SINGLE PARTICLE DETECTION

A Diagnostic Tool for Particle Associated Diseases like Alzheimer’s Disease and

Creutzfeldt-Jakob Disease

Eva Birkmann*

;

, Susanne Aileen Funke**

, Detlev Riesner* and Dieter Willbold*

*Institut fuer Physikalische Biologie, Heinrich-Heine-Universitaet, Universitaetsstr. 1,40225 Dueselorf, Germany

**INB-2, Forschungszentrum Juelich, Juelich, Germany;

These authors contributed equally to this work.

Keywords: Fluorescence correlation spectroscopy, Alzheimer’ disease, Creutzfeldt-Jakob disease, single particle

detection.

Abstract: Neurodegenerative diseases like Alzheimer’s disease (AD), prion diseases and others are progressive and

lethal. High-molecular weight aggregates of the Amyloid-β-peptides (Aβ) or of the misfolded prion protein

(PrP) are found in patients afflicted by AD or prion diseases, respectively. Despite of many attempts, neither

a therapy for recovery, nor an early diagnosis at preclinical stages are available. Psychological tests and

imaging approaches not directly related with a secure disease marker are in use only for late stages of the

disease. The Creutzfeldt-Jakob-disease (CJD), a human prion disease, is caused by accumulation of

aggregates consisting of an abnormally shaped version of PrP. CJD is diagnosed with certainty only by

neuropathology post mortem. In this study a multidisciplinary development of a novel mode of single

particle counting of immobilized Aβ and PrP aggregates as the most direct biomarkers for Alzheimer’s

disease and Prion diseases, respectively, is introduced. For ultrasensitive detection of aggregates, the

suitable instrumentation as well as data acquisition and data analysis are developed using single molecule

detection and advanced laser scanning fluorescence techniques. In the novel assay development effort

biochemistry, detection and analysis were improved to detect single aggregates immobilised on a surface.

First results show the improvement of single particle detection of PrP-aggregates of TSE-afflicted cattle and

hamsters as well as synthetic Aβ-aggregates.

1 INTRODUCTION

In many neurodegenerative diseases e.g. prion

diseases, Alzheimer’s disease, Parkinson’s disease,

Huntington’s Disease, protein aggregates are formed

in the very beginning or in the progress of disease

(Lee et al., 2005). Up to now it is not known, if these

aggregates are causative agents or symptoms of the

respective disease, but many studies show, that the

aggregates or even oligomers of the according

proteins are neurotoxic and therewith a reason of

neurodegeneration. (Selkoe, 2003)

To understand the progression of these diseases,

as well as disease associated or causative

mechanisms and to monitor potential therapeutically

approaches an ultrasensitive tool to quantify these

disease related aggregates is required. A challenge

for the analytic system is to reliably count only those

aggregates or oligomers that consist of the specified

protein or peptide. Monomeric molecules need to be

clearly distinguished because they are present in

healthy organism as well.

We developed a new method, which is able to

count single protein aggregates bound by a capture-

antibody to a surface (surface-FIDA) (Birkmann et

al., 2007). Our new test system is based on

fluorescence correlation spectroscopy (Eigen and

Rigler, 1994). It is quantifying the number and size

of aggregates simultaneously labelled by two

different antibodies for dual colour fluorescence

intensity distribution analysis (2D-FIDA) (Birkmann

et al., 2006). Only aggregates and oligomers but not

monomeric proteins are counted. To increase the

sensitivity, particles were concentrated in the two-

dimensional space by immobilizing it to capture

antibodies on the surface of the slide. Laser beams

are scanning the surface systematically, so even

single particles are detected (Birkmann et al., 2007).

431

Birkmann E., Aileen Funke S., Riesner D. and Willbold D. (2008).

SINGLE PARTICLE DETECTION - A Diagnostic Tool for Particle Associated Diseases like Alzheimer’s Disease and Creutzfeldt-Jakob Disease.

