA fragments were almost eliminated after 10 min, while

more than 80% of the smaller target DNA fragments remained.

When analyzing genomic DNA one cell usually corresponds to one copy of the gene.

In order to describe the sensitivity of the assay, the amount of target DNA was

measured as the number of carrying cells. For this purpose flow cytometry was used

to count the number of cells in one B. cereus agar colony that were used for the DNA

chip-based detection. A typical 24 h B. cereus colony with the average diameter of

3,5±0,5 mm on the nutrient agar plate contained the average (5±1,4) × 107 cells.

Conclusions

When ultrasound is applied to prepare a cell lysate DNA sample from one B. cereus

colony for the electric chip-based DNA detection, the first 2,5 – 5 min of treatment


16

disrupts the bacterial cells and releases the genomic DNA. This is also reflected by an

increasing signal from the chip. However, continued ultrasonication mainly reduces

the DNA fragment sizes and also further increases the signal. This increasing signal is

probably due to the increased diffusion rate of the shortened DNA to the probes. It

may also provide less steric hindrance for the access of probes and hybridization. The

maximal signal is obtained from the target DNA treated with 10 min ultrasonication,

which might be caused by the improved hybridization efficiency of overall much

shortened target DNA strands, even though the total target DNA number declines.

The signal decreasing after 13 min treatment, which was observed in all experiments,

may be due to over-fragmentation of target DNA molecules by the extended

ultrasonication. At this sonication time most DNA fragments were in the range of 200

– 500 bp, while the total length of the targeting site was 51-bp. Thus, it is plausible to

assume that an increasing part of the target DNA received fragmentation in the

hybridization site and therefore did not produce a signal. So for the DNA sensor-

based gene detection, the optimized cell lysate DNA preparation should provide not

only a maximal number of released target DNA, but also shortened sizes without too

much detecting site disintegration of the target DNA.

This work shows that it is possible to analyze the presence of a gene in one single

bacterial colony with only ultrasonication and centrifugation as sample pretreatment.


17

Confirmative electric DNA array-based test for food poisoning Bacillus cereus

(paper II)

In paper I a single-electrode silicon chip was used to detect one of the eight known

pathogenicity coding genes (hblC) of B. cereus. It was shown that the assay principle

could be applied directly to the supernatant of disintegrated cells without any DNA

amplification. To extend the principle of the electric DNA-chip analysis to real

applications, arrays with simultaneous analysis of multiple genes in a sample was

developed and presented in paper 2. In this work, a 16-electrode silicon DNA chip

array was used with the corresponding array analyzer for automatic control of all

assay moments after the application of samples containing the target DNA.

The DNA chip array and its activation by eight toxin sensing probes

Each DNA chip array contained sixteen gold electrode positions (Fig. 11A). These

electrode surfaces were activated by immobilizing capture probes furnished with 5’

thiol group for self-assembling on the gold. 14 of these electrode positions were

functionalized by capture probes for all eight types of toxin-coding genes of B. cereus

(hblA, hblC, hblD, nheA, nheB, nheC, cytK-2, ces) on one chip array by a random

localization. One electrode position was functionalized by the negative control probe

(NC), the sequence of which was neither biotinylated nor relevant to the genomic

sequences of both B. cereus strains ATCC14579 and F4810/72. Another one was

functionalized by the positive control probe (PC), the sequence of which was

equivalent to that of the NC but biotinylated at the 3’ end.

The instrument and assay program

When a chip array was placed into the cartridge of an electrochemical array analyzer

(Fig. 11B), a flow chamber with an internal reaction volume of about 7 µl was built

above the chip surface. Whole detection steps, from introduction of DNA samples to

final electrical signal data readout (Fig. 11C), can be performed automatically with

this instrument. Detection procedure was completed within 30 min and controlled by

the software “MCDDE”. Signal readout was acquired differently from that of the

single–electrode chip in paper 1, which reads amperometric signal after the

completion of the enzymatic reaction. The signal (nA/min) from the DNA chip array


18

Reagent reservoirs

Chip holder

6-in-1 rotary valve

Waste collector

Chip

was read as the initial (2nd – 10th sec.) rate of signal increase under the stop-flow

condition after addition of the substrate pAPP to the flow chamber (Fig. 11D).

