Development of a fluorescence correlation spectroscopy method for the study of biomolecular interactions

DEVELOPMENT OF A FLUORESCENCE
CORRELATION SPECTROSCOPY METHOD FOR
THE STUDY OF BIOMOLECULAR INTERACTIONS
HWANG LING CHIN
(B.Sc.(Hons),NUS)
A THESIS SUBMITTED FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
DEPARTMENT OF CHEMISTRY
NATIONAL UNIVERSITY OF SINGAPORE
2006
This work was perform ed in the Department of Chemistry at the National
Univ ersity of Singapore (NUS), under the supervision of Dr. Thorsten Wohland,
bet ween July 2002 and August 2006, and in the Laboratoire d’Optique Biomédicale
at the Ecole Polytech nique Fédérale de Lausann e (EPFL) under the supervision
of Prof. Theo L asser, between April 2004 an d A pril 2005.
Theresultshavebeenpartlypublishedin:
Hwan g, L. C., and T. Wohland . 2004. Dual-color Fluorescence Cross-correlation
Spectroscopy Using S ingle Laser Wavelength Excitation. Chem . Phys. C hem.
5:549—551.
Hwang, L. C., and T. Wohland. 2005. Single Wav elength Excitation Flu-
orescence Cross-correlation Spectroscopy with Spectrally Similar Fluorophor es:
Resolution fo r B inding Stu dies. J. C hem. P hys. 122: 114708 (1—11).
Hwang,L.C.,M.Leutenegger,M.Gosch,T.Lasser,P.Rigler,W.Meier,and
T. Wohland. 2006. Prism-ba sed M ulticolor Fluo rescence Correlation Spectrome-
ter. Opt. Lett. 31:1310—1312.
Hwan g, L. C., M. Gösch, T. Lasser and T. Wohland. 2006. Simultaneous
Multicolo r Fluorescence Cross-Correlation Spectroscop y to Detect Higher Order
Molecu lar Interactions Using S ingle Wavelength Laser Excitatio n. Biophys. J.
91:715-727
i
Acknowledgements

Adoctoralthesislikethiswouldnothavebeenpossiblewithoutthehelpofmany
people. I would like to ac knowledge thanks to individuals who have contributed
in on e way o r another in helping me com p lete this work.
I would lik e to thank my supervisor Dr. Thorsten Wohland for offering me this
in teresting project and supporting me throughout this researc h. His incredible
patience, in valuable guidance and encouragement have greatly benefited me and
this work. I am also than kfu l to Prof. T heo Lasser w ho supported me du ring
the time I wa s a visiting PhD studen t in his laboratory. His discussions and
suggestions relating to optics w ere o f g reat h elp to my work.
I am grateful to all m y colleagues from the Bioph ysical Fluorescence Labora-
tory in NUS. In particular, Yu Lanlan and Liu Ping who hav e pro vided me with
assistance and comm ents relatin g to chem istry and b iology. I am also grateful to
m y colleagues from the LOB, Michael Gösch for guidance and assistance in get-
ting the o p tical co m ponents for t h is p roject; Marcel L eu ten egger f or his scientific
discussions and proposals that have con tributed to the prism setu p ; Per Rigler
for his nanocon tainer s and discussion s on FCS and chemistry; Ram ach and ra Rao,
Kai Hassler and Jelena Mitic for their friendship and support; A d rian Bac h m a nn ,
An tonio Lopez and Alexandre Sero v for technical help; and Judith Chaubert for
administrative support in Switzerland.
Last but n ot least, I would like to tha nk my parents an d siblings for their love
and concern; and m y boyfriend Kang Yong for his understanding and support that
have been indispensable over t hese years.
ii
Table of Contents
Ac knowledgemen ts ii
Summary vi
List of Tables viii
List of Figures ix
List of Symbols xi
1Introduction 1

2 Theory and Setup 11
2.1 Fluorescence C or relation S pectroscop y . 11
2.1.1 Theautocorrelationfunction 11
2.1.2 Translational Diffusion 17
2.2 Fluorescence C ro ss-cor relation Spectroscopy . . . 19
2.2.1 Thecross-correlationfunction 19
2.2.2 Fittingofmodelstothecorrelationdata 24
2.2.3 Geometryofdetectionvolumes 24
2.2.4 SW-FCCSSetup 25
3 Dual-color SW-F CCS 28
3.1 Introduction 28
3.2 Theory 29
3.3 MaterialsandMethods 31
3.4 ResultsandDiscussion 32
3.4.1 Characterization o f fluorophores 32
3.4.2 SW-FCCSexperimentsofstreptavidin-biotinbinding 37
3.5 Conclusion . 42
4 Resolution of SW-F C CS 43
4.1 Introduction 43
4.2 Theory 44
4.2.1 Receptor-ligandcomplexes 44
4.2.2 TheCross-correlationfunction 48
4.2.3 Thestreptavidin-biotinreceptor-ligandsystem 50
4.2.4 CalculationsofSW-FCCSlimits 52
4.3 MaterialsandMethods 53
iii
4.4 ResultsandDiscussion 54
4.4.1 InfluenceofthedissociationconstantonSW-FCCS 55
4.4.2 InfluenceofimpuritiesonSW-FCCS 55
4.4.3 Influenceofcross-talkandquenchingonSW-FCCS 57

