Citation
Instrumentation Developments And Nanomaterial Engineering For Live Cell Mapping And Bioanalysis

Material Information

Title:
Instrumentation Developments And Nanomaterial Engineering For Live Cell Mapping And Bioanalysis
Creator:
Chen, Yan
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (179 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
Tan, Weihong
Committee Members:
Horenstein, Nicole A.
Cao, Yun Wei
Powell, David H.
Dennis, Donn M.
Graduation Date:
8/7/2010

Subjects

Subjects / Keywords:
Beacons ( jstor )
Binding sites ( jstor )
Cell membranes ( jstor )
Dyes ( jstor )
Fluorescence ( jstor )
Molecular interactions ( jstor )
Molecules ( jstor )
Nanoparticles ( jstor )
Quenching ( jstor )
Receptors ( jstor )
Chemistry -- Dissertations, Academic -- UF
aptamer, binding, bioanalysis, fccs, fcs, fluorescence, fret, kinetics, live, membrane, molecular, mrna, nanoparticle, nanoruler, ribonuclease, set
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
Chemistry thesis, Ph.D.

Notes

Abstract:
Intracellular molecular interaction studies have attracted great interest among cell biologists. However, the intrinsic resolution of conventional optical microscopy only allows the visualization of subcellular localizations. Therefore, a sensitive technique that can observe molecular interactions down to the molecular level is in great need. By extracting information from molecular diffusion, Fluorescence Correlation Spectroscopy (FCS) gives detailed information on molecular interactions, and has become a non-invasive single-molecule-detection technique that can be applied to the intracellular environment with low detection limits. Therefore, the overall direction of this doctoral research was the development of a self-built FCS setup. Combined with nanomaterial engineering, this project has enabled membrane receptor studies and intracellular measurements in live cells. The Fluorescence Correlation Spectroscopy instrumentation was to direct measurement of membrane receptor density of the PTK7 protein, an important cancer marker, in its natural physiological environment on the cell surface. A cellular model using a DNA ligand aptamer was designed for specific receptor targeting and labeling, and was used with FCS to determine receptor densities and distributions on the cell surface. With its intrinsic advantages of direct measurement, high sensitivity, rapid analysis and single-cell measurement, this FCS density estimation approach holds great potential for future applications in molecular interaction studies and density estimations for subcellular structures and membrane receptors. To further understand the structure of PTK7 beyond its spatial distribution and ligand-receptor interactions, a nanoparticle was used as a molecular ruler to measure the distance between two binding sites on the receptor on live cells. Measuring distances at molecular length scales in living systems is a significant challenge. Methods like FRET (fluorescence resonance energy transfer) have significant limitations due to short detection distances and strict orientations. To overcome these limitations and construct a practical nanoruler for measuring distances on live cells, an SET-based nanoruler, using aptamer-gold-nanoparticle conjugates with different diameters, was developed to measure separation distances well beyond the detection limit of FRET. Since application of fluorescence auto-correlation (FCS) to binding analysis is limited to applications in which the binding event significantly reduces the diffusion of the labeled species, the original FCS setup was upgraded to a novel three-channel Fluorescence Cross-Correlation Spectroscopy (FCCS) setup. This lab-built FCCS not only inherits the single-molecule detection capability from FCS, but also further extends its applications for molecular interaction studies by labeling two species with two spectrally distinct fluorophores. This technique has been establied and is now being adapted for real-time monitoring of intracellular mRNA. Another aspect of this research was the development of molecular probes based on nanomaterial engineering Two different types of molecular beacon (MB) probes were designed for enzymatic activity studies and protein inhibition studies. In summary, this research mainly focuses on instrumentation development and nanomaterial engineering for bioanalysis and biomedical applications, especially for cell membrane receptor studies and intracellular measurements. A successful outcome from these studies will lead to a better understanding of biological events and processe. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2010.
Local:
Adviser: Tan, Weihong.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31
Statement of Responsibility:
by Yan Chen.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
8/31/2011
Resource Identifier:
004979879 ( ALEPH )
706489319 ( OCLC )
Classification:
LD1780 2010 ( lcc )

Downloads

This item has the following downloads:


Full Text





provide information about the molecular structure of cell-surface receptors. There is no

easy way to produce a similar panel of monoclonal antibodies in such a short time

without access purified samples of the possible antigens. Compared to 2-D gel

electrophoresis and mass spectrometry for proteomic studies aimed at identifying

proteins, cell-SELEX first produces molecular probes, which can then be to identify the

target proteins. Thus, not only can the selected aptamers be used as molecular probes

for molecular analysis of cancer, but also they can be used as tools for identifying new

biomarkers expressed by tumor cells or other cells in disease status.

Molecular Engineering of Nanoparticles

Nanoparticles (NP)s are defined as particulate dispersions or solid

particles with sizes in the range of 1-100nm. Nanoparticles are of great scientific interest,

as they effectively form a bridge between bulk materials and atomic or molecular

structures. Bulk materials have constant intensive properties regardless of their sizes.

However, this is not the case for nanoparticles, whose properties are size-dependent. In

addition, the properties of materials change as their size approaches the nanoscale,

which makes some specific properties of nanoparticles quite different from bulk

materials, even though they are constructed with the same types of atoms. For example,

it now appears clear that nanoparticles will overcome many of the significant chemical

and spectral limitations of molecular fluorophores. In most cases, these interesting and

unique properties of nanoparticles are mainly due to the large surface area of the

material, which dominates the contributions made by the small bulk of the material.

Thanks to state-of-art synthetic techniques, methods for preparation and handling of









Molecular Diffusion Time

Figure 2-5B shows the molecular weight dependence of the autocorrelation

function of 5 pM fluorescein-dextran complexes. Increased molecular weight makes the

molecules diffuse more slowly, and is reflected in a shift of the characteristic decay

times of the autocorrelation functions to longer delay times.

From these two direct read-outs of the autocorrelation curve, other molecular

parameters, such as the molecular concentration C (mol/L), binding constant Kd (mol-1)

and kinetic rate constants kon (mol-s-1) and koff (s-1), can also be obtained.

FCS for Live Cell Analysis

FCS has proven to be a highly sensitive tool for studying molecular diffusion in

solution. Moreover, thanks to its small focus volume, low LOD's, and noninvasive

detection properties, it has also found wide application in live cell analysis, including the

study of processes such as diffusion in the cellular environment.88'89'90' 91 FCS has also

been applied to measure ligand binding to cell surface receptors of both tyrosine

kinases and G protein-coupled receptor (GPCR) super families,107 and also to observe

some molecular events in cells.103

There are several advantages of using FCS for live cell analysis: (1) FCS

measures biological molecules in the natural cell physiological environment. (2)

Because there is no need for radioactive labeling, molecules can be rapidly and directed

detected inside live cells. (3) Beyond its capability to obtain molecule diffusion

information, FCS also provides detailed kinetics information about moleuclar

interactions (4) Single-cell information and cell profiles can be obtained. This capability

is not possible with other strategies, such as flow cytometry. (5) The femtoliter-sized









Effects of Different Antibody/Dye Labeling Efficiencies on the Calculations of

Binding Site Distances

Different antibody/dye labeling efficiencies were also tested as shown in Figure

5-15. With multiple primary amino groups on the heavy chain, the anti-PTK7 could react

with different amounts of Alexa Fluor 488 dyes during the conjugation. By controlling the

amount of dye added, three different conjugates were prepared, having 2, 4 or 6 dyes,

respectively, on each antibody. As shown in Figure 5-15, results of the fluorescence

quenching experiments were similar to those of the laser intensity experiments. Both of

these studies indicated that the variations in laser source intensity and antibody labeling

efficiency have limited effects on the determination of the binding site distances.

Using Fluorescent Quencher as Energy Acceptors

If the fluorescence energy transfer from the antibody donor to the surface of

aptamer-gold nanoparticle conjugates was in fact acting as a molecular ruler capable of

reporting the distance between the two binding sites, we asked whether it would be

possible to replace the gold nanoparticles with a fluorescent quencher and still be able

to detect the energy transfer.

Two different fluorescent quenchers, Black Hole Quencher I (BHQ) and Dabcyl,

were chosen to replace the gold nanoparticles to investigate whether it is still be able to

detect the energy transfer. As shown in Figure 5-16A, C, an Alexa488-antibody on the

antibody binding site and a fluorescent quencher-linked aptamer sgc8 on the aptamer

binding side, respectively, the flow cytometry results suggested that no significant

fluorescence decrease was observed, indicating that no significant energy transfer

occurred from the fluorescent donor to the quencher on the two binding sites. However,


127









the fluorescence intensity from cell membrane with saturated concentrations of
Alexa488-labeled anti-PTK7 with no gold NP-aptamer conjugate. The red arrow
indicates the mean intensity. The blue curve shows the fluorescence shift in the
presence of gold NP-sgc8 conjugates. The gap between the blue and red arrow
indicates that the fluorescence intensity decreased with the increasing size of the
gold NP conjugates. The green curve represents the fluorescence intensity in the
presence of the same-sized gold NP, but the NP was conjugated with a control
aptamer sequence, TD05, which does not bind the receptor PTK7. No significant
fluorescence shift was shown for the control. The black curve marks the
fluorescence background with cells only, and the yellow curve shows the binding
of control antibody FITC-labeled isotype Mouse IgG2a. (Bottom) Histogram of the
mean fluorescence intensity for the fluorescence quenching assay determined
from the flow cytometry results. All of the experiments for the fluorescence
quenching assay were repeated three times, and the average value was
determined as the mean fluorescence intensity.