Figure 11A. Photo of the DNA chip array with 16 electrode positions and a scheme of the spotting location of the eight sensing probes, which represent three HBL encoding genes: hblA, hblC, hblD, three NHE encoding genes: nheA, nheB, nheC, cytK-2 encoding gene cytK-2 and emetic toxin

encoding gene ces, respectively.

Figure 11B. Photo of the DNA chip array instrument.

Figure 11C. Scheme of the target DNA detection procedure occurred on one electrode position in the flow reaction chamber.

PC

hblC

hblA

hblD nheB nheC

ces NC

hblD

nheA

nheA

nheB

nheC

cytK-2

cytK-2

ces

10 mm

9 mm

pAPP

pAP

extravidin-ALP

target

biotinylated detection probe

capture probe

red-ox

IDA electrodes on one position Electrode width: 800 nm Electrode gap: 400 nm

anode cathode


19

Figure 11D. Illustration of the real-time current from all 16 DNA chip array electrode positions during the 40 s stop flow (X axis: time (sec), scale: 10 s; Y axis: current (nA), range: 0 – 20 nA). The slope of a 8 s current increase from the 2nd second of stop flow was used as the signal (nA/min).

Results

Two B. cereus strains ATCC14579 and F4810/72, reported to be diarrheal and emetic

pathogens, respectively, were applied to PCR for analysis of presence of the hblA,

hblC, hblD, nheA, nheB, nheC, cytK-2 and ces genes. Gel pictures of the amplicons

showed the presence of hblA, hblC, hblD, nheA, nheB, nheC, cytK-2, but not ces, in

the diarrheal strain ATCC14579; but only nheA, nheB, nheC and ces in the emetic

strain F4810/72. This indicates that ATCC14579 contains all genes for HBL, NHE

and cytK-2 toxins, while F4810/72 contains genes for the NHE toxin and the emetic

toxin cereulide.

Analysis of PCR amplicons. Detection program (Table 1) and procedure were

designed and improved to achieve a sensitive genetic sequence-based detection. Total

assay was completed around 30 min without any manual operation, in this 15 min was

spent for target DNA hybridization. All PCR amplicons of the eight genes were

applied for DNA chip array detection. Signals for all amplicons were generated only

from their specific capture probe-spotting electrode positions, respectively; there were

no obvious background signals from non-specific amplicons or negative control

positions.


20

Table 1. Assay program for the electric DNA chip array instrument. step

No.

program step temperature

[°C]

time [sec]

1 Buffer flushing reaction chamber RT 20

2 Temperature of reaction chamber increasing 45 60

3 Sample transfer to reaction chamber 45 6

4 Hybridization in the reaction chamber 45 40

5 Sample renewal 45 25

6 Recycling from step 4 to step 5 for 12 times 45 65 × 12 cycles

7 Buffer washing and cooling down reaction

chamber

38 130

8 Enzyme conjugate transfer to reaction chamber 38 11

9 Binding of enzyme with captured target DNA 38 300

10 Buffer washing 38 130

11 Substrate pAPP transfer to reaction chamber 38 50

12 Stop flow (pAPP), pAP generation, redox recycling

and electric signal readout

38 40

13 Buffer washing RT 60

Analysis of genomic DNA samples. For analysis of presence of the toxin encoding

genes in genomic DNA, cells were disintegrated by ultrasonication according to the

method developed in paper 1. When the DNA chip array was exposed to these

samples, no or very weak signals were obtained, however. This problem was not

observed when working with the single-electrode chip in paper 1. A difference

between these two modes of operation is that the gold electrodes of the specific

positions in the DNA chip array are exposed to the whole cell lysate which was not

the case in the previous work with the single-electrode chip, in which all reactions

except the pAP redox-recycling took place on the magnetic beads in an eppendorf

tube. To investigate if the eliminated signal from the array was caused by problems

with the hybridization or redox-recycling steps, measurements were made with only

the positive control electrode position after exposure to the purified PCR amplicons in

buffer and cell lysates treated with none, two or four times of DNA extraction step.

