Cet ouvrage fait partie de la bibliothèque YouScribe
Obtenez un accès à la bibliothèque pour le lire en ligne
En savoir plus

Development and evaluation of rapid and semi-automated devices for the detection of toxic algae [Elektronische Ressource] / vorgelegt von Sonja Diercks

De
156 pages
DEVELOPMENT AND EVALUATION OF RAPID AND SEMI-AUTOMATED DEVICES FOR THE DETECTION OF TOXIC ALGAEDissertationzur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften -Dr. rer. nat.- im Fachbereich 2 (Biologie/Chemie) der Universität Bremen vorgelegt von Sonja Diercks Bremen, Januar 2007 1. Gutachter: Prof. Dr. G. O. Kirst 2. Gutachter: Prof. Dr. A. D. Cembella Tag des öffentlichen Kolloquiums: Universität Bremen, 2. April 2007 Hiermit erkläre ich, dass ich die vorliegende Dissertation selbständig verfasst und keine anderen als die angegebenen Quellen und Hilfsmittel verwendet habe. Die entnommen Stellen aus benutzen Werken wurden wörtlich oder inhaltlich als solche kenntlich gemacht. Sonja Diercks Table of Contents 1. GENERAL INTRODUCTION………………..………………………………………… 11.1 HARMFUL ALGAL BLOOMS……………………………………………………...... 11.1.1 Associated human illnesses………………………………………………….. 41.1.2 Aquaculture and harmful algal blooms……………………………………... 61.2 MONITORING OF PHYTOPLANKTON………………………………………………. 71.2.1 Methods……………………………………………………………………… 71.2.1.1 Mouse bioassay…………………………………………………………... 71.2.1.2 Methods for the detection of toxins……………………………………… 81.2.1.3 Counting techniques…………………………………………………….... 91.2.1.4 Data buoys and remote sensing using satellites………………………….. 91.2.1.5 Detection of harmful algae using molecular probes or antibodies…………………………………………………………….... 101.3 BIOSENSORS…………………………………………..………………………….. 111.
Voir plus Voir moins

D

EVELOPMENT AND EVALUATION OF RAPID AND SEMI-

AUTOMATED DEVICES FOR THE DETECTION OF TOXIC ALGAE

Dissertation

zur Erlangung des akademischen Grades

eines Doktors der Naturwissenschaften

-Dr. rer. nat.-

im Fachbereich 2 (Biologie/Chemie)

der Universität Bremen

vorgelegt von

Sonja Diercks

Bremen, Januar 2007

1.

2.

Gutachter: Prof. Dr. G. O. Kirst

Gutachter: Prof. Dr. A. D. Cembella

Tag des öffentlichen Kolloquium

s: Universität Bremen, 2. April 2007

it erkläre ich, dass ich die vorliegende Hierm

n selbständig verfasst und keine Dissertatio

anderen als die angegebenen Quellen und Hilfsmittel verwendet habe. Die entnommen Stellen

acht. inhaltlich als solche kenntlich gemrken wurden wörtlich oder eaus benutzen W

Sonja Diercks

Table of Contents

1. GENERAL INTRODUCTION………………..…………………………………………
……………………………………………………......UL ALGAL BLOOMSARMF1.1 H…………………………………………………..1.1.1 Associated human illnesses……………………………………...1.1.2 Aquaculture and harmful algal blooms……………………………………………….ONITORING OF PHYTOPLANKTON1.2 M………………………………………………………………………1.2.1 Methods1.2.1.1 Mouse bioassay…………………………………………………………...1.2.1.2 Methods for the detection of toxins………………………………………1.2.1.3 Counting techniques……………………………………………………....ote sensing using satellites…………………………..1.2.1.4 Data buoys and remolecular probes 1.2.1.5 Detection of harmful algae using m or antibodies……………………………………………………………....…………………………………………..…………………………..IOSENSORSB1.3……………………………..……………………………………...IM OF THESIS1.4 A……………………………………………………………….UTLINE OF THESIS1.5 O1.5.1 Development and adaptation of molecular probes for sandwich ………………………………………………………ntiohybridiza…………………………1.5.2 Design and evaluation of probe sets for toxic algae……………………………………….….1.5.3 Improvement of detection protocol……………..…...1.5.4 Assessment of probe modification for signal enhancement…………………………………1.5.5 Development and evaluation of a biosensor……………………………………………………………………….TIONSUBLICAP2.…………...………………………………………….........PUBLICATIONSIST OF 2.1 L………………………..BLICATIONSTATEMENT OF MY CONTRIBUTION TO THE PU2.2 S2.3PUBLICATION I:COLORIMETRIC DETECTION OF THE TOXIC DINOFLA-
GELLATE ALEXANDRIUM MINUTUM USING SANDWICH HYBRIDIZATION
………………………………………………....... IN A MICROTITER PLATE ASSAY 2.4PUBLICATION II:MOLECULAR PROBES FOR THE DETECTION OF TOXIC
…………………………R USE IN SANDWICH HYBRIDIZATION FORMATS ALGAE FO 2.5PUBLICATION III:ELECTROCHEMICAL DETECTION OF TOXIC ALGAE
WITH A BIOSENSOR………………………………………………………………..

1146777899

10111212

1213141415161617

183955

EIV:UBLICATION P2.6

VALUATION OF LOCKED NUCLEIC ACIDS FOR SIGNAL ENHANCEMENT OF OLIGONUCLEOTIDE PROBES FOR

3.

4.

5.

6.

7.

8.

MICROALGAE IMMOBILIZED ON SOLID SURFACES………………………………....

2.7PUBLICATION V:DEVELOPMENT AND OPTIMIZATION OF A SEMI-

AUTOMATED RRNA BIOSENSOR FOR THE DETECTION OF TOXIC ALGAE…………...

…………………………………………………………………….……..ISYNTHESS

……………DETECTION OF TOXIC ALGAL SPECIESOLORIMETRIC ASSAY FOR THE C3.1

……………

………………………………………….SSESSMENT OF SIGNAL ENHANCEMENTA3.2

D3.3

ETECTION OF TOXIC ALGAL SPECIES USING MULTIPROBE CHIPS AND ………………………………………………………AUTOMATED DEVICE- A SEMI

FUTURE RESEARCH………………………………………………………………....

ARYUMM S……………………………………………………………………………

………………………………………………………………...USAMMENFASSUNG Z

………………………………………………………………………...EFERENCES R

……………………………………………………………………….UNGAGANKS D

71

97

127

127

127

130

130

132

135

137

139

142

151

General Introduction

1.General Introduction

l Algal Blooms Harmfu1.1

Oceans, the origin of life, harbour com

n communities, which play antoplex phytoplank

important role in marine biological ecosystems. Microalgae are the major producers of

biomass and organic compounds in the oceans because of their photosynthetic activity and

lve shellfish (oysters, mussels, c food chain. Filter feeding bivarepresent the base of the aquati

scallops, clams, etc.), the larvae of crustaceans and finfish feed primarily on microalgae

(Hallegraeff 2003). About 5000 species of marine microalgae are known to date (Sournia et

bers that they discolour the proliferate in such high nume 300 species can al. 1991) and som

surface of the sea (Daranas et al. 2001; Hallegraeff 2003) as a so-called bloom (Figure 1).

Figure 1. Bloom of Noctiluca scintillas in October 2002, Leigh, New Zealand (photo: Miriam

Godfrey)

croalgal population activated by suitable iThis is regarded as a sudden increase in the m

45growth conditions so that concentrations of 10 cells per litre can be reached for a certain–10

period of time (Masó and Garces 2006). A bloom can be dominated by a particular species or

itiation of a blooma group of species (Masó and Garces 2006). The in of requires an inoculum

y involve different life stages, e.g., cysts amcells, which can be from several sources and

(Steidinger and Garcés 2006), as well as favourable environmental conditions of temperature,

1

General Introduction

light, nutrients and water salinity (Zingone and Enevoldsen 2000; Daranas et al. 2001).

ental conditions, such as is triggered by inappropriate environmination of a bloomTerm

r and Garcés 2006). nutrient-deplete water, zooplankton predation or advection (Steidinge

Also viral termination of algal blooms of Heterosigma akashiwo,Emiliana huxleyi and

have been observed (Bratbak et al. 1996; Brussaard et al. 2005).Phaeocystis globosa

rous phytoplankton species and alternation eplex life cycles are described for numCom

between dormant, benthic stage and a motile, vegetative existence can take place. Dormant

cysts or resting spores can be formed from many marine phytoplankton species during their

life history and may play a an important role in bloom initiation (Zingone and Enevoldsen

s, reproduce by asexual, llates and diatom2000). Most toxic or harmful species, dinoflage

binary division; however, sexuality can be induced under certain conditions. Morphological

ed during the life zygotes and cysts) are formetes,and physiologically distinct cell types (gam

s of dinoflagellates are annual events; the st algae (Anderson et al. 2003). Bloomocycle of m

ented in the spring. Sexual reproduction often first increase of populations is usually docum

a few hours several in period of vegetative growth and can last fromaoccurs following the m

days. The resulting zygote is usually a restingstage or cyst. Cyst production is also assumed

to be seasonal, because different dinoflagellate species becomese abundant at different tim

abundance within the phytoplankton ume species attain their maximduring the year. Som

during the season spring and, therefore, formcysts in the late spring to early summer (e.g.,

Protoceratium,Alexandrium tamarense). Other species (e.g., Protoperidinium oblongum

y exhibit two annual peaks in abundance and hence two peaks of encystma) mreticulatument

(Harland et al. 2004). Diatoms reproduce by asexual division until cell size reaches a

minimum threshold level, usually below 30%–40% of the dimensions of the maximum cell

size (Aminitiates sexual reproduction, which can be associated with ato et al. 2005). This

increased photoperiod length (Steidinger and Garcés 2006). Life cycle investigations of

s have shown, that, within a population, sexual reproduction is a nearly synchronous diatom

2 to 40 varying from occurs within a restricted size window, with a periodicity event which

years (Mann 1988; Amato et al. 2005). Thick-walled resting cysts are occasionally formed

from diatoms mainly towards the end of a bloom. They settle to the bottom or accumulate at

e dinoflagellate cysts Steidinger and Garcés 2006). Sompycnoclines (Anderson et al. 2003;

can remain viable in the sediments for several years, ready to germinate when conditions

it (Zingone and Enevoldsen 2000; Daranas et al. 2001; Anderson et al. 2003).perm

2

General Introduction

In most cases, the proliferation of microalgae is a normal event and can be beneficial for
consists of harmful algae, it can the bloomaquaculture and fisheries operations. However if ic losses to aquaculture, fisheries and also have a negative effect and cause severe econom (Hallegraeff 2003). Three different types of HABs have been delineated by tourismlessHallegraeff (2003). The first type represents species that produce basically harmbut which can, under exceptional conditions, such as sheltered ter,adiscolorations of the ws that cause indiscriminate kills of fish and invertebrates through dense bloombays, formoxygen depletion (e.g., Noctiluca scintillans). Species that produce potent toxins form the
second type are e.g., species of the genera Alexandrium,Dinophysis or Pseudo-nitzschia.
ans and cause a variety of Their toxins can find their way through the food chain to humgastrointestinal and neurological illnesses. The third type is presented by species that are non-
toxic to humans but harmful to fish and invertebrates by damaging or clogging the gills or gill
) (Hallegraeff 2003). The mulina polylepsisChrysochroPrymnesium parvum,tissue (e.g., on of harmful species, even the most toxic pact of HABs is defined by the concentratiimspecies must occur with a minimum cell concentration to exert a harmful effect (Zingone and
ainlycroalgal species and 97 toxic species (miEnevoldsen 2000). About 200 noxious m (Zingone and Enevoldsen tial to form HABsdinoflagellates) are known to have the potenst Internationale in 1974 at the 1 introduced for the first tim2000; Moestrup 2004), a termful algae can be ms of toxic dinoflagellates (Masó and Garces 2006). Harconference of bloomobserved not only in a single class or in a few genera, but also can be found among six s, dinoflagellates, haptophytes, raphidophytes, cyanophytes and ic groups (diatomtaxonompelagophytes) (Zingone and Enevoldsen 2000).

ena that have occurred throughout recorded history. However, in HABs are natural phenomthe past decades, the public health and economic impacts appear to have increased in
llegraeff 2003). Aaet al. 2001; Hfrequency, intensity and geographic distribution (Daranas worldwide increase of HABs cannot be verified because of missing time series data, but,
ples of HABs have been observed in areas where they were previously erous examnumybe through the transport in ballast water.aunknown (Zingone and Enevoldsen 2000), mParalytic shellfish poisoning (PSP), triggered by bloomA. and Alexandrium tamarenses ofcatenella, was only observed in the temperate waters of Europe, North America and Japan
ented in the Southern Hemisphere (Hallegraeff until 1970. By 2000, it was also well docum2003). The apparent increase of HABs can be explained, on one hand, by an increase of dia and, on the other hand, by escientific awareness, reports in the press and electronic m

3

General Introduction

ing has been observed m fish and shellfish farre. This increase ofincreased aquacultuworldwide and consequently, the reports of harmful algae and human illnesses rise.
ulated by eutrophication activated by s appear to be stime algal bloomAdditionally somdomestic, industrial and agricultural wastes. Also, climatological conditions can have an
ies. The dinoflagellate and PSP-producer ceffect on the spatial distribution of a spein tropical seas, fossil cysts have is presently known to be distributed Pyrodinium bahamenseispheres. Passive introduction of species from perate regions of both hembeen found in tems is stormother areas by transport of cysts in ballast water as well as by currents ande geographic range of a species. (Zingone and considered as an explanation for extending thEnevoldsen 2000; Hallegraeff 2003)

Associated human illnesses1.1.1

an health in different ways. First, the ingestion of seafood Harmful algae can affect humcontaminated with toxins produced by marine microalgae can cause a number of human
ental exposurestic shellfish poisoning described below. Second, environme paralyesses likillnove onshore. rine phytoplankton cells are disrupted by waves as they macan occur when man exposure Reports of skin irritation and respiratory distress have been associated with huments (Backer et al. 2003). to water and aerosols containing toxins and cell fragm

paralytic shellfishfe first recorded cases o – One of thParalytic shellfish poisoning (PSP)poisoning wthe Pacific Northwest s in 1793 when Captain Vancouver and his crew landed in amostly in North Amof the USA (Nishitani and Chewerica and Europe, but also in Malaysia, the P 1988). Early intoxications of humanshilipp have been recorded ines, Indonesia,
ala, China and South Africa (Backer et al. 2003). The first isolated Venezuela, Guatemmetabolite was saxitoxin (STX), whose origin was traced to plankton, and two major groups
and their onset is rapid. Neuronal and mof toxins, saxitoxin and neosaxitoxin, have been identified. PuscularsodiumSP symptom channels are blocked, which prevents s are neurological
scle fibres. A tingling orupropagation of the action potential in nerve axons and skeletal minutes, which gradually spreads to the face bness around the lips is observed within 30 mnumiting and and neck. A prickly sensation in the fingertips, headache, fever, nausea, vomdiarrhoea usually follow. PSP is a life-threatening poisoning syndrome and the most severe
cases result in respiratory arrest within 24 hours of consumption of the toxic shellfish. An antidote is non-existent and if supportive respiratory therapy can be carried out, survivors recover fully (Daranas et al. 2001). PSP toxins are produced by dinoflagellates of the genera

4

General Introduction

Alexandrium,Gymnodinium and Pyrodinium. 1,600 cases of intoxication were reported

osed (Backer et al. cases have been diagnal 900e an additionbefore 1970. Since that tim

2003).

– The first report of DSP originated in 1976 fromDiarrhetic shellfish poisoning (DSP) Japan,

where it caused major problems in the scallop fisheries. The outbreaks in Japan were

correlated with the appearance of the dinoflagellate Dinophysis fortii. Shortly after the

were found to be responsible Prorocentrum lima species and Dinophysisoutbreaks in Japan

afor DSP incidences in Europe (Hallegraeff 2003). The toxin responsible wdes nam

dinophysistoxin (DTX). The principal toxins responsible for incidents DSP are okadaic acid

and its analogs, DTX1 and DTX2. Two other toxin groups, pectenotoxin and yessotoxin, are

also placed in the “DSP” category, because of their co-occurrencewith okadaic acid and

DTX. Pectenotoxin (PTX) is named after the scallop genus Patinopecten from which it was

. Yessotoxins (YTX) are Dinophysis spp.in toxin produced by afirst isolated and is the m

produced by the dinoflagellates Protoceratium reticulatumandLingulodinium polyedrum

s, 2003b). DSP toxins of the okadaic acid group produce gastrointestinal symptom(Quilliam

including diarrhoea, nausea, vomiting and abdominal cramps. The symptoms can begin within

30 minutes after consumption of toxic shellfish and recoverytakes place within three days

without any medical treatment. Symptoms can easily be mistaken for those of bacterial gastric

infections. However, some of the polyether toxins involved may promote stomach tumours

and thus produce chronic problemers (Daranas et al. 2001). In the 1970s s in shellfish consum

re than 8,000 oe 1,300 DSP cases were reported in Japan and mand 1980s, altogether som

cases in Europe. By 2000, the global reports of diarrethic shellfish poisoning had extended to

ailand, Canada, Australia and New Zealand (Hallegraeff 2003). hJapan, Europe, Chile, T

Amnesic shellfish poisoning (ASP)enon was in 1987 – The first recognition of this phenom

ption of blues died and a hundred acute cases occurred after consumwhen three victim

mussels from Prince Edward Island, Canada (Hallegraeff 2003). Domoic acid belongs to a

ino acids called kainoids, a neuroexcitant, and interferes with the group of am

neurotransmission mechanisms in the brain. The diatomPseudo-nitzschia multiseries was

s 2003a). Victimnada (Quilliamaidentified to be causative organism of the incident in C

ptomreported gastrointestinal syminal cramps and diarrhoea, iting, abdoms, such as vom

allyditionption of toxic shellfish. Adours of the consumwhich usually occurred within 24 h

sptomneurological sym can appear, usually within 48 hours. Dizziness, headache,

5

General Introduction

ory loss, respiratory difficulty and coma are also observed memn, short-termdisorientatioorants of(Daranas et al. 2001; Backer et al. 2003). In 1991, brown pelicans and cormPseudo-ulated ASP from ingesting anchovies that had accumCalifornia were victims ofnitzschiaspecies (Silver 2006).Domoic acid has been also isolated fromP. australis,P.
delicatissima,P. multistriata,P. pseudodelicatissima,P. seriata,P. pungens and P. turgidula.
erica and Canada, whereas only low inly restricted to North Amaoic acid are mReports of domw Zealand. (Hallegraeff concentrations have been found in Europe, Australia, Japan and Ne2003)

Aquaculture and harmful algal blooms 1.1.2

reltutches, shellfish production and maricu fishery caBecause of the decrease in wildly in the Asia-Pacific region where seafood experience a worldwide expansion, especialark and the ounts. In Europe, Spain, France, Italy, Denmed in large amproducts are consumNetherlands are the main shellfish producers, with a total production of about one million
fresh and frozen mtonnes in 1997. Mussel production is of great imussels are almost exclusively in Europe (Fernández et al. 2003). In 1998, portance in these areas and the markets for
worldwide production of mariculture fish was about 0.7 million tonnes (Rensel and Whyte
phycotoxins by direct filtration of the algal cells or by feedi2003). Shellfish, such as bivalve molluscs, gastropods, crabs and lobsters, accumng on contaminated organismulates.
Regulation of accumloss from and to the environment as well as by thulation of a particular toxin takes place by balancing toxin intake ande transformation to and from other toxins by
microbial agents. Toxin accumulation rates as well as the rates of toxin loss by filter-feeding
toxic algae are toxin- and species-specific. (Fernández et al. 2003)shellfish fromConsequently, the duration of mmussel industry was shutdown for almarket closurost a yee depends on these rates. In 1984, the Swedish ar because of DSP toxins (Hallegraeff 2003)
and Förlin 2004).slow rate (Svensson ssels depurated at uthat resided in the m

Fish killing microalgae have caused high economiclosses to aquaculture in the last decades.
One example is the massive bloom of Chrysochromulina polylepsis thatoccurred in 1988 in
ark, Norway and Sweden the Skagerrak, the Kattegat, the Belt and the Sound between Denmon and trout (Hallegraeff and caused the deaths of 900 tonnes of fish, including cod, salm2003). Similarly, losses from fish kills amounted to US$95.5 million in Korea and in North
America to US$35 million (Rensel and Whyte 2003). Fish mortality is caused by a variety of
physiological mechanisms. Gill clogging, irritation or mechanical damage to the gill tissue

6

General Introduction

nts. That erasion or to clear the blocked filamleads to the production of mucus to relieve the abcan be followed by blood hypoxia and respiratory dysfunction as the cause of fish death. Other reasons for fish death can be toxigenic reactions to ichthyotoxic reagents, blood a from oxygen tion or gas-bubble traumental oxygen deple environmhypoxia fromhyte 2003). supersaturation (Rensel and W

Monitoring of phytoplankton 1.2

species is important for the prevention of eration of harmful algal Detection and enumtoxication of humans as well as from an ecological and economic point of view. On a global
ptionan poisoning through fish and shellfish consumately 2,000 cases of humscale, approximprogrammare reportedes (e.g. GE each year with a mortality of 15 percenOHAB) at the coastlines all around the world aimt (Hallegraeff 2003). HAB monitoring to prevent
intoxication of humans and animals through the consumption of contaminated seafood.
Additionally, the protection of humans from algal toxins delivered viasea spray or direct
d fish, as well as the such as shellfish anage of living resources,ed. The damcontact is aimeconomic losses to fisherman, aquaculturists and the tourist industry should be minimized
surveillan(Andersen et al. 2003). Monitoce for potential toxic algring programmal species es include, in the m(identification and quantification) and theajority of cases, the
monitoring of toxin content in shellfish. In addition, water temperature, salinity, nutrients,
eters are also observed for paramchlorophyll, water stratification, current circulation and other prediction. bloom

Methods1.2.1

Mouse bioassay 1.2.1.1

tection, analysis and control of toxicity in shellfish in thod for the deeThe traditional mouse bioassay (MBA) (Yasumoto et al. 1978). It is es is the mEuropean monitoring programmcurrently the reference method under EU legislation (Aune et al. 2007). In this method,
shellfish extracts are injected intra-peritoneal into three mice and the mice are monitored over
e, the e framce die within the timia certain period of time. Should at least two of the mshellfish are declared to be unsuitable for human consumption. The maximum permissible
(the whole body or any enotoxins in shellfish level of okadaic acid, dinophysistoxins and pectpart edible separately) is laid down to 160 µg of okadaic acid equivalents/kg, whereas for

7

General Introduction

YTX equivalents/kg (Decision 2002/225/EC) (Mouratidou et of gmyessotoxins this level is 1

nutes (Aoac 1999). If the ie is only monitored for 60 mousal. 2006). For PSP toxins the m

mouse is still alive after this time the sampleis regarded as negative. The detection limit for

it for PSP toxins ately 300 µg/kg of shellfish flesh and if the regulatory limMBA is approxim

lof 800 µg/kg shellfish fesh is reached the harvesting area is closed (Holtrop et al. 2006).

Toxin concentration or toxin type are not quantitatively or qualitatively measured and this

2001). riability (Flanagan et al.aassay is recognised as having poor reproducibility and v

However, the use of animal assays induce ethical problems as even with non-toxic samples

the injection of 1 ml of the acidic extract into the abdomen of a 20 g mouse causes

(Holtrop et al. 2006). This presents the urgent alconsiderable pain and suffering to the anim

need to replace the mouse bioassay with a more suitable monitoring method.

Methods for the detection of toxins 1.2.1.2

ance liquid chromHigh-performatography (HPLC) is a widely used technique for the analysis

and high sensitivity. A wide range t peak resolutionof shellfish toxins that provides excellen

of toxin structures can be separated with this instrument (Quilliam 2003b). The preferred

analytical method is the use of HPLC in comination with UV absorbance detection, which b

2003a). Organic extract of atories (Quilliamhas been used since 1987 in regulatory labor

complex and the toxins have to be extracted using organic shellfish tissue and plankton are

n alkaline oxidation treatm with the HPLC. A pre- or post-columore analysissolvents befent

of a sample for the detection of PSP toxins is required for the fluoremetric detection (Franco

en mn HPLC oxidatioand Fernández 1993; Luckas et al. 2003). A pre-columthod proposed by

Ménard 1991) and Lawrence et al. (1996) Lawrence and Ménard, (1991) (Lawrence and

Pe results but does not separate all PSnsitiv(Lawrence et al. 1996) can produce fast and se

toxins. The method of Oshima (1995) (Oshima1995) can separate all PSP toxins, however, it

order toarate runs ining because of the need to perform three sepis very time-consum

payo 2001).ine all the toxins (Vale and De M. Samdeterm

The analysis of mrine toxins can also be carried out using capillary electrophoresis (CE); it a

2003b). Separation by electrophoresis provides fast and high-resolution separation (Quilliam

is based on differences in solute velocity in an electric field. Thibault et al. (1991) described

PSP toxins. CE is a ination ofthe use of CE with UV detection for the separation and determ

rapid and efficient method that needs only a small volume of sample (Thibault et al. 1991).

ple is required for an effective analysis (Zhao et al. 1997). However, a purified sam

8

General Introduction

Counting techniques 1.2.1.3

Microscope-based methods can identify and quantify microalgae at the species or genus level.

Compound microscopy is a simple and quick method to estimate cell numbers from a drop of

bers below w cell numseawater using counting cells, such as the Sedgewick-Rafter cell. For lo102-104 cells L-1 the cells have to be concentrated before counting. Another possibility for the
quantification of low cell numbers uses an inverted microscope and Utermöhl sedimentation

chambers to concentrate the algae in a sample (Utermöhl 1958). This method can last from a

entation settle cells in the sedimys because of the time needed toafew hours to several dension of e, the fixative used and the linear dimmple volumber, which depends on the sacham

bers can also be counted using quantitative epifluothe cells. Low cell numcerescench as DAPI or ls onto filters and staining. Several stains sucroscopy by concentrating the celim

Acridine Orange, can be used. (Andersen and Throndsen 2003). For the identification of

unicelluar algae, using microscope-based methods, a broad taxonomic knowledge is required,

because toxic and non-toxic strains can belong to the same species and thus are

plex) (John et al. 2005). species comAlexandrium tamarense morphologically identical (e.g.,

1.2.1.4Data buoys and remote sensing using satellites

Marine data buoys are used to monitor plankton as well as physical, chem

ical and

meteorological variables in situ and in real-time. For example, the CytoBuoy (CytoBuoy,

e series Bodegraven, Netherlands), can be used to conduct extended and/or high frequency tim the of phytoplankton distribution and abundance on fixed locations. Several buoys fromthe Norwegian coast and forecasting of upstream uoy System are located along Seawatch B

her new HAB buoy systemayda 2003). Anots can be facilitated (Smbloom identifies species

era for in-flow acquisition (Culverhouse et al. 2006). Recently the using a high speed cam

pling processor (ESP) was introduced (Doucette et al. 2006). The ESP is ental samenvironman electromechanical/fluidic system that collects discrete water samples and concentrates
microorganisms. An automated application of molecular probes is carried out that identifies
s and their gene products (Doucette et al. 2006).croorganismim

Also satellite images are used processes affecting local phytoplankton populations. Sea surface temto achieve understanding of the regional influences of physical perature images can aid

the prediction of transport of noxious phytoplankton. Toxic phytoplankton cannot be

9

General Introduction

with high cell ote sensing. However, detection of a monospecific bloomidentified using rem and ocean-colour acounts at the surface is possible by using species-specific chlorophyll imagery. For example, for the detection of Karenia brevis about 105 cells L-1 are necessary,

which would result in early warning of fish kills but not shellfish toxicity (Franks and Keafer

2003).

Detection of harmful algae using molecular probes or antibodies 1.2.1.5

In the past decade, a variety of molecular methods have been adapted for the detection of

as tools to aid the iew for the use of molecular probes harmful algae. The first rev

presented by Anderson ful algal species was identification of harm(1995) (Anderson 1995).

Today molecular probes are widely applied for the identification of micro-organisms. The

their high target numusual targets for probes are the smber in the all and the large subunit ribosomcell. More or less conserved regions in these genes mal RNA genes, because ofake it

ic levels (Groben et al. 2004). possible to develop probes that are specific at different taxonom

Fluorescencein situ hybridization (FISH) uses a fluorescently labelled probe that is designed

. The probe is hybridized inside the to recognize a specific sequence of a particular organism

intact cells, the ribosomes and cells containing a fluorescently labelled probe can then be

osoi-Tanabe and Sako 2005). FISH allows the croscopy (Hidetected using epifluorescence m

i epifluorescence mrapid detection of different algal groups bycroscopy and even the

separation of closely related and morphologicallysimilar species (Lim et al. 1993; Scholin et

on et al. 1997; Simal. 1996; Scholin et al. 1997; Simon et al. 2000; Groben et al. 2004; Sako

ndwich hybridization assays (SHA) can also aöbe et al. 2006). Sit et al. 2004; Tet al. 2004; Sm

provide the possibility to identify and enumerate toxic algae rapidly. SHA relies on extracted

nucleic acids from cell lysates. A capture probe bound to a solid surface immobilizes the

nd foral RNA atarget ribosom hybrid complex with a second signal probe. An antibody-ms a

iety of the signal probe and reacts with a substrate oal me complex binds to the signenzym

forming a colorimetric product or an electrochemical current (Scholin et al. 1996; Tyrrell et

al. 2002; Metfies et al. 2005). Just recently, the SHA was validated and accepted for

atory use in New Zealand in May 2004 (Ayers rcial laborn for commeinternational accreditatio

et al. 2005). DNA microarrays are used in many applications because of the possibility to

ets in parallel without a cultivation step ber of up to 250,000 different targanalyze a large num

(Lockhart et al. 1996; Graves 1999; Ye et al. 2001). This technology is also used to 2004; Metfies and Medlin 2005b; Ki and Han croalgae (Metfies and Medlinidifferentiate m

10

General Introduction

with special surface croarray consists of a glass-slide i2006; Godhe et al. 2007). A m

properties (Niemeyer and Blohm 1999) and is spotted with many copies of nucleic acids in a

specific pattern, e.g., oligonucleotides, cDNAs or PCR-fragments (Graves 1999). The most

common type of probes used in HAB research are antibodies (Scholin et al. 2003). Atibodiesn

lecules, such as peptides, glycoproteins and toxins. Many of the obind to different m

tested in laboratory but only a few in field developed antibodies for HAB species have been

ny techniques for aary as well as secondary antibodies are applied; however, mstudies. Prim

thod using a fluorescent eploy the indirect-labelling mHAB species identification em

secondary antibody (Mendoza et al. 1995; Cordova and Muller 2002; Scholin et al. 2003;

West et al. 2006). Detection of harmful species employing the polymerase chain reaction

ent of entary strands of nucleic acids. Only a fragm(PCR) is based on the binding of complem

ers that define the size of the e is targeted, based on the use of oligonucleotide primthe genom

fragment as well as the taxonomic specificity of the reaction. PCR requires the extraction of

nucleic acids from the sample, primers and an amplification protocol (Scholin et al. 2003).

Direct quantitative PCR using fluorescent probes was recently used by Bowers et al. (2000) to

species. In this assay, the detection of amplified target DNA required the Pfiesteriadetect

-3’ exonuclease activity tide probes. The 5’- toannealing of fluorescently labelled oligonucleo

of the taq polymerase cleaves the probe and the quencher dye is released from the emitter dye,

which in turn is then able to fluoresce (Bowers et al. 2000). The relative fluorescence is

related the number of free fluorescent molecules in solution and the cycle of fluorescence

detection is directly related to the number oftarget molecules in the initial reaction mixture.

However, sensitivity and specificity of the assays has to be analyzed and the application for

some field samples can be problematic, if sample composition inhibits DNA extraction and

purification (Scholin et al. 2003).

Biosensors1.3

ical recognition with signal transduction for the detection of specific mBiochemlecules is o

quence,eical biosensors. The detection component, such as a probe sbined on electrochemcom

catalyzes a reaction with or specifically binds e or other biomolecules, an antibody, an enzym

to the target of interest. A transducer coms this detection event into a ponent transform

measurable signal. A specific detection of targets in a complex sample is possible. Biosensor

types comprise optical, bioluminescent, thermal, mass and electrochemical recognition (Gau

ental monitoring, biothreat ronmet al. 2005). Various sectors, such as clinical diagnostic, envi

detection and forensics, apply single electrode sensors as well as arrays (Berganza et al. 2006;

11

General Introduction

ultaneous detection ofectrodes enable a simlo et al. 2006; Taylor et al. 2006). Arrays of eLerm

llinbes (Farabuecies with different molecular promultiple spi et al.; Dock et al. 2005).

ples into thevent the need to return sam and therefore circumin situBiosensors can be used

laboratory. Rapid identification of aquatic microorganisms as well as physical and chemical

measurements of the environment are importantto understand coastal dynamics and processes

that can impact marine ecosystems, such as the introduction and spreading of microbial

pollutants and the initiation of HABs (Lagier et al. 2005). Metfies et al (2005) introduced a

biosensor in combination with a hand held device for the detection and identification of the

(Metfies et al. 2005). The biosensor has the Alexandrium ostenfeldiitoxic dinoflagellate

thod for the identification of harmful algae. ek and easy mpotential to serve as a quic

isAim of thes1.4

ent and evaluation of fast and reliable monitoring My thesis was assigned to the developm

methods using molecular technologies. Harmfulalgal species are responsible for fish and

ers through ingesting of contaminated seafood. The shellfish kills and poisoning of consum

detection and enumeration of harmful algal species is important from an ecological and

economic point of view. The current monitoring methods are time consuming and require

challenging icallyare taxonoment. Unicellular algae trained personnel and expensive equipm

and some of them have only few morphological markers for reliable identification. The aim of

this thesis was to design and adapt molecular probes for the identification of toxic algae.

thods developed were adjusted and evaluated to serve as potential early eore, the mFurtherm

s for toxic algae.warning system

1.5 thesisfOutline o

tiontion of molecular probes for sandwich hybridizat and adaptaDevelopmen1.5.1

belongs to the mAlexandrium minutumThe speciesst potent PSP-toxin and other toxin o

A.ento et al. 2005). producers (Taylor and Fukuyo 1998; Chen and Chou 2002; Nascim

equency frcan be observed world-wide and its geographic range as well as its bloomminutum

). Monitoare increasing (Lilly et al. 2005volves the accurate ring of toxic algae in

croscopyiof species by using standard merationmorphological identification and enum

procedures.A. minutum is difficult to distinguish fromother species of the same genus

because it is characterized by minute details of its thecal plates (Taylor et al. 1995). The small

12

General Introduction

and the large subunit ribosomal RNA genes have more or less conserved regions that make it
ben et al. 2004). Molecular probes et specificity (Grog targinpossible to design probes of vary Sandwichal species.e toxic algall percentage of thhave been developed only for a sm

ehybridization mal RNA (rRNA) probes is a suitable tool thods using species-specific ribosomfor the rapid and reliable detection of harmful algae.

InPublication I a commercially available PCR ELISA Dig Detection Kit was adapted for the

ich hybridization in a using sandwAlexandrium minutumdetection of the toxic dinoflagellate

microtiter plate. For the detection ofA. minutum a set of two 18S rRNA probes was

e specificity of the hudwig et al. 2004). Tdeveloped using the ARB software package (L

probes was tested using the microtiter plate assay and also closely related species. An

of this study was to investigate the potential of the modified assay for the additional aim

ination. For the ber determensive cell numdetection of harmful algae without labour-intdetection of A. minutum by means of standard calibration curve the total rRNA concentration

assay and the standard curve were evaluated by using ined. Theper cell had to be determ

ples.spiked water sam

Design and evaluation of probe sets for toxic algae 1.5.2

Phytoplankton communities consist of assemblagesof co-occurring species and the temporal

position in the sea is substantial (Venrick 1999; Figueiras et al. and spatial variability in com

2006). The composition of the harmful algae species in different areas of Europe is complex

,Dinophysis,Alexandriumand several algal genera include toxic species, such as

Gymnodiniumon et al. 1997; John et al. 2003; Moita et al. 2003; (SimPseudo-nitzschia and Chepurnov et al. 2005). Molecular techniques for the detection of toxic algae require the use

ecies.of probes targeting specific genes of the target sp

probe sets for the species-specific identification of the toxic algal speciesPublication IIIn

Gymnodinium catenatumProtoceratium reticulatum, Lingulodinium polyedrum, Prymnesium ,

Pseudo-nitzschia multiseries, P. australis, P. seriata ,parvum, Chrysochromulina polylepisats. An were designed and adapted for the use in sandwich hybridization formand P. pungens

already existing probe set for the genus Pseudo-nitzschia was adapted. Target species as well
as closely related species were utilized for the verification of specificityin the microtiter plate

assay.