A a.- NP- a3) NP- a4) -5 [4nm NP-






















e il i ol a lieb) m NP- W) i3o] NP- / / m mo].

aN N a7) 5 N a8) [o.2o aj*
mec mec sgc mec me m




moom moom mom mom oomm


136









occurs in MBs leads to higher quenching efficiency than other process, such as

fluorescence resonance energy transfer (FRET).65

Fluorescence Resonance Energy Transfer (FRET)

Fluorescence resonance energy transfer (FRET) is an interaction between a

donor chromophore and an acceptor chromophore.15 The donor chromophore, initially in

its electronic excited state, may transfer energy to an acceptor chromophore through

nonradiative dipole-dipole coupling. FRET requires an overlap between the emission

spectrum of the donor with the absorption spectrum of the acceptor, as shown in Figure

1-4. Such an overlap allows the excitation energy to be transferred from the donor to the

acceptor if the two molecules are coupled by a dipole-dipole interaction within a

distance of 10 nm. FRET efficiency defined as Equation (1-3), depends strongly on the

distance between the donor and the acceptor molecules as described in the following

equation:

-6

E= r- (1-3)
1+( )


where the Forster radius Ro is the distance at which energy transfer is 50% efficient,

and r is the distance between the donor and the acceptor.

The efficiency of FRET is dependent on the inverse sixth power of the

intermolecular separation, making it useful over distances comparable to the

dimensions of biological macromolecules. Thus, FRET is an important technique for

investigating a variety of biological phenomena that produce changes in molecular

proximity. When FRET is used as a contrast mechanism, colocalization of proteins and









In static quenching, the quencher forms a non-fluorescent complex (i.e, dark

complex) with the fluorophore in the ground state.15'64 The decrease in the fluorescence

intensity for static quenching is described by Equation (1-2):

F [FQ]
0= 1+K[Q] =+ [FQ] (1-2)
F ([F][Q])

in which K is the formation constant of the dark complex, and [FQ], [F], and [Q] are the

concentrations of the dark complex the fluorophore, and the quencher, respectively.

There are two simple ways to distinguish static quenching from dynamic

quenching, though both of them result in a decrease of emission intensity.15,64 The first

involves lifetime measurements. In static quenching, the lifetime does not change,

because the only observed fluorescence is from the uncomplexed fluorophore, which

has the same lifetime as before quenching. In contrast, in a dynamic quenching

mechanism the lifetime shows the same order of decrease as the intensity. Second,

temperature plays different roles in the two processes. In static quenching, higher

temperature dissociates weakly bound complexes and alleviates static quenching. For

dynamic quenching, higher temperature causes faster diffusion and more quenching.

But in many cases, both static and dynamic quenching processes occur in the same

system.

Static quenching plays an important role in molecular probes. For example, it is

involved in the fluorescence quenching of fluorophores in MBs.65 Many fluorophore/

quencher pairs, including tetramethylrhodamine(TMR)/DABCYL, EDANS/DABCYL,

eosine/DABCYL, fluorescein/TMR and TMR/TMR display absorption spectral changes

when they are brought close together in the hair-pin conformation, indicating the

formation of non-fluorescent complexes in closed-stem MBs.65 The static quenching that









Molecular beacon (MB): molecular switch for signal transduction

Among the various nucleic acid designs, a hairpin shaped molecular beacon (MB)

is one of specific recognition units with an excellent built-in signal transduction

mechanism for detection without separations.14 The hairpin-shaped structures in MBs

are constructed from a stem and a loop (Figure 1-1). One end of the MB stem contains

a fluorophore with a quencher moiety at the other end. The MBs act as primarily closed

switches, bringing the fluorophore/quencher pair together to turn the fluorescence "off".

When hybridized to its target, the MB undergoes a conformational change that opens

the hairpin structure; as a result, the fluorophore and the quencher are separated and

the fluorescence is turned "on". However, digestion of the oligonucleotides enzymes will

also lead to the separation between the fluorophore and quencher, which also makes

the MB switch from the "off" to the "on" state. This problem can be avoided by

introducing artificial nucleic acid bases.39

The MB's signal transduction is based on a fluorescence resonance energy

transfer (FRET) 6,43, described further below in "Fluorescence Methods for Signal

Transduction". The fluorophore and quencher pair can also be customized for any

experimental setup or conditions. These features have led to a wide applicability and

new utility for MBs in areas where conventional nucleic acid probes cannot function

well.44 Since MBs' development, they have provided many exciting opportunities in

DNA/RNA/protein studies.6 45,46 Furthermore, MBs can be designed for (i) selectivity

with single base mismatch identification capability, and (ii) detection without separation

(real-time monitoring in homogeneous solution or living samples), (iii) sensitive









APPENDIX

PROBABILITY CALCULATION OF INTER-RECEPTOR SET INTERACTION
CONTRIBUTION ................. ............ .............. 167

LIST O F REFERENCES .... ............................... ............................. .............. 169

BIOGRAPHICAL SKETCH ................................................. 179








Donor
fluorescence


J(X)


Acceptor
absorption


Wavelength (X)


Figure 1-4. Schematic representation of the FRET spectral overlap integral.




*Py

S \ *P +Py

-"-\. *Py+Py

Pyrene


Py Py
Figure 1-5. The Structure of pyrene (left) and the schematic for the formation of a
pyrene excimer (right).









Measure the Distance between Two Binding Sites using FRET.

Preliminary trials to measure the two binding site distance were conducted using

the FRET design as shown in Figure 5-6. By labeling the two ligands with different dye

pairs (Alexa488-antiPTK7/Cy5-sgc8 or Alexa488-antiPTK7/TMR-sgc8), we monitored

the energy transfer from the donor dye (Alexa Fluor 488 on antiPTK7) to the acceptor

(TMR/Cy5 on sgc8) in Flow Cytometry. In the FRET experiment in the presence of

Alexa488-antiPTK7 and Cy5-sgc8, no significant fluorescence shift was observed in

channel 3 at 680nm with an excitation source at 488nm compared to the Alexa488-

antiPTK7 or Cy5-sgc8 only (Alexa488-antiPTK7 only: 4.30; Cy5-sgc8 only: 3.71; both:

4.53). In the FRET experiment with Alexa488-antiPTK7 and TMR-sgc8 in channel 2 at

580nm excited at 488nm, severe channel leakage was observed from channel 1 to

channel 2 for TMR-labeled sgc8 due to the spectral overlap, which resulted in significant

fluorescence shift in channel 2 for the acceptor only (TMR-sgc8) even without the

presence of fluorescence donor (Alexa488-antiPTK7). Therefore, no significant

fluorescence difference was observed between the FRET signal and TMR-sgc8 only

(TMR-sgc8 only: 8.31; Alexa488/TMR-sgc8: 10.28). All these results have showed the

weak fluorescent signals between the two ligands using the FRET design.



Since no competition was observed between the two ligands on PTK7, which

supported the simultaneous binding of two ligands to the PTK7 receptor on the

membrane. This colocalization phenomenon proved PTK7 as an ideal model system for

the development of a nanoruler to measure binding site distances on live cells. And

preliminary tests with a FRET design by labeling two ligands with different dye pairs


116









Competition Studies to Confirm the Saturation Binding of Aptamer-NP

Conjugates on Cell Membrane.

Competition studies were carried out to confirm the saturation binding of gold

NP-aptamer conjugates on the cell membrane. Cells were first incubated with saturation

concentrations of Alexa488-antiPTK7 (200nM) and different sized gold NP-sgc8

conjugates (8nM for 5.4nm; 4nM for 10.1-42.2nm) for 20 min on ice. After the cells were

washed twice with 700 pL of binding buffer (with 0.1% NaN3), an additional 25 nM Cy5-

labeled aptamer sgc8 was added to the cell surface and incubated for another 20 min

on ice. Then, the cells were washed twice and suspended in 200 pL of binding buffer

(with 0.1% NaN3). Images of the cells stained with Cy5-labeled aptamers were taken

using a 2-mW, 633 nm He-Ne laser as the excitation source and collected by the same

20x objective on the confocal microscope at 680nm. As shown in Figure 5-4, no staining

of Cy5-sgc8 was observed from the cell membrane for cells incubated with Alexa488-

antiPTK7 and gold NP-sgc8 beforehand (c), indicating all the aptamer binding sites on

the cell membrane were saturated with the gold NP-sgc8 conjugates already. However,

for cells incubated with Alexa488-antiPTK7 and the conjugates with control aptamers,

gold NP-TDO5 (d), cells show similar fluorescence staining of Cy5-sgc8 on the

membrane as in cases with no gold NP-aptamer conjugates (a, b). The competition

studies using Cy5-sgc8 confirmed the saturation binding of different sized gold NP-sgc8

conjugates on the aptamer binding sites in the fluorescence quenching experiments.