In Proceedings of the First International Conference on Bio-inspired Systems and Signal Processing, pages 431-436

DOI: 10.5220/0001065104310436

 SciTePress

We report on the successful use of surface-FIDA

as diagnostic tool for prion diseases. The infectious

agents of prion diseases are composed primarily of

the pathogenic isoform of the prion protein

designated PrP

, which is generated by a

conformational change of the cellular isoform PrP

In contrast to its cellular isoform, the pathogenic

isoform PrP

forms insoluble aggregates. Hitherto

accredited prion tests use the PK-resistance of PrP

as a marker for the disease. Because of varying

portions of disease related aggregated PrP, which is

not PK-resistant, these prion tests offer only a

limited sensitivity. Therefore prion protein aggregate

detection, which does not rely on PK-digestion, is

favourable for sensitive diagnosis. It allows

detection of both, PK-resistant and PK-sensitive

PrP

aggregates.

Up to now, we could successfully verify the

novel test system for correct diagnosis of Scrapie-

infected hamsters as well as BSE-infected cattle in

the clinical stages of diseases (Birkmann et al.,

2007). Furthermore, we were able to detect PrP

aggregates in the cerebrospinal fluid of cattle of

BSE-infected cattle for the first time (Birkmann et

al., 2007). During the next steps we will adopt the

highly sensitive test system for diagnosis of human

prion diseases like Creutzfeldt-Jakob disease and

other aggregate related diseases, especially

Alzheimer’s disease.

In this study we apply surface-FIDA to different

disease associated aggregates. First we show the

single aggregate detection of prion protein

aggregates purified from brain homogenates of

Scrapie-infected hamsters and BSE-infected cattle to

demonstrate the principal of surface-FIDA to detect

single particles. In the second part of the work we

show the transfer of surface-FIDA to the detection

of single Aβ aggregates as diagnostic approach for

Alzheimer’s disease. Therefore we compared the

detection of Aβ aggregates in solution with the

application of surface-FIDA.

2 MATERIALS AND METHODS

2.1 Fluorescence Labelling of

Antibodies

Antibodies R1 were kindly provided by S.B.

Prusiner, UCSF, USA (Williamson et al., 1998).

Antibodies 12F10 and Saf32 were obtained from

SpiBio (Massy Cedex, France); antibody D18 was

obtained from InPro (San Francisco, USA). For the

detection of Aβ aggregates, antibodies 6E10 (Sigma

Aldrich, Hamburg, Germany), 8G7, 19H11 and 4G8

(Calbiochem) were purchased.

Antibodies were labelled in free amino groups

via reactive succinimidyl ester groups of Alexa-633

and Alexa-488 (Molecular Probes, Oregon, USA).

For labelling, approximately 50 µg antibodies were

incubated with 5 µg dye in carbonate buffer, pH 8.4

in a total volume of 100 µl for 1 hour. Conjugates

were separated from free dye by gel filtration via

NAP5-column (Pharmacia) with 10 mM TBS, pH

7.2 and 0.2 M NaCl as elution buffer. Labelled

antibodies were stored in the dark at 4°C.

2.2 Fluorescence Correlation

Spectroscopy

In fluorescence correlation spectroscopy (FCS) the

fluorescence intensity is recorded in a very small

volume, i.e. in the femtoliter range. Measurements

were performed with the instrument FCS Olympus

IX 50 (Evotec OAI, Hamburg, Germany) with a

beam scanner unit in dual-colour mode with an

Argon ion laser (excitation wavelength 488/514 nm)

and a helium-neon laser (excitation wavelength 633

nm). The beam scanner unit allows the scanning of

the sample for aggregates. In practice the detection

volume is moved through the sample in horizontal

and vertical dimensions. The beam scanner was used

by moving 1 mm in one direction a rectangular

deviation of 100 µm with a frequency of 50 Hz and

an integration time of 50 µs. A piezo element was

integrated in the optic of the FCS Olympus IX 50,

which allowed a precise z-positioning of the focus in

the 100 nm range.

2.3 Surface-FIDA

The surface-FIDA assay was carried out as

described by earlier (Birkmann et al., 2007).

Briefly, 0.25 – 1 μg capture antibody was

adhesively bound to poly-D-lysine activated glass

surfaces. After blocking the unspecific binding sites

with 10 % fetal calf serum, potentially present

aggregates were bound to the capture by incubating

20 μl of a sample for at least two hours at 4 °C.