13

General Introduction

Improvement of detection protocol 1.5.3

Today biosensors are commonly used in clinical diagnostic, environm

ental monitoring,

biothreat detection and forensics. The advantage of biosensors is the possibility to measure

boratory is unnecessary. Biosensors are usedlaort to the and therefore, sample transpon-site

for the rapid identification of aquatic microorganisms. Metfies et al. (2005) introduced a

biosensor for the identification of the toxic dinoflagellate Alexandrium ostenfeldii for the first

e (Metfies et al. 2005). tim

InPublication III a description and illustrative visualization of the method introduced from

thod to a eMetfies et al (2005) is presented. The aim of this work was to bring up the m

standard for ease of use through others. For this purpose it was necessary to adapt the method

to sensor chips and a measuring device from another manufacturer. Furthermore, the

equipment needed for a complete sample analysis was identified and modifications of

protocols were presented.

Assessment of probe modification fo1.5.4ancementr signal enh

Identification of microbial species with probe-based methodsrequires sensitive and highly

bes. The specificity of the probes depends on the numspecific prober of sequences of the

target gene available in databases. Probes designed from a low number of target species or for

ecies can detect also non-tively unknown or unculturable spa group, which includes rela

targeted species (cross-hybridization). Additionally, many non-targeted species exist whose

ise of probes is necessary because ve not yet been determined. The frequently revasequences h

new sequences are added to databases on a daily basis. The introduction of locked nucleic

acid (LNA) probe technology promises an enhancement of both specificity and sensitivity of

molecular probes (Kongsbak 2002).

Publication IV involved the revision of probes for Alexandrium ostenfeldii and the

conventional mparison of specificity and sensitivity of comlecular probes and LNA modified o

thods, sandwich hybridization on eization mprobes. Two different solid phase hybrid

croarrays, were used for the detection of probe signals. The set of imbiosensors and DNA-

18S-rRNA probes for pact of LNA-probes on the was applied to assess the imA. ostenfeldii

specificity of probes with the biosensor, thus, the sequence of the capture probe was

14

General Introduction

redesigned with locked nucleic acids. Three different species, A. ostenfeldii,A. minutum and
A. minutumodified probes. probes and LNA m, were tested with conventional A. tamutumpreviously showed low cross-hybridization signals (Metfies et al. 2005) and the 18S rRNA sequence of A. tamutum possessed only one mismatch to the capture probe. Five probes, that
croarray. One of the probes targets the imre evaluated with the DNA-target the 18S-rRNA, wesuper kingdom of Eukarya and the other probeseach of these four major phyla of algae: the
locked nucleic acid mChlorophyta, Bolidophyodita, Prymfications were evaluated. nesiophyta andCryptophyta. For each probe, two different

1.5.5Development and evaluation of a biosensor

Monitoring programmes at the world-wide coastlines observe phytoplankton compositions
ouse-bioassay is statutory for lication of the m The appful algal species.and especially harmthe mby HPLC. The monitoring of toxin contamouse-bioassay induces ethical prination of shellfish, whereas toxin determoblems because of the painful procedure for ination is performed
the animals; HPLC, in turn, is a very time-consuming and expensive method. Traditional
methods, such as light microscopy, are time-consuming when numerous samples consisting of
ic knowledge as ny species have to be routinely analyzed and require a broad taxonomamplished using e.g. DNA-ultaneous detection of multiple species can be accomwell. Simmicroarrays with different molecular probes (Metfies and Medlin 2005b). The utilization of
all described methods requires transportation of samples to specialised laboratories and high
in around five working days and therefore, trained staff. The results are achieved withpreventive measures are not always possible.A fast identification of aquatic microorganisms
is realized by the use of biosensors. The in situ investigation of coastal water for the presence
formof different toxic algae could ation and thus, potential shellfish contamprovide a potential early warning ination.tool for monitoring of bloom

InPublication V, the ability and adaptability of a biosensor for the rapid and reliable in situ
detection of toxic algae was investigated. The aim of this study was the design and evaluation of a multiprobe chip and an automated device in order to facilitate the detection of several
species simultaneously. For the design of the multiprobe chip, different materials for
electrodes and the carrier material were tested to obtain accurate signal formation using
lecular probes. An adaptation of analysis and hybridization osandwich hybridization and mprocedures was necessary for the use of the biosensor by layperson. Fore a portable rthermudevice was designed, which performs the analysis in a semi-automated manner.

15

Publication list

2.Publications

List of publications 2.1

s:ing publicationThis doctorial thesis is based on the follow

I.

II.

III.

IV.

V.

SONJADIERCKS,LINDAK.MEDLIN AND KATJAMETFIES
COLORIMETRIC DETECTION OF THE TOXIC DINOFLAGELLATEALEXANDRIUM

MINUTUM USINGSANDWICH HYBRIDIZATION IN A MICROTITER PLATE ASSAY
ittedbmHarmful Algae, to be su

SONJADIERCKS,KATJAMETFIES AND LINDAK.MEDLIN
OLECULAR PROBES FOR THE DETECTION OF TOXIC ALGAE FOR USE IN M

ICH HYBRIDIZATION FORMATSSANDWitted submeJournal of Plankton Research, to b

SONJADIERCKS,KATJAMETFIES AND LINDAK.MEDLIN
ENSORWITH A BIOSTOXIC ALGAE LECTROCHEMICAL DETECTION OF EManual and Guides: Microscopic and molecular methods for quantitative
ittedton analysis, submphytoplank

SONJADIERCKS ANDCHRISTINEGESCHER,KATJAMETFIES,LINDAK.MEDLIN
IDS FOR SIGNAL ENHANCEMENT OF D NUCLEIC ACVALUATION OF LOCKEE

ON SOLID SURFACESOBILIZEDOLIGONUCLEOTIDE PROBES FOR MICROALGAE IMMnology and OceanoLimittedgraphy: Methods, subm

SONJADIERCKS,KATJAMETFIES,STEFFIJÄCKEL AND LINDAK.MEDLIN
DEVELOPMENT AND OPTIMIZATION OF A SEMI AUTOMATED RRNA BIOSENSOR

FOR THE DETECTION OF TOXIC ALGAEitted submeBiosensors and Bioelectronics, to b

16

MOLECULAR TECHNIQUES FOR CLASSICAL AND NTERCALIBRATION OFPublication list

e: the period of timOther publication prepared with contribution of the candiate from

GODHE,A.,AND OTHERS (2007)

I

IDENTIFICATION OF ALEXANDRIUM FUNDYENSE(DINOPHYCEAE) AND ESTIMATION

gae, 6: 56-72.Harmful Al

OF CELL DENSITIES

nse publicatioStatement of my contribution to th2.2

I Publication

entsents were planned together with K. Metfies and L. K. Medlin. The experimThe experim

were carried out by myself and analyzed by myself. The manuscript was written by myself.

II Publication

ed by ents were planned together with L. K. Medlin and K. Metfies and performThe experim

nuscript.amyself. I have analyzed the data and wrote the m

III Publication

The experimed by ents were planned together with L. K. Medlin and K. Metfies and perform

nuscript.amyself. I wrote the m

IV Publication

ents were planned together with K. Metfies, L. K. Medlin and C. Gescher and The experim

carried out from C. Gescher and myself. The manuscript was written equally with C. Gescher.

V Publication

ents were planned together with L. K. Medlin, K. Metfies. S. Jäckel was involved The experim

ents were ent of the lysis buffer. All other experiments for the developmin the experim

performed and analyzed by myself. I wrote the manuscript.

17

Publication I

I:Publication2.3

COLORIMETRIC DETECTIONOF THE TOXIC DINOFLAGELLATEALEXANDRIUM

UMMINUT

USINGSANDWICH HYBRIDIZATION IN A MICROTITER PLATE ASSAYSONJADIERCKS,LINDAK.MEDLIN AND KATJAMETFIES
Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570
anyerhaven, GermBremittedbmHarmful Algae, to be su

Abstract

s is an tal areas and shellfish farmrmful algae in coasaRapid and reliable detection of himportant requirement of monitoring programs. Molecular technologies are rapidly improving
nation ofi. Assays are based on the discrimthe detection of phytoplankton and their toxinsA commercially available PCR ELISA Dig genetic differences within different species. Detection Kit was adapted for the detection of the toxic dinoflagellateAlexandrium minutum
using sandwich hybridization in a microtiter plate. A set of twoprobes for the species-specific
bes was successfully. The specificity of the proA. minutumidentification was developed for demonstrated with the microtiter plate assay.A standard calibration curve for different RNA
assay. Total rRNA was isolated ined for the concentrations and thus cell counts was determfrom three different strains of A. minutum and the mean concentration of RNA per cell of was
ined to be 0.028 ng. The assay and the standard curve were evaluated by using spiked determwater samples. The results demonstrate that the molecular assay was able to detectA.
experimminutuments with spiked natural samcells at different cell counts in the presence of a comples presenta proof of principle of this assay. These tests plex background. The
also provided the necessary specificity tests prior to the probes being adapted to an automated
at (Metfies et al. 2005). biosensor using a sandwich hybridization form

18

Publication I

nIntroductio

(HABs) has increased. Coastal algal blooms fulrrence of harmOver the last decades the occu

systems around the world have had fish kills, outbreaks of shellfish poisonings, deaths of

marine mammals and loss of quality of coastal waters for recreational use. Phytoplankton

population initiated by suitable croalgalis are defined as a sudden increase in the mbloomconditions for growth, and reach cell concentrations up to 104-105 L-1 (Maso and Garces

s are considered harmful: the toxin producers and the 2006). Two types of causative organism

high-biomass producers. Around 4000 marine planktonic microalgaeare described so far. Of

dinoflagellates) and about 200 can be noxious ainlythese, around 97 are toxic species (m

ful/noxious species belong to six (Zingone and Enevoldsen 2000; Moestrup 2004). These harm

s, dinoflagellates, haptophytes, raphidophytes, cyanophytes, and algal groups: diatom

gical and ecological morphological, physiolos ofpelagophytes, which differ greatly in term

the dinoflagellates, 23 species are known to ongcharacteristics (Maso and Garces 2006). Am

produce potent toxins, such as saxitoxins. Saxitoxins are responsible for the life-threatening

option of mparalytic shellfish poisoning (PSP), which can be caused by the consumlluscs that

(Daranas et al. 2001; Chou et al. Alexandrium toxic dinoflagellates of the genus have filtered

2004) as their food source. The identification of the genus Alexandrium bymeans of

, cell size and shape of the apical pore is morphological characteristics, such as general form

s cannot be used alone for difficult and labour-intensive. The morphological characteristic

Alexandrium species identification because of their similarity to other microalgae, and, in

addition, intermediate morphological forms (Cembella and Taylor 1985; Hosoi-Tanabe and

proved monitoring, rapid detection and Sako 2005; John et al. 2005). Consequently, an im

enumeration of toxic algae is crucial. Within the genus Alexandrium,the species Alexandrium

tst poteno, which has been observed world-wide (Lilly et al. 2005), belongs to the mminutum

algal group of PSP-toxin and other toxin producers (Taylor and Fukuyo 1998; Chen and Chou

2002; Nascimento et al. 2005). The geographic range and bloom frequency of A. minutum is

increasing (Lilly et al. 2005). Monitoring methods based on light microscopy are time-

consuming and costly if a large number of samples need to be processed. For the

identification of some species, highly-trained staff and expensive equipment are needed.

hybridization or FISH (Anderson in situ Molecular techniques, such as whole cell fluorescent

and Sako 2005), sandwich hybridization et al. 2005; Hosoi-Tanabe and Sako 2005; Kim

assays or SHA (Tyrrell et al. 2002; Matweyou et al. 2004; Metfies et al. 2005), PCR-based

assays (Penna 1999; Guillou et al. 2002) and monoclonal antibody probes (Anderson et al.

19

Publication I

1999) can identify phytoplankton species. The principle of the sandwich hybridization was

d represents a DNA probe-based atteo et al. 1995; Rautio et al. 2003) anintroduced by (Zamm

micro-algae that uses two species specific oligonucleotide ication ofthod for rapid identifem

al RNA probes targeting ribosomRNA) (Ayers et al. 2005), one to capture the target r(

molecule and the other to carry the detectable signal (Figure 1). Oligonucleotide DNA probes

are designed to bind to complementary sequences of the small or the large subunit ribosomal

RNA algal genes and have a length of 18-25 basee possibility to design probes of h pairs. T

e rRNA regions of thet specificity is possible because of more or less conservedvarying targ

molecule (Groben et al. 2004). It is necessarythat the specificity of probes is extensively

tested, so that false positives are not encountered. The probes must be tested so, that close

and probe neighbours (probe testneighbours (clade tests)s, target sequence close, but

. Such extensive tests require a rapid and phylogenetically unrelated) do not bind to the probe

easy to use format so that the many variations in hybridization conditions and test organisms

hybridization with can be verified as non-reactive. For FISH probes, the dot blot

chemiluminescent detection provides this vehicle for probe specificity testing prior to FISH

applications (Groben and Medlin 2005).

eIn this study a fast and simple m is presented, Alexandrium minutumthod for the detection of

whose principle is based on a sandwich hybridization with the capture oligonucleotide probe

bound to the well of a microtiter plate. The commercially available PCR ELISA Dig

Detection Kit from Roche Diagnostics (Mannheim, Germany) was adapted to the sandwich

hybridization assay as a rapid, cost-effective, easy-to-use method that requires minimal

ic capture probe that prises a biotinylated target specifhandling. The assay presented here com

crotiter plate. Target nucleic acid hybridizes to ibinds to the streptavidin-coated well of a m

plex.lled signal probe hybridises to this comthe capture probe and a second digoxigenin-labe

The detection and visualisation take place via an anti-digoxigenin peroxidase conjugate and

colorimetric substrate. This method provides an easy to use method to test for probe

onitoring of field samples.tial to be used for routine mspecificity and has poten

20

Publication I

Material and Methods

- The algal strains used in this study were cultured under Cultures and growth conditions sterile conditions in seawater-based K-medium (Keller et al. 1987), IMR-medium (Eppley et
al. 1967), F2-medium (Guillard and Ryther 1962; Guillard 1975) and Prov (Provasoli et al.
peratures listed in Table 1. All cultures 1957; Guillard and Ryther 1962; Guillard 1975) at temwere exposed to a photon irradiance rate of 150 µEinstein –200 µEinstein provided by white lamps at a light:dark cycle of 14:10 h.

the different algae cultures prior to Aliquots were taken from -Cell counts of algae cellsbHan Coulter Gmharvesting and counted using the Multisizer 3 Coulter Counter (Beckmany).Diagnostics, Krefeld, Germ

Isolation of RNA - Total RNA was isolated from all algal cultures with the RNeasy Plant Mini
dified for en was mo Qiagany) and the isolation protocol fromKit (Qiagen, Hilden, Germcentrifugatioquality enhancemn time was increased froment. Having applied the cell lysate to the QIAshredder spin colu 2 to 15 minutes to improve separation of supernatant mn, the
fromadding an incubation tim cell debris. The first washing step withe of one minute on the buffer RW1 waRNeasy spin colums repeated twice and mn. Furthermore, the first odified by
wash step with buffer RPE was repeated. A Nanodrop Spectrophotometer (Peqlab, Erlangen,
NA concentration.easure the Rany) was used to mGerm

ented in - Prior to hybridization, the total rRNA was fragmFragmentation of RNAfragmentation buffer (40mM Tris, pH 8.0/100mM KOAc/30mM MgOAc) for 5 minutes at 94
°C and then chilled on ice.

PCR ELISA (DIG DeteELISA Dig Detection Kit fromction) Roche Diagn kit contents and preparation of working solutions ostics (Mannheim, Germany) contains several - The PCR
conjugate dilution buffer, substrate buffer, reagents; however, only the hybridization buffer,, washing tablets and ABTS tablets were D)Anti-digoxigenin-POD conjugate (anti-DIG-POused in these experimentsfor the identification of A. minutum. The microtiter plates are
provided as plate mblocking reagent. The kits are stored at 4 °C. Prior to the experimodules (8 wells each), pre-coated with streptavidin and post-coents, the washing solutioated withn

21

Publication I

e Anti-DIG-in two litre deionized water. Thwas prepared by dissolving one washing tablet

POD is lyophilised and was dissolved in 250 µL of double distilled water.

- The biotinylated capture probe, the digoxigenin-labeled signal probe and the nHybridizatio

positive control were diluted to a concentration of 10 µM prior to hybridization. For the

sandwich hybridization, 4 µL of each probe and different concentrations of rRNA were added

e of 250 µL. A negative control was nal volumfer resulting in a fito the hybridization buf

prepared containing only both probes and hybridization bufwhereas the positive control fer,

included additionally the test DNA (target sequence of the probes). Hybridization solution

containing the RNA, the negative and positive controls were added into the wells of the

microtiter plate and incubated on a shaker for 1 hour at 46 °C.

e Anti-DIG-POD - The Anti-DIG-POD working solution (1 volumIncubation with antibody

es conjugate buffer) was prepared at least one hour prior to the incubation step and 99 volum

perature before use. Subsequent to therate to room tem equiliband stored in dark to

otiter plate were washed with washing solution three rcihybridization, the wells of the m

times. 200 µL of antibody solution were added to each well and incubated for 30 minutes at

37 °C with agitation in the dark. The antibody is directed against the digoxigenin label on the

signal probe.

Incubation of substrate solution - Substrate soed by adding one tablet of was preparlution

ABTS to 5 mL of substrate buffer and stored protected from light. The substrate solution was

allowed to equilibrate to room temperature before use. After the incubation with the Anti-

es with washing solution, 200 µL of e timDIG-POD, the wells were washed again thre

a shaker at 37 °C for cubated in the dark on in the wells andsubstrate solution were filled in

30 mnutes. The hybrids are detected using an anti-digoxigenin antibody conjugated to i

etric product. green colorimhorseradish peroxidase that reacts with substrate to produce a

Reading of microtiter wells- Each well of the microtiter plate was read out at 405 nm using a

stadt,eter (Varian Inc., Darmis Spectromquartz cuvette with a Varian Cary 4000 UV-V

any).Germ

of spiked water samples Preparation the estuary of the men frople was tak- A water sam

Weser River (German Bight) with a natural phytoplankton population as a matrix. The water

22

Publication I

eter, Millipore, USA) to ilter (45 mm diam over a 180 µm nylon fple was pre-filteredsam

ove larger particles, such as zooplankton. Sedimrementation was allowed over night and

rough a 10 µM polycarbonate filter (45 mm d ths filtereple wasubsequently the water sam

diameter, Millipore, Billerica, USA). 500 mL of the supernatant was filtered over a 5 µm

polycarbonate filter (45 mm diameter, Millipore, USA) to collect the remainingmatrix and

spiked with and other algae cells with Alexandrium minutumunts of three different cells co

different cells counts (Table 2). The samples were prepared in triplicate. RNA was isolated

from the samples as described above and analyzed with the microtiter plate assay.

Results

the probe design option within the ARB software FromDesign of oligonucleotide probes -

package (Ludwig et al. 2004) two probes were designed for the sandwich-hybridization that

a database consisting of more (Table 3) fromAlexandrium minutum bind to the 18S rRNA of

than 3000 published and unpublished algal 18S rRNA sequences. Two probes were chosen

next to each other in the target sequence in case the target nucleic acid was degraded and the

, probe AMINC In silicoe length strand of rRNA. sites were no longer accessible from the sam

is specific forA. minutum and has at least one mismatch against A. insuetum and two

mismatches against all other non-target organisms listed in the ARB database. Probe

AMINCNEXT recognizes not only A. minutum, but also A. ostenfeldii,A. tamutum and A.

insuetum. Furthermore, it only has one mismatch against A. affine, but two mismatches

. 1990) was conducted to test the s. A BLAST search (Altschul et alagainst all other specie

available sequences. Positive control and probes ainst alle probes agoverall specificity of th

were synthesized from Thermo Electron Corporation (Ulm, Germany). Thus, from these in-

probe and AMINCNEXT as signal probe. s defined as capture tests, AMINC wasilico

Alexandrium minutum - The specificity of the Specificity of probes probes was tested using

crotiter plate well. Total RNA was isolated from ithe sandwich-hybridization-assay in a m

different strains of the target speciesAlexandrium minutum and more distantly related species

of the genus . The obtained signals were normalised to Gonyaulax spinifera and Alexandrium

a target concentration of 350 ng RNA and compared to one another. Signals were observed

ined for the non-targeted species strains, whereas no signals were determnutumA. mifor all

(Table 4).

23

Publication I

Alexandrium minutum - In a range of 10,000 to 500,000 n per celltioTotal RNA concentra

counts of three strains (AL3T, cells, total RNA was isolated in triplicate from different cell

growth igure 2) at optimumion per cell (Fine the RNA concentratAMP4, AL5T) to determ

conditions, because this correspondsmost closely to bloom development in the field (Ayers et

the different strains show variations in the RNA concentration for the e curves ofhal. 2005). T

es of the different strains show a straight bers. However, all three curvdifferent cell num

proportional development. For each strain, a mean RNA concentration per cell was calculated

cell counts. Strains AMP 4, AL3T and AL5T the RNA concentration of the differentfrom

contained 0.017 ng, 0.027 ng and 0.036 ng RNA per cell, respectively. The mean

concentration of total RNA per cell for the Alexandrium minutum strains was determined to

be 0.028 ng.

Standard Crves of photometer readings to cell countsucrotiter plate assay using a i - The m

sandwich-hybridizationand specific probes for Alexandrium minutum detected hybridization

signals for different RNA concentrations and thus these values could be converted to cell

A.. The photometer readings (Figure 3) for isolated total RNA of three A. minutumbers of num

ean absorbance of 0.0297 forstrains showed a linear increase in signals from a mminutum

10,000 cells to 1.7757 for 500,000 cells. Strain AL5T produced higher probe signals than

strains AL3T and AMP4; however, average values of the tested strains were observed to be in

the same range as the signals for strain AL3T.

thod evaluation, a natural water sample was e - For mMethod application to spiked samples

ulate real samples as closely astaken and spiked with different numbers of cells to sim

possible. The photometer readings from the microtiter plate assay and Alexandrium minutum

A. and field samples spiked with A. minutumpared using a lab culture ofprobes were com

. Signals for 10,000 cells of minutumples were slightly above the for both samA. minutum

background but still measurable (Figure 4). The spiked sample with 50,000 cells of A.

minutum gave a signal of 0.055, which was fourfold lower than the signal of 0.199 for a lab

ple with 100,000 ilar cell concentration. Also the signal for the spiked samculture at a sim

cells was threefold lower than that for the lab culture.

24

Publication I

Discussion

In this study, a new method for the detection of the toxic dinoflagellate Alexandrium minutum

any), Germ Mannheimis presented. The PCR ELISA Dig Detection Kit (Roche Diagnostics,

iin a moat using twcrotiter plate was successfully adapted to a sandwich hybridization form

and the signal probe is otin-labelledi is bdifferently labelled probes. The capture probe

digoxigenin-labelled. The probes used in the sandwich hybridization presented here are

A. minutum.targeted against the 18S-rRNA of Sandwich hybridizations and rRNA targeted

croalgae (Scholin et al. 1996; ie detection of mprobes are used in different applications for th

halloran et al. 2006). . 2005; O'Tyrrell et al. 2002; Ayers et al

ftware package (Ludwig et al. 2004). The oProbes were designed using the software ARB s

was shown using sandwich hybridization in a A. minutumspecificity of the probes for

microtiter plate well. The signals for all A. minutum strains were always significantly above

the signals for the non-target species as predicted by the in-silico tests. Moreover,

Alexandrium species with a single mismatch in the target sequence were not detected with the

et speciespetitor to block these non-targsandwich hybridization even without the use of a com

sing with the capture probe. More distantly related specie hybridisand prevent the RNA from

ing thwere not tested with the assay assumat the species with the fewest number of

mismatches would present the highest possibility of unspecific binding. Distantly related

species have even more mismatches to the probe sequences and probe binding would be

ity to one another in the target 18S-e proximunlikely. The probes were designed to be in clos

rRNA sequence to avoid a loss of signal if the target RNA molecule was degraded.

the totalA. minutum,titer plate assay for To develop a standard calibration curve of the micro

rRNA concentration per cell was determined at optimum growth conditions for three different

strains as this was expected to correspond most closely to bloom development in the field

different(Ayers et al. 2005). A mean concentration of 0.028 ng rRNA per cell was found. The

strains were not synchronised, consequently a part of the culture could have been in the lag or

stationary phase. This calculated rRNA concentration per cell ofA. minutum also

corresponded to that obtained for A. fundyense (data not shown) and A. ostenfeldii (Metfies et

al. 2005). Additionally, similar findings were achieved for different growth conditions for A.

minutum (personal communication L. Carter, Westminster University, London, UK). A

ent rRNA concentrations and consequently different cestandard calibration curve for differll

25

Publication I

counts of A. minutum strains with the microtiter plate method was calculated. The signal
increases with higher RNA concentrations and thus with higher cell numbers. The measured

cells is just above the background and can also be regarded as a A. minutumsignal for 10,000 negative signal. A signal that is clearly dist the background was observed for inguishable from12,500 cells of A. minutum. A low signal with an absorbance of 0.07 presents either 12,500

cells of A. minutum or a very high amount at least 500,000 cells ofA. ostenfeldii orA.
sinterpreted. RNA i with high concentrations of RNA, thus the signal can not be mfundyenseisolation limits the detection method because of high user variability in the ability to isolate
rRNA from the same number of algal cells and thus resulting in lower RNA concentrations

e correct cell ould not reflect thsities of these RNA concentrations wper cell. Signal intennumbers. 10,000 cells ofA. minutum present the smallest possible number of cells for RNA
ited by cell numbers but rather by isolation in this study, however, RNA isolation is not limasurableebers result in the lowest mitations of the extraction kit. But these cell numlim

concentration; otherwise the standard error is too high. Thus, the detection limit of the microtiter plate assay for 12,500 A. minutum cells with an average yield of 0.028 ng RNA per
ply that 50 litres with 250 cells per litre would have to be e would impling volumcell the sameeasured. More work is needed to reduce th m before a reliable detection value isconcentratedit.detection lim

crotiter plate assay using a sandwich hybridization was evaluated with the analysis of iThe m several different species and the fples. Phytoplankton communities often consist ospiked sament withposition in the sea is substantial. The experimporal and spatial variability in comtemspiked samples revealed that for 50,000 and 100,000 cells of A. minutum,the signal was
re. One reason for the lowerber of cells of a lab cultue numlower than the signals for the sam

ounts of natural mple. Large aposition of the samsignals of the natural sample can be the coms to disturb the RNA ent seempling location and this sediment were observed at the samsedimisolation. The concentration of total rRNA may be improvedby changing the RNA isolation

ents should also include the development of an protocol. Therefore, future experim without RNA isolation as described by Tyrrell et al. (2002) and Ayers et independent systeme assay for natural samples, thecrotiter platiwer signals in the mal. (2005). As a result of the locorrelation of signal to cell numbers is limited, only an estimation of cell numbers can be
r analysis with the ent loads are inappropriate foples with high sedimdone. Hence, sammicrotiter plate assay. However, the method presented here using a sandwich hybridization in
amicrotiter plate is reliable, and in comparison to other molecular methods, inexpensive, fast

26

Publication I

and easy to handle. It provides a rapid assay for testing of probe specificity, much in the same

H probes. way that dot blots provide the vehicle for testing probe specificity for FIS

Conclusion

toxic dinoflagellate for the detection of the crotiter plate assay was adaptediA m

Alexandriumminutum using a sandwich hybridization. The assay has the potential to be a fast and reliable

method for the detection of toxic algae by eliminating the need to count algae manually. The

ine up to 30 different samassay takes only two and a half hours to examentsples. The experim

thod. Clearly additional eof this mmples present a proof of principle with spiked natural sa

work is required to improve RNA isolation from natural samples and to optimize the

sensitivity of the method for A. minutumprobes. For the routine testing of probe specificity, it

ocan provide a rapid assay for assessing probe specificity at be and target sequence th the clad

level.

ledgementsAcknow

ce inThe authors would like to thank Sabine Strieben and Megan Crawford for their assistan

ny thanks to Dennis Gowland (North Bay Shellfish aalgae cultivation and harvesting. Also m

Orkney Islands, United Kingdomples. Sonja Diercks was supported by for taking field sam)

eworkthe EU-project ALGADEC (COOP-CT-2004-508435-ALGADEC) of the 6th Fram

of the European Union and the Alfred WeProgramm

Research.

27

gener Institute for Polar and Marine

e

Publication I

References

Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local

403–410. 215:ent search tool. J. Mol. Bio. alingm

Anderson, D. M., D. M. Kulis, B. A. Keafer, and E. Berdalet. 1999. Detection of the toxic

Dinoflagellate Alexandrium fundyense (Dinophyceae) with oligonucleotide and
antibody probes: variability in labelling intensity with physiological condition. J. 870-883. 35:Phycol

A. Keafer, K. E. Gribble, R. Marin, and C. A. Scholin. .Anderson, D. M., D. M. Kulis, B

the Gulf of Maine spp. fromeration of Alexandrium2005. Identification and enumceanography.lecular probes. Deep Sea Research Part II: Topical Studies in Oousing mThe Ecology and Oceanography of Toxic Alexandrium fundyense Blooms in the Gulf

52:of Maine 2467-2490.

s, J. V. Tyrrell, M. Gladstone, and C. A. Scholin. 2005. International Ayers, K., L. L. Rhode

accreditation of sandwich hybridization assay format DNA probes for micro-algae.
1225-1231. 39:New Zeal J. Mar. Fresh

ical variability within the Protogonyaulax ella, A. D., and F. J. R. Taylor. 1985. BiochembmCe

Inplex, p. 55-60. arensis catenella species comtamD. M. Anderson, A. W. White and

c Dinoflagellates. Elsevier. D. G. Baden [eds.], Toxi

atographyance liquid chromodified high-performChen, C. Y., and H. N. Chou. 2002. A m

method for analysis of PSP toxins in dinoflagellate, Alexandrium minutum, and
715-720. 35:shellfish from Taiwan. Food Res. lnt Chou, H. N., Y. M. Chen, and C. Y. Chen. 2004.Variety of PSP toxins in four culture strains

of Alexandrium minutum collected from southern Taiwan. Toxicon 43: 337-340.

Daranas, A. H., M. Norte, and J. J. Fernandez. 2001. Toxic marine microalgae. Toxicon 39:
1101-1132.

Eppley, R. W., R. W. Holmes, and J. D. H. Strickland. 1967. Sinking rates of marine

phytoplankton measured with a fluorometer. J. Exp. Mar. Biol. Ecol 1: 191-208.

Groben, R., U. John, G. Eller, M. Lange, and L. K. Medlin. 2004. Using fluorescently-

ni-iation of phytoplankton diversity – a mNA probes for hierarchical estimlabelled rR

313-320. 79:review. Nova Hedwigia

Groben, R., and L. Medlin. 2005. In Situ Hybridization of Phytoplankton Using Fluorescently Labeled rRNA Probes. Methods in Enzymology, p. 299-310. InA. E. H. R. Elizabeth

28

Publication I

icical Data. AcademA. Zimmer [ed.], Molecular Evolution: Producing the BiochemPress.Guillard, R. R. L. 1975. Culture of phytoplanktonfor feeding marine invertebrates, p. 26-60.
ith and Y. M. H. Chanle [eds.], Culture of Marine Invertebrate Animals. L. Sm.WIns. PresPlenumGuillard, R. R. L., and J. H. Ryther. 1962. Studies of marine planktonic diatoms. I. Cyclotella
nana Hustedt and Detonula confervacea Cleve. Can. J. Microbiol.8: 229-239.
e Denn, M.-A. Cambon-Bonavita, P. Gentien, andGuillou, L., E. Nezan, V. Cueff, E. Erard-LG. Barbier. 2002. Genetic Diversity and Molecular Detection of Three Toxic

French Coasts. Dinoflagellate Genera (Alexandrium, Dinophysis, and Karenia) from153:tProtis 223-238.

Hosoi-Tanabe, S., and Y. Sako. 2005. Rapid detection of natural cells of Alexandrium tamarense and A. catenella (Dinophyceae) by fluorescence in situ hybridization.
319-328. 4:gaeHarmful Alnt of specific rRNA probes to eoben. 2005. DevelopmJohn, U., L. K. Medlin, and R. Grarense species tamof the Alexandriumdistinguish between geographic clades 199–204. 27:plex. J. Plankton Res. comre of Claus, and R. R. L. Guillard. 1987. Media for the cultu.Keller, M. D., R. C. Selvin, Woceanic ultraphytoplankton. J. Phycol 23: 633-338.
Kim, C.-J., and Y. Sako. 2005. Molecular identification of toxic Alexandrium tamiyavanichii
984-991. 4:(Dinophyceae) using two DNA probes. Harmful Algae Lilly, E. L., K. M. Halanych, and D. M. Anderson. 2005. Phylogeny, biogeography, and species boundaries within the Alexandrium minutum group. Harmful Algae4: 1004-
1020.

.Ludwig, Went for sequence data. Nucleic Acids and others 2004. ARB: a software environm 1363-1371. 32:Resatic and s (HAB); problemcroalgae bloomiMaso, M., and E. Garces. 2006. Harmful mf;In Press, Corrected Proor. Pollut. Bull . Maconditions that induce them.doi:10.1016/j.marpolbul.2006.08.006Matweyou, J. A., D. A. Stockwell, C. A. Scholin, S. Hall, V. Trainer, J. Ray, T. E. Whitledge, rRNA targeted probes e of Alexandriumley. 2004. UsA. R. Childers, and F. G. PlumK. A. Steidinger, J. H. InSP events on Kodial Island, Alaska, p. 267-269. to predict PLandsberg, C. R. Tomas and G. A. Vargo [eds.], Harmful Algae 2002. Florida Fish

29

Publication I

iildlife Conservation Command Wssion, Florida Institute of Oceanography, and

ission of UNESCO. mmoental Oceanographic CIntergovernm

ical detection of the Metfies, K., S. Huljic, M. Lange, and L. K. Medlin. 2005. Electrochem

eldii with a DNA-biosensor. Biosens. lagellate Alexandrium ostenftoxic dinof

1349-1357. 20:Bioelectron

ental Toxic Algae, Intergovernmference list ofic ReMoestrup, O. 2004. IOC Taxonom

sion of UNESCO; ioc.unesco.org/hab/data.htmisOceanographic Comm.

entento, S. M., E. L. Lilly, J. Larsen, and S. Morris. 2005. Toxin profile, pigmNascim

position, and large subunit rDNA phylogenetic analysis of an Alexandrium com

minutum (Dinophyceae) strain isolated from the Fleet Lagoon, United Kingdom. J.

343–353. 41:Phycol

O'halloran, C., M. W. Silver, T. R. Holman,and C. A. Scholin. 2006. Heterosigma akashiwo

124-132. 5:ful Algae in central California waters. Harm

(Dinophyceae) species using PCR and Penna, A. M., M. 1999. Identification of Alexandrium

rDNA-targeted probes. J. Phycol 615-621. 35:

Provasoli, L., J. J. A. Mclaughlin, and D. M. R. 1957. The development of artificial media for

rine algae. Arch. Microbiol am 392-428. 25:

eitenstein, S. Molin, and P. Neubauer. 2003. rRautio, J., K. B. Barken, J. Lahdenpera, A. B

Sandwich hybridization assay for quantitative detection of yeast RNAs in crude cell

4. 2:lysates. Microb. Cell Fact

Scholin, C. A., K. R. Buck, T. Britschgi, G. Cangelosi, and F. P. Chavez. 1996. Identification

of Pseudo-nitzschia australis (Bacillariophyceae) using rRNA-targeted probes in

190-197. 35:ats. Phycologia mwhole cell and sandwich hybridization for

e genus Alexandrium otoxigenic dinoflagellatTaylor, F. J. R., and Y. Fukuyo. 1998. The neur

In: General Introduction, p. 381-404. Halimbella and H. D. M. Anderson, A. D. Cem

s.mful Algae BloomG.M. [eds.], Physiological Ecology of Har

ashiwoka aTyrrell, J. V., L. B. Connell, and C. A. Scholin. 2002. Monitoring for Heterosigm

gaeusing a sandwich hybridization assay. Harmful Al 205-214. 1:

acle. 1995. DNA probe Zammatteo, N., P. Moris, I. Alexandre, D. Vaira, J. Piette, and J. Rem

hybridization in microwells using a new bioluminescent system for the detection of

PCR-amplified HIV-1 proviral DNA. Virol. Methods 55: 185-197.

s: a challenge for of harmful algal bloomZingone, A., and H. Enevoldsen. 2000. The diversity

science and management. Ocean Coast Manage 43: 725-748.