4.4.4 InfluenceofreceptorlabelingonSW-FCCS 59
4.4.5 SW-F CCS with spectrally similar fluorophores on the strepta vid in-
biotinsystem 61
4.4.6 Com pa rison of s en sitivities of different fluorophore pair sys-
tems 64
4.4.7 Possible fluorophorepairsforSW-FCCS 65
4.4.8 AcomparisonbetweenFCSandSW-FCCS 66
4.5 Conclusion . 67
5 Multicolor SW -FCCS 6 9
5.1 Introduction 69
5.2 Theory 70
5.2.1 Cross-correlationoftriplespecies 70
5.2.2 Case 1: R + L
g
+ L
y
→ RL
g
+ L
y
74
5.2.3 Case 2: R + L
g
+ L
y
→ RL
y
+ L
g
75

5.2.4 Application of theory to streptavidin-biotin binding system . 75
5.3 MaterialsandMethods 76
5.3.1 Opticalsetup 76
5.3.2 Chemistry 78
5.4 ResultsandDiscussions 78
5.4.1 Characterization o f fluorophoresforSW-FCCS 78
5.4.2 Calibrationexperiments 83
5.4.3 Experimentalresultsofstreptavidin-biotinbinding 84
5.4.4 Correlationsoftriple-colorcomplexes 84
5.4.5 Fittinganalysisoftriple-colorcomplexes 85
5.4.6 Correlations of com plexes w ith alternate ligand bindin g . . . 89
5.4.7 Fittin g analysis of complexes with alterna te liga nd binding . 9 2
5.4.8 LimitationsofSW-FCCS 94
5.4.9 Simulations of cross-correlation amplitudes for differen t re-
actionmodels 95
5.4.10 ApplicationsofmulticolorSW-FCCS 104
5.5 Conclusion 108
6 Prism-based Fluorescence Correlation Spectrometer 110
6.1 Introduction 110
6.2 MaterialsandMethods 112
6.2.1 Prismspectrometer 112
6.2.2 C alibr atio n with a single optic fiber 116
6.2.3 Calibration with an optic fiberarray 117
6.2.4 Correlation ex periments with fiberarray 121
6.3 ResultsandDiscussion 122
6.3.1 Correlationexperiments 122
6.3.2 Designofprismspectrometer 123
6.4 Conclusions 125
iv
6.5 Appendix:Zemaxsimulations 127

7 Conclusions and Outlook 129
Bibliograph y 135
v
Summ ary
The objective of this thesis was to develo p a single laser wavelength fluorescence
cross-correlation spectroscop y m eth od (SW-FCC S ) for the excitation of two or
more fluorescent probes. T he dev elop m ent and t esting of th e method was per-
formed in different stages. The first part of the thesis, from chap ters 2 to 4,
describes the theory and optical setup of SW-FC C S. The experimental implem en -
tation was dem o nstrated w ith the recep tor-ligan d m odel of streptavid in-biotin .
Different fluorophore assays including quantu m d ots, tandem dyes an d organ ic
dyes were tested on the system. T he resolution limit of the SW-FCC S was evalu-
ated with spectrally similar fluorophores. The second part of the thesis in chapters
5 and 6 extended the me thod to multicolor cross-correlation analysis with three
detection c h ann els. This was dem on strated first with conventional optica l filte r
cascades and then with a dispersiv e prism for spectral separation . T h e SW-FC C S
method simplifies the setup considerably without the need for aligning t wo laser
beams or expensive laser systems for two-pho ton excitation.
Cha pte r 1 provides a literature review on single molecu le fluorescence t ech-
niques relating to its application s in biomolecular interactio ns. The fluorophores
and th e recep tor-ligan d binding system used in this thesis wer e also review ed .
Chapter 2 describes the theory and the experimental setup of FC S and dual-
color SW-FCCS.
Chapter 3 investigates the feasibilit y of performing F C CS with a single laser
excitation wavelength. L o ng Stokes shift fluorophores such as tandem dyes, quan-
tum red and quan tum dots were tested on the setup and the strepta vid in-biotin
vi
binding system w as used as a p roof-of-principle. Experimental cross-co rrelation
functions were obtained and their amplitudes fitte d with a bimolecula r binding
model. The fluorophore pair of quan tum red/fluorescein produced a dissociation