After washing twice with PBS buffer (140 mM

NaCl; 2.7 mM KCl; 10 mM Na

HPO

, pH 7.4), the

fluorescence labelled detection antibodies were

applied (0.1 μg/μl) and incubated for 1 h at 20 °C.

After five washing steps with PBST (PBS with 0.1

% (w/w) Tween 20) and two washing steps with

PBS, the measurements were carried out.

BIOSIGNALS 2008 - International Conference on Bio-inspired Systems and Signal Processing

432

2.4 Preparation of Synthetic

Aß-aggregates

Aβ(1-42) was purchased from JPT Peptide

Technologies (Berlin, Germany). For aggregate

preparation Aβ was dissolved in DMSO to 400 µM,

diluted to 66 µM in PBS (140 mM NaCl; 2.7 mM

KCl; 10 mM Na

HPO

) and incubated for five days

at 37 °C. Aggregate formation was monitored using

Thioflavin T (ThT) assay. For that, 10 mM ThT

(Sigma, Hamburg, Germany) was added to the

samples. Fluorescence was monitored with a

microplate reader at excitation and emission

wavelengths of 440 nm and 490 nm, respectively

(Polarstar Optima, BMG, Offenburg, Germany).

For surface-FIDA, the aggregates were diluted

1:10 in pooled CSF of healthy people. CSF was

obtained by Biochemed Services, Winchester,

USA).

3 RESULTS

3.1 Methodical Setup

The new optical method for detection of protein

aggregates is based on fluorescence intensity

distribution analysis (FIDA). For detection of

pathologic protein aggregates single molecule

detection (SMD) combined with quantification of

single aggregates immobilised on a relatively large

surface was employed to achieve high sensitivity

and specificity.

The new method, therefore, is called surface-

FIDA. It is able to count single protein particles

bound to a capture-antibody on the surface (fig. 1a)

(Birkmann et al., 2007). Specific protein-particles

are labelled simultaneously by two different

antibodies for dual colour fluorescence intensity

distribution analysis (2D-FIDA). Among the capture

and both detection antibodies, at least two antibodies

bind the same epitope. Thus, only aggregates but not

monomers are counted. A laser beam scans the two-

dimensional surface systematically in a double-

meander mode. Thus, even single protein-particles

are detected (fig. 1B). By utilising two lasers

simultaneously two different fluorescence labels can

be crosscorrelated. Only if the different labels are

bound to the same aggregate both labels occur to the

same time in small detection focus. A typical

distribution of a coincident signal of double labelled

aggregates is shown in fig. 1C.

3.2 Detection of Pathological Protein

Particles with Surface-FIDA

To observe, if the surface-FIDA setup is able to

detect single aggregates different types of protein

aggregates were tested.

3.2.1 PrP-particles Purified from Brain of

Scrapie Infected Hamsters

Prion Protein aggregates were purified from brain

homogenates of Scrapie infected hamsters in the

clinical state of disease by NaPTA precipitation

(Safar et al., 1998). The antibody R1 (Williamson et

al., 1998) served as capture. The antibodies D13 and

R1 were fluorescence labelled and utilized as

detection probes. Same treated brain homogenates of

healthy hamsters were used as control samples. The

results of 2D-surface-FIDA in different distances to

the surface are shown in fig. 2. In the samples of

Scrapie infected hamsters at all distances between

10 µm and 20µm fluorescence peaks with high

fluorescence intensity in both channels could be

detected. At distances below 10 µm background

signal in the control sample rise and in distances

above 20 µm the signals of the Scrapie samples

descended (data not shown).

Figure 1: A) Scheme of surface-FIDA; B) fluorescence

peak caused by the labelled aggregate: principal of particle

counting; C) 2D-FIDA:Two probes with different

fluorescence labels were used; Only simultaneous binding

of both probes to the aggregates results in the specific

diagonal signal distribution as shown in the plot

SINGLE PARTICLE DETECTION - A Diagnostic Tool for Particle Associated Diseases like Alzheimer’s Disease and

Creutzfeldt-Jakob Disease

433

Figure 2: 2D-surface-FIDA of PrP-aggregates purified

from brain homogenate of Scrapie infected hamsters and

same treated brain homogenate of a healthy control in

different distances to the surface 10-20 µm.