Results and Discussion

The cell membrane receptor Protein Tyrosine Kinase 7 (PTK7 or CCK-4), an

important biomarker receptor for T-cell acute lymphoblastic leukemia (T-ALL) 9, 125, was

chosen as the target molecule for construction of the "SET nanoruler". Monoclonal


114












B 2.0-

1.8-

1.6 -

1.4 -

1.2


5 10 15 20


25 30 35 40 4


NP diameter d (nm)


* Linear Range


2 4 6 8 10 12 14
NP diameter d (In|1 n


"- = Ij I 4 4 1 '", + j 4Il, +q_ +1 i'4 r


Figure 5-11 Binding site distance determination. A) By plotting


4 nanopart
1 vs. nanoparticle


diameter d, a linear relationship was obtained between d=5.4nM to d=18.4nM,
and it reached a plateau with d larger than 18.4nM (d=18.4nM-42.2nM). B) The
expanded plot of linear region.
































138


2.0-

1.8-


F-1
(I)


ii


0.6 I..


18 20
18 20


I T
L









Leukaemia cells) cell lines were obtained from American Type Culture Collection.

CCRF-CEM and HeLa cells were cultured in RPMI 1640 medium (American Type

Culture Collection), with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) and 0.5

mg/ml Penicillin-Streptomycin (American Type Culture Collection, Manassas, VA) at

370C under a 5% CO2 atmosphere. RBL-2H3 cells were cultured in Dulbecco's

modified Eagle's medium (American Type Culture Collection), with 15% fetal bovine

serum (Invitrogen, Carlsbad, CA) at 370C under a 5% CO2 atmosphere. All the cells

were grown in 8-well Nunc chambers (Nalge Nunc Inc., IL, USA) to a density of ~ 1000

cells/well. The cells were washed before and after aptamer incubation with Dulbecco's

phosphate buffer (Sigma) with 5 mM MgCI2.

Aptamer Synthesis

Aptamer sgc8 (5'-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG

TTA GA-3'), aptamer KK1 H08 (5'-ATC CAG AGT GAC GCA GCA GAT CAG TCT ATC

TTC TCCTGA TGG GTT CCT AGT TAT AGG TGA AGC TGG ACA CGG TGG CTT

AGT-3') were synthesized on AB13400 DNA/RNA synthesizer (Applied Biosystems,

Foster City, CA). The aptamer was labeled with 5'-FITC modifier. A DNA library

containing a randomized sequence of 41 nucleotides was used as a control. The

completed sequences were then deprotected in AMA (ammonium hydroxide/40%

aqueous methylamine 1:1) at 65C for 20 minutes and further purified with reverse

phase HPLC (ProStar, Varian, Walnut Creek, CA) on a C-18 column. A Cary Bio-300

UV spectrometer (Varian, Walnut Creek, CA) was used to measure absorbance to

quantify the manufactured sequence.

Antibody

FITC-labeled IgE antibody was purchased from Miltenyi Biotec Company.









them, it would indicate that the aptamers are interacting with their target receptor

specifically. From Figure 3-8, it can be observed that the addition of non-labeled

aptamers did result in reductions of both fluorescence count rate (fluorescence intensity)

on cell membrane (Figure 3-8A) and membrane-bound aptamer density (Figure 3-8B).

After about 90 minutes, the aptamer binding was almost entirely displaced. This proves

that the labeled aptamers recognized the target receptor specifically and did bind to

them on the cell membrane. Control experiments (blue circles) with a randomized

sequence of 41 nucleotides, which have been shown as the negative binding control to

receptor PTK7 in cell-SELEX9, were conducted under the same conditions. No

significant decrease of density and count rates were observed within the first 60 minutes

compared with competition studies using non-labled sgc8. The small decrease after 60

minutes is believed coming from koff from the labeled sgc8 itself with time increased, but

not from specific binding competition. Further experiments with proteinase trypsin

treatment (Figure 3-8C) also indicated that, by removing the target protein receptors

from the cell surface with proteinase and reducing their density artificially, the binding of

aptamers on the membrane was decreased. This, in turn, resulted in the decrease of

membrane-bound aptamer density obtained from FCS detection. Since the aptamer

must have been binding to its target, and not simply undergoing a nonspecific

interaction with the cell membrane surface, this evidence serves to further validate the

results obtained from the above density study. Overall, these results support the use of

FCS as a comprehensive tool for detailed receptor/ligand interaction studies, such as

the determination of binding affinity (Kd), dissociation rate (koff) and other kinetic

parameters, which other density study approaches cannot accomplish.












Figure 5-2 TEM images of different sizes of gold nanoparticles. The average diameter of
NPs for each size was determined by measuring the size of 100 particles from
the TEM images. The red title in each box indicated the mean diameter.


0 50 100 150 200 250 300
Alex488_antiPTK7 (nM)


B
35-


30-


25-
U
U
20-
0
U-


0 5 10 15 20 25
NP Concentration (nM)


Figure 5-3 Ligand binding saturation concentration determinations. A) KD binding curve
for ligand Alexa488-antiPTK7 on CEM cells. B) KD binding curves for 5.4, 10.1
and 18.4 nm gold NP-aptamer conjugates on CEM cells. Red curve: 5.4 nm gold
NP-aptamer conjugates; Blue curve: 10.1 nm gold NP-aptamer conjugates;
Purple curve: 18.4 nm gold NP-aptamer conjugates.


Figure 5-4 Competition studies using Cy5-sgc8 to confirm the saturation binding of gold
NP-aptamer conjugates on the cell surface. (a) Plain cells incubated with 25nM


131


a) Cell only+ 25nM Cy5-sqc8








10LM
7 P111t, I, I P-t,

c) Alexa488-antiPTK7 and Gold NP-Sgc8-,
+ 25nM Cy5-sgc8







1 OW
Ip"t zp"""


b) Alexa488-antiPTK7 + 25nM Cy5-sgc8









7 P111t 11 z P,,,t,,, 10
.LM

d) Alexa488-antiPTK7 and Gold. NP-TDO5
+ 25nM Cy5-sgc8







to
.LM
zp",t zp-h-









According to the literature72, unlike large particle sizes (>20nm) in which ro values

vary a lot for different particle sizes, for gold particle sizes between 5-15nm, a very

similar ro value was obtained around 6-8nm. Therefore, the ro values do not have

significant variance for different sizes for the particle size ranges (5-15nm) that chosen

for the distance calculations and it is reasonable during our calculation to assume ro as

a constant in this range.

1
Since ro is a constant, X and Y obey a linear relationship with slope=- and Y-
2ro


axis intercept=R
ro

The plot of Y versus X in Figure 5-11 is linear between 5.4 and 18.4 nm with

equation:

1 1
( -1)4 = (-0.077 + 0.0072)d(nm) + (2.07 + 0.095) (5-4)


From the slope,

1
ro -- I = 6.49nm
(-0.077nm ')x2

and from the intercept,

R = 6.49nm x 2.07 = 13.4nm

Thus, the obtained distance between the two binding sites is R= (13.41.4) nm, which is

larger than the detection limit for FRET. This explains the weak FRET signal from these

two binding sites obtained in previous FRET studies (Figure 5-7) and proves that

nanoparticle dipole-surface energy transfer can detect larger interaction distances than

dipole-dipole interactions.


123









(34) Uri Raviv; Daniel J. Needleman; Youli Li; Herbert P. Miller; Leslie Wilson; Safinya,
C. R. Proceeding of the National Academy of Sciences of the United States of
America 2005, 102, 11167-11172.

(35) Jae-Hyun, L.; Kyuri, L.; Seungo, M.; Yuhan, L.; Taewan, P.; Jinwoo, C.
Angewandte Chemie International Edition 2009, 48, 4174-4179.

(36) Peisheng, X.; Shi-Yan, L.; Qun, L.; Edward, V.; Jun, R.; Williams, M.; Zhaojie, Z.;
Maciej, R.; Youqing, S. Angewandte Chemie International Edition 2008, 47,
1260-1264.

(37) Davide, P.; Ravi, S.; David, M.; Mathieu, E.; Jean-Paul, B.; Maurizio, P.; Kostas,
K.; Alberto, B. Angewandte Chemie International Edition 2004, 43, 5242-5246.

(38) L. R. Hirsch; R. J. Stafford; J. A. Bankson; S. R. Sershen; B. Rivera; R. E. Price;
J. D. Hazle; N. J. Halas; West, J. L. Proceeding of the National Academy of
Sciences of the United States of America 2003, 100, 13549-13554.

(39) Wang, L.; Yang, C. J.; Medley, C. D.; Benner, S. A.; Tan, W. Journal of the
American Chemical Society 2005, 127, 15664-15665.

(40) Robertson, D. L.; Joyce, G. F. Nature 1990, 344, 467-468.

(41) C. Tuerk; L.Gold Science 1990, 249, 505-510.

(42) Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818-822.