30

Publication I of algae strains used in this study Culture conditions and geographical originTable 1.SpeciesStrainCulturmediumeTemperatureOrigin
Alexandrium minutumAL1VK15°CFraRia degaVigo,Spain, 1987, S.
Alexandrium minutumAMADO6K15°CAustHallegraeffralia, SouthAustralia,
Alexandrium minutumAMITAK15°CAdriatic, Mediteranean Sea
Alexandrium minutumAMP4K15°CSanMediterratiago Fraganean Sea, Spain,
Alexandrium minutumAL1TK15°CGulBeranf of Trieste,Italy, A.
Alexandrium minutumAL3TK15°CGulBeranf of Trieste,Italy, A.
Alexandrium minutumAL5TK15°CGulBeranf of Trieste,Italy, A.
Alexandrium minutumAL8TK15°CGulf of Trieste,Italy, A.
BeranAlexandrium minutumAL9TK15°CGulBeranf of Trieste,Italy, A.
Alexandrium minutumNantesK15°CAtlantic Ocean, France
AlexaAlexandndrium minrium minuuttumumAL 7 V PALMIRA1KK1155°°CCMediterraAtlantic Ocean, Spainnean Sea, Spain
Alexandrium minutumAL 4V K15°CRiade Vigo, Spain,2000,
gaS.FraAlexandrium minutumAL 2V K15°CRiaUchiumde Vigo, Si Bay,pKagaain,wa,Bravo
Alexandrium insuetumCCMP 2082Prov20 °C
Alexandrium sp.CS 001K15 °CJapaScotland, Mn, 1985 S. Yos.Grievehimatsu
Alexandrium tamutumSZNB029K15°CGulMontref of Nasorples,Italy,M.
Alexandrium fundyenseCA 28f215 °C Woods Hole, Institution, D.M. AndeOceanographicrson
Alexandrium tamarenseSZNB 01 IMR15°CGulM. Monf of Naptresolres,Italy1999,
Alexandrium tamarenseSZNB 019IMR15°CGulf of Naples,Italy1999,
rtresoM. MonHarbour, Nova Scotia,ShipAlexandrium ostenfeldiiAOSH 1 K15 °C Canada, A. Cembella
Alexandrium ostenfeldiiCCMP 1773K15 °C Limfjordan, Denmark,
enHansAlexandrium catenellaBAH ME 255IMR15 °C Spain, M. Delgado
Alexandrium tayloriiAY 2T K15 °C BeranLagoon ofMarano, Italy,A.
Gonyaulax spiniferaCCMP409f215°CGulfof Maine, North
erica, 1986R. LandeAmProtoceratium reticulatumf2-Si15°CGermHelgoland,any, M. North Sea,Hoppenrath
Lingulodinium polyedrumIMR15°CNorway, T. Castberg
Prymnesium parvumK-0081K15°CFlade Sø, Denmark
Rhodomonas sp.CCMP 768K22 °CNorth Island,New Zealand,
ChangSouth Pacific, F. 31

Publication I

ts used for spiked samAlgal species and cell counTable 2. ples

SpeciesAlexandrium minutum
Alexandrium ostenfeldiinsendyeuAlexandrium fmutumaAlexandrium tratium reticulatumProtoceLingulodinium polyedrum
Prymnesium parvum.s spdomonaRho

ainStrTAL3773CCMP 1CA 28902SZNB

K-CCMP0081 768

32

countsCell10000, 50000, 100000
0000505002000050000100001009005000001

Publication I

Table 3.Probe

Sequences of probes for

A MIN C

NEXTAMIN C

positive control

(target sequence)

Probe

nceseque

Alexandrium minutumTAGG TTGTCGAA

GCGGA T

TTC C TAA TGA CCA CAA CCC

GCATCC AAA CCT GAC TTC GGA AGG GTT GTG GTC

ATT A

33

Publication I

Alexandrium minutum Specificity of probes for Table 4. SpeciesStrainSignalAver350 µage valueg/µLOD
Alexandrium minutumAL1V+2.3476
Alexandrium minutumAMADO6+2.8662
Alexandrium minutumAMITA+4.9956
Alexandrium minutumAMP4+4.6426
Alexandrium minutumAL1T+5.1715
Alexandrium minutumAL3T+3.2775
Alexandrium minutumAL5T+2.2989
AlexaAlexandndrium minrium minuuttumumALAL9T8T++13..94837521
Alexandrium minutumNantes+2.1611
Alexandrium minutumAL 7 V +2.8885
Alexandrium minutumPALMIRA1+1.8488
Alexandrium minutumAL 4V +1.5897
Alexandrium minutumAL 2V +4.6268
Alexandrium insuetumCCMP 2082-0.0304
Alexandrium sp.CS 001-0.0075
Alexandrium tamutumSZNB029-0.0971
Alexandrium fundyenseCA 28-0.0000
Alexandrium tamarenseSZNB 01-0.0000
Alexandrium tamarenseSZNB 019-0.0351
Alexandrium ostenfeldiiAOSH 1 -0.1215
Alexandrium ostenfeldiiCCMP 1773-0.0201
Alexandrium catenellaBAH ME 255-0.1701
Alexandrium tayloriiAY 2T -0.0161
Gonyaulax spiniferaCCMP409-0.0188

34

Publication I

Figure 1.

Sandwich hybridization

35

Publication I

L600A/ RNgnµ500

400

300

200

100

000000000000000000000000000000
400800120016002000240028003200360040004400480052005600

A. minutumAL3TA. minutumAMP4A. minutumAL5Tcell counts

Total RNA concentration for three Figure 2.different cell counts

36

strains in ng/µL at Alexandrium minutum

Publication I

3.5m3.0 n50 4atecrbanosbA2.0
2.5

1.5

1.0

0.5

0.0000000000000000
40008000200060000000400080002000600000004000800020006000
112223344455
A. minutum AL3TA. minutum AMP4cell counts
A. minutum AL5Taverage value OD

Figure 3. Photometer readings for cell numbers of three different
strains

37

Alexandrium minutum

Publication I

050.40.5040. n50 4tce anarbosbAm0.25
530.030.

020.

510.

010.

500.

000.

10000

50000

lab culturespiked sample

001000

countsA. minutum

u collcestn

Figure 4. Comparison of photometer readings at 405 nm for a lab culture of A. minutumand

ples withspiked sam

A. minutum38

Publication II

II Publication2.4

M

OLECULAR PROBES FOR THE DETECTION OF TXIC ALGAE FOSANDWICH HYBRIDIZATION FORMATS

SONJADIERCKS,KATJAMETFIES AND LINDAK.MEDLIN

R USE OINAlfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570

Abstract

anyerhaven, GermBrem

itted submeJournal of Plankton Research, to b

obes can be used for early and rapid detection of toxic algae species. The rlar pMolecu

sandwich hybridization requires two probes for each species, a capture probe and a nearly

of the toxic algal es-specific identificationnal probe. Probe sets for the specit sigadjacen

speciesProtoceratium reticulatum, Lingulodinium polyedrum, ,Gymnodinium catenatum

Pseudo-nitzschia multiseries, P. australis, ,Prymnesium parvum, Chrysochromulina polylepis

species was Pseudo-nitzschia were designed. A genus probe set for P. seriata and P. pungens

adapted and all probe sets were tested for specificity. The target mlecules for the probe sets o

al RNAs. The specificity of the different probes all subunit ribosomare the large and the sm

al RNA crotiter plate assay with ribosomisets was tested using a sandwich hybridization in a m

misolated frod closely related species. The assayry strains of the target species an laborato

showed the eight probe sets to be highly specific. Detection of one other species, in addition

to the target species, was observed for two of the probe sets. These ten probe sets are valuable

tools for identifying and monitoringdifferent toxic algae. The microtiter plate assay is a cheap

sting probe specificity. eeans of tand effective m

39

Publication II

nIntroductio

cause fish kills and shellfish poisoning. Early Harmful algae can produce powerful toxins that

astal areas and aquaculture is the ml species in coand rapid detection of toxic algast effective o

populations. A variety of detection techniques anative effects on humitigate their negway to m

using molecular probes, such as fluorescencein situ hybridization (FISH) (Scholin et al.

1996; Simon et al. 2000; Smit et al. 2004; Kim and Sako 2005), DNA microarrays (Metfies

and sandwich hybridization assays (SHA)and Medlin 2005a; Metfies and Medlin 2005b)

s et al. 2005) can be applied for this purpose. ers et al. 2005; Metfiey(Scholin et al. 1996; A

Usually, targets for the molecular probes arethe small and the large subunit ribosomal RNA

re or lessobers in the cell and contain mound in high num(rRNA) genes because they can be f

. 2004). The relative conservation of the 18S and 28S gene conserved regions (Groben et al

the species level (Gagnon et al. 1996; Ki and plicate the search for suitable probes atcan com

Han 2006). Specific probes for several algal taxahave been developed recently (Scholin et al.

1999; Tyrrell et al. 2002; John et al. 2003; Kim and Sako 2005; Metfies et al. 2005; Töbe et

algal species is covered. For all percentage of all toxic al. 2006), however, still only a sm

ats (Zammatteo et al. 1995; Rautio et al. 2003) two probes are msandwich hybridization for

needed, and at least one of the probes has to be specific for the target. One of the probes, the

rfaces as in combination with DNA bimmobilized on solid sucapture probe can be iosensors

crotiter plate (see Publication I) and bind to target ill of a m(Metfies et al. 2005) or in the we

RNA or DNA. A second probe, the detection probe, carries the signal moiety and binds near

the binding site of the capture probe. Here, we present the results of the application of 10

probe sets for the detection of the different toxic algal speciesGymnodinium catenatum,

polyedrum, Prymnesium parvum, LingulodiniumProtoceratium reticulatum,

Chrysochromulina polylepis,Pseudo-nitzschia multiseries, P. australis, P. seriata and P.

The probes were tested for Pseudo-nitzschia.d for all species of the genus anpungens

crotiter plate well described in iich hybridization in a mspecificity applying the sandw

Publication I with laboratory strains.

Materials and Methods

ater-basedAll algal strains were cultured under sterile conditions in seaw -ditionsCulture con

dia K (Keller et al. 1987), IMR (Eppley et al. 1967), Drebes (Stosch and Drebes 1964), em

Prov. (Provdasoli et al. 1957; Guillard and Ryther 1962; Guillard 1975), L1 (Guillard an

40

Publication II

(Andersen et al. 1997), f2-lard and Hargraves 1993), DY IVRyther 1962; Guillard 1975; Guil

ith 1968) at Si (Guillard and Ryther 1962; Guillard 1975) and GP%50 (Loeblich and Sm

a light: dark cycle of 14:10 ithately 100 µEinstein wdifferent temperatures and approxim

hours (Table 1).

strains was carried out using Pseudo-nitzschia - Isolation of total rRNA fromRNA-extraction

a, Taufkirchen, Germthe protocol for the Tri Reagent kit (Sigmany). Glass beads (212-300

µm, Sigma, Germany) were also added to the isolation solution to break open the cells with a

A) for 20 bead Mini-Beadbeater (Biospec products, Biospec products Inc, Bartlesville, US

l lysis steps, the Clean up protocol fromseconds. Subsequent to the cel Qiagen (Hilden,

Germany) was used for RNA purification. Total RNA from all otheralgal cultures was

Hilden,isolated according to a modified protocol from the RNeasy Plant Mini Kit (Qiagen,

any). Modifications of thisGermce the quality and quantity of the protocol were done to enhan

arides and proteins. For qualityproved removal of polysacchmextracted rRNA by i

nutes for separation of supernatant and cell ient, the centrifugation step of two menhancem

debris was extended to 15 mes to the RNeasy nutes. Buffer RW1 was applied two timi

nute and then centrifuged. The first wash step with buffer RPE in, incubated for one mcolum

eter (Peqlab, trophotomeasured with a Nanodrop Specwas repeated. RNA concentration was m

any).rmeErlangen, G

ic FISH probes for several species have been previously Specif -Probes and probe synthesis

ity to the developed (Table 2) and were used in this study. A second probe in close proxim

first probe was developed for these probes for the sandwich hybridization. The previously

bination; PNFRAGAwere both used in comPseudo-nitzschiadeveloped probes for the Genus

was adapted as a capture probe and PNEXDELIB as a detection probe. All probes and

positive controls (Test DNA) were synthesized from Thermo Electron Corporation (Ulm,

any).Germ

Sandwich Hybridization using a microtiter plate assay (MTPA) - The probe sets for the

ing a sandwich hybridization in a detection of the different algae were tested for specificity us

microtiter plate assay as described in Publication I. In this assay the capture probe is

biotinylated and the signal probe is digoxigenin-labelled. Prior to the experiments, the

the PCR ELISA Dig Detection Kit from Roche Diagnostics different buffer solutions from

Total rRNA fromany) were prepared for use., Germ(Mannheim the different algae was

41

Publication II

fragmented in a fragmentation buffer (40mM Tris, pH 8.0/100mM KOAc/30mM MgOAc) for
r to hybridization. Biotinylated probes and nutes at 94°C and then chilled on ice prioi5 mdigoxigenin labeled probes at a concentration of 10 µM and different concentrations of rRNA were added to the hybridization buffer. A negative control was prepared containing only probes and hybridization buffer, whereas the positive control contained also Test DNA (synthesized target sequence of both probes). The different hybridization solutions were crotiter plate iadded into the wells of the mand incubated on a shaker for 1 hour at 46°C. crotiter plate were washed with washing iSubsequently to the hybridization, the wells of the msolution. Antibody solution was applied into each well and incubated 30 minutes at 37°C with agitation in the dark. After incubation with the antibody solution, the wells were re-washed with washing solution and substrate solution was filled into the wells and incubated in thenutes. The anti-digoxigenin antibody conjugated to idark on a shaker at 37 °C for 30 mhorseradish peroxidase reacts with substrate to produce a green colorimetric product. The
wells of the microtiter plate were read out at 405 nm using a quartz cuvette with a Varian
any).stadt, Germeter (Varian Inc., DarmCary 4000 UV-Vis Spectrom

Results

Gymnodinium catenatum, - Probes were developed for Probe designPseudo-nitzschia four species,Chrysochromulina polylepis,Prymnesium parvum/patelliferum,Lingulodinium
polyedrum and software package (Ludwig et al. 2004). AdditiProtoceratium reticulatum (Table 3) using the probe design option in ARB onally a BLAST search (Altschul et al. 1990)
the probes against all publically available was conducted to test the overall specificity ofsequences. It was possible to design two specific probes that are located in sufficientproximity for the sandwich-hybridization approach for G. catenatum,L. polyedrum and P.
Pseudo-nitzschiaHowever, it was not possible to design two specific probes for reticulatum.australis,Pseudo-nitzschia pungens,Chrysochromulina polylepis and Prymnesium
parvum/patelliferum.Therefore an unspecific signal probefor these target species was chosen
ly developed specific capture probe. Thus, the ity to the previousthat bind in close proximspecificity of the reaction was determined by the capture probe. The close proximity of the
capture probe and the detection probe minimizes possible degradation effects of the target
ces of each respective probe uenesized target seqnucleic acid. Positive controls are the synthset.

42

Publication II

Specificity of probess tested using a - The specificity of 10 probe sets for toxic algae wa

sandwich hybridization assay in a microtiter plate as described in Publication I with closely

et species and m the targrelated species (Tables 4 and 5). Total rRNA was isolated fromreo

alised to a target obe set. The signals obtained were normtly related species for each prdistan

pared. A probe set for the toxic algal species n of 350 ng RNA and comconcentratio

Chrysochromulina polylepsis was tested with two strains ofC. polylepsis and three closely

related species (Table 4). It showed specific signals only for the target species. The target

showed both G. impudicum and the non-target species Gymnodinium catenatumspecies

e signals for the GCAT probe set, whereaspositivother non-target species showed no signal at

G. catenatumalso showed a signal, but the signal for G. impudicum all (Table 4). However,

gave a signal Lingulodinium polyedrum.G. impudicumwas threefold higher than the one of

for the LPOLY probe set and all non-target species did not (Table 4). The PRETI probe set

Prymnesium(Table 4). Signals for all Protoceratium reticulatumshowed specific signals for

species were achieved with the capture probe PRYM 694 and the detection probe PRYM 694

for specificity with all available probe set was testedPseudo-nitzschia The Genus NEXT.

Pseudo-nitzschiaspecies (Table 5), except P. multiseriesstrain Oroe13 and P. seriata strain

strains AL-93 and SAL-5 P. pseudodelicatissimaCCMP 1309, which were not available.

genus level probes. The species probe sets PSN Pseudo-nitzschiawere not detected with the

Pseudo-re tested with their respectivend PSN SERI weAUS, PSN MULTI, PSN PUNG a

nitzschia target species and with representative strains of the other Pseudo-nitzschia species.

able 5). et species (TSignals of all probe sets were only observed for the targ

Discussion

Probes sets for 10 toxic algal species were developed and tested for specificity using a

microtiter plate assay and a sandwich hybridization. Single probes for some species had

SH. Thus, only a second Ialready been developed and tested for specificity with dot blot and F

te the sandwich hybridization. The combination pleprobe was needed for these species to com

Chrysochromulina for specificity. Our capture probe forof both probes needed to be tested

was developed for FISH and tested for specificity by Simon et al. (1997). The sispolylep

detection probe for C. polylepsis is unspecific; however, in combination with the specific

C. polylepcapture probe only is detected. Although this probe set was only tested with few sis

ples. High ed with further tests e.g., spiked samspecies, its specificity should be confirm

Gymnodiniume set GCAT and the target detection signals were observed with the prob

43

Publication II

catenatum.A threefold lowersignal was determined for the non toxic G. impudicum, a

ssive red tide blooma mworldwide occurring species, that can forms (Fraga et al. 1995). A

signal intensity e presents only 10,000 cells, whereas for the samG. catenatumhigh signal for

at least 250,000 cells of G. impudicum are needed,thus a misinterpretation of signal is

unlikely. The probe set for Prymnesium parvum detected all testedPrymnesium species

rring in marine waters (Pienaar , which is a non-toxic species occuP. nemamethecumincluding

and Birkhead 1994). The majority of Prymnesium parvum blooms have been recorded in

sbrackish waters (Edvardsen and Paasche 1998) and there have not been any reports of bloom

caused by P. nemamethecum(West et al. 2006). In water samples from brackish water, the

detection ofP. nemamethecum cannot be ruled out but seemsunlikely. Some false-positive

results are almost impossible to avoid witha monostrigent hybridization approach, because

the stability of mismatched probe-target hybrids cannot easily be predicted in silico(Loy et al.

2005b).

fThe specificity tests using species o and probes for the different Pseudo-nitzschia the genus

intaining cultures long termaed out to be difficult because of the difficulty in mspecies turn

under laboratory conditions. Consequently only a few representative strains of each species

Pseudo-nitzschiained. The could be examgenus probes were tested with all available strains

and were observed to detect only one of the three P. pseudodelicatissima strains. However,

the 18S gene of the three strainswas sequenced and sequences of all Pseudo-nitzschia strains

red to the probe sequences. The sequences of strains AL-93 and SAL-5 revealed pawere com

two mismatches to the capture probe sequence, whereas no mismatch was found in strain AL-

19. The sequence of AL-19 was identical to that of other P. delicatissima, thus, this strain was

determined to be P. delicatissima rather than P. pseudodelicatissima. Hence, our Pseudo-

nitzschia genus probes are not able to detect P. pseudodelicatissima and the absence of a

signal can be used as a determinate marker for P. pseudodelicatissima, which can be difficult

to separate fromP. delicatissimaat the light microscopic level. The search for suitable probes

the relative conservation of the because of recently evolved speciesecan be difficult for som

18S gene (Gagnon et al. 1996; Ki and Han 2006), therefore a new probe set should be

. Signals of all other probe sets were P. pseudodelicatissimadeveloped for the detection of

a large database, a ly for the target species. Even when probes are designed fromobserved on

frequent revision of probe sequences is necessary because new sequences are added almost

daily to databases.

44

Publication II

Conclusion

Ten probe sets for diffewere designed and eight probe sets proved to rent toxic algal species

ion assay. Two probe sets, GCAT and PRYMbe highly specific in our sandwich hybridizat

694, detect another species in addition to its ta

onitoring of toxic algae usin for the mapplied

rget species. All designed probe sets can be

g solid surface, such as biosensors and the

microtiter plate assay. The microtiter plate assay is a fast and efficient way to test probes for

use in sandwich hybridization m

specificity for FISH probes.

ledgmentsAcknow

in the samhcu

e

t dot blots are used to screen for way tha

The authors would like to thank Sabine Strieben and Megan Crawford for their assistan

ce in

algae cultivation and harvesting. Sonja Diercks was supported by the EU-project ALGADEC

ework Programm2004-508435-ALGADEC) of the 6th Fram(COOP-CT-

e Research.Union and the Alfred Wegener Institute for Polar and Marin

45

of the European e

Publication II

References

Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local

403–410. 215:ent search tool. J. Mol. Bio. alingm

ton, and J. P. Sexton. 1997. Provasoli-Guillard National Center for Andersen, R. A., S. Mor

1-75. 33:Culture of Marine Phytoplankton 1997 list of strains. J. Phycol

s, J. V. Tyrrell, M. Gladstone, and C. A. Scholin. 2005. International Ayers, K., L. L. Rhode

accreditation of sandwich hybridization assay format DNA probes for micro-algae.

39:New Zeal J. Mar. Fresh 1225-1231.

Edvardsen, B., and E. Paasche. 1998. Bloom dynamics and physiology of Prymnesium and

Chrysochromulina, p. 193-208. InD. M. Anderson, A. D. Cembella and H. G.M.

s. Springer. [eds.], Physiological ecology of harmful algal bloom

Eppley, R. W., R. W. Holmes, and J. D. H. Strickland. 1967. Sinking rates of marine

phytoplankton measured with a fluorometer. J. Exp. Mar. Biol. Ecol 1: 191-208.

GFraga, S., I. Bravo, M. Delgado, J. M. Franco, and M. Zapata. 1995. rodinium impudicumy

ing, red tide dinoflagellate. Phycologia sp. nov. (Dinophyceae), a non toxic, chain-form

514-521. 34:

Gagnon, S., D. Bourbeau, and R. C. Levesque. 1996. Secondary structures and features of the

plexan parasite Toxoplasmal RNAs from the Apicomnd 26S ribosom18S, 5.8S aa

129-135. 173:gondii. Gene

Groben, R., U. John, G. Eller, M. Lange, and L. K. Medlin. 2004. Using fluorescently-

ni-iation of phytoplankton diversity – a mNA probes for hierarchical estimlabelled rR

review. Nova Hedwigia 313-320. 79:

Guillard, R. R. L. 1975. Culture of phytoplanktonfor feeding marine invertebrates, p. 26-60.

InW. L. Smith and y. M. H. Chanle [eds.], Culture of Marine Invertebrate Animals.

s. PresPlenum

sis imm StichochryGuillard, R. R. L., and P. E. Hargraves. 1993.m, not a obilis is a diato

234-236. 32:chrysophyte. Phycologia

Guillard, R. R. L., and J. H. Ryther. 1962. Studies of marine planktonic diatoms. I. Cyclotella

nana Hustedt and Detonula confervacea Cleve. Can. J. Microbiol.8: 229-239.

John, U., A. Cembella, C. Hummert, M. Elbrächter, R. Groben, and L. K. Medlin. 2003.

arense and A. tamination of the toxigenic dinoflagellates AlexandriumDiscrim

ostenfeldii in co occurringnatural populations from Scottish coastal waters. Eur. J.

25-40. 38:Phycol.

46

Publication II

re of Claus, and R. R. L. Guillard. 1987. Media for the cultu.Keller, M. D., R. C. Selvin, Woceanic ultraphytoplankton. J. Phycol 23: 633-338.
ly study for parallen. 2006. A low-density oligonucleotide arraaKi, J.-S., and M.-S. Hhybridization of consensus PCR products of detection of harmful algal species using 1812-1821. 21:Biosens. Bioelectron ain.LSU rDNA D2 domKim, C.-J., and Y. Sako. 2005. Molecular identification of toxic Alexandrium tamiyavanichii
984-991. 4:(Dinophyceae) using two DNA probes. Harmful Algae Loeblich, A. R., and V. E. Smith. 1968. Chloroplast pigments of the marine dinoflagellate
5-13. 3:Gyrodinium resplendens. Lipids Loy, A. and others 2005. 16S rRNA Gene-Based Oligonucletide Microarray for Environmental Monitoring of the Betaproteobacterial Order "Rhodocyclales". Appl.
1373-1386. 71:Environ. Microbiol. ent for sequence data. Nucleic Acids and others 2004. ARB: a software environm.Ludwig, W 1363-1371. 32:Resical detection of the Metfies, K., S. Huljic, M. Lange, and L. K. Medlin. 2005. Electrochemeldii with a DNA-biosensor. Biosens. lagellate Alexandrium ostenftoxic dinof 1349-1357. 20:Bioelectronmicrochips for phytoplankton: The fluorescent Metfies, K., and L. K. Medlin. 2005a. DNA 321—327. 79:wave of the future. Nova Hedwigia gtial Use in Assessinal RNA Probes and Microarrays: Their Poten---. 2005b. RibosomMicrobial Biodiversity, p. 258-278. Ina. E. H. R. Elizabeth A. Zimmer [ed.], Methods
icical Data. Academogy,Molecular Evolution: Producing the Biochemolin EnzymPress. sp. nov. Prymnesium nemamethecumPienaar, R. N., and M. Birkhead. 1994. Ultrastructure of 291-300. 30:J. Phycol Provasoli, L., J. J. A. Mclaughlin, and D. M. R. 1957. The development of artificial media for
392-428. 25:rine algae. Arch. Microbiol ameitenstein, S. Molin, and P. Neubauer. 2003. rRautio, J., K. B. Barken, J. Lahdenpera, A. BSandwich hybridization assay for quantitative detection of yeast RNAs in crude cell 4. 2:lysates. Microb. Cell Fact Scholin, C. A., K. R. Buck, T. Britschgi, G. Cangelosi, and F. P. Chavez. 1996. Identification of Pseudo-nitzschia australis (Bacillariophyceae) using rRNA-targeted probes in 190-197. 35:ats. Phycologia mwhole cell and sandwich hybridization for

47

Publication II

Scholin, C. A. and others 1999. DNA probes and a receptor-binding assay for the detection of

Pseudo-nitzschia (Bacillariophycoic acid activity in cultured and eae) species and dom

1356-1367. 35:ples. J. Phycol natural sam

nd L. K. Medlin. 1997. The identification ofdsen, aon, N., J. Brenner, B. EdvarSim

Chrysochromulina and Prymnesium species (Hyptophyta, Prymnesiophyceae) using

provingeans for iminescent oligonucleotide probes: a milumfluorescent or chem

studies on toxic algae. E 393-401. 32:ur. J. Phycol.

ication of three algal groups on, N. and others 2000. Oligonucleotide probes for the identifSim

76-84. 47: whole-cell hybridization. J. Eukaryot. Microbiol by dot blot and fluorescent

Smcrobial ecology. ilar tools and approaches in eukaryotic mit, E. and others 2004. Molecu

Bios Scientific Publishers.

Stosch, H. A. V., and G. Drebes. 1964. Entwicklungsgeschichtliche Untersuchungen an

een. IV. Die Planktondiatomee Stephanopyxis turris - ihre Behandlung und Diatom

ss. Meeresunters. iEntwicklungsgeschichte. Helgol. W 209-257. II:

eration for toxic ated detection and enumtomTöbe, K., G. Eller, and L. K. Medlin. 2006. Au

nesiumtroduction of a new probe for Prymetry and the inalgae by solid-phase cytom

parvum (Haptophyta: Prymnesiophyceae). J. Plankton Res. 28: 643-657.

ashiwoka aTyrrell, J. V., L. B. Connell, and C. A. Scholin. 2002. Monitoring for Heterosigm

gaeusing a sandwich hybridization assay. Harmful Al 205-214. 1:

West, N. J., R. Bacchieri, G. Hansen, C. Thomas, P. Lebaron, and H. Moreau. 2006. Rapid

in natural samples by use of a specificPrymnesium parvumdetection of the toxic alga

860-72:etry. Appl. Environ. Microbiol. monoclonal antibody and solid-phase cytom

868.

acle. 1995. DNA probe Zammatteo, N., P. Moris, I. Alexandre, D. Vaira, J. Piette, and J. Rem

hybridization in microwells using a new bioluminescent system for the detection of

PCR-amplified HIV-1 proviral DNA. Virol. Methods 55: 185-197.

48

Publication II Culture conditions and geographical origin of strains Table 1. SpeciesStrainCulturmediume-tureTempera-Origin
Pseudo-nitzschia australisPS 195 V K15°CRía de Vigo, Baiona, Spain, 05/05/2005,
S. FragaPseudo-nitzschia australisPS 193 V K15°CRía de S. FragaVigo, Baiona, Spain, 05/05/2005,
Pseudo-nitzschia australisPS 191 V K15°CRíade Vigo, (E14B),Spain, 04/05/2005,
Pseudo-nitzschia callianthaCL 187K15 °CS. FragaS. Bates
S. Bates15 °CKCL 190Pseudo-nitzschia callianthaPseudo-nitzschia multiseriesCL 174K15 °C Cardigan River, USA, S. Bates
Pseudo-nitzschia multiseriesCL 195K15 °CDeadmS. Bates an's Harbour, Bay ofFundy, USA,
Pseudo-nitzschia delicatissiPseudo-nitzschia delicatissimamaAL-23AL-63DreDrebesbes115 °C 5 °C NaplesNaples, Italy,, Italy,22004,004, A. A.AmAmatoato
Pseudo-nitzschia delicatissimaAl-86Drebes15°CNaples, Italy,2004, A. Amato
Pseudo-nitzschia fraudulentaAL-104Drebes15 °C Naples, Italy,2005, A.Amato
Pseudo-nitzschia delicatissimaAL-18Drebes15 °C Naples, Italy,2004, A.Amato
Pseudo-nitzschia delicatissimaAL-47Drebes15 °C Naples, Italy,2004, A.Amato
Pseudo-nitzschiapseudodelicatissimaAL-93Drebes15 °C Naples, Italy,2004, A.Amato
Pseudo-nitzschiapseudodelicatissimaSAL-5Drebes15 °C Naples, Italy, 2004, A. Amato
Pseudo-nitzschiapseudodelicatissimaAl-19Drebes15°CNaples, Italy,2004, A. Amato
Pseudo-nitzschia pungens238K15°CS. Kühn, K.Evans
Pseudo-nitzschia pungensOroe 5 K15°CBot. Inst. UniKopenhagen,Denmark, N.
,LundholmPseudo-nitzschia pungens19/8-Thisted 3797K15°CBot. ILundholmnst. U,niKopenhagen,Denmark, N.
Pseudo-nitzschia pungensK15 °C Sylt, Germany
Pseudo-nitzschia pungens708K15°CBristol, UK, K.Evans
Pseudo-nitzschia specK0°CNorthwest Territories, CanadaResolute Passage, Barrow Strait, , R. Smith
Pseudo-nitzschia seriataCCMP 1309K0 °C Northwest Territories, CanadaResolute Passage, Barrow Strait, , R. Smith
North Pacific15 °CKCCMP 281Chrysochromulina ericinaChrysochromulina kappaCCMP 288K20 °CHarbor,Bigelow MaineLaborUSatory docA, M.k, Kelle WerstBoothbay
Chrysochromulina polylepisB11IMR15 °CNorway, B.Edvardsen
Chrysochromulina polylepisB1511IMR15 °C Norway, B. Edvardsen
Phaeocystis globosaK+soil15°CA.Dauelsberg
Gymnodinium nagasakiense(K.PLY 561IMR+soil15°Cnot known
oi)mikimotGymnodinium fuscumCCMP 1677DY IV 15 °C Pond, LaTrobe UnVictoria, Australia, D. Hilliversity, Melbourne,
Gymnodinium simplexCCMP 418K15 °CPlymouth,England, UnitedKingdom
Gymnodinium catenatumGC 12VDreb/IMR20 °CRía de Vigo, Baiona, Spain, S. Fraga
KaGymreninodia papilionanium impudiciumceaeCAWD 91CCMP 2214GP%K502150 °C°CHawValencia,kesBNay,orth NewAtlantic, SZealand,pain, A. HaywI. Broavood
Lingulodinium polyedrumNorwayIMR15 °C Norway, T. Castberg
Ceratocorys horrida CCMP 157L122-26 °CBanda, BandaSea, SouthPacific, South
East Asia, B. Sweeney Ceratium longipesCCMP 1770K15 °C HarBigelowbor, Maine,Labor USatory docA, S.k, L. M WeostrtonBoothbay
49

Publication II

Thecadinium inclinatum

Gonyaulax spinifera

ratium reticulatumProtoceAlexaAlexandndrium minrium minuuttumum
Emiliana huxleyiPrymnesium parvumPrymnesium parvumf. parvum
. patelliferumfPrymnesium parvum

esium nemamethPrymnecumPrymnesium patelliferumPrymnesium patelliferumusFragilariopsis cylindr

CCMP 1890

CCMP 409

SyltAMP43TALCCMP 1516K-0081rv93RL10paK-0252

K-03740082K-

K

L2-Si,f1

f2-SiKKf2KIMRIMR

K1KKK

50

5 °C 1

5 °C1

°C151155 °C °C
20 °C5 °C 115 °C 5 °C 1

C°5C°51C°51 °C 0

mbia, Canada,ulBoundary Bay, British CoonsE. SimGulfof Maine, North America, 1986, R.
LandeHelgoland,North Sea, Germany, M.
enrathHoppntiago Fraga Sea, Spain, SaneanediterraMGulfofTrieste, Italy,A. Beran
lansoL. PPacific,SouthFlade So, Denmark
Bergen,Norway, A. Larsen
Wilson Promontory,Norman Bay,
SCCAPVictoria Australia,es False Bay, South Africa St JamSouthAfricaorwayNBrackish; Thornham, Hunstanton,
Norfolk, England, T. Christensen
KrellA.

Publication II

Sequences of capture and detection probes Table 2.