3.2.2 PrP-particles Purified from Brain of

BSE Infected Cattle

Prion Protein aggregates were purified from brain

homogenates of BSE infected cattle in the clinical

state of disease by NaPTA precipitation (Safar et al.,

2002). The antibody Saf32 (Krasemann et al., 1999)

served as capture. The antibodies 12F10 and Saf32

were fluorescence labelled and utilized as detection

probes. Same treated brain homogenates of healthy

cattle were used as control samples. The results of

2D-surface-FIDA at different distances to the

surface are shown in fig. 3.

3.2.3 Synthetic Aβ-Aggregates

As a first proof of principle, synthetical Aβ

aggregates were used in the assay described above.

First measurements were done in solution without

immobilizing the aggregates. The antibodies 6E10

(N-terminal epitope) and 8G7 (C-terminal epitope)

were fluorescence labelled and used as detection

probes. Experiments were done in PBS. In the

negative control, 0.2 % SDS was used to prevent Aβ

aggregation, as monitored by ThT assay (data not

shown). As expected, only aggregated Aβ resulted in

fluorescence peaks as can be seen in fig. 4a.

Figure 3: 2D-surface-FIDA of PrP-aggregates purified

from brain homogenate of BSE infected cattle and same

treated brain homogenate of a healthy control.

In a next step, Aβ aggregates diluted 1:10 in CSF to

meet realistic conditions were immobilized on the

surface of the slide. Antibody 4G8 (binding to amino

acids 1-17 of Aβ) served as capture. The antibodies

6E10 and 19H11 were fluorescence labelled and

served as detection probes. As both antibodies bind

to the N-terminal part of Aβ, a simultaneous

labelling of Aβ monomers was excluded. As

controls, only CSF without additional Aβ aggregates

was used in the immobilization procedure. The

results of 2D-surface-FIDA are shown in fig. 4b.

The measurements were done at 5 µm distance to

the surface.

4 CONCLUSIONS

The proof of principle for the use of surface-FIDA to

detect aggregates was shown for natural PrP-

aggregates purified from brain of Scrapie infected

hamsters, BSE infected cattle and for synthetic Aβ

aggregates diluted in CSF.

Single particle counting as diagnostic tool is

more sensitive as compared to measuring the

integrated signal of all or many particles. “Single

particle counting” allows measuring of multiple

BIOSIGNALS 2008 - International Conference on Bio-inspired Systems and Signal Processing

434

fluorescence intensity

6H10-alexa 488

fluorescence intensity

19H11-alexa633

control

Aβ aggregates

0 100 200

100

200

0 100 200

100

200

fluorescence intensity

6H10-alexa 488

0 100 200

100

200

100

200

0 100 200

fluorescence intensity

19H11-alexa633

fluorescence intensity

8G7-alexa633

fluorescence intensity

8G7-alexa633

Aβ aggregates

control

fluorescence intensity

6H10-alexa 488

fluorescence intensity

19H11-alexa633

control

Aβ aggregates

0 100 200

100

200

0 100 200

100

200

100

200

0 100 200

100

200

0 100 200

100

200

100

200

fluorescence intensity

6H10-alexa 488

0 100 200

100

200

0 100 200

100

200

100

200

100

200

0 100 200

100

200

100

200

100

200

0 100 200

fluorescence intensity

19H11-alexa633

fluorescence intensity

8G7-alexa633

fluorescence intensity

8G7-alexa633

Aβ aggregates

control

Figure 4: A) 2D-FIDA of synthetically prepared Aβ

aggregates in solution (concentration 3,3 µM). Aβ, kept

from aggregation by 0.2 % SDS, was used as control.

B) 2D-surface-FIDA of synthetically prepared and in CSF

diluted Aβ aggregates (concentration 6µM). CSF without

additional Aβ aggregates was used as control.

Measurements were done at 5 µm distance to the slide

surface.

parameters of the individual particles are recorded

like size, number of epitopes, different epitopes on

the same particle etc. and those parameters can be

used for improvement of specificity.

When the detection of single particles was

carried out in suspension using the dual colour

fluorescence intensity distribution analysis (2D-

FIDA) (Birkmann et al., 2006), it was done in a

small volume taken from a much larger sample

volume by moving the laser detection focus through

a cuvette. Diffusion of the particles and scanning of

the volume were superimposed so that it was

difficult to account quantitatively for all particles in

the sample. Therefore the immobilisation of the

particles on a surface had a major impact of the

sensitivity of the whole assay, because it allows

searching for the particles in a systematical way.