constant similar to the literature va lue whilst QD65 5/fluorescein h ad large erro rs
due to aggregation problem s.
Chapter 4 examines the limitations of the method for measuring dissociation
constants with r espect to vario us parameters such as cross-ta lk, quenching and
sample imp urities. A fluorophore pair consisting of common organic dyes, tetram-
ethylrhodamine/fluorescein, h aving similar excitation and emission spectra, was
experimented with the binding of streptavidin and biotin. Despite the lower signal-
to-noise ratio comp ared with spectrally distinct fluorophore pairs, the method was
able to determine the dissociation constant and stoichiometry of rea ction.
Cha pter 5 extends the SW-FC C S meth odology to m u lticolor detection of three
interacting molecular species. Three fluorescent probes fluorescein or R-phycoerythrin
labeled biotin emits in the green or y e llow c ha nn els respectiv ely; Alexa 647-
R-phycoerythrin labeled strepta v idin (AXSA) emits in the red c hannel. Triple
pair-wise cross-co rrelation s betwe en the thr ee-color chan nels were performed and
binding constants and stoichiometry of binding could be derived. Multicolor SW-
FCC S deliv ers the possibilit y of detecting higher order molecular intera ctions and
molecu lar a ssemblies using a single laser line.
Cha pter 6 c halleng es the con ve ntional FC C S setup by implem enting a disper-
siv e elemen t in the detection path to c hr om a tically disperse the emission light.
The prism-b ased F C Spectrometer was first calibrated with fluorescein and AXSA
with a single optic fiber and then tested for cross-correlations with biotin y lated
rhodamine green nanocontainers and AXSA using an optic fiber array . This novel
w avelength tunable filter-free prism-based FC S pectrometer achieves sim ulta neous
auto/cross-c orrela tion s a nd could be applied for multicolor detec tion.
vii
List of Tables
3.1 Table of fluorescenceyieldsofQR,QD655andBF 34
4.1 Table of fluorescence intensities and yields of fluorescent m o lecu les . 64
4.2 Maximum K
d

/R
t
values with corresponding L
t
/R
t
where the de-
tection thresh old R =1 65
5.1 Molar extinction coefficients and flu orescence yields of BF, BPE
andAXSA 79
5.2 Possible fluorophores and filtersetsforSW-FCCS 82
5.3 Table of best fit values and limits of V
eff
and K
d
89
6.1 Table of dispersion constants of prism material N-BK7 from Schott
Catalog 115
viii
List of Figures
2.1 The autocorrelation function and its c hanges with diffusion time
andsampleconcentration 17
2.2 A typical fluorescencecorrelationspectroscopyopticalsetup 18
2.3 Focigeometryoftwooverlappingdetectionvolumes 26
2.4 The dual-color single wavelength fluorescence cross-correlation spec-
troscopy setup 27
3.1 (A) Fluorescence emission spectra of Q R , QD65 5 and BF. (B)
QuenchingofBF 33
3.2 Change of diffu sion time and number of particles of QR with laser
power 35

3.3 Schem atic dra w ing of flu orescence in t ensit y signal from the green
andreddetectionvolumes 36
3.4 Average count rate and in tensity ratio of QR with varying laser power 37
3.5 Change of diffu sion tim e and number of particles of QD 655 and
fluo rescein w ith laser power . . 38
3.6 Cross-cor relatio n function decrease in amplitude with increasing
BF/QRconcentrationratio 38
3.7 Plots of cross-correlation amplitude and number of particles versus
BF/QRconcentrationratio 40
3.8 Fitting of QR-B F binding c urve and simula tion s o f various K
d
s 41
3.9 Fitting of Q D655-B F binding curve and sim ulation s of various K
d
s. 41
4.1 BindingexperimentsofBFtoTMRSA 56
4.2 Influence of K
d
onthecross-correlationamplitude 56
4.3 Influenceofimpuritiesonthecross-correlationamplitude 58
4.4 Influenceofcross-talkonthecross-correlationamplitude 59
4.5 SensitivityofSW-FCCSdependingoncross-talk 60
4.6 Influence of receptor labeling on the cross-correlation amplitude . . 62
5.1 MulticolorSWFCCSopticalsetup 77
5.2 Absor ban ce, emission spectra and ACFs of B F , B P E and AX SA . . 80
5.3 Cross-cor relatio n fun ctions o f b indin g bet ween AXSA , BF and BPE 86
5.4 Triple pair-wise CCF amplitudes of the positive and negative con-
trolsofBF,BPEandAXSAbinding 87
5.5 CCF amplitudes of with alternate binding of ligands BF, BPE to
AXSA 91

5.6 Simu lation s o f K
d
influenceonpair-wiseCCFamplitudes 93
ix
5.7 Simu lations of g × y CCF amplitudes of positive and negative con-
trols of BF, BP E a nd A XSA binding at different stoichiom etries . . 97
5.8 Sim ulations of triple pair-wise CCF amplitudes of the binding of 1
ligandAand1ligandBperreceptorR 100
5.9 Simu lations of A × B CCF amplitudes of ligands A and B binding
per receptor R at different K
d
scombinations 101
5.10 Simulation s of A × B CCF amplitudes at differen t K
d
sofRAB
binding 102
5.11 Sim ulations of triple pair-wise CCF amplitudes of the binding of 2
ligandsAand1ligandBperreceptorR 105
5.12 Simulation s o f A × B CCF amplitudes of 2 ligands A and 1 ligand
B binding per receptor R at different K
d
scombinations 106
5.13 Simulation s of A × B CCF amp litudes at different K
d
sofRA
2
B
binding 107
6.1 Opticalsetupoftheprism-basedFCSpectrometer 113
6.2 Deviationofaraythroughaprism 114