Figure 12 presents their different blockage effects on the positive control (PC) signals.

The signal generated from the purified amplicons was believed not interfered by any

non-DNA impurities and then used as the reference. The cell lysates without any

DNA extraction step blocked more than 70% of the reference signal; the cell lysates

treated with two times DNA extraction still blocked 40% of the reference signal but to


21

a less degree; the cell lysates treated with four times of that recovered the reference

signal completely and had no blockage effect at all. Slight specific signals from the

gene sensing electrode positions were observed by four times extracted DNA analyte

(5×108 cells). Since the PC signal was originated directly from the assay steps after

enzyme binding to biotinylated detection probes, problems with the hybridization

were not the reason of the signal reduction. Probably the signal level was interfered

by other cellular components except DNA (Gong, Lee et al. 2006), e.g. proteins, lipo-

polysaccharides and so on, which were not or not thoroughly removed from applied

samples treated without or with less than four times extraction.

Figure 12. Blockage effects of differentially treated cellular DNA on the signals from the chip positive control positions. Column 1 represents the reference signal obtained when the purified PCR amplicons in buffer were applied. Column 2 stands for the signal generated after the submission of the cell lysate without any DNA extraction. Column 3 shows the signal obtained from the assay of the cell lysate treated with two times DNA extraction. Column 4 shows the signal after the introduction of the cell lysate treated with four times DNA extraction. The data represent average values from at least three independent replicates.

Due to the low signal generated from extracted DNA analytes, the sample treatment

needed to be improved to enhance the detection limit. Different ultrasound treatment

solutions combined with four times DNA extraction step were studied and compared

(Fig. 13). Clear signals were achieved from samples treated with first 2,5 or 5 min

ultrasonication, followed by another 10 min ultrasonication after DNA extraction

from all sensing probes, but not from those treated with only time of 2,5, 5, 10 or 15

min ultrasonication, nor with first long time ultrasonication of 10 to 15 min and then

another 10 min of treatment after extraction. According to the results in Paper 1, these

data implicated that: large DNA molecules were obtained after first 2,5 or 5 min

ultrasonication, which do not favor signal generation; after their extraction, shortened

DNAs were produced after another 10 min fragmentation. Smaller DNA molecules


22

were generated after 10 to 15 min ultrasonication, but due to their low extraction

efficiency, neither large nor smaller target DNA molecules can be extracted enough to

Figure 13. Comparison of the hblD, nheB, cytK-2 gene signals generated from differentially ultrasound - treated DNA analytes applied for chip array assays.

generate clear signals. The highest and most reproducible signals were achieved when

cellular DNA (5×108 cells) was treated with first 5 min ultrasonication, followed by

four times DNA extraction and another 10 min ultrasonication.

To detect all eight toxin-related genes simultaneously from applied cellular DNA, one

chip array functionalized by eight corresponding probes on the 16 electrode positions

(Fig. 11A) was prepared and applied for both strains. Responses to all eight genes

agreed well with results from gel electrophoresis, except that a signal from the hblA

sensing position was observed also from the emetic strain F4810/72 which did not

contain this gene according to the PCR analysis. The sequence of the hblA probes

must be further investigated. Signal patterns (Fig. 14A-B) for both strains were clearly

generated from 108 cells. There was no proportionally increased signal level generated

when increasing the cell number from 108 to 5×108.

1: 2,5’ cell ultrasonic disruption; 2: 5’ cell ultrasonic disruption;

3: 10’ cell ultrasonic disruption; 4: 15’ cell ultrasonic disruption;

5: 2,5’ cell ultrasonic disruption + 10’ DNA fragmentation;

6: 5’ cell ultrasonic disruption + 10’ DNA fragmentation;

7: 10’ cell ultrasonic disruption + 10’ DNA fragmentation;

8: 15’ cell ultrasonic disruption + 10’ DNA fragmentation.