Probemea nPNEXDELIB

PNFRAGA

PSNAUS A-8 G A-PUNPSN12CPOLY01PRYM694

Used as theDetectionprobe

Captureprobe

Capture probe Capture probe

probeCaptureprobeCapture

nce sequeProbeGCG CAA TCA CTC AAAGAG

ATT CCA CCC AAA CATGGC

AAC GTC GTTCCGCCA AT

GGG CAC CCT CAG TACGAC

AA AGT TTC CCA TTATGACGGTCAG CCGACG CCG AGCGCG

51

tTargePseudo-nGenus 18S itzschia

GenusPseudo-nitzschia 18S

Pseudo-nitzschia australis18S

18SPseudo-nitzschia pungens

Chrysochromulina polylepis18S 28SPrymensium parvum

CitationEller & Medlin, blishedunpuTöbe & Medlin, Eller,blishedunpuin, MedlEller &unpublished &r, TöbeElleMedlin, unpublished 97)(Simon et al. 196)et al. 200(Töbe

Publication II

Table 3. Sequences of the probes and positive control 5’ to 3’
Probe name Probe sequence 5'-3' Target
FNEXTCapture probe: GCATATCTTT TTA AAA GAT TAC CCA Gymnodonium catenatum 18S
Signal probe: GCAT F CTG TCG GAC AAG GTC GTA Gymnodonium catenatum 18S
Capture probe PSNAUS Signal probe: PSNAUSCAA GGT GCT GAC GGA GACPseudo-nitzschia australis 18S
GTANEXTA-17Capture probe: PsnmultGCA TGC GAT CCG CAA TTT 18SPseudo-nitzschia multiseries
A+14Signal probe: PsnmultTCC ATC GCC GCC AAA AGG18SPseudo-nitzschia multiseries
Capture Probe GPUNPSNSignal probe: PSNPUNGCAG ACC AGT ACA GCG CAA Pseudo-nitzschia pungens 18S
ANEXTCapture probe: PSN SERI GAC AGG TTC TCG TGG TCAPseudo-nitzschia seriata 18S
GAT TC ESignal probe: PSN SERI AAT AAA GGA AAC CAA CCA Pseudo-nitzschia seriata 18S
CAAE NEXT polyCapture Probe CSignal probe: CPOLY01GGA GTC AAA AAG GAC TTC Chrysochromulina polylepsis
18SCGNEXT694myrCapture Probe PSignal probPRYM694NEXTe:CGC CAT CCA ACC AGG CTC Prymnesiumparvum/patelliferum28S
Capture probe:GGC CAT CTAAAG CAGLingulodinium polyedrum
LPOLY J AAG18S
Signal probe: LPOLYGCC CAAGAC AAG CCALingulodinium polyedrum
S18GATCCapture probe: PRETITGTAAC TAA TAA AAAProtoceratium reticulatum
S18CAG CCCT TK NEXSignal probe: PRETI KTCC GCG AAAGTCGGGProtoceratium reticulatum
S18CCA AGAA

52

Positive control
TAC GAC CTT GTC TGG CGA CAG GATTAAGTA ATC TTT AAAATT GGC GGA ACG CTCACG TTA CGT CGT CAG CAC CTT G AAA TTG CGG ATC TTTGCA TGC CCT GGAGGC GGC GATAGGGTC GTA CTG CGCGTG CCC TTGCTGTGT ACT GGT GAA TCT GAC CAC TTG TC GAG AAC CTGTCC TTTTGG TTG GTTATTACC TTA TGG GAACGGACT ATA GTC TTGAAG TCC TTT ACT CC CGC GCT CGG CGT CCT CGG CTG GAGGGA TGG ATG GCG CTT CTG CTT TAG C TGGATATG GCC GGCG TTCTT GTCGCC CGA GTTC TTCTT TCG CGGA AGG TTATGCT GTT TTCAA TAGT

Publication II

alSign---++-alSign-----+-

Table 4. Specificity of probes for Chrysochromulina polylepsis,Gymnodinium catenatum,
Lingulodinium polyedrum,Protoceratium reticulatum and Prymnesium parvum
CATGYCPOLSpeciesStrainSignalSpeciesStrainSignal
Chrysochromulina ericinaCCMP 281-Gymnodinium nagasakiensePLY 561-
Chrysochromulina kappaCCMP 288-Gymnodinium fuscumCCMP 1677-
Chrysochromulina polylepisB15+Gymnodinium simplexCCMP 418-
Chrysochromulina polylepisB1511+Gymnodinium catenatumGC 12V+
Phaeocystis globosa-Gymnodinium impudiciumCCMP 2214+
Karenia papilionaceaeCAWD 91-
PRETILPOLYSpeciesStrainSignalSpeciesStrainSignal
Lingulodinium polyedrumNorway+Lingulodinium polyedrumNorway-
Ceratocorys horrida CCMP 157-Ceratocorys horrida CCMP 157-
Ceratium longipesCCMP 1770-Ceratium longipesCCMP 1770-
Thecadinium inclinatumCCMP 1890-Thecadinium inclinatumCCMP 1890-
Gonyaulax spiniferaCCMP 409-Gonyaulax spiniferaCCMP 409-
Protoceratium reticulatumSylt-Protoceratium reticulatumSylt+
Alexandrium minutumAMP4-Alexandrium minutumAL3T-
PRYM 694SpeciesStrainSignal
Chrysochromulina polylepisB1511-
-aobosPhaeocystis gl-Emiliana huxleyi+K-0081Prymnesium parvum.fPrymnesium parvum+rv93RL10pavumpar.fPrymnesium parvum+K-0252patelliferum+ecumesium nemamethPrymn+K374Prymnesium patelliferum+K-0082Prymnesium patelliferum

53

Publication II

PSN MULTI ainStrSpeciesPS 195 V P. australisCL 187thanlliaP. caCL 190thanlliaP. caCL 174P. multiseriesP. multiseriesP. delicatissimaAL-Oroe 1323
P. pseudodelicP. delicatissimaatissimaAl-AL-8963
-5SALatissimaP. pseudodelic91Al-atissimaP. pseudodelicP. pungensP. pungens23Or8oe 5
F. cylindrus

alSign---++--------

Table 5. Specificity of probes for the Genus Pseudo-nitzschia,P. australis,P. multiseries,P.
P. seriata andpungensPSN MULTI schiaGenus Pseudo-nitzSpeciesStrainSignalSpeciesStrainSignal
P. australisPS 195 V +P. australisPS 195 V -
P. australisPS 193 V +P. callianthaCL 187-
P. australisPS 191 V +P. callianthaCL 190-
P. caP. callialliannthathaCL 190CL 187++P. multiseriesP. multiseriesCL 174Oroe 13++
P. multiseriesCL 174+P. delicatissimaAL-23-
P. delicatissimaAL-23+P. delicatissimaAl-86-
P. delicatissimaAL-63+P. pseudodelicatissimaAL-93-
P. delicatissimaAl-86+P. pseudodelicatissimaSAL-5-
P. fraudulentaAL-104+P. pseudodelicatissimaAl-19-
P. delicatissimaAL-18+P. pungens238-
P. delicatissimaAL-47+P. pungensOroe 5 -
P. pseudodelicatissimaAL-93-F. cylindrus-
--5SALatissimaP. pseudodelicP. pseudodelicatissima*Al-19+
+238P. pungensP. pungensP. pungensOrThistedoe537++
+P. pungens+708P. pungens-F. cylindrusPSN PUNG PSN AUS SpeciesStrainSignalSpeciesStrainSignal
P. australisPS 195 V +P. callianthaCL 187-
P. australisPS 193 V +P. callianthaCL 190-
P. australisPS 191 V +P. multiseriesCL 174-
P. callianthaCL 187-P. delicatissimaAL-23-
P. callianthaCL 190-P. delicatissimaAl-86-
P. multiseriesP. delicatissimaAL-CL 17463--P. fraudP. pseudodeliculentaatissimaAL-AL-91304--
P. delicatissimaAL-18-P. pseudodelicatissimaAl-19-
P. pseudodelicatissimaAL-93-P. pungens238+
P. pseudodelicatissimaAl-19-P. pungensOroe 5 +
P. pungens238-P. pungens+
P. pungensOroe 5 -P. pungens708+
-F. cylindrusPSN SERI SpeciesStrainSignal
P. callianthaCL 187-
-CL 174P. multiseriesP. delicatissimaP. delicatissimaAl-AL-8476--
-93AL-atissimaP. pseudodelicP. pungensP. pseudodelicatissimaOrSALoe-55--
-P. pungensP. seriata1309CCMP+
-F. cylindrus* likely misidentified shouldbeP. delicatissima based onsequenceidentity
54

PSN PUNG ainStrSpeciesP. caP. callialliannthathaCL 190CL 187
CL 174P. multiseriesP. delicatissimaP. delicatissimaAl-AL-8263
P. fraudulentaAL-104
39AL-atissimaP. pseudodelicP. pungensP. pseudodelicatissimaAl-23819
Oroe 5 P. pungensP. pungens708P. pungens

alSign--------++++

Publication III

Publication2.5 III

E

LECTROCHEMICAL DETECTION OF TOXIC ALGAE WITH A BIOSENSORSONJADIERCKS,KATJAMETFIES AND LINDAK.MEDLIN
Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570
anyerhaven, GermBremManual and Guides: Microscopic and molecular methods for quantitative phytoplankton
ittedmis, subanalys

nIntroductio

various areas. Glucose detection was one of the firstDNA-biosensors are known fromnyaapplication areas developed for biosensors (Clark 1956). Today, biosensors are used in mners (Hartley and Baeumof infectious organismtificationdifferent areas, such as for the iden2003) and hazardous chemicals, for monitoring of health relevant metabolites or
environmental samples. A new detection methodused for the identification of harmful algae
e 1A) and biosensors. A first prototype was was developed using a hand held device (Figurused to identify the toxic dinoflagellateAlexandrium ostenfeldii (Metfies et al. 2005). A
second prototype manufactured by PalmSens (Houten, Netherlands) was extensively used to
ssa). Biosensors can be produced very cheaply for mprove the biosensors (Figure 1Bimproduction.

cleotide probes that c algae is based on oligonu - Identification of toxiMolecular probeste the small and large subuniRNA. Targets for the probes aralspecifically target ribosomke it aes of the cells, whose conserved and variable regions mrRNA genes in the ribosomic levels (Groben et al. 2004). For possible to develop probes specific for different taxonomlent, the ARB software package is used (Ludwig et al. 2004). Theoreticathe probe developmle inber of sequences of the targeted gene availabicity is dependent on the numprobe specifthe databases. If molecular probes are designed from only a few sequences, there is a danger
s whose sequences are unknown of cross-hybridization to non-targeted species and organismand not in the database. Prior to the analysisof field-samples, molecular probes were tested

55

Publication III

for specificity with cultivated target species and closely related species becausein silicoand

results can show different specificity signals. in situ

Dipsosable sensor-chip and detection principle - The disposable sensorchip consisting of a

carrier material on which is printed a working electrode, where the detection reaction takes

auxiliary electrode (Figure 1B). The working electrode h a reference electrode and anplace,sa

a diameter of 1mm and is made of a carbon paste. A biotinylated probe is immobilised on the

on the cleic acids are detectedidin. The nureaction layer of the working electrode via av

sensor chip via a sandwich-hybridization (Zammatteo et al. 1995; Rautio et al. 2003).The

thod is that one target specific probe, the so-called capture e this munderlying principle of

on the surface of the working electrode. If a target nucleic mobilised via avidinprobe, is im

e detection of the nucleic d to the immobilised probe on the working electrode, thacid is boun

acid takes place via a h the so-called signal to a second target specific probe,ybridization

The digoxigenin probe, that is coupled to digoxigenin (Figure 1C) (Metfies et al. 2005).

adish-peroxidase is added to the sensor chip. Horseradish-specific antibody coupled to horser

peroxodase catalyses the reduction of hydrogen peroxide to water. Reduced peroxidase is

-ampregenerated by in (ADPA), which functions as a mediator. The oxidised inodiphenylam

mediator is reduced at the working electrode with a potential of í150mV (versus Ag/AgCl)

ical signal can only be m(Figure 1D). An electrochemeasured if the target nucleic acid bound

d thus present in the sample to be treated. (Metfies et al. anto both capture and signal probes

2005)

56

Publication III

nd methods Materials a

Laboratory facilities and equipment

e hood for RNA isolation Fumx

Centrifugex

xFilter, 0.5 µm, ISOPORE™, membrane filters, Millipore, Ireland
re, Ireland d funnel, MillipoFrit, flask anx

Mini-Beadbeater™, Biospec products, Biospec products Inc, USA x

Mini-Centrifugex

oshakerThermx

rIncubatox

pump with wash bottle VacuumxBiosensors, Gwent Electronic Materials, Pontypool, UK x

CFreezer -80°x

Chemicals and supplier - The chemicals used in this method are listed in Table 1 with their

suppliers.

e harvesting of cells can be done by centrifugation, the supernatant hTHarvesting of cells -

will be discarded, or by filtration using a filtration device and a hand pump (Figure 2). A
maximum of or 1 x 107 cells can generally be processed with the RNeasy Plant Mini Kit. The

cells can be frozen for long-term storage by flash-freezing in liquid nitrogen and an

70 °C. Another possibility is the storage of cells in RNALater from ediate transfer to –imm

bion (Huntingdon, UK). Am

Preservation and storage - After collecting water samples the algae cells can be stored at

room temperature over several days by using RNALater from Ambion, Huntingdon, UK for a

later RNA isolation. Please read carefter.ully the instructions for using RNALa

IAGEN) (modified protocol)Qi Kit (RNA Isolation with the RNeasy Plant Min

General handling of RNA –Ribonucleases (RNases) are very stable, active enzymes and are

difficult to inactivate; even minute amounts are sufficient to destroy RNA. Use only plastic

ware or glassware where you have first eliminated possible RNase contamination. Glassware

57

Publication III

should be cleaned with a detergent, thoroughly rinsed, and oven baked at 180°C for four or more hours before use. Always wear gloves while handling reagents and RNA samples to ry dusty laboratoace of the skin or fromination from the surfprevent RNase contamle. Keep ever possibkeep tubes closed whenent. Also change gloves frequently and equipm applications. r downstreamoisolated RNA on ice when aliquots are pipetted f

RNA-Isolation

-Mercaptoethanol to the cellsȕAdd 450 µL Buffer RLT with 1.2.Pipet the lysate to glass beads and shredder the lysate in a bead beater two times for 20
seconds3.Pipet the lysate directly onto a QIAshredder spin column (lilac) placed in 2 ml collection
tube, and centrifuge for 15 minutes at maximum speed. Carefully transfer the supernatant
crocentrifuge tube without disturbing the cell-iof the flow-through fraction to a new mllection tube. Use only this supernatant in subsequent steps. debris pellet in the co4.immAdd 0.5 volumediately by pipetting. Do not centrifuge. Continue without delay. e (usually 225 µL) ethanol (96–100%) to the cleared lysate, and mix
5.Apply sample (usually 650 µL), including any precipitate that may have formed, to an
RNeasy mini column (pink) placed in a 2 mlcollection tube. Close the tube gently, and
Discard the flow-through.8000 x g.centrifuge for 15 s at Reuse the collection tube in the next step. Close the tube gently, and wait for ca. 45 n.Add 700 µL Buffer RW1 to the RNeasy colum6.seconds, then centrifuge for 15 s at 8000 x gto wash the column. Discard the flow-
through and collection tube. Repeat step 6 7. collection tube (supplied). Pipet 500 µL ln into a new 2 mTransfer the RNeasy colum8.Buffer RPE onto the RNeasy column. Close the tube gently, and centrifuge for 15 s at n. Discard the flow-through. to wash the colum8000 x gReuse the collection tube in step 9. Repeat step 8 9.n. Close the tube gently, and the RNeasy columoAdd another 500 µL Buffer RPE t10.centrifuge for 2 min at 8000 x gto dry the RNeasy silica-gel membrane.

58

Publication III

11.To elute, transfer the RNeasy column to a new 1.5 ml collection tube. Pipet 30-50 µL

mbrane. Close the tube gently, eRNase-free water directly onto the RNeasy silica-gel m

and centrifuge for 1 min at 8000 x gto elute.

ed by y be performaTo obtain a higher total RNA concentration, a second elution step m12.

using the first eluate (from step 11).

Measure the RNA conce13.ntration

Sandwich Hybridization

A. Coating of Sensor chips

The sensor chips are moistened with 50 µL of carbonate buffer (pH 9.6) (Table 3, Figure 1.

p (Figure 3B, 3C) pum3A) and aspirated of with a vacuum

2.Incubation over night in a moisture chamberat 4 ºC with 2 µL NeutrAvidin (Pierce,

rbonate buffer (Table 3). Storage of the electrodes during this any) in caPerbio, Germ

period in Petri dishes with moist Whatman filters to protect the solutions from evaporation

(Figure 3D)

oved by washing the chips in PBS (pH 7.6) (Table 3, Figure Excessive NeutrAvidin is rem3.

ttached to a wash bottle p a pum3E). Subsequently the chips are dried with a vacuum

The sensors are blocked for one hour at room4.S.Btemperature with 20 µL 3 % casein in P

oved by washing with PBS The casein is rem

es can be stored in a fridge for at least 1 year afterThe NeutrAvidin coated electrod5.

S (pH 7.6). The electrodes are coated with 15 µL of Bincubation with 2 % Trehalose in P

Trehalose solution and dried at 37 °C in an incubator. Before use the electrod

ove the Trehalose. washed with PBS (pH 7.6) to rem

NA-probeobilization of biotinylated DmmB. I

es are

The sensor chips are co6.ated with 2 µL of the biotinylated probe [10 pmol / µL in bead

buffer (Table 3) and incubated for 30 minutes at room temperature

7.the sensors and directly aspirated of to 50 µL of 1x hybridization buffer are added onto

ove excessive probe rem

8.In accordance to the coated electrodes can be stored in a fridge for at least 1 year after

incubation with 2 % Trehalose on PBS (pH 7.6). The electrodes are coated with 15 µL of

es are ior to usage the electrodrTrehalose solution and dried at 37 °C in an incubator. P

ove the Trehalose. washed with PBS (pH 7.6) to rem

59

Publication III

bilized DNA probe, rRNA and dioxigenin labelled DNA C. Sandwich Hybridization of immoprobe9.Fragmentation of rRNA by using a fragmentation buffer (200 mM Tris-Acetate, pH 8.1,
500 mM KOAc, 150 mM MgOA). 10 µL rRNA are added to 2.5 µL fragmentation buffer,
ermoshaker (Figure 3F) and subsequently in a thnutes at 94 °Ciheated for five mediately chilled on ice. imm

de up as shown in Table 2. The positive control aThe hybridization preparation was m10.ensures that the probes are working and the negative control shows the detection of the
used compounds without RNA during the mfour minutes at 94 ºC in a thermoshaker for denaturation and immediately chilled on ice. easurement. This preparation was heated for

11.2 µL of the hybridization solution are applied onto each sensor in triplicate
12.The chips are incubated for 30 minutes at 46 °C in an incubator, cooled down at room
nutesimperature for five tem

D. DetectionA.ove excessive RNThe sensors are washed in 1x POP buffer (pH 6.45) (Table 3) to rem13.14.The sensors are incubated for 30 minutes at room temperature with 1.5 µL Anti-Dig-POD
[7.5 U/mL in PBST] (Table 3).
15.Sensors are separately washed in 1x POP buffer to remove excessive Anti-Dig-POD and
pump dried with a vacuum N-g20 µL of POD substrate are added onto the electrode (POD substrate contains 1.1 m16.ADPA) solved in 110 µL ethanol, 250 µL of ine hydrochloride (Phenyl 1,4-phenylenediam100 mM H2O2 are added and filled up to 25 mL with 1x POP buffer)
ary of the easured (Figure 4). A summThe chip is plugged in the hand held device and m17.used buffers is shown in Table 3 18.Formulas for calculating results - A calibration has to be determined for each probe set to
net species the RNA concentratio (nA) for 1 ng RNA. For each targsityfind the signal intenper cell has to be investigated. Subsequently the cell concentration of the target species in a
als:ical signple can be calculated from the electrochemwater sam

60

Publication III

Letple) nA (probe-signal) = total ng RNA (present in the samthen

eber of cNumrobe-signal)/ ngRNA (per cell) lls = nA (p

Discussion

thod with the hand held device and biosensors is a rapid eical detection mThe electrochemmethod to detect toxic algae in a water sample. Electrodes can be produced in mass. Protocols
even for a scientific ple and easy ical readings of the handheld device are simand electrochemlayperson to use and interpret.

Our initial prototype with many manual steps, has now been defined and improved in the EU-
ated flow and heating ted with an automatomproject ALGADEC so that nearly all steps are auchamber for biosensors for the detection of 14 species in parallel, except the initial sampling
rsensor consists of a disposable sensoNA extraction. The present bio step and Rand filteringture probe een the cap electrodes upon which a redox reaction takes place betw6chip with 1ical detection that isand the signal probe to yield a flow of electrons for an electrochemof cells in water columproportional to the RNA of target captured on thn. Probes for other toxic algae (e.g. e chips and hence proportional to the numAlexandrium minutumber,
etc.) were developed for operating with the hand held device and Gymnodinium catenatumabout 14 different toxic algae can be detected, because a negative and a positive control have
st be reviewed for specificity to new sequence datauto be included in the assay. The probes min defined time intervals, because the current 18S rRNA sequence database is only a small
part of the biodiversity and is always upgraded. For each target species, the RNA concentration per cell has to be investigatedand a calibration curve has to be developed for
each probe set to determine the signal intensityfor the different RNA concentrations to be
ple.bers in the field samable to relate this to cell num

e, which it of the hand held device requires a high sampling volumThe current detection limcan be up to 8-10 litres if the cell counts are expected low. For the isolation of target rRNA a
it with the hand held device for tection limsufficient amount of cells is needed. The de is ~16 ng/µL, with an average yield of 0.02 ng/cell. This equates to Alexandrium ostenfeldii

61

Publication III

ca. 800 cells or a sam

of 6.4 l to get a detectable ameling volump

cells/litre (Metfies et al. 2005).

The manual isolation of RNA is currently th

of rRNA fromount

iting factor of the systeme lim

250

e, because th

concentration and quality needs to be high. Different users could possibly isolate different

amounts of rRNA with different qualities from the same number of algae cells. This results in

ated RNA cell counts. An automdifferent signal intensities, which cannot be compared to

e the quality in rRNA project, will overcomisolation, as developed during the ALGADEC-

als against total rRNA and over the growth nextraction efficiency. A validation of probe sig

is also being conducted to verify ental conditions cycle of the algae under different environm

exthe calibration curves tolate to cells/litre.trapo

ledgementsAcknow

Sonja Diercks was supported by the EU-project ALGADEC (COOP-CT-2004-508435-

of the European Union and the Alfred Wegener ework program of the 6th framALGADEC)

Institute for Polar and Marine Research.

62

Publication III

References

Clark, L. C. 1956. Monitor and control of blood and tissue oxygenation. Trans. Am. Soc.

41-48. Artif. Intern. Organs 2:

Groben, R., U. John, G. Eller, M. Lange, and L. K. Medlin. 2004. Using fluorescently-

NA probes for hierarchical estimlabelled rRni-iation of phytoplankton diversity – a m

313-320. review. Nova Hedwigia 79:

r. 2003. Biosensor for the specific detection of a single neHartley, H. A., and A. J. Baeum

thracis spore. Anal. Bioanal. Chemviable B. an 319-327. 376:

ent for sequence data. Nucleic Acids and others 2004. ARB: a software environm.Ludwig, W

1363-1371. Res32:

ical detection of the Metfies, K., S. Huljic, M. Lange, and L. K. Medlin. 2005. Electrochem

eldii with a DNA-biosensor. Biosens. lagellate Alexandrium ostenftoxic dinof

1349-1357. Bioelectron20:

eitenstein, S. Molin, and P. Neubauer. 2003. rRautio, J., K. B. Barken, J. Lahdenpera, A. B

Sandwich hybridization assay for quantitative detection of yeast RNAs in crude cell

lysates. Microb. Cell Fact 4. 2:

acle. 1995. DNA probe Zammatteo, N., P. Moris, I. Alexandre, D. Vaira, J. Piette, and J. Rem

hybridization in microwells using a new bioluminescent system for the detection of

PCR-amplified HIV-1 proviral DNA. Virol. Methods 55: 185-197.

63

Publication III

ChemTable 1. icals and suppliers

alChemicNeutrAvidin™, biotinbinding protein
D(+)-Trehalose, 99.5 % HPLC
s)base(18Biotin-labelled probeHeDigoxirringgeni-Spern-lam DNbelled prAobe(18 bases)
1x PBS n 20 TweehN-Pydrohechlnyl 1,4-orideCphe12Hnylene12Ndiam2 HCl (ADinePA,N-
Phenyl-1,4-benzenediamine hydrochloride,
1,1-Diphenylhydrazin-hydrochlorid)
PAADntseAnti-Digoxigenin-POD fab fragmHydr(w/w)ogen peroxide solution H2O2, 30%
lanoEthSodium hydrogencarbonatNaHCO3
NaH2PO4 * H2O
NaClCaseinTris (pH8.0)
SDSBSAhanolptoet-MercaȕMini KitRNeasy Plant an filters Whatm, 425 – 600 µm212 – 300 µmbeads,Glass

SupplierFlukaPIERCE, BioChe Perbimio,Germka, Switzerlanyand
ElectronoThermRoThermcheoElectron
anyrmGeo,PIERCE, PerbiGermSigmanya-Aldrich Chemie GmbH,

anyGermMERCK KGaA,

ocheRGermSigmanya-Aldrich Chemie GmbH,
MERCKKGaA,Germany
RdH,n®,Riedel-de HaëLaborchemikalien, GmbH& CoKG,
anyGermMERCKKGaA, Germany
bH,mie Ga-Aldrich ChemSigmanyGermbH,mie Ga-Aldrich ChemSigmGermSigmanya-Aldrich Chemie GmbH,
anyGermSigmGermanya-Aldrich Chemie GmbH,
SigmGermanya-Aldrich Chemie GmbH,
MERCKKGaA,Germany
Qiagen, HildenGermany
Whatman, Brentford,UnitedKingdom
bH,mie Ga-Aldrich ChemSigmanyGerm

64

Publication III

Hybridization preparation Table 2.

the species tion of Detec3.5 µL 4xHybridization buffer
NA µL rR5.71 µL HerringDNA(3480
)ng/µL1 µL DIG marked DNAprobe
(1.4 pM/µL) lliQwateri1 µL m

olive contrNegat3.5 µL 4xHybridization buffer
1 µL HerringDNA(3480 ng/µL)
1 µL DIG marked DNAprobe (1.4
/µL)pMwaterlliQi8.5 µL m

65

ve controlPositi3.5 µL 4xHybridization buffer
1 µL HerringDNA(3480 ng/µL)

1µLTest DNA(36 bases, 1.4 pM/µL)

1 µL DIG marked DNAprobe (1.4
pM/µL)waterlliQi7.5 µL m

Publication III

Buffers for sandwich hybridization on carbon electrodes Table 3.

Buffercarbonate buffer (pH9.6)
10x PBS (pH 7.4)

r""bead buffe

4x hybridizationbuffer

10x POP buffer(pH 6.45)

PBS-BT (pH 7.4)

poundComO3NaHCNaH2PO4 * H2O
H 7.4) NaCl (pNaClTris (pH7.6)
NaClTris (pH8.0)
SDSNaH2PO4 * H2O
H 6.45)NaCl (pPBSBSATWEEN 20(pH
4)7.

ConcentrationM50 m10..554M M
3 M0. M.10 M3.0M m0804%0.M5.0 M 11x] % [w/v.10

05 % 0.v][v/

66

Publication III

Figure 1.Sens device, (B) Sensor chip of original prototype, (C) Sandwich (A) Palm

hybridization, (D) Principle of redox-reaction

67

Publication III

Figure 2.

viceeent and filtration d Filtration equipm

68

Publication III

Figure 3.

(B) Pump and washbottle, (C) Drying of (A) Applying of buffer onto the electrode,

chips, (D) Petri dish with Whatman f

oshakerTherm

ilter

and electrodes, (E) Washing of chips, (F)

69

Publication III

Figure 4.

Measuring of chips with hand held device

70

Publication IV

IV Publication2.6

E

VALUATION OFLOCKED NUCLEIC ACIDSFOR SIGNAL ENHANCEMENTFOOLIGONUCLEOTIDE PROBES FOR MICROALGAE IMMOBILIZEDON SOLID

SURFACES

SONJADIERCKS ANDCHRISTINEGESCHER,KATJAMETFIES,LINDAK.MEDLIN

Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, 27570
anyerhaven, GermBremittedgraphy: Methods, submnology and OceanoLim

Abstract

onitoring of croarrays are powerful tools for species detection and miBiosensors and mmicroorganisms, e.g., phytoplankton. A reliable identification of microbial species with
thods requires highly specific and sensitive probes. The introduction of LNA eprobe-based ment of both specificity and ises an enhancem(locked nucleic acid) probe technology promsensitivity of molecular probes. In this study, we compared the specificity and sensitivity of
conventional molecular probes and LNA modified probes in two different solid phase
hybridization methods; sandwich hybridization on biosensors and on DNA-microarrays. In
combination with the DNA-microarrays, the LNA-probes displayed an enhancement of
sensitivity, but also more false-positive signals. In combination with the biosensor, the LNA
probes could show neither signal enhancement nor discrimination of only one mismatch. In
ined cases, the conventional DNA probes showed equal or better results than the all examLNA probes. In conclusion, the LNA technology mayhave great potential in methods that use
probes in suspension and possible in gene expressions studies, but under certain solid surface-prove signal intensity.ns they do not imhybridization applicatio

71

Publication IV

nIntroductio

ngel (Koshkin et al. 1998a; Koshkin et eted by WLNA (locked nucleic acids) were first presen

bicyclic anishi (Obika et al. 1998) and their co-workers. They are a class ofal. 1998b) and Im

RNA analogs with exceptionally high affinitiesand specificities toward their complementary

lecules (Koshkin et al. 1998b; Singh et al. 1998). They can be oDNA and RNA target m

substituted for any conventional nucleic acid in any synthetic oligonucleotide. It is possible to

ny of the conventional nucleic oligonucleotides by replacing antional of conveenhance the Tm

acid in the oligonucleotides with a LNA (Singh et al. 1998). Thus, the use of LNAs could

significantly increased mismatch discrimination(Kauppinen et al. 2003). In modified nucleic

thylene bridge connects the 2’-oxygen and the 4’-carbon (Parekh-Olmedo et al. eacids, a m

ination of the ribose and ational determ2002) and consequently produces higher conform

tion (Braasch am backbone in a 3P-endo conforincreased local organization of the phosphate

tson-Crick base pairing (Koshkin et al. 1998b) aore, LNAs obey Wand Corey 2001). Furtherm

plemand thus, are easy to imistry (Kauppinen ent into standard oligonucleotide synthesis chem

offer new potentials for use in DNA/RNA oligo recognition based et al. 2003). LNAs

methods because of certain enhanced properties over normal nucleic acids. According to

(Kongsbak 2002), they could be used in any hybridization assay as a modified probe or

y. They are used with standard reagents and er to increase specificity and reproducibilitprim

protocols, have the same solubility as DNA or RNA, low toxicity, can makechimeras with

DNA or RNA, are obtainable fromrey 2001) and are o industrial companies (Braasch and C

004). The only disadvantage is that they are ngel 2enot affected by nucleases (Vester and W

acids. Because of these re expensive than conventional nucleic omuch menhanced properties,

e.g., gene expression e their first introduction,ny applications sincaLNAs have been used in m

ielsen and Kauppinen 2002), genotyping assays (Jacobsen et al. 2002a; Jacobsen profiling (N

hybridization (Silahtaroglu et al. 2003; Silahtaroglu et al. et al. 2002b), fluorescence in situ

2004; Wienholds et al. 2005; Kloosterman et al. 2006; Kubota et al. 2006), real-time PCR

es (Vester et al. elshoj et al. 2005; Sun et al. 2006) DNAzym(Ugozzoli et al. 2004a; Humm

e2004; Vester et al. 2006) and other mthods.

Because of these successful applications of LNA-modified probes, their use in species

identification in sandwich hybridization and microarray assays should be evaluated. LNA

s of low hybridization efficiency and cross e problemmodified probes could possibly overcom

72

Publication IV

the target hybridization of probes to closely related non-target species, often separated from

atch.smiase mspecies by a single b

Molecular probes are widely applied for the identification of micro-organisms, e.g., toxic
algae. They are applied in comection techniques: Fluorescence in ination with a variety of detb

situ hybridization or FISH (Scholin et al. 1996; Scholin et al. 1997; Simon et al. 2000; Smit et
ion assays or SHA (Scholin et al. 1996; and Sako 2005), sandwich hybridizatal. 2004; Kim

iMetfies et al. 2005) and DNA mcroarrays (Metfies and Medlin 2005a; Metfies and Medlin

2005b). The small and the large subunit ribosomal RNA genes are the usual targets for
rmolecular pber in the cell and they contain more or se there is a high target numobes, becauking it possible to develop probes that are specific at different aless conserved regions, m

ic levels (Groben et al. 2004). Probe specificity is dependent on the number oftaxonom

databases. Cross-reactions can occur with eted gene available insequences of the targ a low quence of the probe is designed fromunknown non-targeted species if the target senumber of sequences or the group is relatively unknown or unculturable and there are many

ined. Even when a probe is yet been determnon-targeted species whose sequences have not designed from a large database, it is necessary to revise probe sequences frequently because

denteility has been documost daily to databases. Genetic variabnew sequences are added almamong geographically dispersed strains of the same species (Scholin et al. 1994), making

specific probes design even more challenging if global strains have not be sampled. One
uction is to choose the best sequence from in probe design and constrportant problemimhybridizationlly identify the target. Excellent in-situseveral possibilities that could theoreticaeters, suchcorrelate well with in-silico paramresults of any probe does not always appear to perature (Graves 1999). It is not possible to predict which lting temeas G–C content or m

probes will work well under all hybridization conditions. Sometimes probes that work well in

dot blot and FISH formats do not work at all in a microarray format (Metfies and Medlin,
unpublished).

an portant fromThe identification of phytoplankton, especially of harmful algae species, is im

point of view. Certain harmful algae have the potential to produce ciecological and econom

toxins that have the capability to seriously harm, or even kill, other organisms or even humans

erous monitoring ed. Numediaries in the food chain, such as mussels, are consumif interm

programful harms are established along all coastlines around the world for the detection of

algae. The European Union demands the monitoring of the coastlines for toxin-producing

73

Publication IV

phytoplankton and toxins in mussels by the member states (Directive 91/492d/EC and

Commission Decision 2002/225/EC). Cell detection methodology based on light microscopy

can be tedious and time-consuming when large numbers of samples need to be processed

ained personnel and y require highly trae species mroutinely, and identification of som

liable species identification and long-tereent (Tyrrell et al. 2002). Rexpensive equipmm

monitoring are difficult to achieve by traditional methods, because unicellular algae are

onging to the samically challenging with toxic and non-toxic strains beltaxonome species. In

have been adapted for the identification ofthodsethe past decade, a variety of molecular m

crobial species, which are often lacking in distinct morphological features. Molecular im

natural phytoplankton populations ul alternative in the study ofidentification is a very usef

ent of a molecular probe-(Guillou et al. 1999b). In our lab, we are working on the developm

atingcroarray for the detection of harmful algae and for estimimbased biosensor and a DNA-

hidden biodiversity. In particular, we focus on those species that have the potential to harm

ent by the production of potent toxins. the environm

The two solid-phase mcroarrays for phylogenetic analyses and ithods described here: DNA me

using target specific undance of algal species asure the abean rRNA-biosensor, are used to m

probes bound to a surface.

rRNA biosensor - The detection method using a rRNA-biosensor was successfully introduced

elecular moby (Metfies et al. 2005) as a mthod for the detection and identification of the toxic

. It utilizes sandwich hybridization (SHA) with a Alexandrium ostenfeldiidinoflagellate

e target RNA or DNA and a second signal probe that carries the capture probe that binds to th

iety and binds near the binding site of the capture probe. A third additional probe, osignal m

ing site of the two other probes to modify the the so-called helper probe, binds near the bind

lecule so that the signal probe can easily form its heteroduplex. osecondary structure of the m

This region usually consistsof approximately 50 bps leaving little for probe manipulation

plicated by the should the probes not work properly. The search for suitable probes is com

relative conservation of the 18S gene at the species level (Gagnon et al. 1996; Ki and Han

not been rigorously evaluated because only hyper-variable2006). More variable genes have

jority of the gene unknown and open for non-aregions have been sequenced leaving the m

specific binding. The detection is measured electrochemically by the PalmSens instrument

and its PSLite software (Palm Instruments,Houten, Netherlands) and was adapted from the

original biosensor presented by Metfies et al. (2005).