6.3 Dependence of angular dispersion and lateral displaceme nt on w ave-
lengths 116
6.4 Em ission spectra of BF, RPE and AXSA and their ACF s measu red
ontheFCSpectrometer 118
6.5 Schem atic draw ing of spots of different wavelengths imaged onto an
optic fibercore 120
6.6 Calculated normalized transmission of optic fibers v ersus wavelength122
6.7 ACFs and CCFs of the binding of biotinylated RhG nanocontainers
andAXSA 124
6.8 Zema x software configu rations for the design of the prism-based
fluo rescence co rrelation spectrometer. . . 127
6.9 Zema x sim ulation of the detection path of the prism-ba sed FC Spec-
trometer 127
6.10 Zem ax simulations of spot images produced by the prism-based
FCSpectrometer 128
6.11WavelengthinputdatafortheZemaxsimulations 128
x
List of Symbols
α apex angle of prism
α
i
relative fluorescence yield
χ
2
Chi squ ared to describe goodness-of-fit
∆θ(λ) angular dispersion as a function of w avelength
δC(r, t) function of concen tration fluctuations at positions r and time t
∆y(λ) lateral disp lacem ent as a function of wavelength
η flu orescence yield of a molecule, product of Q and κ
η

i
s
j flu orescen ce yield of species s where s = L, L
g
,L
y
,R,RL in detec-
tion channels ij where i = j for ACF or i 6= j for CCF
κ detection e fficiency of the optical instrument
λ
ref
reference w avelength
λ
em
emission wavelen gth
ρ densit y of m olecule
σ
i
standard deviation of data point i
τ correlatio n time
τ
d
diffusion time
τ
trip
triplet life time
θ de flection angle o f light ray passing through p rism
C, < C > concentration, average concen tr ation
C
L

,C
R
,C
RL
time-dependen t average concentration s of ligand, receptor and com-
plex
CEF(r) collection efficiency function
D diffusion coefficien t
f focal length of o ptical lens
xi
f(r, r
0
,τ) concentration correlation function
F
i
(t),F
j
(t),δF fluorescence signals in detector chann els i and j as a function of
time, fluorescence intensity fluctuations
f
L
g
,f
L
y
proba bility of binding of green, yellow ligand
F
trip
fraction of m o lecules in triplet state
G(τ),G(0) correlation function, correlation function am plitu de

G
+
×
(0),G
−
×
(0) cross-correlation amplitude of the positiv e con trol, negative con-
trol, where × can be g × r, y × rorg × y
G
−
×
(0) cross-correlation function amp litude of the negative control
I(r) intensity d ist ribu tion function
J
1
Bessel fu n ction of the first kind of order 1
K geom etric ratio of axial to radial dimensions of the observa tion
volume
k Boltzm an n’s co n stant
K
d
,K
max
d
,K
min
d
dissociation consta nt, m ax imu m and minimum dissociation con-
stan t
L, L

g
,L
y
ligand, g r een ligand, yellow ligand
M magnification of optical system
M mole cula r mass
MDF(r), MD F molecule detection func tion
N,N
t
n umber of molecules, total number of molecules
n, n
t
,n
∗
number of b indin g ligands, total number of bind ing sites per re-
ceptor, number of bright ligands
N
A
Avogadro’s number
n
g
,n
y
number of green, yellow ligands
P laser power
p
L
(u),p
R
(v) prob ab ility o f ligand or recept or to have u or v number of fluo-

rophor es, respectively
Q product of the absorption coefficien t and the m olecular quantum
yield of a flu orescent species
q
L
,q
R
quenchin g factor of ligand or receptor
R detection th reshold
xii
R receptor
r, r
0
,r radial coordinate, coordinate vector
RL, RL
n
receptor-ligand complex
T (λ) transmissio n intensity as a function of waveleng th
u, U number of fluoro pho res attached to ligand, number of labeling sites
on ligand
V v olume of spherical molecule
v, V number of fluoroph ores attac h ed to receptor, nu mber o f labeling
sites on receptor
V
eff
effectiv e observatio n volume
w
o
beam waist radius of laser beam
Y

i
mole fraction of mo lecu lar s pecies i
z
o
distance along optical axis where the laser in tensity has dropped
to 1/e
2
of its m aximum
∗
p
L
,
◦
p
L
,
∗
p
R
,
◦
p
R
probab ility of binding a bright (∗)/dark (◦) ligand or receptor
A C F autocorrelation function
APC allophycocy anine
APD avalanche photodiodes
AX SA Alexa Fluor 64 7-R -p hyc oerythrin-streptav idin
BF bi otin-fluorescein or biotin-4-flu orescein
BPE R-ph ycoerythrin biotin-XX conjugate

CCD c h arge-coupled device
CCF cross-correlation function
CdSe Cad m iu m Selenide
CHO chinese ham ter o vary
CLSM con focal laser scanning microscopy
CM O S c h arged metal-o xide semiconductor
cpm count ra tes per mo lecule or counts per molecule and second
cw contin uous wav e
Cy2, 5, 7 cy anine dy e 2, 5 7
xiii
DN A deoxyribonu cleic acid
EGF R epiderma l grow th factor receptor
F CCS flu orescen ce cross-correlation spectroscop y
FCS fluorescen ce correlation spectroscop y
FFS fluorescence fluctuation spectroscop y
FIDA flu orescen ce intensity d istribution analysis
FILD A flu orescence intensity lifetim e d istribu tion analysis
FIM DA Fluorescen ce intensity m ul tiple distribution analysis
FLIM flu orescence lifetime imaging microscopy
FP fluorescen t protein
FRET Förster resonance energy transfer
FWHM full width half maximum
GFP g reen fluorescent pro tein
HeLa HeL a cell is an immortalized cell line. These are h um a n epithelial
cells from a fatal cerv ical ca rcinoma
Her2 Huma n epid erm al gr owth factor receptor 2
ICCS image cross-correlation spectroscop y
ICS image correlation spectroscopy
mRFP monomeric red fluorescent protein
NA n umerical aperture