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8


23

Figure 14. Analysis of presence of the toxin-related genes from the diarrheal strain ATCC14579 (A) and the emetic strain F4810/72 (B). Signals were detected from 1×108 (white columns) and 5×108 (grey columns) cells applied, respectively.

Conclusions

A fast and automated method for the multiplex detection of toxin-related genes can be

realized by the application of the electrochemical DNA chip arrays and properly

designed assay program. One DNA chip array can be activated by multiple gene-

sensing probes and applied for the detection of these genes simultaneously. The whole

instrumentation automates detection procedure and handles final data in a 30 min

assay procedure. Proper DNA analyte preparation from the bacteria is very important.

Ultrasound was introduced to disrupt cells, release and fragment cellular DNA. The

fragmentation is important to increase the signal. Contrary to previous results in paper

1, only ultrasonic disruption treatment is not enough to prepare suitable DNA analyte,

because other cellular compounds in the cell lysate block the pAP redox recycling at

the chip electrode surface (data not shown). Currently, the DNA extraction step is

therefore needed. Due to low extraction efficiency for small DNA molecules, large

cellular DNA are mainly obtained in the extraction and therefore the extracted DNA

B.

A.


24

must are fragmented by a second ultrasonicaion. The need of the DNA extraction

significantly reduces the versatility of those electric DNA chip arrays. It increases the

assay time by about 20 min and is not included in the automation program. Therefore

it is important to further study the fouling mechanisms to circumvent this extraction

and enable direct application of ultrasonicated cells to the assay.


25

Concluding remarks

For the electric DNA chip-based gene detection, targeting DNA strands presented

with a proper form is very important. Ultrasound, as a simple and fast cell disruption

method, performs different functions on cellular DNA strands as the function of time

applied. Dominant function of the first minutes treatment is to open bacterial cells and

release most of the DNA; afterwards, the DNA is fragmented, large DNA molecules

are gradually shortened to smallest ones with average sizes of 400-bp and without

obvious loss of intact target sequence sites; extended treatment will overfragment or

damage DNA strands. The difference between the BBSH-based and electrical chip

array-based DNA sensing systems is: BBSH-based signal generation is insensitive to

the interference from non-nucleic acid cellular components, thus ultrasonicated crude

bacterial lysates can be directly applied for assays; however, DNA chip array-based

sensing electrode surfaces are subject to blockage of signal transduction from cellular

components. Therefore it is necessary that released large DNA molecules are first

extracted out of the crude cell lysate before being fragmented, and then subjected to

the assay.

By applying the 16-sensing electrode DNA chip array instrumentation and the

optimized detecting procedure, the analysis of eight toxin-related genes can be

completed around 30 min simultaneously and automatically when DNA analytes are

ready for assays. The detection limit is currently 108 bacterial cells, which is enough

for analyzing a large B. cereus colony from an agar plate.

The DNA chip described in this thesis should provide considerable improvements to

current food and clinical analysis of pathogenic microorganisms. Current methods are

almost exclusively based on an initial enrichment culture on partly selective media

followed by confirmative analysis. These analyses mostly involve further cultivations

revealing biochemical properties of the cells. To increasing extent also immunological

analysis and PCR are used. Compared to confirmative cultivations, which typically

takes one day, the DNA chips provide much faster result and also much more accurate

description of the pathogenic capacity of the organisms. Compared to immunological

analysis, the DNA chip-based assay gives more accurate results since the

immunological properties are mostly only indirectly correlated to the pathogenicity.


26

Compared to PCR, which gives the same type of information as the electric DNA

chips, the latter provides a simpler and more rapid assay.


27

Abbreviations

B. cereus Bacillus cereus

HBL haemolytic toxin B and L complex

NHE nonhaemolytic toxin complex

NRPS non-ribosomal peptide cereulide synthetase

ces (cerNRPS) gene of cereulide non-ribosomal peptide synthetase

cytK cytotoxin K protein

PlcR pleiotropic regulator

ss- single stranded-

ds- double stranded-

bp base pair

SPR surface plasmon resonance

QCM quartz crystal microbalance

BBSH bead-based DNA sandwich hybridization

IDA interdigitated array

extravidin-ALP extravidin – alkaline phosphatase conjugate

pAPP p – aminophenyl phosphate monosodium

pAP p – aminophenol

RT room temperature (22 - 24°C)

PC positive control

NC negative control


28

Acknowledgements

First I would like to acknowledge the financial support from the European

Community (project LSHB-CT-2004-512009, eBIOSENSE) and the FORMAS

(Swedish Research Council for Environment, Agricultural Sciences, and Spatial

Planning).