(43) Fang, X.; Li, J. J.; Tan, W. Analytical Chemistry 2000, 72, 3280-3285.

(44) Piatek, A. S.; Tyagi, S.; Pol, A. C.; Telenti, A.; Miller, L. P.; Kramer, F. R.; Alland,
D. Nat Biotech 1998, 16, 359-363.

(45) Drake, T. J.; Tan, W. Appl Spectrosc. 2004, 58, 269-280.

(46) Nitin, N.; Santangelo, P. J.; Kim, G.; Nie, S.; Bao, G. Nucl. Acids Res. 2004, 32,
e58-.

(47) Conrad, R.; Keranen, L. M.; Ellington, A. D.; Newton, A. C. Journal of Biological
Chemistry 1994, 269, 32051-32054.

(48) Macaya, R. F.; Schultze, P.; Smith, F. W.; Roe, J. A.; Feigon, J. Proceedings of
the National Academy of Sciences of the United States of America 1993, 90,
3745-3749.

(49) Famulok, M.; Mayer, G. Curr Top Microbiol Immunol. 1999, 243, 123-136.

(50) Hicke, B. J.; Marion, C.; Chang, Y.-F.; Gould, T.; Lynott, C. K.; Parma, D.;
Schmidt, P. G.; Warren, S. Journal of Biological Chemistry 2001, 276, 48644-
48654.


171









stands for the total number of bound aptamers. Assume that the waist of the confocal

volume ellipsoid is situated on the cell membrane such that the highest fluorescence

intensity is given. In this case, the surface area on the membrane covered by the focus,

where all the bound aptamers are located, will be a circular area with a radius equal to

Wxy (shown in the magnified diagram of Figure 3-1). Therefore, the receptor density

estimated in the confocal volume can be expressed as the total number of bound

aptamers divided by the area covered:

N-r
Density = -2 (3-2)


Fifty HeLa cells were investigated using aptamer sgc8 to carry out the

experimental measurements and determine the PTK7 receptor density on the cell

surface per unit area by applying the above formula. The 0.4fL ellipsoid shaped

confocal volume from a 1.2mW laser beam with half axes wxy=0.22 pm and wz=1.56 pm

was projected onto each cell membrane. The circular area covered by the focus volume

could be estimated to be nrx(wxy)2= nrx(0.22pm)2= 0.15pm2. Within this tiny covered

area, the number of specific receptors that occupied the area could be interpreted by

the amount of specifically tagged aptamers at the saturation condition. As determined

from the earlier binding affinity study, a 3.0nM aptamer concentration of FITC-sgc8 was

used to saturate all the PTK7 binding sites on HeLa cells. Autocorrelation curves were

obtained from the membranes of fifty individual cells. An average of Nxr=84 bound

aptamer sgc8 was obtained with a variation of 14, which indicated that there were an

average of 84 bound aptamers that occupied the 0.15pm2 covered area in the confocal

volume. By applying these parameters into formula (3-2), receptor density is obtained. A

Gauss-shaped distribution of the receptor density is shown in Figure 3-5. A mean









emission beams from the three channels are transmitted through the first dichroic mirror

to reach the second dichroic. The second dichroic mirror separates the emission

wavelength from each channel to reach its respective APD detector. It is a dichroic long

pass mirror (550 DCXR) that reflects any wavelengths below 550nm and transmits

wavelengths above 550nm (Chroma, Bellows Falls, VT).

Filters: A bandpass interference filter is placed before the pinhole for each

detector. For channel 488nm, a bandpass filter with a center wavelength (CWL) at

520nm and a full width at half maximum (FWHM) of 40nm (52020nm) is used. For

channel 543nm, a bandpass filter (59020nm) is applied. And for channel 632.8nm, a

bandpass filter (68030nm) is used. These bandpass filter have special coatings on the

surface and are designed to transmit light only within a small wavelength window. They

effectively cut off the interference light and reduce the nonspecific background, while

maximizing the transmission of the emissions at specific wavelengths to the detector.

Pinholes: The fluorescence from the emission filter is focused onto a pinhole in

the image plane. Light originating from the focus passes through the pinhole aperture,

while light from other regions is preferentially blocked. The pinhole was introduced into

the instrument by focusing the fluorescence onto a multimode fiber with a 50 pm internal

diameter, where the fiber face acts as the pinhole. Therefore, changing the fiber allows

easy alteration of the pinhole aperture size without the need for substantial realignment

of the instrument.









encouragement and help during my difficult times. I appreciate Micheal Marvos and Lu

Peng for spending many difficult hours with me on the FCCS constructions. I am

thankful to Meghan B. O'Donoghue for being there whenever I need any help or advice.

I thank Dr. Xiaolan Chen for her help in particle synthesis and functionalization. I thank

Dr. Yanrong Wu and Patrick Conlon for assisting me with the pyrene molecular beacon

synthesis. Also I would like to thank Zhi Zhu for the all the helpful discussions.

It has been a great privilege to spend several years in the Tan research group,

and its members will always remain dear to me. I would like to thank Dr. Xiaoling Zhang,

Dr. Ronghua Yang, Dr. Dihua Shangguan, Dr. Zhiwen Tang, Dr. Marie Carmen Estevez,

Dr. Wenjun Zhao, Dr. Zehui Cao, Dr. Josh E. Smith, Dr. Colin Medley, Dr. Prabodhika

Mallikaratchy, Dr. Kwame Sefah, Dr. Liu Yang, Dr. Joseph Phillips, Dr. Youngmi Kim, Dr.

Dosung Sohn, Dr. Karen Martinez, Dr. Hui Chen, Dr. Jilin Yan, Dr. Zeyu Xiao, Dr.

Kelong Wang, Ling Meng, Jennifer Martin, Pinpin Sheng, Dalia Lopez Colon, Hui Wang,

Suwussa Bamrungsap, Xiangling Xiong, and others for their friendship and help. Each

of them has made this journey very enjoyable and pleasant.

Also, I am deeply grateful to my boyfriend, Fei Huang, for being a wonderful

companion, friend, and lover. I thank him for being with me to persevere during the bad

times, and celebrating the good times.

Finally, I owe a huge debt of gratitude to my grandma and my parents for their

unfailing love, encouragement, support and companion. Their great personality, endless

love and constant guidance make me who I am. I really appreciate their unselfish

dedication all this time, especially for my beloved grandma. She will live in my heart,

every moment, forever and ever.









lowest-energy vibrational state of S1. As a result, fluorescence emission energy is

usually independent of the excitation energy. In addition, because a small fraction of

energy absorbed is lost during vibrational relaxation, there is an energy difference

between the energy absorbed and emitted during the fluorescence process. Therefore,

emission of a specific molecule usually appears at longer wavelengths than the

absorption wavelength, which results in a "Stoke's shift" between the two peaks (Figure

1-3B). More importantly, this "Stoke's shift" allows the spectral separation of the

excitation photon from the emission photon for sensitive studies.

The signal transduction techniques used in fluorescence-based design consider

how to associate the target recognition event with changes in the relative rates of

fluorescence, which result in change in the internal conversion, external conversion, and

inter-system conversion. In addition, if the target binding event changes the electronic

structure of the fluorophore, changes in fluorescence excitation/emission wavelengths

will be observed and can be used to signal the target binding event.

The specific mechanisms of four different signal transduction approaches used in

the Tan laboratory are described below. They are fluorescence quenching, fluorescence

resonance energy transfer (FRET), surface energy transfer (SET) and excited-state

dimer formation.

Fluorescence Quenching

In addition to internal and external conversions, there are other non-radiative

processes by which excited state molecules relax to the ground state. Because these

occur without photon emission, they are termed fluorescence quenching processes.

Quenching and dequenching upon interaction with a specific molecular biological target









of cell suspension bound with Alexa488-anti-PTK7 and aptamer-gold NP conjugates

was dropped on a thin glass slide placed above a 20x objective on the confocal

microscope and then covered with a coverslip. Imaging of the cells was performed on

an Olympus FV500-IX81 confocal microscope. A 5-mW, 488-nm Argon laser was the

excitation source for the Alexa Fluor 488 dyes throughout the fluorescence quenching

experiments. The objective used for imaging was a XLUMPLFL20XW x20 water-

immersion objective with a numerical aperture of 0.95 (Olympus).

Saturation Binding Concentration Determination.

In order to monitor the fluorescence quenching phenomenon between the two

binding sites on live cell membrane, both Alexa488-labeled anit-PTK7 and aptamer-gold

NP conjugates (aptamer sgc8 or TDO5) were added to the cell surface at saturation

concentrations to saturate all the aptamer and antibody binding sites on the cell

membrane. The saturation concentrations for Alexa488-antiPTK7 were determined as

following: varying concentrations of Alexa488-labeled anti-PTK7 (5 nM 300 nM) were

incubated with CCRF-CEM cells (1 x 106) at 4C to prevent receptor internalizations3 for

20 min in the dark in a 200-pl volume of binding buffer containing 20% fetal bovine

serum (FBS). Cells were then washed twice with 700 pL of the binding buffer with 0.1%

sodium azide, suspended in 200 pL of binding buffer with 0.1% sodium azide, and

subjected to flow cytometric analysis within 15 min. The fluorescence was determined

with a FACScan cytometer (BD Immunocytometry Systems) by counting 10,000 events.