74

Publication IV

Probes for the rRNA biosensor (Table 1) - AOST1 (the signal probe), AOST2 (the capture

are 18S-rRNA probes designed by (Metfies et al. probe), and their helper oligonucleotide, H3,

alized signals for 2005) and were tested for specificity with dot blot and SHA. Although norm

all non-target organism the signals from are significantly higher thanA. ostenfeldiis, there is a

low cross hybridization to A. minutum, whichhas 2 mismatches to the capture probe. An

proved protocol for the isolation of algal RNA with the Qiagen RNeasy Plant Mini Kit,im

Hilden only enhances this cross reaction. The recently described non-toxicAlexandrium

tamutum(Montresor et al. 2004) presents a single mismatch to the capture probe for A.

specificity of this probe.its of, thus challenging the limostenfeldii

DNA-Microarray-A DNA-microarray consists of a glass-slide with special surface properties

of nucleic acids, e.g., oligonucleotides, ny copies aeyer and Blohm 1999a) and m(Niem

cDNAs or PCR-fragmnts spotted on it (Graves 1999) in a specific pattern . It is a widely e

used routine tool in many applications because it offers the possibility to analyze a large

ber of up to 250,000 different targets in parallel without a cultivation step (Lockhart et al. num

1996; Graves 1999; Ye et al. 2001). Nucleic acids are fluorescently labelled before

croarray scanner (Derisi et al. 1997). ihybridization and they are detected afterwards with a m

Many functional genomic methods benefit fromthis technology, such as genome expression

profiling, single nucleotide polymorphism detection and DNA resequencing (Lipshutz et al.

t al. 2005; Broe1999; Kauppinen et al. 2003; Ji and Tan 2004; Yap et al. 2004; Al-Shahrour et

et al. 2006; Gamberoni et al. 2006). DNA-microarray technology is also used to differentiate

croalgae (Metfies and Medlin 2005a; Ki and Han 2006; Metfies et al. 2006), fish (Kappel im

plies et e al. 2002; Peplies et al. 2003; Peplies et al. 2004a; Pet al. 2003) and bacteria (Loy et

al. 2004b; Lehner et al. 2005; Loy et al. 2005a; Peplies et al. 2006).

–Four out of five probes used here (Table 2) were previously Probes for the DNA-microarray

icroarray (Metfies and Medlin 2005a). The fifth probe, Crypto B, evaluated on the DNA-m

recognizes all pigmonad algae. It could be shown that these probes work ented cryptom

specifically with microarrays, but there was potential for enhancement of the signal-to-noise-

ratios because these probes gave low signals and thus w

ent with LNAs.enhancem

75

re good candidates for signal-e

Publication IV

Materials and Procedures

- ditionsCulture conater-basedAll algal strains were cultured under sterile conditions in seawtein – 200 µEinstein dia (Eppley et al. 1967; Keller et al. 1987) at 15 °C and 150 µEinsemwith a light: dark cycle of 14:10 hours (Table 1).

RNA-extraction - Total RNA was isolated from all algalcultures with the RNeasy Plant Mini
Kit (Qiagen, Hilden, Germany) with modifications of the protocol to enhance the quality of
the RNA. This involved a centrifugation of 15minutes instead of two minutes to achieve an
es to the is. Buffer RW1 was applied two timproved separation of supernatant and cell debrimRNeasy column, incubated for one minute and then centrifuged. The first wash step with
easured with a Nanodrop NA concentration was mbuffer RPE was repeated. RSpectrophotometer (Peqlab, Erlangen, Germany). All of these changes increased the removal
NA extracted. oteins to improve quality and quantity of the rRrarides and pof polysacch

ental clones was isolated frommplate DNA from the environ- The temDNA-extractionstrains was extracted frombacteria by using the Plasmid Mini Kit (Qiagen, Hilden, Germ pure cultures with the DNeasy Plant Mini Kit (Qiagenany). DNA from the algal , Hilden,
any).Germ

The entire 18S gene (1800 bp) from the target DNA was PCR Amplification of 18S rRNA -ers 1F (5'-AAC CTG GTT GAT CCT GCC AGT-R primplified with universal specific PCam) without the polylinkers ACC TAC- 3'- TGA TCC TTC TGC AGG TTC) and 1528R (5'3'(Medlin et al. 1988). The PCR protocol was: 5 min 94°C, 2 min 94°C, 4 min 54°C, 2 min
(Eppendorf, Ha72°C, 29 cycles and 7 mmburg, Germiany). A 250 bp fragmn 72°C. All PCR experiment of the TATA-box binding protein-gene ents were done in a Mastercycler
(TBP) of Saccharomycescerevisiae was amplified with the primers TBP-F (5'-ATG GCC
') and TBP-R-Biotin (5'-TTT TCA GAT CTA ACC TGC GAT GAG GAA CGT TTA A-3ACC C- 3') and used as a positive control in the microarrayhybridization experiments. The
TBP amplification protocol was: 5 min 94°C, 1 min 94°C,1 min 52°C, 1 min 72°C, 35 cycles,
10 min 72°C. All PCR-fragments were purified with the QIAquick PCR purification (Qiagen,
odifications of the protocol to enhance the quantity of the PCR-any) with mHilden, Germfragments. The elution with the elution buffer EB (Step 8) was performed twice with the same

76

Publication IV

eterasured with a Nanodrop Spectrophotomebuffer. The concentration of the DNA was many).(Peqlab, Erlangen, Germ

Biotin-Labelling of the purified PCR- fragments - For the enhancement of signal intensities
any) was used. entas, St. Leon-Rot, Germthe Biotin DecaLabel DNA Labeling Kit (Femwas carried out over night (17 to 20 hours) to entLabelling of 200 ng of purified PCR-fragmmaximize biotin incorporation into the PCR-fragments. After labelling the purification was
done with the MinElute PCR Purification Kit (Qiagen, Hilden, Germany) with modifications
ents as above. Concentration of the CR-fragme Pof the protocol to enhance the quantity of thDNA was measured with a Nanodrop Spectrophotometer (Peqlab, Erlangen, Germany).

Probe synthesis –All probes and helper oligonucleotide probes and positive and negative
controls were synthesized from Thermo Electron Corporation, Ulm, Germany. The locked
nucleic acids were synthesized from Exiqon (Vedbaek, Denmark). The position of the LNA-
ation from Exiqon but they were regularly residues within the sequence is proprietary informmal nucleic acids.ong norinterspersed am

rRNA biosensor– A set of two specific 18S-rRNA probes (AOST1 and AOST2, Table 2) was used Probe set to assess the impact of LNA-probes on the specificity of probes with the biosensor. The
ed nucleic acids as a Exiqon with lockfromsequence of capture probe AOST2 was redesigned shorter oligonucleotide to maintain the identical melting temperature as the conventional
probe AOST2. Three different probes, LNA 65, LNA 66 and LNA 67, were synthesized with ination with AOST1. Probe AOST2 hasba biotin-label and were used as signal probes in com) of 66 °C, AOST1 of 64.3 °C, LNA 65 and LNA 66 of 65°C and lting temperature (Tmea me control was not modified with LNAs. LNA 67 of 60°C. The positiv

rThe specificity of the LNA probes using the rRNA biosenso -Algal strains and templatesable 1) and the non-target strains, (TAlexandrium ostenfeldiiwas tested with the target strain SZNB029. Alexandrium tamutum AL3T and Alexandrium minutum

The biotinylated capture probes (AOST2, Immobilization of the probes on the sensor chip -e sensor chips as described by (Metfies et LNA 65, LNA 66; LNA 67) were immobilized on thal. 2005). The working electrode was pretreated with Carbonate buffer (50 mM NaHCO3, pH

77

Publication IV

l] (Pierce Biotechnology, m/g9.6) following which incubation with NeutrAvidin [0.5 m

was carried out. Excessive NeutrAvidin was Rockford, USA) for at least 4.5 hours at 4 °C

oved fromremith PBS (BupH phosphate saline the working electrode by washing the sensor w

pack, Pierce Biotechnology, USA). Subsequently, the working electrode was blocked with

S. The B temperature and afterwards washed in P3% [w/v] casein in PBS for 1 hour at room

in bead buffer (0.3 M NaCl/0.1M Tris, pHMprobes were dissolved at a concentration of 10 µ

7.6) prior to immobilization on the electrodes for 30 minutes at room temperature. All

oincubation steps were carried out in a mber to avoid evaporation. Unbound probe isture cham

was removed from the electrode by washing with hybridization buffer (75mM NaCl/20mM

Tris, pH 8.0/0.04% SDS).

Hybridization - Prior to hybridization the total rRNA was fragmented in fragmentation buffer

(40mM Tris, pH 8.0/100mM KOAc/30mM MgOAc) for 5 minutes at 94°C. The hybridization

M NaCl/20mMxture for the detection of rRNA contained 1x hybridization buffer (75mim

ol/µL dig-labeled probe DNA, 0.1 pmL herring spermTris, pH 8.0/0.04% SDS), 0.25 µg/µ

AOST1 and rRNA at different concentrations. Negative control and positive controls contain

NA. Incubation for 4 mwater and Test-DNA, respectively, instead of rRnutes at 94°C of thei

xture was carried out to denature the target nucleic acid. Subsequently 2 µL ihybridization m

ofthe mixture was applied to the working electrode and the sensor was incubated for 30

minutes at 46°C. The hybridization was accomplished in a moisture chamber to avoid

evaporation. Afterwards, the sensor chips were washed with POP buffer (50mM NaH2PO4 ×

evaporation. Afterwards, the sensor chips were washed with POP buffer (50mH NaM2

M NaCl).H2O, pH 7.6/100m

Detection - The sensor chip was incubated for 30 minutes at room temperature with an

digoxigenin coupled to horseradish-peroxidaseplex directed against the comeantibody-enzym

l in PBS, pH 7.6/0.1% e solution (7.5 U/mOD). 1.5 µL of the antibody-enzym(Anti-DIG-P

se wa was added onto the electrode. Excessive enzymBSA [w/v]/0.05% Tween 20 [v/v])

removed by washing the sensor with POP buffer; subsequently the sensor chip was inserted

into the PalmSens (Palmnts BV, Houten, Netherlands), 20 µL of the substrate e Instrum

solution (4-aminophenylamine hydrochloride [44 µg/ml]/0.44% ethanol [v/v]/0.048% H2O2

[v/v]/50mM NaH2PO4 × H2O/100mM NaCl) was applied to the working electrode and an

easured for 10 seconds at a potential ted that was directly mical signal was generaelectrochem

ration.s of equilibersus Ag/AgCl) after 8 secondillivolt (vof -147 m

78

Publication IV

ents were carried at four ST probe experimO The LNA probe and the AExperimental setup -

.perature), 55°C, 60 °C and 65°Cal hybridization temdifferent temperatures: 46 °C (norm

Each LNA probe and the AOST2 probe were tested using the rRNA of the target and non-

ed three replicates for tain A hybridization experiment contarget species at each temperature.

Unclear results were repeated totive control.detection of target RNA, and a negative and posi

ta. The maverify the dean value of the signals was calculated and the standard derivation was

ined with the following formula:determ

Microarray

nx¦¦22()x

nn1

n

ve probes evaluated in this publication target the 18S-The fiProbe set DNA microarray -

and one for each of these four m of Eukarya arRNA: one for the super kingdomjor phyla of

yptophyta. The probe lengths of rnesiophyta and Calgae: the Chlorophyta, Bolidophyta, Prym

16-20 base pairs (Table 3). Euk1209, Chlo 02, Boli 02, varied fromthe conventional probes

qon with two different locked nucleic acid 02 and Crypto B were processed by ExiPrym

modifications, LNA2 or LNA3 varying in the number of LNAs/probe and the length and in

the methylation of Cytosine. The positive control was not modified with LNAs.

Algal strains and templates -The tests of the LNA probes using the microarray-format were

carried out with PCR-fragments amplified from two uncultured, environmental clones and

two algal strains (Table 3) as target strains. Four strains from the genus Alexandrium (A.

catenellaBAHME217,A. ostenfeldiiBAHME 136, A. ostenfeldii AOSH1 and A. minutum

Nantes) were used as non-target strains.

Microarray production inolink at the 5'-croarray had a C6/MMT ami- The probes for the m

lide A” slides (Peqlab end of the molecule and were spotted onto epoxy-coated “Nexterion S

any). The oligonucleotides were diluted to a final GermBiotechnologie GMBH, Erlangen,

concentration of 1µM in 3x saline sodium citrate buffer and printed onto the slides with the

any)München, GermbH,riter Pro (Bio-Rad Laboratories Gmpin printer VersArray ChipW

rado, USA). The probes and split pins (Point Technologies, Inc., Coloobilized on were imm

n. at 60°C and stored at -20°C. ithe slides with a baking procedure of 30 m

79

Publication IV

- The hybridization solution contained a hybridization bufferStandard hybridization protocol

(1M NaCl/10 mM Tris, pH 8/ 0,005% Triton X-100/ 1 mg/ml BSA/ 0.1 µg/µL HS-DNA), the

entration of 11.25ng DNA per µL and the positive ent in a final concbiotin-labeled PCR-fragm

control, the 250 bp PCR-fragment TBP fromS. cerevisiae with biotin-labeled primers in a

7ng DNA per µL. First, 1 hour pre-hybridization was carried out at final concentration of 4.

n at 94°C and for i58C with 2xSTT buffer. The hybridization solution was denatured for 5 m

even dispersal of hybridization solution between the chip and the coverslip, a volume of 30

µL was injected under a Lifter Slip cover slip (Implen, München, Germany). The slides were

ber with the hybridization id chamhybridized as follows: 1 hour hybridization in a hum

rwards with 2x and 1x saline perature of 58°C, washing aftesolution at a hybridization tem

sodium citrate (2 × SSC/10 mM EDTA/0.05% SDS; 1 × SSC/10 mM EDTA) for 15 min

each. In all microarray hybridization experiments,the chip contained four replicates of each

with the perfectly seprobe in four individual arrays. These hybridizations were done four tim

ions were repeated twice. et hybridizations, the hybridizattched targets. For the non-targam

- The bound PCR-Staining stained subsequently with Streptavidin-CY5ents werefragm

bridization buffer at a final concentration of any) in hy Biosciences, Stadt, Germersham(Am

100 ng /ml. The staining took place for 30 min.at room temperature in a humid chamber.

Excess staining moieties were removed by washing twice with 2x saline sodium citrate for 5

min. and once with 1x saline sodium citrate for 5 min.

Scanning and quantification of Microarrays - The fluorescent signals of the microarrays were

anner (Molecular Devices Cooperation, Sunnyvale USA)scanned with a GenePix 4000B sc

and the obtained signal intensities were analyzed with the GenePix 6.0 software (Molecular

ulamDevices Coperation, Sunnyvale USA). The signal to noise-ratio was calculated with a for

according to (Loy et al. 2002) and all ratios were normalized on the signal of the TBP positive

ove.ise-ratios was calculated as abean value of the signal-to-nocontrol. The m

Assessment

rRNA Biosensor - The PalmSens was adapted for the biosensors using a control chip with a

etric detection technique pere (nA). In this study, an amperomfixed resistance of 2682 nanoam

easuremwas used with m seconds. At the recommendation of Palment duration of 10

Instruments, the time equilibration of 8 seconds was programmed into measuring method,

80

Publication IV

ent duration of 18 seconds, 8 seconds longer than with the easuremeans a total mwhich m

Inventus Biotec Gm005). The redox-reaction goes to bH potentiostat used by Metfies et al (2

completion and then signals decrease over the measurement time because of the limited

is lower after 18 ssubstrate amount. Consequently, the signal intensityeconds than after 10

easured by Metfies et al (2005), all the signals presented seconds. Compared to the signals m

an those in Metfies et al. (2005). in this study are about 600 nA lower for the AOST probes th

The hybridization temperature for both Alexandrium ostenfeldii probes was optimized in the

of AOST low the calculated Tmpresent assay to 46°C (Figure 1A). This is around 20 °C be

l 25°C below its theoreticaprobes. Hybridization reactions can be carried out at a Tm

calculation because the rate of DNA annealing is maximal at 20-25°C below its melting

temperature. Hybrids formed from completely homologous nucleic acids will be thermally

ed9). However, if hybridizations are performstable under these conditions (Howley et al. 197

at temperatures significantly below the theoretical Tm, the probes could crosshybridize to

Alexandrium ostenfeldii non-target nucleic acids. The AOST probes gave a signal for of 680

A. at 605 nA. However, A. minutumnA and also showed high cross hybridization signals for

tamutum, having only one mismatch to AOST2 was not detected by the AOST probes, thus it

is possible to discriminate target from non-target with a single base pair mismatch. All three

Aals at 46°C for the different species (Figure 1nLNA probes showed almost no sig). Only

LNA 66 showed a weak signal for A. ostenfeldii. Also the positive control signals were about

twofold lower for LNA 65 and about 2.7 x lower for LNA 66 and LNA 67 than for the AOST

perature for the LNA al hybridization temprobes, which can be explained by the suboptim

probes and their melting temperature. It seems that LNA probes do not have the same

thod.ehybridization properties as conventional probes in this m

Metfies et al 2005 showed that a temperature of 55°C results in higher hybridization signals

but at this temthan at 46°Ces were non-specific (Figure 1B). Only LNA 67 perature, all prob

perature of the signals at a hybridization temilar togave very low signals for all species sim

of about 65 °C; LNA 67 has46°C. Probes AOST1/AOST2, LNA 65 and LNA 66 have a Tm

a Tm of 60°C. A hybridization temperature of 55°C should be the optimum temperature for

the first three probes. We maintained uniformtemperatures and salt concentrations in the

washing buffers in order to compare the performance of the LNAs against optimal conditions

on temperature of 60°C (Figure 1C) the AOST for the unmodified probes. At hybridizati

probes were specific for A. ostenfeldii and showed no signals for the other species, but the

81

Publication IV

signal intensity was lower than at 46 °C. All three LNA probes detected A. ostenfeldii and A.

minutum. The AOST probes detected all three species at a hybridization temperature of 65°C

(Figure 1D), but the signals for A. ostenfeldii and A. minutum were quite low and there was a

high signal forA. tamutumsimilar to the signals obtained at 55°C. LNA probe 65 was specific

. This was the only specific signal that we detected. A. ostenfeldiiat 65°C and detected only

but high A. minutum and A. ostenfeldiiLNA probes 66 and 67 showed only low signals for

. The properties of the LNA probes should enhance the signal intensity A. tamutumsignals for

atches but we obtained exactly the opposite smiinate the mat higher temperatures and discrim

A.results. All three LNA probes show non-specific signals at 46°C, 55°C and 60°C for

ostenfeldii.

stability (data not also tested for long termreeFor the use on an rRNA biosensor the probes w

ents with LNA shown). Probes without LNAs are stable over a year. During the experim

probes were unstable after probes on the biosensors, it was observed that the LNA

im a few weeks of storage.obilization after onlym

Microarraycroarrayi - For this hybridization study, previously published and mProbe development/design

ic levels, so it is challenging to design tested probes were used. They all target higher taxonom

probes to achieve better specificity and sensitivity that can recognize all taxa belonging to the

not show sufficient moderately well but dotarget group. The selected probes are working

sensitivity for use in routine applications and monitoring of phytoplankton because cell counts

in field samples are often not high and taxonomic groups with low abundance cannot be

detectable.

Validation of results in the hybridization protocol - The results of the microarray

ents indicated that both LNA probes showed with specific PCR-fragmhybridization (Figure 2)

e best, except for CryptoB, the probe for the ed thsity. LNA2 performincreased signal inten

approx. froment variedCryptophyceae, where LNA3 had the highest result. Signal enhancem

old higher signals ind Bolidophyceae to 8.5-flts in the Cryptophyceae an4.5-fold higher resu

hyceae.the Chlorop

Validation of results using non- target hybridizations - lts,e resuparison to the abovIn com

probes with non-target odifiedsignals of the hybridization of the conventional and LNA-m

82

Publication IV

ed specifically with e conventional probes workonstrated that thalgae species (Figure 3) dem

weak cross hybridization with non-related species. All probes, both conventional and LNAs,

showed positive enhanced signals with the Eukaryotic probe, as they should but there was no

not presented. All LNA probes showed cross e enhancement and these data are pattern to th

Alexandriumhybridization signals with non-target DNA. Hybridizations with 27 other strains

e tendency (data not shown). all showed the sam

inative potential of the LNAsperature to enhance the discrimIncrease of hybridization tem

was already tested with the biosensor and the LNA modified probes did not perform as

croarray protocol was not modified any further. ihus the mconventional oligonucleotides. T

andard protocols are with target DNA using st the hybridizationsEven though the results from

ising with increase in signal-to-noise-ratios, inprom the hybridization with non target DNA,

ty. For further clarification, the an unacceptable lack of pecificithe LNA probes show

mismatches of the probes to the sequences of the four Alexandrium strains are shown in Table

4. The differences span from 2 to 9 base pairs. Theoretically, it is impossible for these DNAs

to bind to these probes.

Discussion

In this study, we tested and evaluated the use of LNA probes in two solid-phase hybridization

methods. Although there have been many publications on enhancement of probe or

ohybridization signals with LNA mdified probes, there has been no rigorous testing of these

s showed no signal found that LNA probeeprobes using known target sequences. W

osensor. Only one of thod using the rRNA bieent in the sandwich hybridization menhancem

the tested Lperature of 65°C. Using NA probes showed specific signals at a hybridization tem

eals than thcroarray, the LNA probes could enhance the sensitivity and gave higher signithe m

ith non-conventional probes using only target DNA but unfortunately, unspecific binding w

etarget DNA also was enhanced. These results were surprising because in other mthods the

to enhance the signals and to d an abilityfied probes show great potential anodiLNA m

thod (Silahtaroglu et al. 2004; eprove specificity, accuracy and sensitivity in the whole mim

Wan et al. 2006; Kubota et al. 2006; Sun et al. 2006). Results enholds et al. 2005; Kloostermi

thods using LNA probes cannot be easily compared to the results presented in e other mfrom

this study, because of the different experimental setups, such as in situ hybridizations in

ents,rimhybridization (FISH), in situ hybridization). In FISH expece in situtissues (fluorescen

83

Publication IV

the LNA probes using human-specific repetitive elements were very efficient (Silahtaroglu et

al. 2003; Silahtaroglu et al. 2004). To evaluate the potential possibilities and abilities of LNA

probes, more experiments with more methods are necessary. A comprehensive and ultimate

evaluation of the potential of LNA probes cannot be conducted here because only a small

subset of probes were tested in two different solid phase based hybridization techniques with

n protocols. It is likely that the increased signals seen in the use of our standard hybridizatio

these studies result from non-sepcific binding which cannot be documented because the target

odifiedrd protocols developed for our unmand non-target sequences are unknown. The standa

rriate foecific hybridization temperatures are appropprobes on multiprobe chips at sp

monitoring of phytoplankton. By choosing other salt concentrations in combination with other

different. Further peratures, the signals of the LNA probes could behybridization tem

optimization experiments are only appropriate for the use of only one LNA probe at a time,

because different LNA probes can have different hybridization temperature optima.

ple,s occurred using LNA probes. For examAdditionally to unspecific binding, other problem

gae are prepthe biosensors for the monitoring of the toxic alared in advance of application.

ve to be stable and need to give the sameabes on the biosensors he proBecause of this, th

ith the LNA probes, this application was not nths of storage. Wosignals after several m

possible.

Signal enhancement of both methods, biosensors and microarrays, has been achieved

by changing substrate concentration for the biosensor and by reducing the background noise

croarrays, signal enhancemiwith the help of other blocking solutions. In the case of the ment

can be accomplished by using labelling kits that incorporate multiple labels to a target.

ledgmentsAcknow

The authors would like to thank Annick Sawala (University of Durham, United Kingdom) for

obes were designed and provided e LNA prhents. Ther assistance in the hybridization experim

ark and paid by the EU projects by Exiqon A/S, Bygstubben 9, 2950 Vedbaek, Denm

ALGADEC and FISH & CHIPS. Christine Gescher and Sonja Diercks were supported by the

2003-505491) and ALGADEC (COOP-EU-projects FISH&CHIPS (GOCE-CT-CT-2004-

of the European Union and the Alfredork programwe508435-ALGADEC) of the 6th fram

te for Polar and Marine Research.gener InstitueW

84

Publication IV

References

Al-Shahrour, F., R. Diaz-Uriarte, and J. Dopazo. 2005. Discovering molecular functions

significantly related to phenotypes by coming gene expression data and biological bin

information. Bioinformatics21: 2988-2993.

NA): fine-tuning the Corey. 2001. Locked nucleic acid (LBraasch, D. A., and D. R.

1-7. 8:istry & Biology recognition of DNA and RNA. Chem

Miller. 2006. Identifying gene Broet, P., V. A. Kuznetsov, J. Bergh, E. T. Liu, and L. D.

ongexpression changes in breast cancer that distinguish early and late relapse am

aticsormuncured patients. Bioinf 1477-1485. 22:

own. 1997. Exploring the Metabolic and Genetic Control BrDerisi, J. L., V. R. Iyer, and P. O.

680-686. 278:ic Scale. Science of Gene Expression on a Genom

Eppley, R. W., R. W. Holmes, and J. D. H. Strickland. 1967. Sinking rates of marine

phytoplankton measured with a fluorometer. J. Exp. Mar. Biol. Ecol 1: 191-208.

Gagnon, S., D. Bourbeau, and R. C. Levesque. 1996. Secondary structures and features of the

aplexan parasite Toxoplasmal RNAs from the Apicomnd 26S ribosom18S, 5.8S a

129-135. 173:gondii. Gene

Gamberoni, G., S. Storari, and S. Volinia. 2006. Finding biological process modifications in

cancer tissues by mining gene expression correlations. BMC Bioinformatics7: 6.

17:e of age. Trends Biotechnol Graves, D. J. 1999. Powerful tools for genetic analysis com

127-134.

Medlin. 2004. Using fluorescently-and L. K.Groben, R., U. John, G. Eller, M. Lange,

labelled rRni-iation of phytoplankton diversity – a mNA probes for hierarchical estim

313-320. 79:review. Nova Hedwigia

and ,RGuillou, L., M.-J. Chrétiennot-Dinet, L. K. Medlin, H. Claustre, S. Loiseaux-De Goe¨

onas: A New Genus with two species belonging to a new D. Vaulot. 1999a. Bolidom

368–381. 35:hyceae (Heterokonta). J. Phycol algal class, The Bolidop

Guillou, L., S.-Y. Moon-Van Der Staay, H. Claustre, F. Partensky, and D. Vaulot. 1999b.

Diversity and Abundance of Bolidophyceae (Heterokonta) in Two Oceanic Regions.

4528-4536. 65:Appl. Environ. Microbiol.

thod for detecting ed mrael, M. Law, and M. A. Martin. 1979. A rapiHowley, P. M., M. A. Is

and mapping homology between heterologous DNAs. J. Biol. Chem.254: 4876-4883.

.mNAs. J. Biol. Cheology between heterologous Dpping homaand m

85

Publication IV

elshoj, L., L. P. Ryder, H. O. Madsen, and L. K. Poulsen. 2005. Locked nucleic acid Humminhibits amplification of contaminating DNA in real-time PCR. BioTechniques38:
605-610.Jacobsen, N. and others 2002a. LNA-enhanced detection of single nucleotide polymorphisms
e100. 30:oprotein E. Nucleic Acids Research lipin the apo---. 2002b. Genotyping of the Apolipoprotein B R3500Q Mutation Using Immobilized
657-660. 48:Locked Nucleic Acid Capture Probes. Clin ChemJi, L., and K.-L. Tan. 2004. Mining gene expression data for positive and negative co- 2711-2718. 20:icsatregulated gene clusters. Bioinform. 2003. Microarray-based identification of sternhagen, and D. H. BlohmeKappel, K., H. V. Wp-a Chieggs and larvae from fish species common in the North Sea, Dechemany.Technology Meeting, Frankfurt, Germ 24-32. :icsa GenomKauppinen, S. and others 2003. LNA Microarrays in Genomics. Pharmre of Claus, and R. R. L. Guillard. 1987. Media for the cultu.Keller, M. D., R. C. Selvin, Woceanic ultraphytoplankton. J. Phycol 23: 633-338.
ly study for parallen. 2006. A low-density oligonucleotide arraaKi, J.-S., and M.-S. Hhybridization of consensus PCR products of detection of harmful algal species using LSU rDNA D2 domain.Biosens. Bioelectron 21: 1812-1821.
Kim, C.-J., and Y. Sako. 2005. Molecular identification of toxic Alexandrium tamiyavanichii
984-991. 4:(Dinophyceae) using two DNA probes. Harmful Algae Kloosterman, W. P., E. Wienholds, E. De Bruijn, S. Kauppinen,and R. H. A. Plasterk. 2006.
In situ detection of miRNAs in animalembryos using LNA-modified oligonucleotide
27-29. 3:probes. Nature Methods Kongsbak, L. 2002. LNA: Fine-tuning of primers and probes. LNA 01: 1.
ngel. 1998a. Novel convenient syntheses of LNA eKoshkin, A. A., V. K. Rajwanshi, and J. W 4381-4384. 39:edron Letters[2.2.1]bicyclo nucleosides. TetrahKoshkin, A. A. and others 1998b. LNA (Locked Nucleic Acids): Synthesis of the adenine, cytosine, guanine, 5-methylcytosine, thymine and uracil bicyclonucleoside monomers,
oligomerisation, and unprecedented nucleic acid recognition. Tetrahedron54: 3607-
3630.proved In Situ Hybridization achi, and H. Harada. 2006. ImKubota, K., A. Ohashi, H. Im-Incorporated DNA Probes. Appl. Environ.idcEfficiency with Locked-Nucleic-A 5311-5317. 72:iol.Microb

86

Publication IV

croarray for identification of Enterococcus iLehner, A. and others 2005. Oligonucleotide m 133-142. 246:iol. Lett. MS Microbspecies. FEaral, D. A. Caron, and E. F. Delong. 1993. Application of rRNA-based , E. L., L. A. AmLim59:rine nanoplanktonic protists. Appl. Environ. Microbiol. aprobes for observing m1647-1655.Lipshutz, R. J., S. P. Fodor, T. R. Gingeras,and D. J. Lockhart. 1999. High density synthetic
20 - 24. 21:oligonucleotide arrays. Nature Genet.onitoring by hybridization to high-density Lockhart, D. J. and others 1996. Expression m 1675-1680. 14:oligonucleotide arrays. tection of ene-Based DLoy, A. and others 2002. Oligonucleotide Microarray for 16S rRNA GeAll Recognized Lineages of Sulfate-Reducing Prokaryotes in the Environment. Appl. 5064-5081. 68:Environ. Microbiol. ---. 2005. 16S rRNA Gene-Based OligonucleotideMicroarray for Environmental Monitoring
71:crobiol.iof the Betaproteobacterial Order "Rhodocyclales". Appl. Environ. M1373-1386.Medlin, L. K., H. J. Elwood, S. Stickel, and M. L. Sogin. 1988. The characterization of 491-499. 71:plified eukaryotic 16S-like rRNA-coding regions. Gene atically amenzym Medlin. 2006. Meeting Report: .o, C. Gualerzi, G. Muyzer, and L. KMetfies, K., M. BerzanMolecular Ecology Workshop. Detection of Microbial Biodiversity in Environmental
Samples, Camerino, Italy, September 19-21, 2005. Protist 157: 247-250.
ical detection of the dlin. 2005. ElectrochemeMetfies, K., S. Huljic, M. Lange, and L. K. Mth a DNA-biosensor. Biosens.exandrium ostenfeldii witoxic dinoflagellate Al 1349-1357. 20:Bioelectronkton: The fluorescent microchips for phytoplanMetfies, K., and L. K. Medlin. 2005a. DNA 321—327. 79:wave of the future. Nova Hedwigia gtial Use in Assessinal RNA Probes and Microarrays: Their Poten---. 2005b. Ribosom.], H. R. Elizabeth A. Zimmer [edA. E.InMicrobial Biodiversity, p. 258-278. ical Data. ology,Molecular Evolution: Producing the BiochemMethods in Enzymic Press. AcademMontresor, M., U. John, A. Beran, and L. K. Medlin. 2004. Alexandrium tamutum sp. nov.
398-40:. J. Phycol (Dinophyceae): A new nontoxic species in the genus Alexandrium411.croarraysie use of LNA Oligonucleotide M Kauppinen. 2002. ThNielsen, P. S., and S. 1-3. 17:ANProvides Superior Sensitivity and Specificity in Expression Profiling. L

87

Publication IV

ie International 1999. DNA Microarrays. Angewandte Chem.eyer, C. M., and D. BlohmNiem

2865-2869. 38:Edition

ctural features of the duplexes containing Obika, S. and others 1998. Stability and stru

N-type conformation, 2'-O,4'-C-nucleoside analogues with a fixed

5401-5404. 39:thyleneribonucleosides. Tetrahedron Letters em

ec. 2002. Targeted Nucleotide Exchange in iParekh-Olmedo, H., M. Drury, and E. B. Kmtides Containing Locked by Short Oligonucleoisiae Directedyces cerevSaccharom 1073-1084. 9:istry & Biology mNucleic Acids. Che

ization Strategies for DNA Peplies, J., F. O. Glockner, and R. Amann. 2003. Optim

Targeting OligonucleotideNA-Microarray-Based Detection of Bacteria with 16S rR

1397-1407. 69:Probes. Appl. Environ. Microbiol.

Peplies, J., F. O. Glockner, R. Amann, and W. Ludwig. 2004a. Comparative Sequence

Analysis and Oligonucleotide Probe Design Based on 23S rRNA Genes of

- 57327:biolAlphaproteobacteria from North Sea Bacterioplankton. Syst. Appl. Micro

580.

Microarray Platform Manz. 2006. A DNA .und, F. O. Glockner, and WPeplies, J., C. Lachm

ent-of Freshwater SedimBased on Direct Detection of rRNA for Characterization

iol.karyotic Communities. Appl. Environ. MicrobRelated Pro 4829-4838. 72:

Peplies, J., S. C. K. Lau, J. Pernthaler, R. Amann, and F. O. Glockner. 2004b. Application and

validation of DNA mrineacroarrays for the 16S rRNA-based analysis of mi

638-645. 6:bacterioplankton. Environ. Microbiol

R. Buck, T. Britschgi, G. Cangelosi, and F. P. Chavez. 1996. Identification Scholin, C. A., K.

of Pseudo-nitzschia australis (Bacillariophyceae) using rRNA-targeted probes in

whole cell and sandwich hybridization for 190-197. 35:ats. Phycologia m

Scholin, C. A., M. Herzog, M. Sogin, and D. M. Anderson. 1994. Identification of group- and

strain-specific genetic markers for the globally distributed Alexandrium

(Dinophyceae). 2. Sequence analysis of a fragment of the LSU rRNA gene. J. Phycol

999-1011. 30:

antification of Pseudo-nitzschia australis in Scholin, C. A. and others 1997. Detection and qu

nol. Oceanogr NA rRNA-targeted probes. Limcultured and natural populations using L

1265-1272. 42:

Silahtaroglu, A. N., H. M. Pfundheuppinen. Tommerup, and S. Ka Koshkin, N.ller, A. A.

odified oligonucleotides are highly efficient as FISH probes. m2004. LNA-

32-37. 107:e Research Cytogenetic and Genom

88

Publication IV

sSHing with locked nucleic aciderup, and H. Vissing. 2003. FI TommSilahtaroglu, A. N., N.

(LNA): evaluation of different LNA/DNA mixmers. Mol. Cell. Probe17: 165-169.

on, N. and others 2000. Oligonucleotide probes for the identification of three algal groups Sim

47: whole-cell hybridization. J. Eukaryot. Microbiol by dot blot and fluorescent 76-84.

ngel, and P. Nielsen. 1998. LNA (locked nucleic acids): ehkin, J. W, A. A. KosSingh, S. K.

synthesis and high-affinity nucleic acid recognition. Chem. Commun.: 455 - 456.

crobial ecology. icular tools and approaches in eukaryotic mit, E. and others 2004. MoleSm

Bios Scientific Publishers.

Sun, Z., L. Zhou, H. Zeng, Z. Chen, and H. Zhu. 2006. Multiplex locked nucleic acid probes

for analysis of hepatitis B virus mutants using real-time PCR. GenomicsIn Press,

.Corrected Proof

ashiwoka aTyrrell, J. V., L. B. Connell, and C. A. Scholin. 2002. Monitoring for Heterosigm

tion assay. Harmful Alusing a sandwich hybridiza 205-214. 1:gae

e genotyping Ugozzoli, L. A., D. Latorra, R. Pucket, K. Arar, and K. Hamby. 2004. Real-tim

istrycleic acids. Analytical Biochemuwith oligonucleotide probes containing locked n

143-152. 324:

seVester, B. and others 2006. Locked nucleoside analogues expand the potential of DNAzym

19. 7:C Molecular Biology Mto cleave structured RNA targets. B

rensen, B. R. Babu, S. Douthwaite, and J. WøVester, B., L. B. Lundberg, M. D. Sngel. 2004. e

Improved RNA cleavage by LNAzyme derivatives of DNAzymes. Biochem. Soc.

37-40. 32:Trans.

ffinity Targeting ofngel. 2004. LNA (Locked Nucleic Acid): High-AeVester, B., and J. W

Complementary RNA and DNA. Biochemistry43: 13233-13241.