Thank Ulle, my supervisor, for giving me the opportunity to work in the exciting

research field of biosensors as a PhD student, and for all the guidance to my work,

papers and this thesis.

Thank Magda, my co-supervisor, for guiding me into the research work from the very

beginning, for all fruitful discussions and valuable suggestions, and also for the fun of

gossip time.

Thank Bruno Elsholz and Joerg Albers, the EU project partner from the Frauenhofer

Institute of Silicon Technology, for all co-operations and generous helps for the chip

spotting and the instrument.

Thank Fredrik for the help in using flow cytometer, Ela for the help in protein

analysis, Kaj for helps in taking microscopic pictures of the Bacillus cereus and the

chip array. Thank all my colleagues in our department, the present and the former, for

all your helps inside and outside the lab.

Hope all of you would still accompany me to the end of my PhD study!


29

References Agaisse, H., M. Gominet, et al. (1999). "PlcR is a pleiotropic regulator of extracellular

virulence factor gene expression in Bacillus thuringiensis." Molecular Microbiology 32(5): 1043-1053.

Albers, J., T. Grunwald, et al. (2003). "Electrical biochip technology - a tool for microarrays and continuous monitoring." Analytical and Bioanalytical Chemistry 377(3): 521-527.

Andersson, M. A., R. Mikkola, et al. (1998). "A novel sensitive bioassay for detection of Bacillus cereus emetic toxin and related depsipeptide ionophores." Applied and Environmental Microbiology 64: 1338-1343.

Beecher, D. J. and J. D. Macmillan (1991). "Characterization of the components of hemolysin BL from Bacillus cereus." Infect. Immun. 59(5): 1778-1784.

Boon, E. M., D. M. Ceres, et al. (2000). "Mutation detection by electrocatalysis at DNA-modified electrodes." Nature Biotechnology 18(10): 1096-1100.

Boon EM, Salas JE, Barton JK (2002). "An electrical probe of protein-DNA interactions on DNA-modified surfaces." Nature Biotechnology 20(3): 282-286.

Broude, N. E. (2002). "Stem-loop oligonucleotides: a robust tool for molecular biology and biotechnology." Trends in Biotechnology 20(6): 249-256.

Bunimovich, Y. L., Y. S. Shin, et al. (2006). "Quantitative Real-Time Measurements of DNA Hybridization with Alkylated Nonoxidized Silicon Nanowires in Electrolyte Solution." J. Am. Chem. Soc. 128(50): 16323-16331.

Chee, M., R. Yang, et al. (1996). "Accessing genetic information with high-density DNA arrays." Science 274(5287): 610-614.

Cheek, B. J., A. B. Steel, et al. (2001). "Chemiluminescence detection for hybridization assays on the flow-thru chip, a three-dimensional microchannel biochip." Analytical Chemistry 73(24): 5777-5783.

Cheung VG, Morley M, et al. (1999). "Making and reading microarrays." Nature Genetics 21: 15-19.

Choma, C. G., P. Einar (2002). "The enterotoxin T (BcET) from Bacillus cereus can probably not contribute to food poisoning." FEMS Microbiology Letters 217(1): 115-119.

Dietrich, R., M. Moravek, et al. (2005). "Production and Characterization of Antibodies against Each of the Three Subunits of the Bacillus cereus Nonhemolytic Enterotoxin Complex." Appl. Environ. Microbiol. 71(12): 8214-8220.

Dietrich, R., C. Fella, et al. (1999). "Production and characterization of monoclonal antibodies against the hemolysin BL enterotoxin complex produced by Bacillus cereus." Applied and environmental microbiology 65(10): 4470-4474.