A green laser at 488nm with an excitation voltage 750V was used as the excitation

source. The FITC-labeled isotype Mouse IgG2a antibody was used as a negative

control to determine nonspecific binding. All of the experiments for binding assay were


112









FCS is exquisitely sensitive, provides access to a multitude of measurement

parameters in real time, is noninvasive and offers rapid temporal and high spatial

resolution. Through analysis of minute spontaneous fluorescence fluctuations, the

autocorrelation function provides an ideal method for studying diffusion and dynamics of

proteins at nanomolar concentrations in living cells. Coupled with imaging, FCS can be

used on the subcellular level to morphologically dissect biological processes in different

compartments, such as dendritic spines,81 that have been previously intractable to

biochemical experiments. FCS measurements also allow the determination of both cell-

to-cell and position-to-position (within the same cell) variability. Many techniques

measure average values of a seemingly 'homogeneous' ensemble, in which the

heterogeneity of a process in a cell population or within the same cell is lost. This

information may be biologically important to understand the extent of variation in cellular

responses. Besides the sensitivity improvements, FCS also changes the mode of

observing molecular interactions from the change of fluorescence intensity to the

difference in molecular diffusion. This different approach can greatly simplify molecular

probe design for signaling molecular interactions. The detailed theories and

instrumentations of FCS are discussed in Chapter 2.









labeled species. Therefore, to overcome this limitation, the original FCS setup was

upgraded to a novel three-channel dual-color Fluorescence Cross-Correlation

Spectroscopy (FCCS) setup in our lab. The difficulties of overlapping three femtoliter-

sized observation volumes were overcome by careful alignments and calibrations during

the setup. This lab-built FCCS not only inherits the single-molecule detection capability

from FCS, but also further extends its applications for molecular interaction studies by

labeling two species with two spectrally distinct fluorophores. By taking advantage of its

co-localization capabilities and the high sensitivity and low detection limit properties, we

are now further adapting this technique for real-time monitoring of intracellular gene

expression levels.

Besides instrumentation development, nanomaterials have been used to

construct a nanoruler to study the detailed structures of individual membrane receptor

by mapping its binding site distances on live cell membrane. This SET-based nanoruler

uses aptamer-gold-nanoparticle conjugates with different diameters to monitor the

distance between two binding sites on a receptor in the natural physiological

environment of the cell surface. This nanoruler has been proved to successfully

measure separation distances well beyond the detection limit of FRET (~10 nm). This is

a significant development because many membrane proteins will change their

conformations if separated from the cell membranes for in-vitro studies, resulting in an

inability to measure the real binding site distances in these proteins. Thus, for the first

time, we have successfully developed an effective SET nanoruler for live cells with long

distance, simple construction, fast detection and low fluorescence background.167


163









Penicillin-Streptomycin (American Type Culture Collection, Manassas, VA) at 370C

under a 5% CO2 atmosphere. Cells were washed before and after incubation with wash

buffer [4.5 g/L glucose and 5 mM MgCI2 in Dulbecco's PBS with calcium chloride and

magnesium chloride (Sigma-Aldrich, St. Louis, MO)]. Binding buffer used for selection

was prepared by adding yeast tRNA (0.1 mg/ml; Sigma-Aldrich, St. Louis, MO) and BSA

(1 mg/ml; Fisher Scientific, Pittsburgh, PA) to wash buffer to reduce background binding.

Flow Cytometric Analysis. Saturation concentration of Alexa488-labeled anti-PTK7

(200 nM, Figure 5-3A) was incubated with CCRF-CEM cells (1 x 106) at 40C to prevent

receptor internalizations146 for 20 min in the dark in a 200-pL volume of binding buffer

containing 20% FBS. Cells were then washed once with 700 pL of the binding buffer,

then incubated with saturation concentrations of different sizes of gold NPs (5.4nm: 8nM;

10.1nM-42.2nm: 4nM, saturation concentration determination shown in Figure 5-3B) or

with 4nM 15nm silica NP aptamer conjugates for another 20 min. Then cells were

washed twice with 0.1% sodium azide, suspended in 200 pL of binding buffer with 0.1%

sodium azide, and subjected to flow cytometric analysis within 15 min. The fluorescence

was determined with a FACScan cytometer (BD Immunocytometry Systems) by

counting 10,000 events. A green laser at 488nm with different excitation voltages (650V,

700V, 750V) was used as the excitation source. The FITC-labeled isotype Mouse IgG2a

antibody was used as a negative control to determine nonspecific binding. The cell

fluorescence background was determined from samples with cells only as the

fluorescence background for later calculations.

Confocal Imaging Analysis. For confocal imaging, treatment steps for fluorescence

quenching experiments were the same as described in Flow Cytometric Analysis. 30 pL


111













A


Sample


Cover slide


Obhicliie





Lens


Dichroic mirror


Computer


Pinhole


Figure 2-4 FCS experimental set-up.101


Number of molecules N


-10nM
1nM
0.5nM
- 0.1nM
0.05nM
0.01OnM


1
g(O) oc
N


- Fluorescein-dextran 10KDa
- Fluorescein-dextran 40KDa
Fluorescein-dextran 500KDa
- Fluorescein-dextran 2000KDa


1 10 100 1000 10000 U.u I u.' I u lu
t (msec) t (msec)


Figure 2-5 FCS for molecular diffusion studies in solution. (A) Autocorrelation curves
with different concentrations of dye Rhodamine 123. (B) Autocorrelation curves
for fluorescein-dextran complexes of different molecular weights.


S6


Lens
-A









LIST OF REFERENCES


(1) Bacia, K.; Kim, S. A.; Schwille, P. Nat Meth 2006, 3, 83-89.

(2) Chen, Y.; Munteanu, A.; Huang, Y.-F.; Phillips, J.; Zhu, Z.; Mavros, M.; Tan, W.
Chemistry A European Journal 2009, 15, 5327-5336.

(3) Chen, Y.; Yang, C. J.; Wu, Y.; Conlon, P.; Kim, Y.; Lin, H.; Tan, W.
ChemBioChem 2008, 9, 355-359.

(4) Chen, Y.; Huang, J.; Wang, H.; Zhu, Z.; Tan, W. submitted 2010.

(5) Bruce, A.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walters, P. Molecular
Biology of the Cell; Garland Science: New York and London, 2002.

(6) Tan, W.; Wang, K.; Drake, T. J. Current Opinion in Chemical Biology 2004, 8,
547-553.

(7) Yang, C. J.; Jockusch, S.; Vicens, M.; Turro, N. J.; Tan, W. Proceeding of the
National Academy of Sciences of the United States of America 2005, 102,
17278-17283.

(8) Tan, L.; Li, Y.; Drake, T. J.; Moroz, L.; Wang, K.; Li, J.; Munteanu, A.; Yang, C. J.;
Martinez, K.; Tan, W. The Analyst 2005, 130, 1002-1005.

(9) Shangguan, D.; Li, Y.; Tang, Z.; Cao, Z. C.; Chen, H. W.; Mallikaratchy, P.; Sefah,
K.; Yang, C. J.; Tan, W. Proceedings of the National Academy of Sciences 2006,
103, 11838-11843.

(10) Daniel, M.-C.; Astruc, D. Chemical Reviews 2003, 104, 293-346.

(11) Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H. Nat Mater 2005, 4,
435-446.

(12) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-
609.

(13) Omid C. Farokhzad; Jianjun Cheng; Benjamin A. Teply; Ines Sherifi; Sangyong
Jon; Philip W. Kantoff; Jerome P. Richie; Langer, R. Proceeding of the National
Academy of Sciences of the United States of America 2006, 103, 6315-6320.

(14) Tyagi, S.; Kramer, F. R. Nat Biotech 1996, 14, 303-308.

(15) Lakowicz, J. R. Principles of fluorescence spectroscopy, 2nd ed.; Springer:
Heidelberg, 1999.

(16) Rigler, R. Journal of Biotechnology 1995, 41, 177-186.


169









solution (pH 7.4 0.1) with 11.9 mM P043-, 137mM NaCI and 2.7 mM KCI. The reaction

mixture was then allowed to react for 12 h. Unbound oligonucleotides were

subsequently removed by centrifugation and resuspension of the pellet (14,000 rpm,

twice). DNA-coated gold NPs were resuspended, and the concentrations were

determined by measuring the absorptions using a Cary Bio-300 UV spectrometer

(Varian, Walnut Creek, CA). Condensed to a concentration of 10 nM, the DNA-gold NP

conjugates were stored in a 1 x PBS solution (pH 7.4 0.1, 11.9 mM PO43-, 137mM

NaCI and 2.7 mM KCI).