NM R nuclear m agnetic resonance
PAID photon arrival-time inten sity d istrib ution analysis
PBS phosphate buffer solution
PCH photon countin g h istogram
PCR polymerase chain reaction
PE ph ycoerythrin
PMT photomultiplier tubes
PSF poin t spread function
QD, Q D 655 quan tum dots, quan tu m dot 655
xiv
QR qua ntum red
RhG r hodamine green
RICS raster imag e correlation spectroscop y
RPE R-phy coerythrin
SMD s ingle molecule detection
SPT single particle tracking
SW -FC CS single wavelen gth fluorescence cro ss-correla tion spectroscopy
T temperature
TIR F tota l-intern al-reflection fluorescence
TM R , T M RSA tetramethylrhodamine dye, tetramethylrh odamine-labeled strep-
tavidin
ZnS Zinc Sulphide
xv
Chapter 1
Introduction
Life is based on molecular processes that are essential for the structure and func-
tion of all living organ ism s. B iomolecular in terac tions between proteins, n ucle ic
acids and small molecules are responsible for co m p lex b iolo gical p rocesses. By
studying these biomolecular in teractions, life scientists hope to better u nderstan d
and predict cellular mechanism s and function s. Bioc hem ists have mad e huge ad-

vances in protein sequenc ing and genomic analyses of living organ ism s , pain ting
a network of interactio ns in a cell. But to resolve the underlying in terac tion s
inv o lved in complex biological processes, it requires more than the identification
with biochem ical methods. With recen t advan ces in single molecule techn iques,
it becomes possible to in vestigate the biomolecula r in tera ction s that give rise to
higher order biological phenomena. This empowers biologists and bioph ys icists
to stu dy the mech anism s and fun ctions in biological p rocesses such as immune
response, neu rophy siologica l pr ocess and sign al transdu ction.
Conventional ensemble tec hniques used for investiga ting biomole cular interac-
tions include yeast two-h ybrid screenings, immunoprecipitation and mass spec-
trometry. Structure determinatio n m eth ods such as X-ray crystallography an d
NMR provide additional information on binding sites and molecular conformation.
Howe ver, these tech niq ues used for analyzing nuc leic acids and protein m olecu les
require relativ ely large amount s and concen tra tions of sample. In addition, exper-
1
Chapter 1 Intr oduction
iments have to be performed occasionally under non-physiological conditions. In
recen t decades, the adva ncem ent of instrumenta tion ha ve led to the emergence of
biophysical techn iques capable of probing single molecules on surfaces and solu-
tions in real-time. B y focusing on an indiv idua l molecule in space and time, such
analyses provide quantitative information of force properties, conformational dy-
namics, molecular interactions and temporal c hanges with its microenvironmen t
that could otherwise be hidden in ensemble experiments. Molecular dynamics can
be studied without ha v ing to bring the ensem b le population into a non-equilibrium
state. Futhermo re, because of the small measurem ent v olum e needed for samp le
assa y s, the high spatial resolution of single-molecule methods enables them to sort
and examine rare molecular ev en ts or subpopulations that exist only in highly lo-
calized regio ns in the cell.
Onetypeofapproachtosinglemoleculedetection(SMD)techniquesisthe
optical method based on fluorescence detection. F luorescence tec hniqu es are non-

in vasive and non-destru ctive to samples. They can be performed in real-time at
ambient or physiolo gical temperatures. T h eir versa tility with the molecular envi-
ronment implies that they be applied in vitro or in vivo . B y labeling the object
of interest with a fluo roph ore and illum inating a sm a ll observation volum e with a
focused laser beam coup led with interference filters and sensitiv e detectors suc h
as cooled cha rge-coup led device (C C D ) cameras, photomultip lier tubes (PMT ) or
a valanche photodiodes (APD), the signal-to-noise ratio can be greatly increased
o ver bac kg round scattering and cellular autofluorescen ce. Fluorescence microscopy
techniques include epi-illumination wide-field imaging that has been applied in
single particle tra cking (SPT), confocal laser scan ning microscop y (CLSM ), total-
internal-reflection fluorescen ce (TIRF), Förster resonance energy transfer (FRET )
and fluorescence lifetime im aging microscopy (FLIM ). Besides being able to visu -
alize and mon itor intrace llular and memb ra ne dynamics with precise spatial local-
ization, protein-protein and protein-n ucleic acid interactions can also be probed.
Various SMD methods and its applications, in particular molecula r inte raction s,
2
Chapter 1 Intr oduction
have been described in several review s [1—7].
A widely used SM D method for measurin g molecular in teractions are FRET
and FRE T -based techn iques such as F L IM . FRE T process involves the r esonan ce
energy transfer between a single pair of donor and acceptor fluorophore with o v er-
lapping emission and excitation spectra respectively [8]. FRE T efficiency depends
on dipole-dipole intera ctio ns and molecular distance (in verse sixth po we r) and
is used as a spectroscopic ruler on a scale of 1-10 nm [9]. C o mbining FRET
and TIRF imaging, the dimerization and activation of EGFR on cell mem branes
were revealed [10]. Alternating la ser excita tion was used to improve signal-to-
background ratio and to study the transcription mechanism by RNA polymerase
[11, 12]. FLIM, on the other hand, measures the characteristic lifetime of a fluo-
rophore (nanosecond range) [13, 14]. F R ET -F L IM imag ing observes the reduction
of donor fluorescence lifetime as sho w n in the association of EGFR in live cells