In the near future, we will develop surface-FIDA

into an ultrasensitive diagnostic assay for particle

associated disease, especially CJD and Alzheimer’s

disease. Such an assay will allow early diagnosis of

AD and CJD using a minimally invasive approach in

the living patient. In addition, such a diagnostic tool

will be crucial for on line monitoring of disease

progression and progress of a therapeutic approach.

Table 1: Sensitivity and Specificity Characteristics of

surface-FIDA.

sensitivity specificity

Concentration of

particles on two

dimensional surface

Simultaneous binding of

three probes (one capture,

two detection probes)

Single particle detection Adjustable washing steps

Reproducible and

complete counting of all

aggregates by surface

scanning

Detection of protein

aggregates only, no

monomers

ACKNOWLEDGEMENTS

This work was supported by the Ministry for

Environment and Nature Protection, Agriculture and

Consumer Protection of North Rhine-Westphalia

(grant VI-1-17.90.01), Forschungs- und

Innovationfonds of the Heinrich-Heine-University of

Duesselsdorf and EU NoE Neuroprion. We

gratefully acknowledge Dr. Prusiner for supplying

the antibody R1, D13 and the scrapie brain material

with negative controls. We would also like to thank

Dr. M. Groschup, Friedrich-Loeffler-Institut (FLI),

Institute for Novel and Emerging Infectious

Diseases, for supplying BSE material of infected

cattle and negative controls.

REFERENCES

Birkmann, E., Schäfer, O., Weinmann, N., Dumpitak, C.,

Beekes, M., Jackman, R., Thorne, L., Riesner, D.,

2006. Detection of prion particles in samples of BSE

and scrapie by fluorescence correlation spectroscopy

without proteinase K digestion. Biol. Chem. 387, 95–

102.

Birkmann, E., Henke, F., Weinmann, N., Dumpitak, C.,

Groschup, M., Funke A., Willbold, D., Riesner, D.

2007. Counting of single prion particles bound to a

capture-antibody surface (surface-FIDA)

Vet. Microbiol. (2007), in press

Eigen, M. and Rigler, R. (1994). Sorting single molecules

– application to diagnostics and evolutionary

biotechnology. Proc. Natl. Acad. Sci. USA 91,

5740–

5747.

Krasemann S, Jurgens T, Bodemer W., 1999. Generation

of monoclonal antibodies against prion proteins with

an unconventional nucleic acid-based immunization

strategy. J. Biotechnol. 73, 119-29.

Lee C, Yu MH., 2005 Protein folding and diseases.

J Biochem Mol Biol. 2005; 38(3), 275-80

Safar, J., Wille, H., Itri, V., Groth, D., Serban, H., Torchia,

M., Cohen, F.E., Prusiner, S.B., 1998. Eight prion

SINGLE PARTICLE DETECTION - A Diagnostic Tool for Particle Associated Diseases like Alzheimer’s Disease and

Creutzfeldt-Jakob Disease

435

strains have PrP(Sc) molecules with different

conformations Nat. Med. 10, 1157-1165.

Safar, J.G., Scott, M., Monaghan, J., Deering, C.,

Didorenko, S., Vergara, J., Ball, H., Legname, G.,

Leclerc, E., Solforosi, L., Serban, H., Groth, D.,

Burton, D.R., Prusiner, S.B., and Williamson, R.A.,

2002. Measuring prions causing bovine spongiform

encephalopathy or chronic wasting disease by

immunoassays and transgenic mice. Nat. Biotechnol.

20, 1147-1150.

Selkoe, D.J. 2003, Folding proteins in fatal ways

Nature 426; 900-904

Williamson, R.A., Peretz, D., Pinilla, C., Ball, H.,

Bastidas, R.B., Rozenshteyn, R., Houghten, R.A.,

Prusiner, S.B., Burton, D.R., 1998. Mapping the prion

protein using recombinant antibodies. J. Virol. 72,

9413-8.

BIOSIGNALS 2008 - International Conference on Bio-inspired Systems and Signal Processing

436