Preparation of aptamer-functionalized silica nanoparticles. 3-Aminopropyl(3-

oxobutanoic acid)-functionalized silica NPs (~15 nm (DLS), 2.5 % (w/v) in DMF) were

purchased from Sigma-Aldrich (St. Louis, MO) and used as the control NPs for

fluorescence quenching experiments. The particles were first washed with double-

distilled water three times (centrifuge rate: 14,000 rpm) to remove dimethylformamide

(DMF) solvent and resuspended in DI water to a final concentration of 200 nM. Before

DNA conjugation, the carboxylic acid groups of silica NPs were activated with 1-ethyl-3-

(3-dimethylaminopropyl) carbodiimide (EDC) and N-Hydroxysuccinimide (NHS) by

adding 1.0 ml of carboxylic acid-functionalized silica NPs to 2.0 mL of 1 x PBS buffer

(pH 7.4) containing 6.5 mg of EDC and 5.8 mg of NHS. After 1.0 h of stirring at 25 oC,

the particles were centrifuged and redispersed in 1 x PBS buffer (pH 7.4) to give the

NHS ester-terminated silica NPs. Excess amino-labeled aptamers (5'-NH2-sgc8 and 5'-

NH2-TDO5) (200 pl, 1 mM) were added and incubated with the silica NPs under gentle

shaking overnight at 4 C. (Aptamer:NP = 1:1000 molar ratio) The conjugates were


109









receptors and identifying binding sites on the cell membrane surface, which could also

be recognized as potential biomarkers124. Nowadays, more and more aptamers have

been generated as specific probes for molecular signatures on the cell surface.

However, there are only limited ways to fully study and use them for biological studies.

Here, we combine aptamer recognition with the highly sensitive FCS tool to perform a

ligand-receptor interaction study on the cell membrane. This technique yields the

information required to estimate receptor density and distribution, thus extending the

potential applications of aptamers generated from cell-SELEX, as well as FCS as an

effective biophysical tool for cell membrane surface study

Using an in vitro cell-SELEX procedure9' 12122, we selected an aptamer, sgc8,

towards T-cell ALL CCRF-CEM cell line. We also successfully elucidated its target

protein, human protein tyrosine kinase-7 (PTK7) by using this newly selected

aptamer125. PTK7 has been discovered to be highly expressed on the cell membrane in

a series of leukemia cell lines124. It is recognized as a potential cancer biomarker,

having a role characteristic of tumors, i.e., signal amplifier or modulator126. In order to

demonstrate FCS as an effective approach for mapping receptor densities on live cells,

we chose two different cell types with different expression levels of PTK7 on the cell

membrane as proof of principle. This is the first study using FCS to estimate the density

of membrane receptor PTK7 on different cell types through aptamer/receptor

interactions.

Experimental Section

Cell Lines

CCRF-CEM (human leukaemia cells), HeLa (cervix adenocarcinoma) K-562 cells

(CCL-243, human chronic myelogenous leukemia) and RBL-2H3 (Rat Basophilic









(128) Erickson, J.; Goldstein, B.; Holowka, D.; Baird, B. Biophysical Journal 1987, 52,
657-662.

(129) Meseth, U.; Wohland, T.; Rigler, R.; Vogel, H. Biophys. J. 1999, 76.

(130) Kinjo, M.; Rigler, R. Nucleic Acids Res. 1995, 23, 1795-1799.

(131) Rauer, B.; Neumann, E.; Widengren, J.; Rigler, R. Biophys. Chem. 1996, 58.

(132) Schwille, P.; Oehlenschlager, F.; Walter, N. G. Biochemistry 1996, 35, 10182-
10193.

(133) Schuler, J.; Frank, J.; Trier, U.; Schafer-Korting, M.; Saenger, W. A. Biochemistry
1999, 38, 8402-8408.

(134) Chen, C.; Ridzon, D. A.; Broomer, A. J.; Zhou, Z.; Lee, D. H.; Nguyen, J. T.;
Barbisin, M.; Xu, N. L.; Mahuvakar, V. R.; Andersen, M. R.; Lao, K. Q.; Livak, K.
J.; Guegler, K. J. Nucl. Acids Res. 2005, 33, e179-.

(135) Shingara, J.; Keiger, K.; Shelton, J.; Laosinchai-Wolf, W.; Powers, P.; Conrad, R.;
Brown, D.; Labourier, E. RNA 2005, 11, 1461-1470.

(136) Weiss, S. Science 1999, 283, 1676-1683.

(137) Zhuang, X.; Bartley, L. E.; Babcock, H. P.; Russell, R.; Ha, T.; Herschlag, D.;
Chu, S. Science 2000, 288, 2048-2051.

(138) Yildiz, A.; Forkey, J. N.; McKinney, S. A.; Ha, T.; Goldman, Y. E.; Selvin, P. R.
Science 2003, 300, 2061-2065.

(139) Blanchard, S. C.; Kim, H. D.; Gonzalez, R. L.; Jr., P., Joseph D. ; Chu, S.
Proceeding of the National Academy of Sciences of the United States of America
2004, 101, 12893-12898.

(140) Liu, G. L.; Yin, Y.; Kunchakarra, S.; Mukherjee, B.; Gerion, D.; Jett, S. D.; Bear,
D. G.; Gray, J. W.; Alivisatos, A. P.; Lee, L. P.; Chen, F. F. Nat Nano 2006, 1, 47-
52.

(141) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P.
Nat Mater 2008, 7, 442-453.

(142) Sonnichsen, C.; Reinhard, B. M.; Liphardt, J.; Alivisatos, A. P. Nat Biotech 2005,
23, 741-745.

(143) Bene, L.; Szentesi, G.; Matyus, L.; Gaspar, R.; Damjanovich, S. Journal of
Molecular Recognition 2005, 18, 236-253.


176









anti-PTK7 was observed as the sizes of aptamer-NP conjugates grew from d=5.4nm

(92.8% fluorescence intensity compared to no NPs) to d=18.4nm (35.9% fluorescence

intensity compared to no NPs). While the aptamer-NP conjugates reached a diameter of

18nm, the fluorescence quenching reached a fluorescence plateau with similar intensity

as the IgG2a control, and larger NP conjugate sizes (18.4nm-42.2nm) remained similar

fluorescence quencher efficiencies. This d=18nm converging point not only indicated

the situation of maximum fluorescence quenching, but also provide the information of

the minimum diameter of gold NPs on the aptamer site whose surface can reach the

dye molecules on the antibody sites. Therefore, the distance between the dye molecule

and the gold NP reached minimum to produce the maximum fluorescence quenching.

Therefore, a fluorescence plateau was obtained and gold NP conjugates with diameters

larger than 18nm gave a similar degree of fluorescence quenching. All these

quantitative fluorescence quenching efficiency results against the change of gold NPs

on the aptamer binding sites were adopted into the further calculations and discussions

in the following part "Binding Site Distance Determination" to further obtain the distance

between the two binding sites on the cell membrane.

Binding Site Distance Determination

As illustrated in Figure 5-7, the relationship of the binding site distance (R), the

NP diameter (d) and the distance from the dye molecule to the gold NP surface (r) is

given by:

d
r=-R (5-1)
2

Since the binding site distance R is fixed, increasing the diameter of the gold

nanoparticle will result in closer proximity between the fluorescence donor and the


121









LIST OF TABLES


Table page

3-1 Comparison of Kd Value of IgE/Receptor Complexes and IgE Receptor
Density on RBL-2H3 Cells Using Different Methods........................................ 88

5-1 Summary of Properties for Gold Nanoparticles of Different Sizes.................... 145

5-2 Sum mary of DNA sequences .................................... ............. ....... ........ 145

5-3 Summary of aptamer-functionalization for gold nanoparticles....................... 146

6-1 Probes and oligonucleotides used in Ribonuclease H kinetic study ................. 161

6-2 Kinetic parameters of pyrene beacon assays for E. coli RNase H.................... 161









of the ellipsoid-shaped volume element is defined by the half-axis length (zw) and half-

axis radius (wxy).

The principle of use of fluorescence fluctuations to obtain molecular diffusion

information is illustrated in Figure 2-3. When a molecule (represented by a red square in

Figure 2-3A) enters the excitation volume element (represented by the blue region), it

emits photons which can be detected by the photon detector in FCS. And depending on

the random path of the molecule, it may remain a while, emitting more photons (maybe

even crossing the entire detection volume element), or, alternately, it may quickly exit

the volume element having emitted only a single photon to the detector. This

randomized behavior can be recorded and analyzed by FCS. Diffusion of fluorophores

into and out of the focus volume alters the local concentration of the fluorophores inside

the tiny focus volume, contributing to the spontaneous fluorescence intensity

fluctuations (Figure 2-2B). After obtaining the fluorescence fluctuations, FCS compares

the fluorescence intensity measured at time t, 6I(t), with that of a later time (t + r),

(t + r), over the mean fluorescence intensity , yields the normalized intensity

autocorrelation function G(r):


G( < (t)I(t + r) > (2-1) 104
G(r) = 1+ (2-1)
2

In order to use this equation for the evaluation of experimentally obtained

autocorrelation functions, Equation (2-1) has been derived to a single-component

solution. The resulting equation is considering only diffusion along the axial dimensions

of the laser beam. 91,100,105,106









scattering system normally requires the dye and the particle surface to locate in close

proximity (<10nm) in order to have the scattering occurred; in longer distance range,

fluorescence quenching effect will be the dominant factor in the interactions.143 In

addition, it was reported that light scattering effect only dominates in smaller sized

nanoparticles (<2nm) due to the insufficient overlap for fluorescence energy transfer. In

contrast, particles with greater than a 2nm diameter have higher probability and

incidence of fluorescence quenching.144 Therefore, there's a great limitation on the

choice of nanoparticle sizes for the ruler constructions. All these restrictions make the

detection of interactions between large complexes on live cells extremely difficult.