[15, 16]. Howe ver, a major disadvantage of FRET is the sensitivity to dye orien ta -
tion, whic h may induce artefacts that may cause misin t erpreta tion s in molecular
interactions.
Another group of fluorescen ce methods monitor the fluorescence intensity fluc-
tuations of single m olecu les m ov in g in and ou t of a confined illuminated volum e.
These methods kno wn as fluorescen ce fluctuation spectroscopy (F F S) p rovide in -
forma tion that lie hidden in the fluctuating signal such as dynamic processes,
c hem ical kinetics or molecular in t eractions [17]. Convention ally, cor relation func-
tions of the intensity fluctuation s are calculated to give the n umber of particle s
and th e averag e r esidence tim e spent in the detection v o lume. Recently, other
meth ods have been develo ped based on the distribution of fluorescence in tensity
to extract information not measurable with cor relation functio ns. Photon count-
ing histogram (PC H ) or fluorescence in tensity distribution analysis (FIDA ) ha ve
emerged at the same time from independen t researc h groups to determine the
fluo rescence brigh tn ess p a ram eter and d istingu ish d ifferen t species according to
their molecular brightness [18, 19 ]. FIDA has been applied in high th ro ug hp ut
3
Chapter 1 Intr oduction
screening to mea sure binding a ssays [20—22], and PC H has been u sed to probe
ligand-pr ote in binding [23] and protein oligomeriz ation in live cells [24]. Exten-
sions to P C H /FIDA include 2D-F IDA [25] and dual-color PCH [26] where two
detectors monitor different e m ission polarization or em ission w avelen gths. Flu -
orescence int ensity multiple distribution analysis (FIM DA) [27] and fluorescen ce
intensity lifetime distribution analysis (FILDA) [28 ] combin es the measure m ent
of molecular brightness and diffusion time or fluo rescence lifetime respectiv e ly.
A multidimensional method known as photon arrival-time in tensity distribution
analysis (PAID) measures the photon arriva l t im e intervals in stead of counting
photons at fixed time in tervals. It w as introduced to sim ultan eou sly extract dif-
fusion time, molecula r brightne ss and occupanc y in multiple detection chann els
[29].

One of the firstFFSmethodstobeintroducedbyElson,MagdeandWebbin
the 1970s w a s fluorescence correlation spectroscopy (F C S) [30]. The theory was
established to use intensity fluctu ations of fluorescent particles diffusing through a
focused laser beam, to char acterize translational diffusion coefficients and chem ical
rate constan ts [31—34 ]. T he improvem ent of this tech nique to single-molecule sen-
sitivit y w as ac hieved b y using a confocal microscope system with a high numerical
aperture objective and single photon countin g a valanc he photodiodes as detectors
[35, 36]. Since then, it has become an increasingly popular technique for the study
of dynamics at therm odynamic equilibrium. Besides the ability to determine the
concentration, diffusion chara cteristics [37], rotational diffusion [38—41] and vari-
ous p rocesses such as flo w [42] and chemical reactions [43, 44], FCS has also been
used to measure receptor-ligand interactions in s olution and on cell membra nes
[45—47] and enzymatic turnovers [48]. Photodynamic properties of c hemical dyes
[49] and flu orescent proteins (FPs) [50, 51] ha ve been studied and applied in the
detection of pH cha nges in cells [52].
The concept of FC S is based on the correlation analysis of fluorescence fluc-
tuations in a confin ed observation volume. T he sensitivity of this t ech niqu e to
4
Chapter 1 Intr oduction
detect binding of two or more components depends on the relativ e change in mass
upon binding. For a m ulti-com ponent system consisting of reactan ts and products
labeled with the sam e flu orescent dy e, the on ly wa y of differentiating the product
fromthereactantiswhentheproducthasamolecularmassthatdiffer s from the
reactants by at least a factor of 4 [53]. This in turn shifts the correlation curve to
higher diffusion times by at least a factor of 1.6 giv en by the Stok e s-E instein equa-
tion for spherical diffusing particles [54]. B y separately labeling the reactan ts with
differently emitting fluoro pho res, the labels can be simultaneo usly excited with two
different laser lines and detected in separate channels. The signa ls from both de-
tector chan nels are cross-correlated and the doubly labeled products can be easily
distinguished from the singly labeled reactants independent of their mass. Earlier