Wienholds, E. and others 2005. MicroRNA Expression in Zebrafish Embryonic Development.

310-311. 309:Science

Yap, Y., X. Zhang, M. Ling, X. Wang, Y. Wong, and A. Danchin. 2004. Classification

al and tumor tissues based on thbetween norme pair-wise gene expression ratio. BMC

72.4:Cancer

Ye, R. W., T. Wang, L. Bedzyk, and K. M. Croker. 2001. Applications of DNA microarrays

in microbial systems. J. Microbiol. Meth 47: 257-272.

iol. Meth s. J. Microbcrobial systemiin m

89

Publication IV

Algae culturesTable 1.

SpeciesAlexandrium minutum
Alexandrium minutum
mutumaAlexandrium tAlexandrium ostenfeldiiAlexandrium ostenfeldiiAlexandrium ostenfeldii.fPrymnesium parvumpatelliferumRhinomonas reticulateAlexandrium catenella

inaStr3TALNantes029SZNB 1 SHAOCCMP 1773BAH ME613527PLY358PLYBAH ME721

eCulturmediumKKKKKK

KMRIIMR

Temperature5 °C 115 °C 15 °C5 °C 15 °C15 °C1

15 °C °C155 °C1

90

OriginGulfofTrieste, Italy,A. Beran
n, France Atlantic OceaGulShipfoHfarNabourples,, Nova Scotia, Italy, M. MoCntresoranada, A.
ellabmCeLimfjordan, Denmark,Hansen
BiologischeAnstalt Helgoland,Germany

Plymouth Culture Collection, UK
Plymouth Culture Collection, UK
BiologischeAnstalt Helgoland,Germany

Publication IV

Table 2. the probes, the helper oligonucleotide H3, p Sequences ofcontrol for the biosensor

name ProbeSignal probe: AOST1

2 AOST probe:Capture

Capture probe: LNA65

Capture probe: LNA66

Capture probe: LNA67

Helper oligonucleotide:
H3Test DNA (positiverolcont

CAAProbe sequeCCC TTC nceCCAATA GTC AGG T

GAA TCA CCAAGG TTCCAAGCAG

AAT CAC CAAGGT TCAA

AGG TTC CAAGCAG

CCA AGG TTC CAAG

GCATAT GAC TAC TGG CAG GAT C

GAA CCT TGG TGA TTC CTGC TTG ACCT GACTAT TGGGAA GGG TTG

91

tTargeAlexandriumostenfeldiiCCMP 1773AlexandriumostenfeldiiCCMP 1773AlexandriumostenfeldiiCCMP 1773AlexandriumostenfeldiiCCMP 1773AlexandriumostenfeldiiCCMP 1773AlexandriumostenfeldiiCCMP 1773

e and negativsitivo

Source(Metfies et al. 0052)(Metfies et al. )0052ExiqonExiqonExiqon(Metfies et al. )0052(Metfies et al. )0052

e

Publication IV

croarrayi Probe Sequences for the mTable 3.

Source et al. 1993)(Lim(Sim(Guilloonu et al. et al. 2000)1999a)

Probe name Probe sequenceTargetSource
Euk1209GGGCATCACAGACCTGAll Eukaryotes 18S(Lim et al. 1993)
Chlo02CTTCGAGCCCCCAACTTTChlorophytceae HE001005.53*(Simon et al. 2000)
Boli 02 TACCTAGGTACGCAAACCBolidophyceaeHE001005.51*(Guillou et al. 1999a)
f.mrvuesium paPrymnPrym 02GGAATACGAGTGCCCCTGACpatelliferumPLY 527**(Simon et al. 2000)
Crypto B ACGGCCCCAACTGTCCCTRhinomonas reticulataPLY 358**Medlin,unpublished

and Medlin 2005a)(Metfies

Prym 02GGAATACGAGTGCCCCTGACpatelliferumPLY 527**(Simon et al. 2000)
Crypto B ACGGCCCCAACTGTCCCTRhinomonas reticulataPLY 358**Medlin,unpublished
Positivecontrol (PC) ATGGCCGATGAGGAACGTS. cerevisiae, TBP (Metfiesand Medlin 2005a)
eativNegcontrol (NC)TCCCCCGGGTATGGCCGC(Metfies andMedlin2005a)
*Environmental clone from EU FP5- Project PICODIV, ** Plymouth Culture Collection, UK

92

Publication IV

Table 4.Chlo 02

li 02Bo

Prym 02

Mismatches of the probes to the Alexandrium strains in base pairs (bp)

Crypto B

HME217A. catenella BA

3 bp

b9p

5 bp

3 bp

E136MHA .ostenfeldii BA

2 bp

8 bp

5 bp

3 bp

93

1A.ostenfeldii AOSH

2 bp

8 bp

5 bp

3 bp

i. mA Nantes nutum

2 bp

5 bp

5 bp

3 bp

Publication IV

1800AC °46160014001200A)n (algnSi
10008006004002000TSOA

5 6ALN

6 6ALN

6ALN7

1800CC °60160014001200A)n (algnSi
10008006004002000AOSTLNA 65LNA 66LNA 67
A. ostenfeldiiA. minutumA. tamutumpositive control

B180055 °C
01600140012001000800600400200TSAO

D180065 °C
01600140012001000800600400200TSAO

A 6NL5

5A 6NL

6A 6NL

NL6A 6

7A 6NL

7A 6NL

Signal intensity of the rRNA-biosensor. Four different probes were tested at fourFigure 1. peratures and with three different species. (A) 46°C, (B) 55°C, (C) different hybridization temNA for all tested species was 450 ng/µL. The60°C, (D) 65°C. The concentration of the rRrks the only specific LNA probe.aasterisk m

94

Publication IV

018

016

014

012

010esoiN/nalgSioiatR-
80

60

40

20

0

conventional Probe

Euk1209 hybridized with all targets

CryptoB hybridized withR.reticulata

2ANL

Chlo02 hybridized with the Chlorophyte clone

Prym02 hybridized with P. parvum f. patelliferum

NL3A

Boli02 hybridized with the Bolidophyte clone

hybridization with gnal/Noise-Ratios of all fifteen probes in comparison fromi SFigure 2.

specific PCR-fragments for each setof probes. The black line represents the value of 2 for the

e threshold for a true signal. signal-to-noise ratio, defining th

95

Publication IV

2li0oB

Boli02_LNA2

Boli02_LNA3

lhCo02

Chlo02_LNA2

Chlo02_LNA3

AB5804,5704603,55032,540iSoiatR-esoiN/alng1,520
2301100,500Boli02Boli02_LNA2Boli02_LNA3Chlo02Chlo02_LNA2Chlo02_LNA3
140D60C12050100408030oiatR-esoiN/algniS
602040102000CryptoBCryptoB_LNA2CryptoB_LNA3Prym02Prym02_LNA2Prym02_LNA3
A. catenellaBAHME217A. ostenfeldii BAHME136A. ostenfeldii AOSH1A. minutum NANTES

Figure 3. (A) Signal/Noise-Ratios of the set ofthree Boli02 probes in comparison from
(B) Signal/Noise-Alexandrium. the genus ents fromhybridization with unspecific PCR-fragmparison with hybridization with unspecific lo02 probes in comRatios of the set of three ChPCR-fragments from the genus Alexandrium.(C) Signal/Noise-Ratios of the set of three
theents fromCryptoB probes in comparison with hybridization with unspecific PCR-fragmgenusAlexandrium.(D) Signal/Noise-Ratios of the set of three Prym02 probes in comparison
with hybridization with unspecific PCR-fragments from the genus Alexandrium.The black
al-to-noise ratio, defining the threshold for a true e of 2 for the signresents the valuline repsignal.

96

Publication V

Publication2.7 V

DEVELOPMENT ANDOPTIMIZATION OF A SEMI AUTOMATEDRRNA BIOSENSOR

FOR THE DETECTION ALGAEOF TOXIC

SONJADIERCKS1,KATJAMETFIES1,STEFFIJÄCKEL2 AND LINDAK.MEDLIN1
1Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, 27570

Abstract

anyerhaven, GermBrem2Hochschule Anhalt, Bernburger Str. 55, 06366 Köthen, Germany

itted submeBiosensors and Bioelectronics, to b

In order to facilitate the monitoring of toxic algae, a multiprobe chip and a semi-automated

rRNA biosensor for thein situdetection of toxic algae were developed. Different materials

terial were tested using single electrode sensors and afor the electrodes and the carrier m

sandwich hybridization that is based on species specific rRNA probes. The biosensor consists

of a multiprobe chip with an array of 16 gold electrodes for the detection of up to 14 target

species. The multiprobe chip is placed inside an automated hybridization chamber, which in

e waterproof case with reservoirs for different solutions. A turn is placed inside a portabl

peristaltic pump moves the reagents into the flow cell containing the multiprobe chip. For use

was successfully developed and mlof the device by layperson, a lysis protoconual rRNAa

nually. Theaple filtration has to be done misolation is no longer required. Only water sam

algae cultures and field was evaluated using isolated total rRNA fromstand-alone system

samples. The device processed automatically the main steps of the analysis and completed the

electrochemical detectionof toxic algae in less than two hours in comparison to other routine

least a day for analysis.thods that need atemonitoring m

97

Publication V

nIntroductio

.ic source for fishery aquaculture and tourismportant economCoastal areas are an im

Aquaculture is an increasinglportant industry world-wide as a source of food and y im

employment. Planktonic algae are critical food for shellfish and fish and thus, in most cases,

amenon and beneficial for aquaculture and ing is a natural phenomrine phytoplankton bloom

ing is regarded as a sudden increase in wild fisheries operations. Marine phytoplankton bloom

itions and cell concentrations cangrowth cond by suitablethe population and can be activatedreach up to 104-105 L-1 (Masó and Garces 2006). However, algal bloomscan also pose a

ns that can find their threat, because about 80 or even more algal species produce potent toxi

way through the food chain via shellfish (e.g. oysters, mussels)and fish to humans

includes a Alexandriumrine dinoflagellate a(Hallegraeff 2003). Among the toxic algae, the m

number of species producing saxitoxin and potent neurotoxins, which are responsible for

so certain Pseudo-nitzschia ssp. produce a paralytic shellfish poisoning (Penna 1999). Al

nesic shellfish poisoning (Scholin et al. 1999; Masó and Garces mneurotoxin, which causes a

2006). World-wide monitoring programs have been introduced to observe phytoplankton

composition. Monitoring of toxic algae by means of traditional methods, namely light-

microscopy, can be time-consuming if many samples have to be routinely analyzed. Reliable

entification requiresspecies idtrained personnel to carry out the analysis and expensive

equipment (Tyrrell et al. 2002; Ayers et al. 2005), because unicellular algae are taxonomically

challenging and some of them have only few morphological markers. Various molecular

on of phytoplankton, such as whole cell thods are used up to date for the identificatiem

fluorescentin situ hybridization (Anderson et al. 2005; Hosoi-Tanabe and Sako 2005; Kim

and Sako 2005), PCR-based assays (Penna 1999; Guillou et al. 2002) and sandwich

thod for e al. 2002; Ayers et al. 2005). A rapid and potential mhybridization assays (Tyrrell et

the detection of toxic algae was introduced by Metfies et al. (2005) using sandwich

hybridization on a biosensor and molecular probes that specifically targeted the rRNA of

toxic algae. Electrochemical biosensorscombine biochemical recognition with signal

ngle electrode sensors ilecules (Gau et al. 2005). Sotransduction for the detection of specific m

entalas well as arrays are known from various sectors like clinical diagnostic and environm

me detection of biochemonitoring. Biosensors have been applied for thcal substances as well i

o et al. 2006; Taylor et al. s like bacteria (Berganza et al. 2006; Lermcro-organismias of m

poral and spatial unities consist of different species and the tem comm2006). Phytoplankton

variability in composition in the sea is substantial. The simultaneous detection of multiple

98

Publication V

species can be accomplished using arrays of electrodes with different molecular probes. There

are examples for on-site monitoring of toxic algae, such as the environmental sampling

lecular techniques for the oprocessor (Doucette et al. 2006; Silver 2006). However, m

sportation of samples to specialisedl algae usually require tranmonitoring of harmfu

laboratories. The same applies to conventional methods. As a consequence, results are usually

e sample and therefore pobtained within five working days after receiving theeventivreasures are not always possible. m

In this regard, a system with two majorparts was developed during the EU-project

ALGADEC: a multiprobe biosensor with the aim to detect specific compositions of toxic

algae simultaneously in combination with a hand-held device for the in situ analysis. The

hybridization method involves a capture probe,immobilised on the working electrode surface

the target organism as well as a secondof a biosensor that binds to rRNA isolated from

iety. An odigoxigen-labelled probe that also binds to the rRNA but carries the signal m

e complex directed against digoxigenin is added and incubated. Aantibody-enzym redox-

reaction takes place after substrate additionand the resulting electrical current can be

stat.tioeasured with a potenm

We present here the testing of all components in the biosensor and the optimisation of the

sis of toxic algae. analyin-situprotocol for

Materials and Methods

Probe sets - One set of capture and signal 18S-rRNAprobes (AOST1 and AOST2, (Metfies et

e compare th, was used toAlexandrium ostenfeldiial. 2005), Table 1), specific for

), UK). ance of carbon sensors and gold sensors (Gwent Electronic Materials (GEMperform

AlexandriumAnother set of 18S-rRNA probes (AMINC and AMINC NEXT), specific for

ents using , was developed previously (see publication II) and used for the experimminutumdifferent lysis buffers and the adaptation of the multiprobe chip to the semi-automated device.

The probes and the positive controls were synthesized from ThermoElectron Corporation,

Ulm, Germany.

99

Publication V

Single electrode chips

Immobilization of probes on carbon sensors - The immobilization ofthe biotinylated capture

ing to a is study was done accordrs used in thprobe (AOST2) on single electrode carbon senso

istureocubation steps were carried out in a metfies et al. 2005). All inprevious protocol (Mration. The surface of the carbon working electrode was pretreatedber to avoid evapochamwith carbonate buffer (50 mM NaHCO3, pH 9.6) that was followed by an incubation with
l] (Pierce Biotechnology, Rockford, USA) for at least 4.5 hours at 4 m/gNeutrAvidin [0.5 m

°C. Subsequently, the sensor was washed with PBS (BupH phosphate saline pack, Pierce

the working electrode ove excessive NeutrAvidin. For blocking, Biotechnology, USA) to rem

erature and afterwards the pm tewas incubated with 3% [w/v] casein in PBS for 1 hour at room

sensors were washed in PBS. Prior to the application on the electrodes, the probes were
ead buffer (0.3 M NaCl/0.1M Tris, pH 7.6) to achieve a concentration of 10 µM.diluted in bFor the immobilization of the probes on the electrodes, the sensors were incubated for 30
minutes at room temperature. Unbound probe was removed from the electrode by a washing

M Tris, pH 8.0/0.04% SDS).step with hybridization buffer (75mM NaCl/20m

Immobilization of probes on gold sensors - The immobilization of thiolated probes on single

was first introduced by s done according to a modified protocol that aelectrode gold sensors w

r to the immobilization of the probes onto the Carpini et al. (2004) (Carpini et al. 2004). Prio

lated probes were dissolved at a concentration of 10 µM in 0.5 gold working electrode the thio

mol/L phosphate buffer. The gold working electrode surface was incubated with a probe for at
least 16 hours at room temperature. During all incubation steps, the sensors were stored in a

moisture chamber to protect the solutions from evaporation. In order to minimize the non-
ent with 6- and the probes, a post treatmteraction between the gold surfacespecific in

lution) was carried out for 1 hour. Excessive CH; 1mmol/L aqueous sorcapto-1-hexanol (Memprobe and MCH were removed by washing the sensor with 2x saline sodium citrate buffer.

PBS and 2% [w/v] Trehalose in - The sensors were coated withStorage of coated sensorsat 37°C. Afterwards coated sensors can be stored at 4°C.nutesiately 30 mdried for approxim

Hybridization of test DNA on single electrode sensors - The hybridization mixture for the

M Tris, pH Cl/20mM Nadetection of test-DNA contained 1x hybridization buffer (75m

DNA, 0.1 pmol/µL dig-labelled probe AOST1 S), 0.25 µg/µL herring spermD8.0/0.04% S

100

Publication V

and 0.1 pmol/µL test-DNA (positive control) as target for the probes. The negative control ied out by incubatingrget nucleic acid was carrcontains no test-DNA. Denaturation of the tathe hybridization mixtures for 4 minutes at 94°C. 2 µL ofthe mixture was applied to the
nutes at 46°C. The biosensors were iworking electrode and the sensor was incubated for 30 mmber during hybridization to prevent evaporation. Subsequently, the stored in a wet chasensors were washed with POP buffer (50mM NaH2PO4 × H2O, pH 7.6/100mM NaCl).

Electrochemical detection with single electrode sensors - An antibody-enzyme complex
(Anti-DIG-POD, 7.5 U/mldirected against the digoxigenin coupled to horseradish-peroxidase e the single electrodtoin PBS, pH 7.6/0.1% BSA [w/v]/0.05% Tween 20 [v/v]) was applied oneperature. Unbound antibody-enzymnutes at room temisensor and incubated for 30 moved by washing the sensor with POP buffer and the sensor was inserted s remplex wacominto the measurement device, PalmSens (Palm Instruments BV, Houten, Netherlands). 20 µL
of substrate solution (4-aminophenylamine hydrochloride (ADPA) [44 µg/ml]/0.44% ethanol
[v/v]/0.048% H2O2 [v/v]/50mM NaH2PO4 × H2O/100mM NaCl) was added to the working
easured for 10 seconds at a ical signal was directly melectrode and the resulting electrochementserimillivolt (versus Ag/AgCl) after 8 seconds of equilibration. All exppotential of -147 mwere carried out in triplicate, the mean value of the signals was calculated and the standard
derivation was determined with the following formula:

Multiprobe chips

nx¦¦22()x
nn1n

Spotting of multiprobe chips - Multiprobe chips were either hand-spotted or spotted with a
GEM. Hand-spotted chips were non-contact dispenser (Biodot Ltd., Chichester, UK) froml/L phosphate buffer) and ocovered with 10 µL of thiolated capture probe (10 µM in 0.5 mincubated as described above. 10 µL of MCH solution were added and incubated for one oved by washing the sensor with 2x hour, subsequently, unbound probe and MCH were remsalinand washede sodium again with 2x salin citrate buffer. The me soultipdiumrobe chip citrate buffer. Multiprobs were blocke chiped with 10 µL of 5% [w/v] BSAs were biodotted by
immobilising 0.05 µL thiolated capture probe per electrode and adding of 0.05 µL of MCH

101

Publication V

ash steps and blocking of the surface was carried out as previously after incubation. Wdescribed. The multiprobe chips were subsequently coated with 10 µL 2% [w/v] Trehalose in
ent.PBS buffer and dried for storage and shipm

xture using test-i - The hybridization memical detectionHybridization mixture and electrochtion for the multiprobe chip wereDNA (positive control), antibody solution and substrate soluxture and antibody solution ie of 10 µL hybridization mprepared as described above. A volumwas applied each time onto the chip to cover the whole electrode array. Electrochemical
detection was carried out by placing the multiprobe chip into a substrate reservoir that
easured using a ical signals were mochemharboured the substrate solution. The electrmultiplexer, which can measure eight electrodes simultaneously, and the PalmSens detector
(Palm Instruments BV,Houten, Netherlands).

Semi automated Device

Culture conditions - TheAlexandrium minutum strain AL3T was cultured under sterile
dia K (Keller et al. 1987) at 15 °C and 120 µEinstein with aeconditions in seawater-based ments, the cells were counted using the light: dark cycle of 14:10 hours. Prior to the experimbH Dia(Beckman Coulter GmMultisizer 3 Coulter Counter any).gnostics, Krefeld, Germ

Total rRNA-extraction - The RNeasy Plant Mini Kit (Qiagen, Hilden,Germany) was used to
with modifications of the protocol to Alexandrium minutumte the total RNA fromisolaenhance the quality and quantity of the RNA by removal of polysaccharides and proteins
paration of supernatant and cell debris, the seprovedent of an imcontent. For the achievemcentrifugation step oftwo minutes was extended to 15 minutes. The washing buffer RW1 was
applied twice to the RNeasy columstep with buffer RPE was repeated. RNA concentration was mn, incubated for one minute and centrifueasured with a Nanodrop ged. The first wash
.any)eter (Peqlab, Erlangen, GermSpectrophotom

Total rRNFragmentation of total rRNA from Alexandrium minutum - Alexandrium fromAminutum was fragmented in fragmentation buffer (40mM Tris, pH 8.0/100mM KOAc/30mM
nutes at 94°C prior to hybridization.iMgOAc) for 5 m

102

Publication V

Testing of different combinations of lysis buffer and hybridization buffers - Two different lysis
al lysis ination of the optimbuffers and hybridization buffers were tested for the determrobe chals on the multipproperties and hybridization signip. Lysis buffer 1 (Table 2) was the second lysis buffer RLT was taken (Kingston 1998) and prepared after Kingston (1998)from the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). In combination with the two
e 4x hybridization hlysis buffers two different hybridization buffers were tested (Table 2). Tedbuffer was described by Metfies et al. (2005) and the second hybridization buffer, namentsple buffer, was introduced by Scholin et al. (1999) (Scholin et al. 1999). The experimsamwere carried out using 400,000 cells of 600 µL of 4x hybridization buffer and sampleAlexandrium minutum and 450 µLbuffer were added to the different lysis of the lysis buffers.
solutions, respectively. Cell debris was removed by filtration through a 0.45 µm filter
entation buffer were added to mgNC NEXT and fra(Millipore, USA). Detection probe AMImuthe lysis-hyltiprobe chips with imbridization solutionmobiliseds, incubated for 5 mi capture probe AMINC. Negative and positivnutes at 94 °C and applied onto e controls
A. minutumwere prepared as described above and total rR and also hybridised for comparison of the signals.NA was isolated from the same cell counts of

xture was i - The hybridization mHybridization and analysis in semi-automated deviceprepared as described above, but the amount was amplified. Multiprobe chips consisted of an
immobilised AMIN probes on all 16 working electrodes. The adjustment of the device was
tion with. HybridizaA. minutumconducted using Test-DNA as target of the probes for different concentrations of target rRNA fromA. minutum followed instead of the target-DNA.
Final adjustments of hybridization mixture and the lysis buffer 1 were carried out using
.A. minutum500,000 cells of

Results

The signals of the electrochemical detection are measured with negative values, but for
simplification of analysis, the signals are multiplied by –1 unless otherwise noted.

Sensor design using single electrode sensors

Comparison of electrochemical signals of carbon and gold sensors - In order to determine the
for the working electrodes on the sensors, two terialamost efficient and cost effective m

103

Publication V

different materials, carbon and gold were tested to compare signal intensity and the

effectiveness of probe immobilization (AOST2). Additionally the signals were compared to

the signals shown by Metfies et al. (2005) with carbon sensors from a different manufacturer.

aterials different melectrodes withe controls detected on The achieved signals for the positiv

and sensors from different manufacturers are comparable being in the range of ~1500 nano

ampere (nA) (Figure 1). However, the signal intensity of the negative control for the different

surface materials varied highly. The carbon sensor from Metfies et al. (2005) showed the

lowest signal with 78 nA, whereas for the carbon sensor from GEM a signal of 190 nA was

obilizationore the imme gold sensor showed a very high signal of 611 nA. Therefhachieved. T

ized to reduce the background noise of the gold sensors.protocol for gold sensors was optim

n of immobilizaOptimizatio tionrotocol for gold sensors -pization of the The optim

immobilization protocol was carried out by adding a surface blocking step to the protocol

H. Two Cent with Mobilization of the probe (AOST2) and the treatmsubsequent to the imm

different blocking reagents, casein and bovine serum albumin (BSA), known from the

reduce thefor their attributes toined literature for their blocking properties were exam

background noise of the gold surface. As a control, gold electrodes with no blocking were

hybridized. The blocking with 3% casein in PBS was accomplished at room temperature for 1

the negative control to 281 nA but also reduced the signal hour, and could reduce the signal of

igure 2). Different concentrations of BSA, 3%, 5% and 10% of the positive control to 1168 (F

r, were applied to the gold sensors and incubated for 1 hour at 46°C.in 4x hybridization buffe

eal a decrease of signal of the negative control regardless which ents revAll treatm

provemconcentration of BSA is used, but 3% BSA and 5% BSA showed the strongest iment.

control of the gold sensors blocked with 5% BSA and the signals of the positive Additionally

10% BSA increased about 200 nA. In consideration of these results, the 5% BSA blocking

ents.solution was chosen for the further experim

Long term stability of sensors - Long term stability of carbon and gold sensors was tested by

bilization of the probes (AOST2) onto the ocoating the sensors with Trehalose after imm

working electrode. The sensors were stored at 4 °C and hybridised with target-DNA and the

nths. Signal intensity decreased from freshly odetection probe (AOST1) after 4, 6 and 12 m

prepared carbon sensors with 1416 nA to 798 nA for carbon sensors stored over 12 months at

4 °C (Figure 3). Also the signals for gold sensors decreased from 1711 nA to 1282 nA.

104

Publication V

Optimization of the substrate concentrations - The enhancement of signals intensity was

rent concentrations of substrate (POD) by EM) and diffeined using carbon sensors (Gexam

varying the concentration of the mediator 4-aminophenylaminehydrochloride (ADPA) and of

the hydrogen peroxide (H). Figure 4 shows that an increase of signal was achieved from O22

1530 nA of normal POD substrate to 3971 nA of 6.6 mg ADPA and 600 mM of H2O2 by
increasing concentrations of ADPA and H2O2, simultaneously. The highest signal was
obtained with 6.6 mg of ADPA and 600 mM of H2O2, however, also the signal of the negative

control increased from 38 nA to 203 nA.

robe chipt of a multipDevelopmen

- A disposable multiprobe chip was designed from iSiTECbe chipDesign of the multipro

any) with the size erhaven, GermbH (BremGmof a conventional glass slide and produced by

GEM (UK). The multiprobe chip consisted of a carrier material that contains 16 gold working

bined counter/reference electrod the size of 1.5 mm and a comelectrodes, each withee abov

the electrode array (Figure 5). Working and counter/reference electrodesare encircled with a

dielectric layer. The stems of the electrodes fit to a typical connecting strip.

Eg electrode (e.g. W - Every second workinSignal transmission between working electrodes

2, 4, 6) of a multiprobe chip with plastic carrier material was spotted by hand with thiolated

probe. Signals were detected only for the spotted working electrodes (Figure 6); non-coated

ission between the 62 nA to 129 nA. There was no signal transmelectrodes gave signals from

edelectrodes. The signals are in average 3x lower than the signals for the single electro

sensors because of the smes.diameter of the electrodaller

Selection of carrier material for multiprobe chips - Two different carrier materials (plastic

ic) were chand ceramosen for comparison of spotting properties and signal intensities.

Additionally two variations of the ceramic were tested, a plain ceramic material and ceramic

e 7 shows the signal intensity for the different carrier er. Figurwith a hydrophobic polym

materials. Plastic showed signals from 716 nA up to 1099 nA with a mean signal of 913 nA,

ean of 937 728 nA to 1324 nA with a mic with hydrophobic polymer signals fromthe ceram

nA whereas the plain ceramic showed the lowest signals from 421 nA to 1296 nA with a

mean of 602 nA. The plastic material showed a higher stability of probe drops during

spotted with a biodot because of the hydrophilic ic cannot be spotting, whereas the plain ceram

105

Publication V

er underneath a hydrophobic polymic with ic (data not shown). Ceramproperties of the ceram

probe drop electrodes shows good stability ofthe workingt shown). During the s (data no

was found to be difficult to cut into the correct size and terialaents the plastic mexperim

spotted with probes because of material plasticity.

utomated devicet of a semi-aDevelopmen

Development of lysis protocol - The current protocol using a kit for total RNA isolation

atedi-autom crucial for the use of the semplification isrequires trained personnel and sim

erties and the signdevice. Two different lysis buffers were tested for their lysis propal

formation in combination with two different hybridization buffers. For comparison of the

signals negative and positive controls as well as hybridization with target rRNA were carried

t for the different uout (Figure 8). The signals of all 16 electrodes were averaged o

experiments and compared. All experiments with lysis/hybridization buffer combinations and

total rRNA showed similar signals. 4x hybridization buffer in combination with lysis buffer 1

ean signal with 554 nA, whereas in combination with RLT buffer from achieved the highest m

the Qiagen Kit, the lowest signal (365 nA) were detected. Sample buffer in combination with

ilar signal of 518 nA to the 4x hybridization buffer/lysis buffer 1 RLT buffer showed a sim

combination. Sample buffer with lysis buffer 1 achieved a mean signal of 462 nA.

t and adDevelopmenjustment of semi-automated devicei-automated portable device, - A sem

named ALGADEC, was developed by iSiTEC GmbH (Bremerhaven, Germany) and the

Alfred Wegener Institute (Bremerhaven, Germany) during the EU-Project ALGADEC

(Figure 9). The device contains reservoirs for antibody, substrate and washing buffers as well

cell unit and an additional inlet for applying the as a flow cell unit for hybridization. The flow

peratures during the analysis procedure. d tem the requireples can be heated and cooled tosam

A peristaltic pump moves the reagents through the flow cell and finally into the waste

reservoir (Figure 10). The main steps of the analysis process can be executed automatically in

developed for the varying processes (e.g., ent device. A flow chart waseasuremthe m

hybridization, wash steps, antibody incubation and measurement) and pump times were

adapted. Adjustment of the semi-automated device was conducted using multiprobe chips

with the probe set forAlexandrium minutum and Test-DNA as target for the probes. The

ltiprobe chip was inudisposable m cell unit before analysis was started.serted into the flow

During measurement of the electrochemical reaction, the signals from the working electrodes

106

Publication V

with probes are recorded by a microcontroller unit. Process data can be visualized with
bH if a PC is connected to the system. Graphic special software programmed by iSiTEC Gmeasured values are storedresults and the mportable ALGADEC deviceon the hard disc. The ypad, display, power supply ande with a built in ke systemcan be operated as a stand-alonmemory card. A waterproofed case protects the system and allows its use under adverse
conditions.

- Hybridizations with two differentn of target RNA on multiprobe chipsHybridizatioconcentrations of target rRNA fromA. minutum; a negative and a positive control were
carried out in the semi-automated device. The measurements were started when washing
ely 150 seconds oftathe flow cell unit. After approximbuffer was still present in measurement, substrate buffer arrives in the unit and was pumpedcontinuously through it.
of the reaction was saturationRedox-reaction takes place and the signals decreased; however,observed after approximately 500 seconds. The highest signals were found for the positive
enteasuremean signal of 265 nA and for all electrodes after 500 seconds of mcontrol with a ment point, signals for the negative control (Figure 11, easureme m(Figure 11, A). At the samoncentration (Figure 11, D) werelow RNA cigure 11, C) and B), high RNA concentration (Fobserved from 104 nA, 201 nA and 106 nA, respectively.

Hybridization of dissolvedcells on multiprobe chips - 500,000 cells fromAlexandrium
xed with hybridization solution and analyzed in the i were dissolved in lysis buffer, mminutuments (Figure 12,easurem the mdevice. Both analyses display higher signals at the beginning ofectrodeslean signals of all 16 eA+B), than the experiments described above. However, the mof the analyses at 500 seconds were found to be 158 nA and 148 nA, respectively.

Discussion

rsterials for sensoa - Two mDesign of sensors and comparison of immobilization protocolswere tested and an immobilization protocol for gold sensors was developed and tested.
ld surfaces was already established (Carpini et al. 2004; es to goImmobilization of probewere adapted to the gold sensors with somthodseMannelli et al. 2005) and the described mmodifications. The signal formation of a gold or carbon covered surface was similar and
stability tests showed that the signals of carbon and parison of long termefficient. Signal comorage about 45 % and 26 %, respectively, but ral month of stgold sensors decreased over seve

107

Publication V

storage enables ed better and achieved higher signals. Long termstored gold sensors perform

with higher substrate entse of use. Experimthe production and coating of sensors in advanc

2.2 mg ced signals. A substrate concentration ofns revealed the potential of enhanconcentratio

ADPA and 200 mM of H2O2 would be sufficient for a twofold signal increase. However, the

immobilization protocols for the different sensors have advantages and disadvantages

concerning costs. One advantage of the carbon sensors is the lower price of the carbon paste

ating with expensivethe benefit that the coparison to the gold paste. Gold sensors have in com

n that thiolated DNA probes bind itted giveNeutrAvidin can be omdirectly to the surface of

col, thustoobilization steps in the pror immthe gold. Because the gold sensors required fewe

-storage, the gold produced higher values during long termnufacturing costs and areducing m

ent of a biosensor.sensors were chosen over the carbon sensors for the further developm

Development of multiprobe chip - A multiprobe chip was designed from iSiTEC GmbH with

16 gold electrodes, that can detect 16 different target species. The chip was developed with

the size of a conventional glass slide, which offers the possibility to use automated dispensing

ips are easy to handle because of their ore, the chs for the spotting of probes. Furthermsystem

size and can be stored in standard boxes. The size of the working electrodes was reduced in

comode area and consequently the e single electrode sensors to decrease the electrparison to th

ission between the electrodes was ount of reagents needed for analysis. Signal transmam

terials for the ained. Different carrier massessed and only background noise was determ

electrodes were investigated for signal formation and probe spotting properties. Plastic

e signals terial showed high signals and can be spotted with probes by hand, however, tham

from electrodes spotted with probes by hand were irregular and unacceptable. Automated

spotting with a dispenser requires accurately cut chips and the plastic material was easily

bent, making this material unsuitable. To overcome this problem, the ceramic carrier material

s. It could be accurately cut and thus be spotted with an ltiprobe chipuwas chosen for the m

automated dispenser. The experiments with the ceramic chips showed lower signals than the

plastic material because the hydrophilic surface hampered the spotting. Finally the addition of

hydrophobic polymer overcame this last problem. However, during the manufacturing of

and the ated dispenser for probe spotting occurred difficulties using the automs,these chip

multiprobe chips for our device tests presented here were hand-spotted. We anticipate that in

that all spotting difficulties encountered here rcial use emthe production of these chips for com

e.will be overcom

108

Publication V

Development of a semi-automated device - The methods described in the first part of our
s. The proposed use of our device the algal cellstudy involve the isolation of total rRNA fromby inexperienced users meant that we needed to simplify the rRNA extraction method. A lysis
NA isolation. The vent manually rRully developed to circumprotocol was successfbination of our 4x Hybridization buffer with lysis buffer 1 resulted in high signals and comated the detection of d steps for the automcan be inexpensively produced. Thus, all requiretoxic algae were achieved. A portable device was developed during the EU-project
in the field (e.g., on ships or shores) a stand-alone system which can be used asALGADEC,as well as in the laboratory. The device is easy to handle even for laymen and sample analyses
with all required steps can be performed automatically in less than two hours. Only the water
sample has to be filtered by hand by the userand placed in the inlet of the device. Data are
stored in the microcontroller unit or, if attached to a PC, can be analyzed directly. Multiprobe
Alexandriumlls frome ALGADEC device were tested using isolated RNA and cechips and thminutumseconds of measure and the data were comment because pasaturation of the reaction was observed. Hybridizations withred. The signals for comparison were chosen after 500
t. were carried ouA. minutumtwo different concentrations of target rRNA, high and low, fromined for low and high concentration of rRNA; aClearly distinguishable signals were determ the range of the negative control and was n resulted in signals inlow rRNA concentratio. A high rRNA concentration A. minutumit of the probes for consequently at the detection limgave mean signal of 201 nA. When compared to hybridization signals for dissolved cells of A.
minutumhigh quality decreased signals (m originated from about 260,000 cells, ean signal 150 nAwhereas th) can be observed. The isolated rRNA with a e filtered cell lysate of 500,000
h can disturb the hybridization cells contained also the proteins and polysaccharides, whicimmense. Additionally a field sample withPseudo-nitzschia cells from the Orkney Islands,
Pseudo-ltiprobe chip coated with the genus probe for u, was tested with a mUnited Kingdomnitzschia (data not shown). The analysis revealed a strong positive signal forPseudo-
nitzschia. Hence, the semi-automated device in combination with multiprobe chips can also
ples.e analysis of field saml used for thbe successfu

nized and the detectio has to be optim - The sensitivity of the systemForthcoming researchlimit must be reduced, because when a cell countof the toxic algal cells is reached, then the
fisheries are closed. We must have a detection limit far less than this number to meet
monitoring requirements. To meet these requirements, several adaptations must be made. The
spotting of the multiprobe chips with probes has to be automated to achieve a regular signal

109

Publication V

formation. We plan to spot different probes, i.e. species onto the chip, thus chips specific for

different geographic areas can be developed. Several specific probe sets for toxic algae have

been developed and need to be adapted to the chips. Furthermore, the sensors must be

calibrated for each probe set to convert the electronic signal into concentration of toxic cells

with the help of the software.