Therefore, alternative methods are needed for constructing molecular rulers with a still

larger detection range for these cellular systems.

One such option involves fluorescence measurements which have wide

applications in cellular systems as well as in vivo systems. Recently, several groups66-74

have reported the nanoparticle surface energy transfer (SET) where energy transfer

flows from a donor molecule to a nanoparticle surface at a much slower decay rate than

the dipole-dipole energy transfer in FRET, with a 1/d4 distance dependence.66 While

SET is similar to FRET, in that the interaction is dipole-dipole in nature, it is

geometrically different from FRET because an acceptor nanoparticle has a surface and

an isotropic distribution of dipole vectors to accept energy from the donor, leading to a

dipole-surface resonance mechanism.67-70 This arrangement effectively breaks the

inherent detection barriers of FRET, thereby increasing the probability of energy transfer

and ultimately enhancing the efficiency of SET over FRET. The intensity quenching

mechanism via coupling of the oscillating electronic dipole of a dye to a metal surface


105









CHAPTER 6
ENGINEERING A LIGHT-SWITCHING MOLECULAR BEACON (MB) FOR
RIBONUCLEASE H KINETIC STUDY

Introduction


RNase H is a ribonuclease that specially degrades the RNA strand in a RNA-

DNA hybrid to produce 3'-hydroxyl and 5'-phosphate terminated products. It is a non-

specific endonuclease and catalyzes the cleavage of RNA via an endonucleolytic

mechanism 148, aided by an enzyme-bound divalent metal ion. However, DNA strands

or unhybridized RNA strands will not be degraded. The enzyme is involved in several

important cellular processes including DNA replication, DNA repair and transcription. 149

Members of the RNase H family can be found in nearly all organisms, from archaea and

prokaryota to eukaryota. RNase H also has wide applications in molecular biology and

biotechnology in terms of its unique cleaving property. Retroviral RNase H, a part of the

viral reverse transcriptase enzyme, is an important pharmaceutical target, as it is

absolutely necessary for the proliferation of retroviruses, such as HIV. Inhibitors of this

enzyme could therefore provide new drugs against diseases like AIDS. E. coli RNase H

usually requires at least 6 base pairs of RNA-DNA hybrids as substrates to bind and

cleave effectively in solution, while the hybrid length required in living cells may be

somewhat greater. 150 The complete digestion of poly(rA):poly(dT) with E. coli RNase H

yields oligoribonucleotides with varying chain lengths, ranging from monomers to

hexamers. 151 In order to understand more about these functions and processes, and

more importantly, to screen new drugs against retroviruses, it is necessary to develop a

fast, real-time, sensitive and isotope-label free system to assay the cleavage activity of

RNase H. A number of traditional methods have been used to assay the enzyme


147










m Tripsin Treatment


> 0.8-

W 0.6-






S5 10 15 20 25 30 35 40

Time (min)
Figure 3-8 Competition studies and Trypsin treatment. (A) Time course of fluorescence
count rate for the displacement of FITC-labeled sgc8 binding by non-labeled
sgc8 on the membrane of HeLa cells. Cells were first incubated with 3.0nM of
FITC-sgc8 for 40 min at 4C. A 1000-fold molar excess of non-labeled sgc8 was
then added to compete against the labeled sgc8 in binding to the target receptor
PKT7. Fluorescence count rates were monitored for 90min. Blue squares are
competition studies with a control sequence, which is a non-labeled DNA library
containing a randomized sequence of 41 nucleotides. HeLa cells were first
incubated with 3.0nM of FITC-sgc8 for 40 min at 4C. A 1000-fold molar excess
of non-labeled control sequence (library) was then added to compete the binding
of labeled sgc8 to the target receptor PKT7. (B) Time course of normalized
membrane-bound aptamer density change of displacement of FITC-labeled sgc8
binding by non-labeled sgc8 on HeLa cell membrane. Same reaction condition
was used as (A). Changes in membrane-bound FITC-sgc8 density were
monitored for 90min. Blue circles are competition with the control sequence
(library). (C) Time course of normalized density change with trypsin treatment on
HeLa cell membrane. Cells were first incubated with 3.0nM of FITC-sgc8 for 40
min at 4C to saturate all the receptor binding sites. Then, a 10x diluted trypsin
solution was added. Changes in membrane-bound FITC-sgc8 density were
monitored for 90min at 37uC.
monitored for 90m in at 37 C.









differentiate bound aptamers from free ones. The corresponding fluorescence intensity

fluctuations are shown on the bottom in Figure 3-2. Stable fluorescence fluctuations

show that no photobleaching effects had occurred during the entire detection time in the

detection volume. Control experiments with a randomized sequence of 41 nucleotides

(Library, Figure 3-2B), which have been shown as the negative binding control to

receptor PTK7 in cell-SELEX9, show similar diffusion times for aptamers free in solution

and on cell membranes. It indicates only specific interactions between aptamers and

membrane receptors will give the response of increased diffusion time on cell surface.

Specificity of Binding by the Aptamers

The specificity of aptamer/receptor interactions, as demonstrated by the specific

labeling of target receptors by aptamers, provides the basis for studying receptor

density using FCS. Results shown in Figure 3-3 demonstrate the specific binding of

aptamer sgc8 in different concentrations to its membrane-bound receptors in different

cell types. The formation of complexes between the receptor and aptamer is identified

by the change in the autocorrelation function in the positive target cells, which is not

shown in the negative control cells that lack PTK7 expressions on the cell membrane9'
127

Binding Affinity

Studying the density and distribution of membrane receptors at the molecular level

in living biological specimens has been challenging. The above experiments have

shown the efficacy of FCS in terms of ligand binding specificity to membrane receptors.

However, since FCS is also capable of measuring detailed molecular interaction

between ligands and receptors, this property was applied to study the binding affinities

of aptamer-receptor interactions in their native physiological environment on the cell









RNase H. For determination of the Michaelis-Menten kinetic parameters, the beacon

assay concentration was varied from 25nM to 500nM, in the range of the previously

reported Km values for RNase H. 160,161 Dye calibration curve using SYBR

GreenERTM is applied to determine the accurate concentration of the hybrid. SYBR

GreenERTM is a double-strand oligonucleotide binding dye that can quantitatively

differentiate duplex from single strand oligonucleotides after hybridization. In all kinetic

experiments, initial rate measurements were obtained in the first 30 second, with an

enzyme concentration (6nM, 50units/ml). All experiments were done at 37C and

repeated for 2-3 times. Average values were used for calculation. Data in Table 6-2 and

Table 6-3 were obtained by performing curve fitting to the Michaelis-Menten equation

using OriginPro 7.0 (Microcal Software Inc.).

Results and Discussion

Here, we describe the molecular engineering of a light-switching excimer beacon

probe for RNase H activity monitoring. Some spatially sensitive fluorescent dyes, such

as pyrene 15160, 162, 163 and BODIPY FI 161,164 can form excimer upon proximity of an

excited-state molecule with another ground-state molecule. The excimer results in an

emission wavelength shift to a longer wavelength compared to the monomer. The

formation of excimer between two pyrene molecules that are connected by a flexible

covalent chain, such as a DNA chain, is highly useful to probe spatial arrangement.