cross-correlation systems have made use of light scattering or a comb ination with
fluo rescence to m ea sure their cross-correlation functions and determine rotational
diffusion and association-dissociation kinetics [5 5, 56]. In dual-beam fluorescence
cross-correlation spectroscop y, the setup consisting of t wo spatially separated focal
points has been applied to characterize flow systems [57]. Althoug h the concep t
of dual-color fluorescence cross-correlation spectroscop y (F CC S ) has been pro-
posed for biotec h nolo gic al a pp licatio n s [58 ], it was first experimentally rea lized
by Schwille et al. to measu re nu cleic acid hyb r idizat ions [59, 60]. The poten tial
of this technique to effectiv ely measure biom olecular intera ctions has expanded
its applications to detecting PCR complexes [61, 62], monitoring enzyme kinetics
[63, 64] and measuring prote in-D N A in tera ction s [65]. FCC S h as been app lied in
liv e cell measure m ent s (for review s, see [66, 67]) to probe the endocytic pathway
of bacterial c h olera toxin la beled with Cy2 and Cy5 dyes on d ifferent subunits of
the same holoto x in [68]. F P -ba sed cross-correlation analysis in liv e cells have been
recen tly reported where green fluo rescent protein (GF P ) wa s fused to monom er ic
red fluorescent pr otein (m R F P ) w ith a caspase-3 r ecogn ition linker . Caspase-3
activation was detected throug h the d ecrease of th e cross-correlation amplitu de
when the cells undergo apoptosis and protease clea vage [69]. A nother in vivo
5
Chapter 1 Intr oduction
application of FC CS is the study of protein-protein in teractions of transcription
factors Fos and Jun fused with FPs [70]. ICS/ICCS is a variation of F C S/F CCS
that rapidly captures a time-series of images with CLSM to determine the distrib-
ution and co-localization of biomo le cu le s in liv e cells or cell membranes [71—7 3 ]. It
is a ver y useful method to investigate motilit y of large r structures such as pro tein
clusters. H owever, its temporal resolution is limited by the image acquisition time
of the microscope [74, 75]. Raster image correlation spectroscopy (RICS) ac hieves
the temporal resolution of FCS b y rapidly measuring many focal points in the cell
during the raster-scan mode of the CLSM [76].
The first dual-color fluorescence cross-correlation experimen ts on a single mole-

cule level were performed w ith two lasers at d ifferent wavelen gths [59]. Although
this approach improves the detection sensitivity of interacting particles compared
to FCS, the requirement of matching two laser beams to the same focal spot makes
it experimentally challengin g. The mism atch of laser excitation volu m e s also led
others to dev elop new methods of aligning two laser beams to the same excitation
v o lume using a prism [77] an d alternative excitation methods usin g a multiline
laser [78]. Two-p hoto n excitation laser sources have been used to o vercom e the
difficulty of aligning two laser beams to the same excitation volum e and has re-
cently found severa l applications in solution measurements of p roteolytic cleavage
[63]. Increased axial resolution from a m ore confined focal spot reduces background
fluo rescence and photo bleachin g makin g it suitable for in vivo studies [79, 80] such
as in trac ellula r calm odulin and calmodulin-k ina se II b ind in g [81, 82]. Recently,
t wo-photon excitation has ac hieved the excitation of up to three dy es simultane-
ously to perform triple-color coincidence analysis [83]. Ho wever, the high cost of
a high power femtosecond laser so urce and relativ ely lower emission rates, thus
lower signal-to-no ise ratio, limit its potential applications. Pulsed inte r leaved laser
excitation [84] that is faster than the timescale of diffusion has been implemented
to elimina te cross-talk for spectrally similar fluorescent proteins, e.g. CFP- and
YFP -co nnexin fusion p r otein s in the m emb ran es of liv e HeLa cells [85]. A less
6
Chapter 1 Intr oduction
expensive and simp ler optical setup has been suggested . This in vo lves a system
of two or more fluoroph ores excited at the same w avelength but emit at distinctly
separate emission w avelengths. Ho wever, till date, no adequate system has been
proposed [64, 86]. With in creasing demand for m u ltiplex detection, the detection
setup will become increasingly complex with more optical components in tegra ted.
Hence, a gratin g-based d etection u nit has been dev eloped to replace the series of
dich roic m irrors and bandpass filters, offering a wavelength t un able setu p with
multicolor detection [87]. A lthough commercial laser scanning m icrosco pes can
now be combined w ith FC (C )S for cell imaging and spectroscopy [88], the ability