Conclusion

ltiprobe chip with 16 gold electrodes was designuA m

ed and adapted for the use in a

ated device was developed utomi-aore, a portable semsandwich hybridization assay. Furtherm

that automatically processed the main steps of the analysis and facilitated the electrochemical

an two hours. The device can be used by laymae in less th toxic algdetection ofen because a

ent of a lysis protocol. The nual RNA isolation is not longer required with the developmam

iprobe chip and the ALGADEC device can beltuproof of principle was presented here. The m

used as stand-alone system in the field and will contribute to monitoring programs to provide

an early warning system for the aquaculture and tourist sectors who are most affected by toxic

algal blooms.

ledgmentsAcknow

excellent the EU-Project ALGADEC formThe authors would like to thank all partners fro

cooperation and valuable discussions in the development of the multiprobe chips and the

s supported by the EU-project ALGADEC (COOP-CT-aALGADEC device. Sonja Diercks w

k Programme of the European Union and the ewor2004-508435-ALGADEC) of the 6th Fram

eAlfred Wsearch.egener Institute for Polar and Marine R

110

Publication V

References

Anderson, D. M., D. M. Kulis, B. A. Keafer, K. E. Gribble, R. Marin, and C. A. Scholin.

eration of Alexandrium2005. Identification and enum the Gulf of Maine spp. from

ceanography.cal Studies in Olecular probes. Deep Sea Research Part II: Topiousing m

The Ecology and Oceanography of Toxic Alexandrium fundyense Blooms in the Gulf

2467-2490. 52:of Maine

s, J. V. Tyrrell, M. Gladstone, and C. A. Scholin. 2005. International Ayers, K., L. L. Rhode

accreditation of sandwich hybridization assay format DNA probes for micro-algae.

1225-1231. 39:New Zeal J. Mar. Fresh

Berganza, J., G. Olabarria, R. García, D. Verdoy, A. Rebollo, and S. Arana. 2006. DNA

microdevice for electrochemical detection of Escherichia coli O157:H7 molecular

m.doi:10.1016/j.bios.2006.09.028rkers. Biosens. Bioelectron a

i. 2004. Oligonucleotide-modifiedCarpini, G., F. Lucarelli, G. Marrazza, and M. Mascin

lified sensing of nucleic acids. Biosens. pe-am-printed gold electrodes for enzymscreen

167-175. 20:Bioelectron

Doucette, G. J. and others 2006. A domoic acid immunosensor onboard the Environmental

Saote, sub-surface phycotoxin detection, p. le Processor: the first steps toward rempm

for the cietyternational So65-65, 12th International Conference on Harmful Algae. In

study of harmful algae.

Ma, H. WGau, V., S.icalng, J. Tsukuda, J. Kibler, and D. A. Haake. 2005. Electrochema

molecular analysis without nucleic acid amplification. Methods37: 78-83.

Guillou, L. and others 2002. Genetic Diversity and Molecular Detection of Three Toxic

French Coasts. Dinoflagellate Genera (Alexandrium, Dinophysis, and Karenia) from

Protis 223-238. 153:t

Hallegraeff, G. M. 2003. Harmful algal blooms: a global overview, p. 25-49. InH. G.M., D.

mM. Anderson and A. D. Ceella [eds.], Manual on Harmful Marine Microalgae. b

ific and Cultural Organization. United Nations Educational, Scient

Hosoi-Tanabe, S., and Y. Sako. 2005. Rapid detection of natural cells of Alexandrium

tamarense and A. catenella (Dinophyceae) by fluorescence in situ hybridization.

319-328. 4:gaeHarmful Al

re of Claus, and R. R. L. Guillard. 1987. Media for the cultu.Keller, M. D., R. C. Selvin, W

oceanic ultraphytoplankton. J. Phycol 23: 633-338.

traphytoplankloceanic uton. J. Phycol

111

Publication V

Kim, C.-J., and Y. Sako. 2005. Molecular identification of toxic Alexandrium tamiyavanichii

984-991. 4:(Dinophyceae) using two DNA probes. Harmful Algae

In E. 1998. Preparation and Analysis of RNA. .Kingston, RF. M. Ausubel et al. [eds.],

iley & Sons, Inc. lecular biology. John WoCurrent protocols in m

S. Alegret, and M. I. Pividori. 2006. In situ poy, J. Barbé, S. Hernández, o, A., S. CamLerm

DNA amplification with magnetic primers for the electrochemical detection of food
.ios.2006.08.048doi:10.1016/j.bpathogens. Biosens. Bioelectron

Mannelli, I., M. Minunni, S. Tombelli, R. Wang, M. Michela Spiriti, and M. Mascini. 2005.

e development of affinity biosensors. Direct immobilization of DNA probes for th

rkshop on Surface oistry. Proceedings of the International WemBioelectroch

Modification for Chemical and Biochemical Sensing, SMCBS'2003 66: 129-138.

atic and s (HAB); problemcroalgae bloomiMasó, M., and E. Garces. 2006. Harmful m

In Press, Corrected Proor. Pollut. Bull . Maconditions that induce themf;

.doi:10.1016/j.marpolbul.2006.08.006

ical detection of the dlin. 2005. ElectrochemeMetfies, K., S. Huljic, M. Lange, and L. K. M

th a DNA-biosensor. Biosens.exandrium ostenfeldii witoxic dinoflagellate Al

Bioelectron 1349-1357. 20:

PCR and (Dinophyceae) species using. Identification of AlexandriumPenna, A. M., M. 1999

615-621. 35:rDNA-targeted probes. J. Phycol

Scholin, C. A. and others 1999. DNA probes and a receptor-binding assay for the detection of

Pseudo-nitzschia (Bacillariophycoic acid activity in cultured and eae) species and dom

1356-1367. 35:ples. J. Phycol natural sam

Silver, M. W. 2006. Protecting Ourselves from Shellfish Poisoning. American Scientist 94:

316-325.

Taylor, A. D., J. Ladd, Q. Yu, S. Chen, J. Homoloa, and S. Jiang. 2006. Quantitative and

simultaneous detection of four foodborne bacterial pathogens with a multi-channel

752-758. 22:SPR sensor. Biosens. Bioelectron

ashiwoka aTyrrell, J. V., L. B. Connell, and C. A. Scholin. 2002. Monitoring for Heterosigm

1:gaetion assay. Harmful Alusing a sandwich hybridiza 205-214.

gaetion assay. Harmful Alusing a sandwich hybridiza

112

Publication V

Table 1. Sequences of capture and detection probes and positive control forAlexandrium

and ostenfeldii

A. minutum name Probeprobe: AOST1Detection

2 AOST probe:Capture

Test DNA AOST (positive control

Capture probe: AMIN C

be: AMIN C Detection proNEXTTest DNA AMIN (positive control)

ProbeCAA CCC sequenceTTC CCA ATA GTCAGG T

GAA TCA CCAAGG TTCCAAGCAG

CTGC TTG GAA CCT TGG TGA TTC ACCT GACTAT TGGGAA GGG TTG

GCTT GGATTGAA GTC AGG

TTC C TAA TGA CCA CAA CCC GCAAGG GTTGTTCC AAA CCT G GTC ATT AGAC TTC GGA

113

tTargeAlexandriumS 18ostenfeldiiAlexandriumS18ostenfeldii

AlexandriumS 18minutumAlexandriumS 18minutum

CitationMetfies et al. 5)200(Metfies et al. 5)200(Metfies et al. 5)200(Publication I

Publication I

Publication I

Publication V

Contents of lysis buffer 1 and hybridization buffers Table 2.

Buffers

11buffer 1, pH Lysis

4x Hybridization buffer,

Sample buffer, pH7.5

pH

8

alChemic

4 M guanidin-isothiocyanat

citrate sodiumM25 m

0.5 % Sarcosyl [w/v]

0.3 MNaCl

TrisM80 m

04% S0.SD

100 mM Tris

EDTA M17 m

5 MGuanidine isothiocyanate

amide8.35% Form

114

Publication V

)Any (tisnetn ignalSi

0200

0150

0100

050

0

Carbon sensors (Metfies

et al. 2005)

Carbon sensors GEM

negative control

lortno cevitispo

Gold sensors GEM

Figure 1. Comparison of signal intensity of carbon and gold sensors was done using probes

AOST2 and AOST1 and test-DNA

115

Publication V

)An (ytisnetn ialgniS

0200

0150

0100

050

0

withoutblocking

3% Casein

S B3%A

ASB%5

negative controlpositive control

Reduction of background signal by using casein and bovine serumFigure 2.

lated probe AOST2blocking solutions on gold sensors coated with the thio

116

10% BSA

in as album

Publication V

00200015)Any (tisnetn ialgniS1000

050

0

hserf

hnto m4

carbon sensors

hntom6

gold sensors

12 month

htnMo

ith 2% Trehalose and stability of carbon and gold sensors after coating wLong termFigure 3.

alstervstorage at 4°C over indicated in

117

Publication V

004500400035)A (nitysnte inlangiS2500
0030

2000

0015

0010

050

0

DPO

1A, ADPg m1.2O2 HM200 m

A, ADPg m1.12 HM400 m2O

A, ADPg m2.22O2 HM100 m

A, ADPg m2.2200 m2O2 HM

A, ADPg m2.22O2 HM400 m

negative controlpositive control

ent by varying Signal enhancemFigure 4. substrate concentrations

118

g m4.4A, ADP2O2 HM400 m

A, ADPg m4.42O2 HM600 m

.6A, ADPg m62O2 HM600 m

Substrate solutions

Publication V

Figure 5. robe chltipuMip with 16 gold working electrodes

119

Publication V

700600y (tisnetna igniS)An400
500

300

200

100

0

1

2

3

4

5

6

7

8

9

01

11

12

13

41

15

16

Working electrode

ission between working electrodes with every second electrode coated Signal transmFigure 6.

lated probe AOST2with thio

120

56789

Figure 7. Comparison of hand-spotted multiprobe chips with different carrier materials

01

11

21

61

4151

31

Working electrode
Ceramiccarrier material with hydrophobic polymer

Ceramic carrier material

Plastic carrier material

1234

0

050

)0150nA(ytisnetn ialgniS
0100

Publication V

121

Publication V

00140012An(ytinsetn ignalSi)800
0010

060

040

020

0

Negative
control

lpmSaebuffer/RLT

Sample4x Hyb.4x Hyb.
buffer/lysisBuffer/RLTBuffer/lysis
buffer 1buffer 1

ANR

Positive
control

Figure 8. Determination of optimal signal formation using two different lysis and

hybridization buffers and probes for Alexandrium minutum

122

Publication V

Figure 9. le ALGADEC device ated portabi-automSem

123

Publication V

Figure 10.

Flow chart of semi-automated device

124

Publication V

1000800600

400ndsoceS6 1 8WEWE00015 7 1WEWE8004 6 1WEWE6003 5 1

WEWE2 1 4WEWE1004ndsoceS 8WE6 7 1WEWE5 6 1WEWE4 1 5WEWE

3 1 4WEWE2 1 3 3WE 1WEWEWE
200200 2WE0 1WE 2WE1 1WE
00 1WE 9WE 1WE0 1WE
07-E001.00+E000.7-0E0.0-1y7-0E0.0-2tisne7-0E0.0-3tn ial7-0E0.0-4gniS7-0E0.0-57-0E0.0-67-0E0.0-77-0E0.0-87-0E0.0-907-E001.00+E000.7-0E0.0-1-0E0.0-2y7tins7-0E0.0-3etn i7-0E0.0-4lang7-0E0.0-5Si7-0E0.0-67-0E0.0-77-0E0.0-87-0E0.0-9
DB1000

1000800600400

sdnocSe6 1 8WEWE5 1 7WEWE8004 1 6WEWE6003 1 5WEWE2 1 4400WEWE

1ndsoceS6 1 8WEWE5 1 7WEWE4 1 6WEWE3 1 5WEWE2 1 4WEWE

1 3WE 1WEWE 3 1WE
200200 2WE0 1WEWE 20 1WE
00777777777777 9 1 977 1777070070700-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-+-0WEWEEEEEEEEEEEWEWE-+

E00.1AE00.0E00.-1E00.-2ytisE00.-3netnE00.-4 ialgnE00.-5iSE00.-6E00.-7E00.-8E00.-900.1CE00.000.-100.-2yti00.-3nsetn00.-4 ilan00.-5gSi00.-600.-700.-800.-9l, (C)ontrocl, (B) Negativee contrositivated device. (A) Poi-autom the sem
HybridisatioFigure 11.ltiprobe chips inuto the mets onn of different targion of 7.32 µg, (D) Target RNA with a final concentration of 4.95 µg Target RNA with a final concentrat

125

Publication V

0001

008

600

400

002

000+00E0.-070E1.0-
B0001

00+00E0.008

600

400

002

0

A70-E00.-1)-070E2.0-nA(yti-070E3.0-nsetn ila-070E4.0-ngSi

70-E00.-2)Any (t70-E00.-3isnetn i70-E00.-4gnaliS-070E5.0-

70-E00.-56 1 8ndsWEWEoceS5 1 7WEWE4 1 6WEWE3 1 5WEWE2 1 4WEWE1 1 3WEWE0 1 2WEWE-07 9 10EWEWE6.0- 9ndsWEoceS6 1 8WEWE5 1 7WEWE4 1 6WEWE3 1 5WEWE2 1 3WEWE1 1 2WEWE

70-E00.-60 1 1WEWEAlexandrium minutumated device. (A+B) 500,000 lysed cells of -automie semin thn of lysed cells onto the multiprobe chips HybridisatioFigure 12.

126

Synthesis

Synthesis3.

Colorimetric assay for the detectio3.1gal species ln of toxic a

The composition of phytoplankton communities in Europe includes several genera of toxic
algal species, such as Alexandrium,Dinophysis,Gymnodinium and Pseudo-nitzschia (Simon
s. Monitoring program. 2003; Moita et al. 2003; Chepurnov et al. 2005)et al. 1997; John et alaim at the rapid and reliable detection of harmful algae in coastal areas and shellfish and fish
ation of harmful species using standard s. Morphological identification and enumermfarmicroscopy procedures are time-consuming and a broad taxonomic knowledge is required.
For exampleAlexandrium minutum is characterized by minute details of its thecal plates and
e genus (Taylor et al. 1995). thus, is difficult to distinguish from other species of the samalall subunit ribosomlecular probes, that target the large or smoMolecular technologies and mecause the applications proving the detection of phytoplankton, bidly imRNA (rRNA), are rapwn to the fferences of the different species. Doination of the genetic diare based on the discrimpresent day the development molecular probes is limited to only a small percentage of the
al species.ic algdifferent tox

In Publication I the detection of the toxic dinoflagellateAlexandrium minutum was
o probes are needed conducted by the use of molecular probes in sandwich hybridization. Twat (Zammatteo et al. 1995; Rautio et al. 2003), and at least ich hybridization formin the sandwone of the probes has to be specific for the target. The so-called capture probe is immobilized
on solid surfaces as in cotarget RNA or DNA. A simbgnal moination with DNA biosensors (Metfiiety is covalently bound to a second probe, which binds in es et al. 2005) and binds to
close proximity to the binding site of the capture probe. A commercially available PCR
18S ribosomELISA Dig Detection Kit was adapted for the al RNA probes for the species-specificuse of sandwich hybridiz identification was develoation. A set of two ped forA.
nd the signal probe digoxigenin-labelled for The capture probe was biotin-labelled aminutum.ficity was successfully the application in the sandwich hybridization. Probe specidemonstrated with the microtiter plate assay; because the signals for all A. minutum strains
non-target species. It was also were always significantly higher than the signals for the pointed out that Alexandrium species with a single mismatchin the target sequence were not
thod, achieved signals need to be eonitoring mdetected. For the application of the assay as a mcorrelated to cell numbers. Bloom development in the field is expected to correspond most

127

Synthesis

um growth conditions (Ayers et al. 2005). Hence, total rRNA concentration closely to optimper cell of A. minutum was determined at optimum growth conditions for three different
ilar findings were s found. Simaean concentration of 0.028 ng rRNA per cell wstrains and a mA.and for different growth conditions of A. ostenfeldii Metfies et al. (2005) for achieved fromminutum(personal communication L. Carter, Westminster University, UK). Consequently,
isolated rRNA concentration is possible. A the bers fromthe calculation of the cell num was investigated for the assay, resulting in a good A. minutumstandard calibration curve for correlation of signal with rRNA concentration.Thus, cell numbers can be calculated from the
l cells toples were spiked with different algaity of the assay. Natural water samtenssignal inevaluate the potential of the microtiter plate assay for the monitoring of field samples. The
results demonstrate that the molecular assay was capable of detectingA. minutum cells at
parison to a plex background. However, in comdifferent cell counts in the presence of a comr of cells, lower signals were achieved for the spiked e numbepure culture with the samposition of the ion signals could be the comt hybridizatples. One reason for the differensamsample, because large amounts of sediment were observed at the sampling station. Sediment
ation and the isolation protocol needs to be modified. the RNA isols to disturbseemailable PCR ELISA Dig Detection Kitercially ave commNevertheless, the adaptation of thising proof of t a proments with spiked natural samples presenwas successful and the experimprinciple.

In this study a method for the detection of the toxic dinoflagellate Alexandrium minutum
thod has the ecrotiter plate assay was established. This miusing 18S rRNA probes and a mpotential to be a fast and reliable method for the detection ofinating the toxic algae by elima rapid assay was developed for the routine ore,need for manual algae counts. Furtherme way ence level much in the sam target sequtesting of probe specificity at both the clade and screen for specificity of FISH probes.t blots are used tothat do

The microtiter plate assay developed in Publication I was applied for further species-specific
identification of toxic algal species in Publication II. Probe sets for the toxic species
odinium polyedrum, Prymnesium Protoceratium reticulatum, Lingul,Gymnodinium catenatumPseudo-nitzschia multiseries, P. australis, P. seriata ,parvum, Chrysochromulina polylepisre than 3000 published and o were designed from a database consisting of mand P. pungense species had already been obes for somunpublished algal 18S rRNA sequences. Single preans of dot blot and FISH (Simon et al. 1997; Töbe city by mdeveloped and tested for specifi

128

Synthesis

et al. 2006). Hence, for these only a second probe was designed to complete the sandwich

hybridization and the combination of both of the probes needed to be tested for specificity. A

species was previously designed and needed to be Pseudo-nitzschiaprobe set for the genus of

all subunit ribosomal RNA genes are the targets for the different madapted. The large and the s

probe sets. The sandwich hybridization in themicrotiter plate assay was applied for the

specificity tests. Target rRNA was isolated from laboratory strains of the target species and

closely related species. A total of ten probe sets for different toxic algal species were designed

and tested and eight probe sets proved to be highly specific. Two probe sets with the target

species detected one non-target species in Prymnesium parvum and Gymnodinium catenatum

addition to the target species, respectively the non-toxic G. impudicum and P. nemamethecum.

The specificity of the probes is dependent on thber of sequences of the targeted gene e num

a large database, such as the available in databases. However, even if a probe is designed from

used database, it is almost impossible to avoid the occurrence of some false-positive results

with a monostringent hybridization approach. The in silico prediction of the stability of

mismatched probe-target hybrids is difficultand influenced by many factors, such as the

number of mismatches, the nature of the mismatching nucleotides, the position of the

mismatches in the probe target site, and possible stacking interactions of nucleotides adjacent

is study using specific atches (Loy et al. 2005b). The protocol applied in thsmito the m

crotiter plate assay as well as for the iized for the mperatures is optimhybridization tem

biosensor described in Publication V. However, specific identification of target organisms is

still possible with the probes sets for the species described above. The probe set for

r only 10,000 cells, whereas for o presents significant high signals fGymnodinium catenatum

are G. impudicumnon-target species e signal intensity at least 250,000 cells of the the sam

needed, thus a misinterpretation of signal is unlikely. The detection of P. nemamethecum in

brackish waters with the probe set forP. parvum cannot be ruled out but seems to be unlikely,

because the majority of P. parvum blooms have been recorded in brackish waters (Edvardsen

and Paasche 1998) and there have not been any reports of bloom

(nemamethecumest et al. 2006). W

P.s caused by

onitoringThe ten probe sets presented in this study are valuable tools for identifying and m

different toxic algae and can be adapted to the multiprobe chip and the semi- automated

biosensor presented in Publication V. Furthermore, the specificcapture probes can also be

adjusted to other molecular methods using ribosomal RNA probes, such as the DNA-

e PCR or FISH.croarray technology, real-timim

129

Synthesis

In summary, a commercially available PCR ELISA Dig Detection Kit was successfully

n of the toxadapted for the detectio by application of Alexandrium minutumic dinoflagellate

ean concentration of total rRNA peral 18S probes and sandwich hybridization. The mribosom

n curve for different RNA concentrations and ined and a standard calibratiocell of was determ

thus cell counts was investigated for the assay. Additionally the assay was able to detect A.

plex background. This unts in the presence of a com cells at different cell cominutum

represents the potential to serve as a fast and reliable mfor the detection of toxic algae thode

ore, the assay showed the nually. Furthermainating the need to count algae mby elim

,Gymnodinium catenatumspecificity of 10 additional probe sets for the toxic algal species

polyedrum, Prymnesium parvum, umProtoceratium reticulatum, Lingulodini

Chrysochromulina polylepis,Pseudo-nitzschia multiseries, P. australis, P. seriata and P.

pungens and the genus Pseudo-nitzschia.

hancementAssessment of signal en3.2

DNA-biosensors are commonly used in clinical diagnostic (e.g., glucose detection),

sicals), identification of infectious organismonitoring (e.g., hazardous chemental menvironm

and forensics. Biosensors are also commonlyused for the rapid identification of aquatic

microorganisms. The identification of the toxic dinoflagellate Alexandrium ostenfeldii using a

biosensor was presented by Metfies et al. (2005). The advantage of biosensors is displayed by

the in situ use and therefore the circumvention of sample transport to the laboratory.

Therefore, this technique is suitable for the application in monitoring programmes, because of

the simple use and analysis. Another potential method is the DNA-microarray-technology,

250,000 probes (Lockhart et al. 1996) and can ultaneous analysis of up tothat enables the sim

thod to analyse samples from comebe used as a ments (Metfies and Medlin plex environm

2004; Medlin et al. 2006). However, the reliable identification of harmful algal species with

thods requires highly specific and sensitive probes and high quality nucleic eprobe-based m

acids.

InPublication III the method for the electrochemical detection of toxic algae is presented,

the steps were described in detail and visualized for easy application by inexperienced users.

difications were established to the protocol described by Metfies et al. (2005). oSeveral m

nufacturers were located for the production of the single electrode aFirst of all, different m

130

Synthesis

sensors and the ment of the biosensors. In addition, the provement device for the imeasurem

total RNA isolation protocol from Qiagen (Hilden, Germany) was modified to increase the

rem and quantity of the extractedprove qualityarides and proteins to imoval of polysacch

rRNA. The improved quality of the rRNA led to an enhanced signal intensity of the

electrochemical measurements. The improved method was tested with spiked field samples in

her signal was also a higPublication IVan intercalibration workshop (Godhe et al. 2007). In

determined for undesired cross-hybridization of A. ostenfeldii probes to Alexandrium minutum

parison to the data previously presented by Metfies et al. (2005). This represents no in com

A. ostenfeldiidrawback of the probe detection of is a toxic species A. minutum, because also

and needs monitoring. On the multiprobe chip, presented in Publication V; a combination of

probes will facilitate the differentiation of both species. The introduction of locked nucleic

acids (LNAs) promises an enhancement of both specificity and sensitivity of molecular

probes (Kongsbak 2002). LNAs have shown their potential in mny applications, such as a

gene expression profiling, genotyping assays, fluorescencein situ hybridization and real-time

PCR (Jacobsen et al. 2002b; Nielsen and Kauppinen 2002; Silahtaroglu et al. 2003; Ugozzoli

ent of probe or hybridization signals hancemet al. 2004b). Many publications describe the en

with LNA modified probes, but there has been no rigorous testing of these probes using

known target sequences. The specificity and sensitivity of conventional molecular probes and

LNA modified probes were compared in Publication IV by application of sandwich

hybridization on biosensors and on DNA-microarrays. Three different species, A. ostenfeldii,

A. minutum and A. tamutum, were tested with conventional and LNA modified probes on the

wasA. tamutum also A. minutumbiosensor. In addition to the cross-hybridization signal for

tested for signal formation, because of only one mismatch in the 18S rRNA sequence to the

(2003) suggested that the use of LNAs could significantly capture probe. Kauppinen et al.

ination (Kauppinen et al. 2003). Previously, several probes were atch discrimsmiincrease m

croarray butisuccessfully adapted to the DNA-m, because of low signal intensities, an

LNAs was evaluated. The present study ent of the signal-to-noise-ratios usingenhancem

osensor,ent using the rRNA birevealed that the LNA probes showed neither signal enhancem

smatch. iination of only one mnor discrimThe DNA probes showed equal or better results in

all experiments using the biosensor, whereas LNA probes could enhance the sensitivity of the

e conventional probes. However, unspecific croarray and gave higher signals than thim

s also enhanced. In conclusion the LNA probes do not binding with non-target DNA wa

improve signal intensity under at these solid surface-hybridization applications. Other

potential application for signar could be the variation of ent of the biosensohanceml en

131

Synthesis

substrate concentration and the reduction of background noise with blocking solutions. Signal

enhancement in case of the microarray can be accomplished by using labelling kits that

or also the reduction of background noise.els to a targetltiple labuincorporate m

NA isolation protocol was iIn summary, the total rRproved and the hybridization procedure m

for the electrochemical detection of toxic algal species was described in detail and

illustratively visualized for easy application by inexperienced users. Furthermore, locked

re tested with known target sequences and the specificity and enucleic acid probes w

sensitivity was compared to signal formation of conventional molecular probes. The

anced and thods could not be enhen signals for both of the tested solid surface mhybridizatio

the conventional DNA probes showed equal or better results.

tomatedDetection of toxic algal species using multiprobe chips and a semi-au3.3

device

The increasing demand for fast monitoring techniques emerges frompoisoning incidences

ing up to five ple analysis takand economic losses, which cannot be foreseen because of sam

ouse-thod for shellfish flesh analysis, the mee statutory mworking days. In addition, th

bioassay, induces ethical problems. HPLC as well as traditional light microscopy methods are

time-consuming and need high trained personnel. Furthermore, the samples have to be

transported to specialized laboratories for analysis. A potential tool for bloom formation and

in situ ination is provided by the thus potential shellfish contamtigation of coastal waterinves

ples for on-site monitoring of toxic ic algae. There are examfor the occurrences of different tox

et al. 2006; Silver 2006).pling processor (Doucetteental samalgae, such as the environm

In order to facilitate thein situ monitoring of toxic algae, a multiprobe chip and a semi-

e in situ detection of toxic algae were developed and ated rRNA biosensor for thautom

evaluated inPublication V. Simultaneous detection of different species canbe accomplished

using arrays of probes, such as microarrays(Metfies and Medlin 2004). A multiprobe chip

multaneous detection of up to 14 algal target with an array of 16 gold electrodes for the si

r a simospecies was designed in this study. Fplification of sensor handling the standard size of

can be stored in standard boxes. Different a conventional glass slide was chosen and thus

materials for the electrodes and the carrier material were tested in order to achieve a

132

Synthesis

atically. The mass production and with probes autommultiprobe chip that can be coated

consequently the probes vance of use are able to decrease the costs; of sensors in adcoating

rface and need to give samhave to be stable on the electrode suonths signals after several me

ents showed that the sensors are stable over a year, storage experimof storage. Long-term

thodet morder to facilitate a cost efficienhowever a signal decrease of 26 % was observed. In

parison to the single electrode sensors in coms reduced athe size of the working electrodes w

eased and the amount of nce, the electrode area is decre Metfies et al. (2005). Hused from

reagents needed for analysis is reduced.

So far, all monitoring methods demand high trained personnel for sample analysis.

Experienced users are needed for the isolation of total rRNA from the different algal species

for analysis using the biosensor. Scholin et al. (Scholin et al. 1999) reported the use of crude

in sandwich hybridization assays. ForPseudo-nitzschiaogenates for the detection of cell hom

the use of the biosensor by layperson an adaptation of analysis and hybridization procedures

l was successfully developed, thus mwas required. An easy to use lysis protoconual rRNA a

isolation is no longer necessary, only water sample filtration has to be performed manually.

For the simultaneous detection of several toxic species, a multiprobe chip with 16 working

ated, portable device, which is easy to handle even for electrodes was generated. An autom

laypersons, was designed and extensively tested in combination with the multiprobe chip and

molecular probes for Alexandrium minutum. Isolated RNA and cells fromAlexandrium

were analyzed with the devminutumice and the data was compared. It was observed that after

reaction takes place. Clearly distinguishable ent a saturation of the easurem500 seconds of m

signals were determined for low and high concentration of rRNA and when compared to

, decreased signals were observed. The nutumA. mihybridization signals for dissolved cells of

ple. The isolated rRNA signal variations can be explained by the quality of the analyzed sam

sate contains still proteins and polysaccharides, de cell ly quality, whereas the cruhad a high

which can disturb the hybridization. During a demonstration of the device to mers,mssel faru

alyzed and resulted in a strong cells was anPseudo-nitzschiaple containing a field sam

positive signal forPseudo-nitzschia. Hence, the device is able to contribute to monitoring

for the aquaculture and tourist sectors, which s to provide an early warning systemprogram

are affected by toxic algal blooms the most. The probes presented inPublication I and

Publication II can be adapted to the use on the multiprobe chip, thus area chips for different

uregions in Erope can be developed.

133

134

Synthesis

In summary, a stand-alone, semi-automated system in combination with multiprobe chips was

developed. A multiprobe chip with 16 gold electrodes was designed and adapted for the use in

ined and the

stability of the sensors was exam

a sandwich hybridization assay. Long-term

nual RNA a

sensors found to be stable over a year. A lysis protocol was adjusted and m

ed total rRNA

isolation is not longer required. Analysis of different concentrations of isolat

clearly distinguishable signals, but lower signals for the cell

and crude cell lysates revealed

i-

the semours with two h

d in less thaneples was perform

lysate. The analysis of all sam

thods that need at least a day e

onitoring m

other routine m

ated device in comparison toautom

for analysis.

Future Research

Future Research 4.

ented in this study. thods for toxic algal species were preseonitoring mTwo potential mHowever, both methods can be improved throughseveral measures. The microtiter plate assay
lples and showed a signa was applied for natural water samPublication Ipresented in s to ent seements using pure laboratory cultures. Sedimparison to experimreduction in comat total rRNA cannot be isolated disturb the RNA isolation; additionally it was observed th spp., using a conventional Kit but can be using aPseudo-nitzschias, such as diatomfromphenol-chloroform method (Publication II). An improvement of the existing protocol has to
ent of an independent system without RNA isolation, such as ined and the developmbe exammethod described by Scholin et al. (1999), should be included in further experiments. The
RNA concentration per cell has to be determined for every target species at optimum growth
conditions, because this correspondsmost closely to bloom development in the field (Ayers et
bers.on curves allow the correlation of signal to cell numubsequently, calibratial. 2005). SDetection limits of each probe set for the different toxic species have to be identified. Final
ples.titer plate assay should include field samcroitest of the m

The sensitivity of the semi-automated system presented in Publication V has to be optimised
and probes developed in Publication I andPublication II need to be adapted to the
multiprobe chip to allow the development of chips specific for different geographic areas. For
this requirement, the probes have to be dispensed onto the multiprobe chips automatically to
trations are emphasizedation. Recommended action cell concenmachieve a regular signal forhyte 2003), thus the different probes sel and Wful algal species (Renrent harmfor the diffehave to be examined for their detection limit. The detection limit needs to be below the
allowed cell numbers limits to meet monitoringrequirements.Furthermore, the sensors have
to be calibrated for each probe set to convert the electronic signal into concentration of toxic
ples need to be tested for samently, field the software. Subsequcells with the help ofevaluation. A total of 17 different probe sets can be applied to the multiprobechips, however,
known today (Moestrup 2004) and the number is increasing. about 97 toxic species are ent of new probe sets and ents should include the developmConsequently, further experimpresented in this study is a prototype and hastheir adaptation to the biosensor. The biosensor to be improved in terms of system integration and maintenance for commercial purpose.
s for continuous analysis of ore, the device could also be integrated into buoy systemFurtherm

135

Future Research

coastal waters. Finally, the biosen

sor can also be

for several o adapted

ter or for clinical diagnostics.acrobial pathogens in widetection of m

136

ehtr field

s, su

ch as the

arymmSu

Summary5.

This doctorial thesis aimed at the developmentand evaluation of fast and reliable monitoring

methods using molecular technologies. The detection of harmful algal species in coastal areas

and shellfish farms is an important requirement of monitoring programs, because of their

inated seafood and for ers through ingesting contamresponsibility for poisoning of consum

thods include the statutory application of the eonitoring mfish and shellfish kills. Current m

inationof shellfish, toxin determinationmonitoring of toxin contammouse-bioassay for the

using HPLC and standard light microscopy. The methods are time-consuming, expensive and

prove therequire high trained personnel. Molecular technologies using probes can im

detection of phytoplankton.

rmful algae is presented by an assay that is thod for the detection of haeThe first potential m

ferent species. The PCR ELISA ination of the genetic variation of the difbased on the discrim

d was adapted for the detection of the toxicrcially available anemDig Detection Kit is com

using sandwich hybridization. Sandwich hybridization Alexandrium minutum dinoflagellate

requires two probes for each species, a capture probe and a nearly adjacent signal probe. A set

of two probes for the species-specific identification was designed and were found to be highly

specific. The mean concentration of total rRNA per cell was determined from three different

and found to be 0.028 ng. A standard calibration curve for different A. minutumstrains of

plesblished for the assay. Spiked water samRNA concentrations and thus cell counts was esta

nstrated the ability ofomewere used to evaluate the assay and the standard curve. The results d

the assay to detectA. minutum cells at different cell counts in the presence of a complex

thod for the detection of ebackground. The assay has the potential to be a fast and reliable m

toxic algae by eliminating the need to count algae manually. The microtiter plate assay was

applied for further species-specific identification of the toxic algal speciesGymnodinium

,catenatum

Protoceratium reticulatum, Lingulodinium polyedrum, Prymnesium parvum, Chrysochromulina polylepis,Pseudo-nitzschia multiseries, P. australis, P. seriata and P.