Drummond, T. G., M. G. Hill, et al. (2003). "Electrochemical DNA sensors." Nature Biotechnology 21(10): 1192-1199.

Ehling-Schulz, M., M. Fricker, et al. (2004). "Identification of emetic toxin producing Bacillus cereus strains by a novel molecular assay." Fems Microbiology Letters 232(2): 189-195.

Ehling-Schulz, M., M.-H. Guinebretiere, et al. (2006). "Toxin gene profiling of enterotoxic and emetic Bacillus cereus." FEMS Microbiology Letters 260(2): 232-240.

Elsholz, B., R. Worl, et al. (2006). "Automated detection and quantitation of bacterial RNA by using electrical microarrays." Analytical Chemistry 78(14): 4794-4802.

Fagerlund, A., A. Ween, et al. (2004). "Genetic and functional analysis of the cytK family of genes in Bacillus cereus." Microbiology-Sgm 150: 2689-2697.

Fermanian, C., C. Lapeyre, et al. (1997). "Diarrhoeal toxin production at low temperature by selected strains of Bacillus cereus." Journal of Dairy Research 64: 551-559.

Finlay, W. J. J., N. A. Logan, et al. (1999). "Semiautomated metabolic staining assay for Bacillus cereus emetic toxin." Applied and Environmental Microbiology 65(4): 1811-1812.

Fodor, S. P. A., J. L. Read, et al. (1991). "Light-directed, spatially addressable parallel chemical synthesis." Science 251(4995): 767-773.


30

Gabig-Ciminska, M., A. Holmgren, et al. (2004). "Electric chips for rapid detection and quantification of nucleic acids." Biosensors & Bioelectronics 19(6): 537-546.

Gabig-Ciminska, M., Y. L. Liu, et al. (2005). "Gene-based identification of bacterial colonies with an electric chip." Analytical Biochemistry 345(2): 270-276.

Ghelardi, E., F. Celandroni, et al. (2002). "Identification and characterization of toxigenic Bacillus cereus isolates responsible for two food-poisoning outbreaks." Fems Microbiology Letters 208(1): 129-134.

Guinebretiere, M. H., A. Fagerlund, et al. (2006). "Rapid discrimination of cytK-1 and cytK-2 genes in Bacillus cereus strains by a novel duplex PCR system." Fems Microbiology Letters 259(1): 74-80.

Haggblom, M. M., C. Apetroaie, et al. (2002). "Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced under various conditions." Applied and Environmental Microbiology 68(5): 2479-2483.

Hansen, B. M. and N. B. Hendriksen (2001). "Detection of enterotoxic Bacillus cereus and Bacillus thuringiensis strains by PCR analysis." Applied and Environmental Microbiology 67(1): 185-189.

Hansen, B. M., P. E. Hoiby, et al. (2003). "The Bacillus cereus bceT enterotoxin sequence reappraised." Fems Microbiology Letters 223(1): 21-24.

Hansen, K. M., H. F. Ji, et al. (2001). "Cantilever-based optical deflection assay for discrimination of DNA single-nucleotide mismatches." Analytical Chemistry 73(7): 1567-1571.

Hoang A. Ho, K. D., Maurice Boissinot, Michel G. Bergeron, Robert M. Tanguay, Denis Boudreau, Mario Leclerc (2005). "Direct Molecular Detection of Nucleic Acids by Fluorescence Signal Amplification." J. AM. CHEM. SOC. 127: 12673-12676.

Homs, W. C. I. (2002). "DNA Sensors." Analytical Letters 35(12): 1875-1894. Horwood, P. F., G. W. Burgess, et al. (2004). "Evidence for non-ribosomal peptide synthetase

production of cereulide (the emetic toxin) in Bacillus cereus." Fems Microbiology Letters 236(2): 319-324.

Ivanova, N., A. Sorokin, et al. (2003). "Genome sequence of Bacillus cereus and comparative analysis with Bacillus anthracis." Nature 423(6935): 87-91.