of the setup to perform multicolor cross-correlations will depend on the stabilit y
of alignm ent of sev era l lasers to the sam e focal spot.
Fluores cent probes play an important role in distinguishing the target mole-
cule from the bac kgrou nd light such as scattering or autofluorescence. With the
recent advent of long Stokes shift fluorophores such as quantum dots, tandem dyes
and Me ga Sto kes dyes [89], multicolor im ag ing usin g a single laser waveleng th for
excitation has been achieved with quantum dots [90]. Quantum dots a re semi-
conducto r nanocrystals made of Cadmiu m Selenide (CdSe) which has been coated
with an additional semiconductor shell of Zinc Sulphide (ZnS) to improv e the op-
tical pro perties of the material. This c ore-sh ell material is further coated w ith
a polymer shell [91] or other ligands [92] that allo w the materials to be conju-
gated to biologica l mo lecules. Quantum dots ha ve the unique o ptica l property of
size-dependen t emission w avelengths [93]. O ther benefits of quan tum dots include
long-term photostability, high quan tum yield, multiple labeling with several col-
ors, and single wavelength excitation for all colors. Q uantum dot conjugates h ave
found recent applications in live cell im aging of membran e receptors, Her2 and
other cellular targets [94] and imaging in live anima ls [95]. Sin gle molecule studies
have also revealed blinking characteristics [96], longer fluorescence lifetimes [97],
brightness and size properties [9 8]. Bec ause of its lon g Stok es shift, multicolor
FCS experimen ts have been perform ed to detect h eterog eneities in lipid bilaye r
7
Chapter 1 Intr oduction
membranes [99], comb ined with submicro m eter fluidic c hannels for isolation and
detection [100] and used to measure the binding constan ts of quan tum dot-labeled
strepta v idin-b iotin with two -pho ton excitation [101]. For extensiv e reviews of
quantum dots on biological app licatio ns, see [102—104].
Phyco bilipro tein -based tandem dye s h ave also been used for mu lticolor detec-
tion with single laser wavelen gth and were first app lied in flo w cyto me try and
cell sorting in fluo rescence immunoassa ys [105]. As most clinical flo w cytom eters
use only single laser excitation, there is a constant need for m ore fluorophores

that can be simulta neously used to measure more than t wo parameters in a sin-
gle cell. Phyco bilip rotein s, a class of ligh t-h arvestin g proteins that enhances the
efficiency of photosy nth esis are found in many species of algae [106]. Phycobilipro-
teins have high ex t inc tion coefficien ts and quantu m yields. The molecu lar sizes
can be large, with R-p hy coerythrin (RP E ) at 240 kDa contain ing 34 bilin fluo-
rophores but this does not seem to interfere with its experimental a pplica tions
[105]. With its high molar absorption coefficient at a broad range of absorbance
wavelengths between 470 and 550 nm, phy coerythrin (PE) can be coupled as an
energy d on or to a range of potential accepto r molecules, includin g Alloph ycocy a -
nine (APC, λ
em
=660nm) [106, 107], Cyanine dy es (Cy5, λ
em
=670nmor Cy7,
λ
em
= 767 nm) [108] and Alexa Fluor dy es (Alexa Fluor 647, λ
em
=667nm)
[109]. When excited at an excitation wave length of 488 nm, energy transfer of the
tandem dyes produces large Stok es shifts with emission w avelengths that can be
easily resolved from PE (λ
em
=575nm)orfluorescein emission (λ
em
=518nm)
[110]. Three-color immun ofluorescence analysis of cells was performed with flow
cytometry [111] and this h as since a dvanced to the capab ilit y of measuring up to
12 differen t colors [112]. The developmen t of the tandem dyes has significantly
enhan ced the capabilities of sin gle-laser excitation flo w cytometer s for performing

m u ltiparametric analysis and higher throughput screening, and can be extended
to other single molecule applications, including m ulticolor fluorescence microscopy
and spectroscopic tec h niques [113] su ch a s FC S/FCCS.
8
Chapter 1 Intr oduction
The aim of this work is to develop a FC C S tec h niqu e that u ses only a sin-
glelaserlinefortheexcitationofmultipleflu orescent p robes. This method is
called single wavelen gth fluorescence cross-correlation spectroscop y (SW-FCC S ).
Fluoroph ore a ssays inclu ding small organic dye s, quantum dots and ta ndem dy es
are tested on the setup . As a proof-of-principle, model recepto r-ligand binding
system streptavidin-biotin is investigated for molecular in tera ctions. Av idin is a
tetrameric protein found in egg white and streptavid in is a similar protein (Str epto-
myces avidinii ) isolated from a bacterium . The precise function of these proteins
are still uncertain. However, the (strept)avidin-b iotin binding comp lex is known
to ha ve the highest affinity in teraction between a protein and ligan d (dissociation
constant K
d
=10
−15
M) [114, 115]. Strepta vidin consists of four identical subunits,
each arranged as a structure of eight-stranded, sequen tially connected, an tipar allel
β sh eets as determined by X-ray crystallography. A single vitamin biotin molecule
binds in poc k ets at the ends of each of the β barrels, thus ha v ing a stoichiom et ry
of streptavidin:b io tin as 1:4. In the absence of biotin, the binding poc ke t con tain s
five water molecules to main tain a defined structure. Upon binding of biotin, the
bound water molecules are displaced b y biotin and binding is induced b y h y dro -
gen bonding and van der Waals interactions and the ordering of two surface loops
[116]. These structural and biochemical factors produce a hig h affinit y binding
and high activation energy for dissociation for the almost irreversible interaction
of strepta vidin -biotin [117]. The applications of the (strept)avidin-biotin system

has been w ell-established i n th e life scien ces in immun oa ssay a nd DNA probes
[118, 1 19 ]. R e cent ly, it ha s been ex ten de d to m ed ical applications for localization
and imaging of cancer cells, and biophysics where it has shown to be a standard
model to t est new techniqu es designed to study molecula r intera ctions [120, 121].
Fluorim etric assays have been previously conducted for the quan tification of avidin
and streptavidin with biotin-fluorescein and biotin-4-fluorescein conjugates [122].
Binding of biotin-4-fluorescein to strepta vidin was reported to be comparable to
D-biotin in terms of high affinity, fast association and non-cooperative interaction
9