. Probe sets were designed to target the large or the Pseudo-nitzschia and the genus pungens

sms. The specificity of the different probes sets was tested al RNA geneall subunit ribosom

al RNA isolated from laboratory strains of the target species and closely related with ribosom

AT and Cspecies. Eight probe sets proved to be highly specific in the assay. Two probe sets, G

species. The designed probe sets s, in addition to the target PRYM 694, detect one other specie

137

arymmSu

i-nitoring of toxic algae and can also be adapted to the semole tools for the mare valuabautomated biosensor. The mfor use in a sandwich hybridization foricrotiter plate assay mats, similar to the way that dot blots are used to is an effective and fast method to test probes
ecificity for FISH probes.screen for sp

ically achieved using a biosensor and chemThe detection of toxic algae can also be electrole transport to the pmeasure on-site and sasandwich hybridization. Biosensors can mlaboratory is unnecessary. The protocol introduced by Metfies et al. (2005) using a biosensor odified and illustrated for easy application was mAlexandrium ostenfeldiifor the detection of by inexperienced users. The modifications included the adaptation of single electrode sensors
and the measurement device from different manufacturer as well as the total RNA isolation
tensity of the protocol. Improved quality of the rRNA led to an enhanced signal inelectrochemical measurements. An enhancementof both the specificity and sensitivity of
As). The s (LNobes can also be achieved by introduction of locked nucleic acidrmolecular plecular probes and LNA modified probes were ospecificity and sensitivity of conventional mcompared in two different solid phase hybridization methods; sandwich hybridization on
biosensors and on DNA-microarrays. Conventional molecular probes and LNA probes that
target Alexandrium ostenfeldii were examined for signal formation in combination with the
biosensor. In addition to A. ostenfeldiialsoA. minutum and A. tamutumwere tested for cross-
hybridization. However, signal enhancement for A. ostenfeldiicould not be observed.
Furthermore, the LNA capture probes could not discriminate only one mismatch in the 18S
ddition, the conventional probes showed a higher cross-. In aA. tamutumnce of rRNA sequeeviously presented by Metfies parison to the data pr in comA. minutumhybridization signal for et al. (2005), because of the higher quality of the rRNA. However, both species, A. ostenfeldii
bination with the DNA-d. In com are toxic and need to be monitoreA. minutumandmicroarrays, the LNA-probes displayed an enhancement of sensitivity, but also more false-
DNA probes showed equal or better results positive signals. In summary, the conventional prove signal intensity under certain solid A probes. LNA technology could not imthan the LNplications.surface-hybridization ap

In addition to the microtiter plate assay and the single electrode assay, a multiprobe chip and a
detection of toxic algae were developed. The in situ i-automated rRNA biosensor for the semdesign of the multiprobe chip with an array of 16 gold electrodes was conducted by testing
different materials for the electrodes and the carrier material with the help of single electrode

138

arymmSu

sensors. The multiprobe chip can detect up to 14 target species using the previously designed

ted, portable device was designed and ammolecular probes. An easy to handle, auto

extensively tested in combination with the multiprobe chip and molecular probes for

Alexandrium minutum. A peristaltic pump moves the reagents from the reservoirs into the

in steps of the analysis are processes automatically. ahybridization/fluidic chamber, thus the m

Furtherml was successfully developed for use of the device by otocorore, a lysis p

nual rRNA isolation is no longer required. The device was ainexperienced staff and m

m algae cultures and clearly distinguishable evaluated using isolated total rRNA and cells fro

ined. The stand-alone systemsignals were determo hours ples in less than tw can analyse sam

and can be applied in the field. Thus, the device and the multiprobe chip have the potential to

serve as an early warning system for the aquaculture and tourist sectors.

Zusammenfassung6.

In dieser Dissertation wurden schnelle und verlässliche Monitoring-Methoden m

it Hilfe

wertet. Aufgrund der Vergiftungen von molekularer Techniken entwickelt und be

inierte Fische und Meeresfrüchte, sowie von Fisch- und enten durch kontamKonsum

dlichen Algenarten in Küstenzonen und in Schalentiersterben, ist der Nachweis von schä

Zuchtgebieten für Meeresfrüchte und Fische eine wichtige Voraussetzung für Monitoring-

Programme. Die derzeitigen Monitoring-Methoden beinhalten den gesetzlich

inierten Meeresfrüchten, vorgeschriebenen Maus-Bioassay für die Überwachung von kontam

ikroskopie. Die den chromatographischen Toxin-Nachweis sowie die Standard-Lichtm

beschriebenen Methoden sind zeitaufwendig, teuer und verlangen die Erfahrung von

geschultem Personal. Der Nachweis von Phytoplankton kann durch den Einsatz von

molekularer Techniken und Sonden deutlich vereinfacht und verbessert werden.

Ein auf der Unterscheidung von genetischer Vriation der verschiedenen Arten basierender a

ethode für schädliche Algen dar. Ein Assay stellt eine erfolgsversprechende Nachweism

chweis der toxischen handelsüblicher PCR ELISA Dig Detection Kit wurde für den Na

ndwich Hybridisierung angepasst. Für jede durch SaAlexandrium minutumteDinoflagella

Spezies werden eine Fänger-Sonde und eine benachbarte Signal-Sonde in der Methode der

wurde ein Satz von zwei Sonden für die Sandwich Hybridisierung verwendet. Folglich

artspezifische Identifikation entworfen und als höchstspezifisch nachgewiesen. Anschließend

Konzentration pro Zelle anhand von drei t-rRNA-wurde die durchschnittliche Gesam

139

Suarymm

verschiedener Stämme von A. minutumerfasst und auf 0.028 ng ermittelt. Für den Assay

ntrationen und den wurde eine Standard-Kalibrierungskurve für verschiedene RNA- Konze

ne Bewertung des Assays und der Standard-illzahlen erstellt. Eekorrespondierenden Z

Kalibrierungskurve wurde mit Hilfe Algen-beimpfter Wasserproben durchgeführt. Die

A.Ergebnisse zeigen eindeutig, dass der Assay in der Lage ist, eine verschiedene Anzahl an

Zellen in einemminutumnd nachzuweisen. Der beschriebene Assay hat ru komplexen Hinterg

chweis von giftigen n Methode für den Nadas Potential einer schnellen und verlässliche

sätzlichuieden werden kann. ZAlgen, wodurch eine aufwendige, manuelle Zellzählung verm

wurde der Mikrotiterplatten-Assay für die artspezifische Identifikation der toxischen

AlgenartenGymnodinium catenatum,Protoceratium reticulatum,Lingulodinium polyedrum,

Pseudo-nitzschia multiseries, P. australis, ,rysochromulina polylepisPrymnesium parvum, Ch

P. seriata and P. pungensendet. Sonden-Sätze verwPseudo-nitzschia und der Gattung

alen RNA-Gene iten der ribosomwurden entwickelt, die an die große oder kleine Untereinhe

binden. Die Spezifität der verschiedenen Sonden-Sätze wurde mittels isolierter ribosomaler

eelarten und nah verwandter Arten getestet. Acht der Sonden-SätzRNA von betrachteten Zi

sch. Zwei Sonden-Sätze, GCAT und PRYM 694, Assay als höchstspezifierwiesen sich in dem

weisen zusätzlich zur Zielartauch eine andere Art nach. Die entwickelten Sonden stellen ein

rkzeug für das Monitoring von toxischen Algen dar und können auch für den ewertvolles W

halbautomatischen Biosensor verwendet werden. Der Mikrotiterplatten-Assay ist eine

effektive und schnelle Technik für die Überprüfung der Sondenspezifität für Sandwich

Hybridisierungs-Formate, vergleichbar mitder Ermittlung der Spezifität von FISH-Sonden

durch Dot Blots.

it Biosensoren und isch mDer Nachweis von toxischen Algen kann auch elektrochem

Sandwich Hybridisierung erfolgen. Biosensoren können auch vor Ort für Messungen

ällt der Transport von Proben inverwendet werden, folglich entf)s Labor. Metfies et al (2005

ilfe eines it H mAlexandrium ostenfeldiipräsentierte ein Protokoll für die Detektierung von

den Zugang für Laien odifiziert und bildlich dargestellt wurde, ummBiosensors, welches nun

zu vereinfachen. Die Modifizierungen beinhalten die Anpassung von Einzelelektroden und

eines Messgerätes von unterschiedlichen Herstellern, sowie des RNA-Isolationsprotokolls.

Die erhöhte Qualität der rRNA führte zu verstärkten Signalintensitäten der elektrochemischen

Messung. Eine Verstärkung von Spezifität und Sensitivität von molekularen Sonden kann

auch durch die Verwendung von Locked Nucleic Acids (LNAs) erreicht werden. Die

Spezifität und Sensitivität von herkömmlichen molekularen Sonden und LNA-modifizierten

Sonden wurde methoden, der n Festphasen-Hybridisierungsmit Hilfe von zwei verschiedene

140

arymmSu

Sandwich Hybridisierung auf Biosensoren und den DNA-Mikroarrays, verglichen. In

Kombination mit dem Biosensor wurden die herkömmlichen molekularen Sonden und die

LNA Sonden mitAlexandrium ostenfeldii als Zielart auf ihre Signalbildung untersucht.

Zusätzlich zu A. ostenfeldiiwurden auchA. minutum und A. tamutumauf Kreuz-

A. ostenfeldiiverstärkung für t. Es konnte jedoch keine SignalHybridisierung überprüf

beobachtet werden. Des Weiteren war mit den LNA-Fänger-Sonden eine Unterscheidung von

nur einer Fehlpaarung in der 18S-Sequenz von A. tamutum nicht möglich. Darüber hinaus

ies et al (2005) vorgestellt wurden, eine Vergleich zu den Daten, die von Metfwurde im

erhöhte Kreuz-Hybridisierung für A. minutum mit den herkömmlichen Sonden festgestellt.

e Qualität der rRNA begründet werden. Jedoch en kann durch die verbessertDieses Phänom

sind beide Arten toxisch und macht werden. Die LNA-Sonden zeigen in ssen überwü

hreit den DNA-Mikroarrays eine erhöhte Sensitivität, jedoch zusätzlich mbination mKom

falsch-positive Signale. Zusammengefasst zeigen die herkömmlichen Sonden gleiche oder

bessere Ergebnisse als die LNA-Sonden. Die LNA-Technologie konnte unter diesen

Festphasen-Hybridisierungsanwendungennicht die Signalstärke erhöhen.

latten Assay und der Einfachelektroden-Anwendung, wurden Mikrotiter-PZusätzlich zu dem

atischer rRNA Biosensor für den vor Ort ein Mehrfach-Sonden-Chip und ein halbautom

it einer onden-Chips mNachweis von giftigen Algen entwickelt. Das Design des Mehrfach-S

Reihe von 16 Goldelektroden wurde durch den Test von verschiedenen Materialien für die

Elektroden und des Trägermaterials mit Hilfe von Einfachelektroden bestimmt. Der

Mehrfach-Sonden-Chip kann mit den vorher entwickelten molekularen Sonden bis zu 14

atisiertes und tragbares Gerät wurde sches, sowie automArten nachweisen. Ein prakti

entwickelt und ausgiebig in Kombination mit dem Mehrfach-Sonden-Chip und molekularen

Sonden für ert die Lösungen aus pe beförd getestet. Eine PeristaltikpumAlexandrium minutum

mmer, wodurch alle ungskaden Vorratsbehältern in eine Hybridisierungs- bzw. Ström

atisch auhritte der Analyse automcwichtigen Ssgeführt werden. Zusätzlich wurde erfolgreich

tes von Laien entwickelt, welches eine ein Lyse-Protokoll für die Anwendung des Gerä

t-rRNA und nuelle rRNA-Isolation erübrigt. Das Gerät wurde anhand isolierter Gesamam

bewertet und eindeutige, unterscheidbare Signale wurden Zellen aus einer Algenkultur

ermittelt. Dieses autonome System kann Proben in weniger als zwei Stunden analysieren und

rät und der Mehrfach-Sonden-eit weisen das G Freiland angewendet werden. Somauch im

Chip Potential für die Anwendung als Frühwarnsystem im Aquakulturbereich und dem

ussektor auf.Tourism

141

References

cesReferen7.

Alelio, and M. Montresor. 2005. Life cycle, size reduction ato, A., L. Orsini, D. D’Amudo-nitzschia Psepatterns, and ultrastructure of the pennate planktonic Diatom 542-556. 41:phyceae). J. Phycol a (Bacillariodelicatissimvoldsen, and D. M. Anderson. 2003. Harmful algal monitoring Andersen, P., H. Eneprogramme and action plan design, p. 627-647. InG. M. Hallegraeff, D. M. Anderson
and A. D. Cembella [eds.], Manual on Harmful Marine Microalgae. United Nations
n.Cultural Organizatio Scientific and Educational,G. M.Inbers, p. 99-129. ating cell numAndersen, P., and J. Throndsen. 2003. EstimHallegraeff, D. M. Anderson and A. D. Cembella [eds.], Manual on Harmful Marine
tific and Cultural Organization. ScienMicroalgae. United Nations Educational, Anderson, D. M. 1995. Identification of harmful algal species using molecular probes, an
emerging perspective, p. 3-13. InP. Lassus, G. Arzul, E. Erard, P. Gentien and C.
entation-nique et DocumAlgae. Tech [eds.], Harmful MarineMarcaillouLavoisier/Intercept Ltd. G.Inthodologies, p. 165-189. eAnderson, D. M., Y. Fukuyo, and K. Matsuoka. 2003. Cyst mbella [eds.], Manual on Harmful M. Hallegraeff, D. M. Anderson and A. D. CemOrganization.Marine Microalgae. United Nations Educational, Scientific and Culturalthod, 17th ed. ethod 959.08-paralytic shellfish poison, biological meAoac. 1999. Official Mists.lysis. Association of official Analytical Chemthods of AnaeOfficial MAune, T., S. Larsen, J. A. B. Aasen, N. Rehmann, M. Satake, and P. Hess. 2007. Relative pared with okadaic acid, based on acute toxicity of dinophysistoxin-2 (DTX-2) comintraperitoneal toxicity in mice. Toxicon49: 1-7.
s, J. V. Tyrrell, M. Gladstone, and C. A. Scholin. 2005. International Ayers, K., L. L. Rhodeaccreditation of sandwich hybridisation assay format DNA probes for micro-algae.
1225-1231. 39:New Zeal J. Mar. Fresh iology, public ing, A. D. Rowan, and D. G. Baden. 2003. EpidemBacker, L. C., L. E. Flemhealth and human diseases associated with harmful marine algae, p. 723-749. InG. M.
Hallegraeff, D. M. Anderson and A. D. Cembella [eds.], Manual on Harmful Marine
tific and Cultural Organization. ScienMicroalgae. United Nations Educational,

142

References

Berganza, J., G. Olabarria, R. García, D. Verdoy, A. Rebollo, and S. Arana. 2006. DNA microdevice for electrochemical detection of Escherichia coli O157:H7 molecular
.doi:10.1016/j.bios.2006.09.028rkers. Biosens. Bioelectron am.d D. Wolder, P. A. Rublee, anBowers, H. A., T. Tengs, H. B. Glasgow, Jr., J. M. BurkhPCR Assays for Rapid Detection of eent of Real-TimOldach. 2000. DevelopmPfiesteria piscicida and Related Dinoflagellates. Appl. Environ. Microbiol.66: 4641-
4648.Bratbak, G., W. Wilson, and M. Heldal. 1996. Viral control of Emiliania huxleyi blooms? J.
75-81. 9:Marine Syst Brussaard, C. P. D., B. Kuipers, and M. J. W. Veldhuis. 2005. A mesocosm study of
ics: I. Regulatory role of viruses in bloom Phaeocystis globosa population dynam 859-874. 4:control. Harmful AlgaeChen, C. Y., and H. N. Chou. 2002. A modified high-performance liquid chromatography
method for analysis of PSP toxins in dinoflagellate, Alexandrium minutum, and
715-720. 35:shellfish from Taiwan. Food Res. lnt Chepurnov, V. A. and others 2005. Sexual reproduction, mating system, chloroplast dynamics
and abrupt cell size reduction inPseudo-nitzschia pungens from the North Sea
379-395. 40:(Bacillariophyta). Eur. J. Phycol. sequencing of the large-subunit ller. 2002. Use of PCR and partial uCordova, J. L., and I. MrRNA gene to identify Alexandrium catenella (Dinophyceae) from the South of Chile. 343-350. 1:gaeHarmful Alonitoring and ent for in situ mCulverhouse, P. F. and others 2006. HAB Buoy: a new instrumearly warning of harmful algal bloom events. Afr. J. Mar. Sci 28: 245-250.
Daranas, A. H., M. Norte, and J. J. Fernandez. 2001. Toxic marine microalgae. Toxicon 39:
1101-1132.neus, and T. Ruzgas. 2005. A Dock, E., A. Christenson, S. Sapelnikova, J. Krejci, J. Eml for screen-printed eight-electrode arrays. Anal.steady-state and flow-through celChim. Acta531: 165-172.
Doucette, G. J. and others 2006. A domoic acid immunosensor onboard the Environmental
le Processor: the first steps toward rempmSaote, sub-surface phycotoxin detection, p. for the cietyternational So65-65, 12th International Conference on Harmful Algae. Instudy of harmful algae.

143

References

Edvardsen, B., and E. Paasche. 1998. Bloom dynamics and physiology of Prymnesium and
Chrysochromulina, p. 193-208. InD. M. Anderson, A. D. Cembella and H. G.M.
s. Springer. [eds.], Physiological ecology of harmful algal bloomG. Marrazza, and M. Mascini. Disposable F., F. Lucarelli, I. Palchetti, Farabullini,ous analysis of different bacterial food ultaneical genosensor for the simelectrochemcontamFernández, M. L., S. Shuminants. Biosensors and Bioelectronway, and J. Blancoics. 2003. ManagemIn Press, Corrected Proofent of shellfish resources, p. .
657-692.InHarmful Marine Phytoplankton. United Nations Educational, G. M. Hallegraeff, D. M. Anderson and A. D. Cembella [eds.], Manual on Scientific and Cultural
Organization.Figueiras, F. G., G. C. Pitcher, and M. Estrada. 2006. Harmful algal bloom dynamics in
Ecology of Harmful Alrelation to physical processes, p. 127-138. gae. EcologiIncal Studies. Springer-Verlag. E. Graneli and J. T. Turner [eds.],
Flanagan, Acytotox.icity assay for the detec F., K. R. Callanan, J. Donlon, R. Palmtion and differentiaer, A.tion of two fam Forde, and M. Kane. 2001. A ilies of shellfish
1021-1027. 39:toxins. Toxicon Franco, J. M., and P. Fernández. 1993. Separation of paralytic shellfish toxins by reversed n reaction and atography, with postcolumance liquid chromphase high performfluorimetric detection. Chromatographia35: 613-620.
Franks, P. J. S., and B. A. Keafer. 2003. Sampling techniques and strategies for coastal G. M. Hallegraeff, D. M. Anderson and A. D. In p. 51-76. s,phytoplankton blooms Educational,ella [eds.], Manual on Harmful Marine Microalgae. United NationbmCeation.l Organizd CulturaScientific anicalng, J. Tsukuda, J. Kibler, and D. A. Haake. 2005. Electrochema Ma, H. WGau, V., S.molecular analysis without nucleic acid amplification. Methods37: 78-83.
lecular techniques for oGodhe, A. and others 2007. Intercalibration of classical and m fundyense (Dinophyceae) and estimation of cell n of Alexandriumidentificatio 56-72. 6:ful Algaearmdensities. H17:e of age. Trends Biotechnol Graves, D. J. 1999. Powerful tools for genetic analysis com127-134. Medlin. 2004. Using fluorescently-and L. K.Groben, R., U. John, G. Eller, M. Lange, labelled rRreview. Nova Hedwigia 79:NA probes for hierarchical estim 313-320. ation of phytoplankton diversity – a mini-

144

References

Hallegraeff, G. M. 2003. Harmful algal blooms: a global overview, p. 25-49. InH. G.M., D.
ella [eds.], Manual on Harmful Marine Microalgae. bmM. Anderson and A. D. Ceific and Cultural Organization. United Nations Educational, ScientHarland, R., K. Nordberg, and H. L. Filipsson. 2004. The seasonal occurrence of dinoflagellate cysts in surface sediments from Koljo Fjord, west coast of Sweden - a
107-117. 128:note. Rev. Palaeobot. Palyno hiney, and N. Dennison. 2006. Can general anaesthesia be used Holtrop, G., J. Petrie, J. Mcel 336-347. 47: Toxicon for the Paralytic Shellfish Poison bioassay?Hosoi-Tanabe, S., and Y. Sako. 2005. Rapid detection of natural cells of Alexandrium tamarense and A. catenella (Dinophyceae) by fluorescence in situ hybridization.
319-328. 4:gaeHarmful Al and others 2002. Genotyping of the Apolipoprotein B R3500Q Mutation Using .Jacobsen, N 657-660. 48:es. Clin Chem Locked Nucleic Acid Capture ProbImmobilizedJohn, U., A. Cembella, C. Hummert, M. Elbrächter, R. Groben, and L. K. Medlin. 2003.
arense and A. tamination of the toxigenic dinoflagellates AlexandriumDiscrimostenfeldii in co occurringnatural populations from Scottish coastal waters. Eur. J.
25-40. 38:Phycol.John, U., L. K. Medlin, and R. Groben.2005. Development of specific rRNA probes to
arense species tamof the Alexandriumdistinguish between geographic clades 199–204. 27:plex. J. Plankton Res. com 24-32. :icsa GenomKauppinen, S. and others 2003. LNA Microarrays in Genomics. Pharmly study for parallen. 2006. A low-density oligonucleotide arraaKi, J.-S., and M.-S. Hhybridization of consensus PCR products of detection of harmful algal species using LSU rDNA D2 domain.Biosens. Bioelectron 21: 1812-1821.
Kongsbak, L. 2002. LNA: Fine-tuning of primers and probes. LNA 01: 1.
Lagier, M. J., C. A. Scholin, J. W. Fell, J. Wang, and K. D. Goodwin. 2005. An
electrochemical RNA hybridization assay for detection of the fecal indicator
1251-1261. 50: Escherichia coli. Marine Pollution Bulletinbacteriumination of paralytic atography determLawrence, J. F., and C. Ménard. 1991. Liquid chromshellfish poisons in shellfish after prechromatographic oxidation. J. Assoc. Offic.
1006-1012. 74:Anal. ChemLawrence, J. F., B. Wong, and C. Ménard. 1996. Determination of decarbamoyl saxitoxin and
atographyaphic oxidation and liquid chromatogrits analogues in shellfish by prechrom 1111-1115. 79:with fluorescence detection. J. AOAC Internat.

145

References

S. Alegret, and M. I. Pividori. 2006. In situ poy, J. Barbé, S. Hernández, o, A., S. CamLermDNA amplification with magnetic primers for the electrochemical detection of food
ios.2006.08.048doi:10.1016/j.bpathogens. Biosens. Bioelectron .hy, and Lilly, E. L., K. M. Halanych, and D. M. Anderson. 2005. Phylogeny, biogeograpspecies boundaries within the Alexandrium minutum group. Harmful Algae4: 1004-
1020.aral, D. A. Caron, and E. F. Delong. 1993. Application of rRNA-based , E. L., L. A. AmLim59:rine nanoplanktonic protists. Appl. Environ. Microbiol. aprobes for observing m1647-1655.onitoring by hybridization to high-density Lockhart, D. J. and others 1996. Expression m 1675-1680. 14:oligonucleotide arrays. Loy, A. and others 2005. 16S rRNA Gene-Based Oligonucletide Microarray for Environmental Monitoring of the Betaproteobacterial Order "Rhodocyclales". Appl.
1373-1386. 71:Environ. Microbiol. Luckas, B., C. Hummert, and Y. Oshima. 2003. Analytial methods for paralytic shell poisons,
p. 191-209. InG. M. Hallegraeff, D. M. Andersonand A. D. Cembella [eds.], Manual
on Harmful Marine Microalgae. United Nations Educational, Scientific and Cultural
Organization.r sequence data. Nucleic Acids ent fo and others 2004. ARB: a software environm.Ludwig, W 1363-1371. 32:Res life A discussion of the diatomy didn't Lund see sex in Asterionella?hMann, D. G. 1988. Went.], Algae and the Aquatic EnvironmF. E. Round [ed.Incycle in nature, p. 385-412. Biopress.atic and s (HAB); problemcroalgae bloomiMasó, M., and E. Garces. 2006. Harmful mf;In Press, Corrected Proor. Pollut. Bull . Maconditions that induce them.doi:10.1016/j.marpolbul.2006.08.006karyotic006. Picoeuiltshire, and K. Valentin. 2Medlin, L. K., K. Metfies, H. Mehl, K. H. Wlecularree Moe Series Site as Assessed by Th TimPlankton Diversity at the Helgoland 53-71. 52:crobial Ecol iMethods. Me use ofh., V. Lopez-Rodas, S. Gonzalez-Gil, A. Aguilera, and E. Costas. 1995. TMendoza, Hpolyclonal antisera and blocking of antibodies in the identification of marine
and ra against Gymnodiniumic antisedinoflagellates: species-specific and clone-specif 103-115. 186:olcAlexandrium. J. Exp. Mar. Biol. E

146

References

ical detection of the dlin. 2005. ElectrochemeMetfies, K., S. Huljic, M. Lange, and L. K. Mith a DNA-biosensor. Biosens.exandrium ostenfeldii wtoxic dinoflagellate Al 1349-1357. 20:Bioelectronphytoplankton: The fluorescentcrochips for iMetfies, K., and L. K. Medlin. 2004. DNA m 321—327. 79:wave of the future. Nova Hedwigia ssessingeir Potential Use in Aal RNA Probes and Microarrays: Th---. 2005. RibosomMicrobial Biodiversity, p. 258-278. Ina. E. H. R. Elizabeth A. Zimmer [ed.], Methods
icical Data. Academogy,Molecular Evolution: Producing the Biochemolin EnzymPress.ental Toxic Algae, Intergovernmference list ofic ReMoestrup, O. 2004. IOC Taxonom.sion of UNESCO; ioc.unesco.org/hab/data.htmisOceanographic CommMoita, M. T., P. B. Oliveira, J. C. Mendes, and A. S. Palma. 2003. Distribution of chlorophyll
associated with coastal upwelling plumes off central catenatumnodiniuma and Gym S125-S132. 24:icaPortugal. Acta OecologMouratidou, T., I. Kaniou-Grigoriadou, C. Samara, and T. Kouimtzis. 2006. Detection of the
ssels during a diarrhetic shellfish poisoning (DSP)urine toxin okadaic acid in mamical and immunological em biological, chaikos Gulf, Greece, usingepisode in Thermmethods. Science of The Total Environment366: 894-904.
entd S. Morris. 2005. Toxin profile, pigm, anento, S. M., E. L. Lilly, J. LarsenNascimposition, and large subunit rDNA phylogenetic analysis of an Alexandrium comminutum (Dinophyceae) strain isolated from the Fleet Lagoon, United Kingdom. J.
343–353. 41:Phycolcroarraysie use of LNA Oligonucleotide M Kauppinen. 2002. ThNielsen, P. S., and S. 1-3. 17:ANProvides Superior Sensitivity and Specificity in Expression Profiling. LNiemeyer, C. M., and D. H. Blohm. 1999. DNA Microarrays. Angew. Chem. Int. Ed. 38:
2865-2869.onitoringacific Coast states: m., and K. K. Chew. 1988. PSP toxins in the PNishitani, Lprogramms and effects on bivalve industries. J Shellfish Res.7: 653-669.
Oshima, Y. 1995. Postcolumn derivatization liquid chromatographic method for paralytic
528-532. 78:shellfish toxins. J. AOAC Internat.Quilliam, M. A. 2003a. Chemical methods for domoic acid, the amnesic shellfish poisoning
(ASP) toxin, p. 247-265. InG. M. Hallegraeff, D. M. Anderson and A. D. Cembella
[eds.], Manual and Harmful Marine Microalgae. United Nations Educational, ation.l Organizd CulturaScientific an

147

References

---. 2003b. Chemical methods for lipophilic shellfish toxins, p. 211-245. InG. M. Hallegraeff,

Marine Microalgae..], Manual on Harmful bella [edsD. M. Anderson and A. D. Cem

ific and Cultural Organization. United Nations Educational, Scient

rRautio, J., K. B. Barken, J. Lahdenpera, A. Beitenstein, S. Molin, and P. Neubauer. 2003.

Sandwich hybridisation assay for quantitative detection of yeast RNAs in crude cell

4. 2:lysates. Microb. Cell Fact

Rensel, J. E., and J. N. C. Whyte. 2003. Finfish mariculture and harmful algal blooms, p. 693-

bella [eds.], Manual on G. M. Hallegraeff, D. M. Anderson and A. D. CemIn722.

s Educational, Scientific and CulturalNationalgae. United Harmful Marine Micro

Organization.

Hosoi-Tanabe, and A. Uchida. 20Sako, Y., S.04. Fluorescence in situ hybridization using

nd rapid identification of the toxic dinoflagellates ple arRNA-targeted probes for sim

Alexandrium tamarense and Alexandrium catenella. J. Phycol 40: 598–605.

R. Buck, T. Britschgi, G. Cangelosi, and F. P. Chavez. 1996. Identification Scholin, C. A., K.

of Pseudo-nitzschia australis (Bacillariophyceae) using rRNA-targeted probes in

190-197. 35:ats. Phycologia mwhole cell and sandwich hybridization for

Scholin, C. A. and others 1999. DNA probes and a receptor-binding assay for the detection of

eae) species and domPseudo-nitzschia (Bacillariophycoic acid activity in cultured and

1356-1367. 35:ples. J. Phycol natural sam

natural red and-nitzschia australis in cultu quantification of Pseudo---. 1997. Detection and

populations using LNA rRNA-targeted probes. Limnol. Oceanogr 42: 1265-1272.

Scholin, C. A., E. Vrieling, L. Peperzak, L. L. Rhodes, and P. Rublee. 2003. Detection of

G. M. InHAB species using lectin, antibody and DNA probes, p. 131-163.

Hallegraeff, D. M. Anderson and A. D. Cembella [eds.], Manual on Harmful Marine

tific and Cultural Organization. ScienMicroalgae. United Nations Educational,

Silahtaroglu, A. N., N. Tommerup, and H. Vissing. 2003. FISHing with locked nucleic acids

(LNA): evaluation of different LNA/DNA mixmers. Mol. Cell. Probe17: 165-169.

Silver, M. W. 2006. Protecting Ourselves from Shellfish Poisoning. American Scientist 94:

316-325.

N.,on,Simdsen, and L. K. Medlin. 1997. The identification of J. Brenner, B. Edvar

nesiophyceae) usingyptophyta, Prymnesium species (Hmulina and PrymChrysochro

eans for iminescent oligonucleotide probes: a milumfluorescent or chemproving

393-401. 32:r. J. Phycol. ustudies on toxic algae. E

148

References

Simon, N. and others 2000. Oligonucleotide probes for the identification of three algal groups

76-84. 47: whole-cell hybridization. J. Eukaryot. Microbiol by dot blot and fluorescent

Smayda, T. J. 2003. Environmental monitoring with examples from Narragansett Bay, p. 595-

bella [eds.], Manual on G. M. Hallegraeff, D. M. Anderson and A. D. CemIn625.

s Educational, Scientific and CulturalNationalgae. United Harmful Marine Micro

Organization.

Smcrobial ecology. icular tools and approaches in eukaryotic mit, E. and others 2004. Mole

Bios Scientific Publishers.

nyamSournia, A., M. Chretiennot-Dinet, and M. Ricard. 1991. Marine pyhtoplankton: how

e world oceans. J Planktospecies in th 1093-1099. 13:n Res

Steidinger, K. A., and E. Garcés. 2006. Importance of life cycles in the ecology of harmful

imE. Granéli and J. T. Turner [eds.], Ecology of Harmful Algae. Incroalgae, p. 37-49.

udies. Springer-Verlag. tEcological S

lipid breakdown for portance ofSvensson, S., and L. Förlin. 2004. Analysis of the im

elimination of okadaic acid (diarrhetic shellfish toxin) in mussels, Mytilus edulis:

results from a field study and a laboratory experiment. Aquatic Toxicology

418.

66: 405-

Taylor, A. D., J. Ladd, Q. Yu, S. Chen, J. Homoloa, and S. Jiang. 2006. Quantitative and

elannens with a multi-chultaneous detection of four foodborne bacterial pathogsim

SPR sensor. Biosens. Bioelectron 752-758. 22:

Taylor, F. J. R., and Y. Fukuyo. 1998. The neurotoxigenic dinoflagellate genus Alexandrium

Halim: General Introduction, p. 381-404. InD. M. Anderson, A. D. Cembella and H.

mful Algae BloomG.M. [eds.], Physiological Ecology of Hars.

Taylor, F. J. R., Y. Fukuyo, and J. Larsen. 1995. Taxonomy of harmful dinoflagellates, p.

G. M. Hallegraeff, D. M. Anderson and A. D. Cembella [eds.], Manual on In283-317.

Harmful Marine Microalgae. IOC UNESCO.

Laycock. 1991. Analysis of paralytic shellfish poisons .Thibault, P., S. Pleasance, and M. V

483-501. 542:atography A by capillary electrophoresis. Journal of Chrom

eration for toxic ated detection and enumTöbe, K., G. Eller, and L. K. Medlin. 2006. Autom

nesiumtroduction of a new probe for Prymetry and the inalgae by solid-phase cytom

parvum (Haptophyta: Prymnesiophyceae). J. Plankton Res. 28: 643-657.

Tyrrell, J. V., L. B. Connell, and C. A. Scholin. 2002. Monitoring for Heterosigmashiwoka a

205-214. 1:gaetion assay. Harmful Alusing a sandwich hybridiza

gaetion assay. Harmful Alusing a sandwich hybridiza

149

References

e genotyping Ugozzoli, L. A., D. Latorra, R. Pucket, K. Arar, and K. Hamby. 2004. Real-tim

143-324:with oligonucleotide probes containing locked nucleic acids. Anal. Biochem

152.

en Phytoplankton Methmmnung der quantitativöhl, H. 1958. Zur VervollkoUtermodik. Mitt.

8. 1-39:Int. Ver. Limnol.

ination of paralytic shellfish toxins in yo. 2001. DetermpaVale, P., and M. A. De M. Sam

ated pre-columPortuguese shellfish by autom 561-571. 39:n oxidation. Toxicon

Venrick, E. 1999. Phytoplankton species structure in the central North Pacific 1973-1996:

variability and persistence. J. Plankton Res. 21: 1029-1042.

W06. Rapid as, P. Lebaron, and H. Moreau. 20st, N. J., R. Bacchieri, G. Hansen, C. Thome

in natural samples by use of a specific Prymnesium parvumdetection of the toxic alga

72:etry. Appl. Environ. Microbiol. -phase cytommonoclonal antibody and solid 860-

868.

Yasumoto, T., Y. Oshima, and M. Yamaguchi.1978. Occurrence of anew type of shellfish

poisoning in the Tohuku District. Bull. Jap. Soc. Sci. Fish 1249-1255. 44:

Ye, R. W., T. Wang, L. Bedzyk, and K. M. Croker. 2001. Applications of DNA microarrays

in microbial systems. J. Microbiol. Meth 47: 257-272.

acle. 1995. DNA probe Zammatteo, N., P. Moris, I. Alexandre, D. Vaira, J. Piette, and J. Rem

hybridisation in microwells using a new bioluminescent system for the detection of

PCR-amplified HIV-1 proviral DNA. Virol. Methods55: 185-197.

. 1997. Analysis of domZhao, J., P. Thibault, and M. A. Quilliamers in acid and isomoi

268-276. 18:seafood by capillary electrophoresis. Electrophoresis

allengehs: a cvoldsen, O. 2000. The diversity of harmful algal bloomZingone, A., and H. Ene

for science and management. Ocean Coast Manage 43: 725-748.

ent. Ocean Coast Manage nagemafor science and m

150

Danksagung

8.sagungDank

Als erstes möchte ich mich bei Linda Medlin bedanken, die es mir ermöglicht hat diese

Labor und ihrer Arbeitsgruppe durchführen. Vielen Dank für Deine Dissertation in ihrem

ch fortzubilden iagen, die Geduld und die Zeit, für die Möglichkeit mrUnterstützung in allen F

und Vorträge zu halten. Danke.

Weiterhin möchte ich mich herzlich bei Herrn Prof. Dr. rer. nat. Gunter O. Kirst and Herrn

Dissertation bedanken. inerebella für die Begutachtung mProf. rer. nat. Allan. D. Cem

Ich möchte mich ganz herzlich bei den jetzigen und ehemaligen Mitgliedern der

Arbeitsgruppe von Linda Medlin sowie anderer Gruppen bedanken, für die tolle

Zusammenarbeit, die vielen anregenden Diskussionen, den Rat und die Hilfe bei vielen

aFragen und natürlich nicht zu vergessen für die Motivation wenn mhr ging: el so gar nichts m

, Monica, Andrea, Dick, Georgia, Megan, rin, Shinya Sara, Bank, KatKlaus, Ines, Uwe,

e.d ChristinSabine, Helga, Jessica, Steffi, Kerstin, Katja un

Ganz herzlicher Extra-Dank gehört meinerwie minen Eltern Karin und Johann Diercks, soe

a Käthe und Kai Horn für ihr unvergleichliches Vertrauen in mOm

nde Unterstützung.erwähreimm

ch und ihre i

Diese Disseration entstand im Rahmen des EU-Projektes ALGADEC (COOP-CT-2004-

s demmenprogra508435-ALGADEC) des 6. Rahmr Europäischen Union und durch die

egener-Instituts für Polar- und Meeresforschung.Förderung des Alfred-W

151

Un pour Un
Permettre à tous d'accéder à la lecture
Pour chaque accès à la bibliothèque, YouScribe donne un accès à une personne dans le besoin