Jaaskelainen, E. L., M. M. Haggblom, et al. (2004). "Atmospheric oxygen and other conditions affecting the production of cereulide by Bacillus cereus in food." International Journal of Food Microbiology 96(1): 75-83.

Kagan Kerman, M. K. a. E. T. (2004). "Recent trends in electrochemical DNA biosensor technology." Measurement Science and Technology 15: R1–R11.

Kara, P., K. Kerman, et al. (2002). "Electrochemical genosensor for the detection of interaction between methylene blue and DNA." Electrochemistry Communications 4(9): 705-709.

Kawamura-Sato, K., Y. Hirama, et al. (2005). "Quantitative analysis of cereulide, an emetic toxin of Bacillus cereus, by using rat liver mitochondria." Microbiology and Immunology 49(1): 25-30.

Lechner, S., R. Mayr, et al. (1998). "Bacillus weihenstephanensis sp. nov. is a new psychrotolerant species of the Bacillus cereus group." International Journal of Systematic Bacteriology 48: 1373-1382.

Lee, T. M. H. and I. M. Hsing (2002). "Sequence-specific electrochemical detection of asymmetric PCR amplicons of traditional Chinese medicinal plant DNA." Analytical Chemistry 74(19): 5057-5062.

Liao, J. C., M. Mastali, et al. (2006). "Use of electrochemical DNA biosensors for rapid molecular identification of uropathogens in clinical urine specimens." Journal of Clinical Microbiology 44(2): 561-570.

Liu, R. H., J. N. Yang, et al. (2004). "Self-contained, fully integrated biochip for sample preparation, polymerase chain reaction amplification, and DNA microarray detection." Analytical Chemistry 76(7): 1824-1831.


31

Lubin, A. A., R. Y. Lai, et al. (2006). "Sequence-specific, electronic detection of oligonucleotides in blood, soil, and foodstuffs with the reagentless, reusable E-DNA sensor." Analytical Chemistry 78(16): 5671-5677.

Lund, T. and P. E. Granum (1996). "Characterisation of a non-haemolytic enterotoxin complex from Bacillus cereus isolated after a foodborne outbreak." FEMS microbiology letters 141(2-3): 151-156.

Mantynen, V. and K. Lindstrom (1998). "A rapid PCR-based DNA test for enterotoxic Bacillus cereus." Applied and Environmental Microbiology 64(5): 1634-1639.

Minunni, M., S. Tombelli, et al. (2005). "Detection of fragmented genomic DNA by PCR-Free piezoelectric sensing using a denaturation approach." Journal of the American Chemical Society 127(22): 7966-7967.

Nakamura, L. K. (1998). "Bacillus pseudomycoides sp. nov." International Journal of Systematic Bacteriology 48: 1031-1035.

Ouassila, O., C. Thierry, et al. (2006). "The Production of Bacillus cereus Enterotoxins Is Influenced by Carbohydrate and Growth Rate." Current Microbiology V53(3): 222-226.

Ozkan, D., A. Erdem, et al. (2002). "Allele-specific genotype detection of factor V Leiden mutation from polymerase chain reaction amplicons based on label-free electrochemical genosensor." Analytical Chemistry 74(23): 5931-5936.

Paananen, A., R. Mikkola, et al. (2002). "Inhibition of human natural killer cell activity by cereulide, an emetic toxin from Bacillus cereus." Clinical and Experimental Immunology 129(3): 420-428.

Panda, S., T. K. Sato, et al. (2003). "An array of insights: application of DNA chip technology in the study of cell biology." Trends in Cell Biology 13(3): 151-156.

Peng, H., C. Soeller, et al. (2006). "Electrochemical detection of DNA hybridization amplified by nanoparticles." Biosensors & Bioelectronics 21(9): 1727-1736.

Priest, F. G., M. Barker, et al. (2004). "Population structure and evolution of the Bacillus cereus group." Journal of Bacteriology 186(23): 7959-7970.

Pruss, B. M., R. Dietrich, et al. (1999). "The hemolytic enterotoxin HBL is broadly distributed among species of the Bacillus cereus group." Applied and Environmental Microbiology 65(12): 5436-5442.

Rhodehamel, E. J., S. M. Harmon, et al. (2001). "Bacillus cereus and Bacillus cereus diarrheal enterotoxin." In: Bacteriological Analytical Manual online, U. S. FDA. CFSAN. Chapter 14 – 15.

Rosenquist, H., L. Smidt, et al. (2005). "Occurrence and significance of Bacillus cereus and Bacillus thuringiensis in ready-to-eat food." Fems Microbiology Letters 250(1): 129-136.

Schena, M., D. Shalon, et al. (1995). "Quantitative monitoring of gene-expression patterns with a complementary-DNA microarray." Science 270(5235): 467-470.

Schoeni, J. L. and A. C. L. Wong (2005). "Bacillus cereus food poisoning and its toxins." Journal of Food Protection 68(3): 636-648.

Slamti, L., S. Perchat, et al. (2004). "Distinct mutations in PlcR explain why some strains of the Bacillus cereus group are nonhemolytic." Journal of Bacteriology 186(11): 3531-3538.

So-Jung Park, T. A. T., Chad A. Mirkin (2002). "Array-Based Electrical Detection of DNA with Nanoparticle Probes." Science 295: 1503-1506.

Stenfors, L. P., R. Mayr, et al. (2002). "Pathogenic potential of fifty Bacillus weihenstephanensis strains." Fems Microbiology Letters 215(1): 47-51.

Thaenthanee, S., A. C. L. Wong, et al. (2005). "Phenotypic and genotypic comparisons reveal a broad distribution and heterogeneity of hemolysin BL genes among Bacillus cereus isolates." International Journal of Food Microbiology 105(2): 203-212.

Umek, R. M., S. W. Lin, et al. (2001). "Electronic detection of nucleic acids - A versatile platform for molecular diagnostics." Journal of Molecular Diagnostics 3(2): 74-84.

Wang, J. (2000). "SURVEY AND SUMMARY: From DNA biosensors to gene chips." Nucl. Acids Res. 28(16): 3011-3016.


32

Wang, J. (2002). "Portable electrochemical systems." TrAC Trends in Analytical Chemistry 21(4): 226-232.

Wang, J. (2006). "Electrochemical biosensors: Towards point-of-care cancer diagnostics." Biosensors and Bioelectronics 21(10): 1887-1892.

Wirtz, R. W., C.; Germishuizen, W. A.; Pepper, M.; Middelberg, A. P. J.; Davies, A. G. (2003). "High-sensitivity colorimetric detection of DNA hybridization on a gold surface with high spatial resolution." Nanotechnology 14: 7-10.

Zezza, F., M. Pascale, et al. (2006). "Detection of Fusarium culmorum in wheat by a surface plasmon resonance-based DNA sensor." Journal of Microbiological Methods 66(3): 529-537.

Zhang, Y. C., A. Pothukuchy, et al. (2004). "Detection of similar to 10(3) copies of DNA by an electrochemical enzyme-amplified sandwich assay with ambient O-2 as the substrate." Analytical Chemistry 76(14): 4093-4097.

Zwirglmaier, K., W. Ludwig, et al. (2004). "Improved method for polynucleotide probe-based cell sorting, using DNA-coated microplates." Applied and Environmental Microbiology 70(1): 494-497.

Fykse, E. M., J. S. Olsen, et al. (2003). "Application of sonication to release DNA from Bacillus cereus for quantitative detection by real-time PCR." Journal of Microbiological Methods 55(1): 1-10.

Mann, T. L. and U. J. Krull (2004). "The application of ultrasound as a rapid method to provide DNA fragments suitable for detection by DNA biosensors." Biosensors & Bioelectronics 20(5): 945-955.

Gong, P., C. Y. Lee, et al. (2006). "Hybridization behavior of mixed DNA/alkylthiol monolayers on gold: Characterization by surface plasmon resonance and P-32 radiometric assay." Analytical Chemistry 78(10): 3326-3334.

Electric DNA arrays for determination of pathogenic ...

Documents

Transcript of Electric DNA arrays for determination of pathogenic ...