Similar to FRET, the spatially dependent property of excimer formation can be used as

a signal transduction in the development of effective molecular probes. This unique

technique is especially useful for the design of MBs, which can undergo conformational

change upon target binding. By attaching pyrene molecules 162 to both ends, an excimer


151

































2.0-

1.8-

1.6-
1 1
- 1 1.4-

1.2-

1.0-


SALe 488-aiiliPTK7(4)750v
ALex488-antiPTK7 (4) 700v
A ALe>48-antiPTK7 (4) 650v


5 10 15 20 25 30 35 40 45
NP Diameter (nm)




C ALex488-antiPTK7 i44 750v
ALex488-antiPTK7 141 700v
A ALex488-antiPTK7 i4I 650v

Red: .77r
ue: ,073r0 988--0,


U X i I I I
5 10 15 21
NP Diameter (nm)


Figure 5-14 Effects of different laser intensities on fluorescence quenching efficiency. A)
Histogram of the fluorescence quenching assay with different laser excitation
intensities (Red: 750V; Green: 700V; Blue: 650V) from flow cytometry analysis.
Experimental procedures were the same as described in Fig. 1. In all these
experiments, the labeling efficiencies of Alexa Fluor 488 dyes on the anti-PTK7
were kept the same, and there were four dyes on each antibody according to the
labeling protocol. All of the experiments for fluorescence quenching assay were
repeated three times, and the average value was determined as the mean
fluorescence intensity. B) Relationship between fluorescence quenching
efficiency and gold nanoparticle diameter. Fluorescence quenching efficiencies
were determined from fluorescence intensity in Fig. 6A using the following


141









measurement is accomplished by focusing an excitation laser beam onto the sample

and then monitoring the fluorescence fluctuations derived from the focal region of the

laser beam (Figure 3-1). Diffusion of fluorophores into and out of the focal volume alters

the local concentration of the fluorophores, thus giving rise to spontaneous fluorescence

intensity fluctuations. Correlating the fluorescence intensity fluctuations at time t, 51(t),

with that of a later time (t+T), 61(t+T), over the mean fluorescence intensity <1>, yields the

normalized intensity autocorrelation function G (T):

G(r) = 1+ < 8t)(t + ) > (2-1)104
<1>2

Then, we are able to monitor the real-time aptamer/receptor interactions on the

cell membrane by tracking the diffusion of the fluorescent-labeled aptamers into and out

of the focal volume. More specifically, a derivative of G (T), which assumes a single

component solution and only considers diffusion along the 3D axial dimensions (wxy and

wz) of the laser beam, yields equation (2-2):

1 1 1


SDTD
l l l (2-2) lO4

Z; 'rz


The amplitude of G (T) depends on the absolute number of molecules, N,

occupying the observation volume. A higher number of molecules in the observation

volume (higher concentration) results in a lower correlation amplitude. Mean diffusion

time iD describes the average time it takes for a molecule to diffuse through the

observation volume. It works as a characteristic fingerprint for specific molecules in

specific diffusion states, as molecules with different molecular weights take different









other. As shown in Figure 5-7, the distance, R, from the fluorophore on the antibody

binding site to the center of the nanoparticle on the aptamer binding, is equivalent to the

distance between the two binding sites on the PTK7 receptor to be measured. In order

to simplify the model, R was divided into two parts: the distance r, from the fluorophore

to the surface of the nanoparticle, and the distance from the surface of the particle to its

center, which is the radius of the particle, d/2, with R= r + (d/2). Therefore, as the size of

the gold nanoparticle (d) is varied, the distance from the fluorophore to the particle

surface (r) changes accordingly. Although some variations in position around the center

point of the aptamer binding site can potentially occur, one million cells were counted

each time to cancel out these variances. In addition, a series of gold nanoparticles of

different sizes were adopted to fit in the binding pockets to avoid this steric effect.

Therefore, by controlling the sizes of the gold nanoparticles, the distance from the

fluorophore molecule to the surface of gold nanoparticles could be manipulated, and the

relationship between the size variations of the gold nanoparticles and the change in the

energy transfer efficiency could be evaluated.

First, gold nanoparticles of different sizes were prepared with precisely controlled

diameters from 5 nm, 10 nm, 13 nm, 15 nm, 18 nm, 20 nm, 25 nm, and 31 nm to 42 nm,

as described in Experimental Section. The sizes were controlled by using different

amounts of reducing agent (sodium citrate) and were monitored afterwards by TEM

imaging. The average diameter of each size was obtained by measuring the sizes of

100 particles from the TEM images with a narrow size distribution (shown in Table 5-1).

In order to obtain the specific SET information from the two binding sites on the same

receptor, the nanoparticle size range (d=5.4nm-42.2nm) was chosen to avoid


118









other molecules can be imaged with spatial resolution beyond the limits of conventional

optical microscopy.

Surface Energy Transfer (SET)

Recently, several groups66-74 have reported the phenomenon of nanoparticle

surface energy transfer (SET), in which energy transfer flows from a donor molecule to

a nanoparticle surface at a much slower decay rate than the dipole-dipole energy

transfer in FRET, the l/d4 distance dependence of SET is described in the Equation (1-

4): 66

-4


1+( )
L-+r) (1-4)


where the Forster radius Ro is the distance at which energy transfer is 50% efficient,

and r is the distance between the donor and the surface of the metal particle.

While SET is similar to FRET, in that the interaction is dipole-dipole in nature, it

is geometrically different from FRET because an acceptor nanoparticle has a surface

and an isotropic distribution of dipole vectors to accept energy from the donor, leading

to a dipole-surface resonance mechanism.67-70 This arrangement effectively breaks the

inherent detection barriers of FRET (~10 nm), thereby increasing the probability of

energy transfer and ultimately enhancing the efficiency of SET over FRET.

The intensity quenching mechanism via coupling of the oscillating electronic

dipole of a dye to a metal surface with loss of energy was developed by Chance, Prock

and Silbey,75 and by Persson and Lang76 for metals. Aside from bulk systems, recent

attempts have conjugated different lengths of dsDNA onto the metal nanoparticle









substrate for the RNase H cleavage, has a low fluorescence background at 485nm

(Figure 6-2). After the addition of the enzyme, only the RNA strand will be cleaved from

the duplex 148, which sets free the DNA beacon. The restoration of the hairpin structure

brings the pyrene moieties back together and gives a dramatic fluorescence

enhancement at the excimer emission at 485nm, and the real-time fluorescence

monitoring is shown in Figure 6-3. This critical step is the one we used for the RNase H

kinetic parameters study. Comparing to normal fluorophore-quencher labeled MB

design, whose fluorescence signal will be quenched at this step; the pyrene beacon

assay results in a fluorescence increase and thus has a much higher sensitivity. It

brings a lot more convenience to the calculation with the signal enhancement in the

design. Also, Scheme 5-1 reveals another important advantage of the light-switching

excimer signaling approach over traditional gel electrophoresis experiments: real-time

detection without separation. Because only the cleaved duplex gives excimer emission,

the uncleaved targets do not have to be separated from the solution for detection. In

addition, since it is a real-time detection, the fluorescence assay gives us a much

clearer picture for the enzyme cleavage activity and also minimizes the inconvenience

of being discontinuous in experiment manipulation. In order to confirm that the signal

enhancement comes from the cleavage of the RNA strand by the RNase H, a

complementary DNA is added to the solution after the enzyme cleavage. The free

beacon hybridizes with the complementary DNA and opens up its hairpin structure

again as shown by the dramatic decrease in excimer emission. (Figure 6-3)

Optimization of cDNA lengths


153









transported across the cell membranes to recognize bioactive substances. This offers a

chance to recover the real recognition event in vivo. Owing to these advantages, the

combination of DNA molecular design and with a variety of nanoparticles has been

widely applied in the interdisciplinary fields of chemistry, biology and medicine.63

In summary, the combination of functional nucleic acid molecules and

nanoparticles offers significant advantages for bioanalysis and molecular recognition

studies. The high surface-area/volume ratio of nanoparticles not only provides an

effective platform for nucleic acid probes for molecular recognition, but can also

increase the loading efficiencies of molecular probes to enhance the detection

sensitivity. In addition, nanoparticles can not only work as molecular probe carriers, but

can also act as reporters to simplify the design of molecular probes. In addition, the

easy handling of DNA strands and different modification strategies of nanoparticles

provide a vast platform to achieve molecular recognition. Finally, with the help of

nanoparticles, nucleic acids can escape digestion by, for example, cellular nucleases,

and be transported across the cell membranes to recognize bioactive substances. This

offers a better chance for applying nucleic acid probes for in vivo applications. Therefore,

the combination of DNA molecular design and different nanoparticles has been widely

applied in chemistry, physics and medical research.

Fluorescence Methods for Signal Transduction

Fluorescence is a widely used tool for a variety of investigations in biochemical,

medical, and chemical research due to its high sensitivity, nondestructive nature, and

multiplexing capabilities.15 This section presents different types of fluorescence

mechanisms for signal transduction for use in the design of molecular probes.









Preparation of Aptamer-Functionalized Gold Nanoparticles ...................... 108
Preparation of Aptamer-Functionalized Silica Nanoparticles ..................... 109
Antibody Labeling .............. .... ..................................... ........ ....... 110
Cell Culture .................. ....... .................. 110
Flow C ytom etric A nalysis......... ............................................. ............... 111
Confocal Im aging A nalysis............................................... 112
Saturation Binding Concentration Determination......... .............................. 112
Competition Studies to Confirm the Saturation Binding of Aptamer-NP
Conjugates on Cell Mem brane .............. .... ......................................... 114
Results and Discussions.......................... ............ .................... .................... 115
Competition Studies between Aptamer Sgc8 and Antibody Anti-PTK7 on
P T K 7 ............. ............ ... ... ... ...... .. .. .. ....... .. ..... .. ........................ 1 1 5
Measure the Distance between Two Binding Sites using FRET.................... 116
SET Nanoruler Construction ............................. .......... .... ............... 117
Binding Site Distance Determ ination............................................... 121
Potential Effects of Steric Hindrance on the Fluorescence Quenching......... 124
Effects of Different Excitation Intensity and Labeling Efficiency on Distance
D term nation ............................ .... ................. ......... ........ ................... 125
Effects of Different Antibody/Dye Labeling Efficiencies on the Calculations
of Binding Site Distances ................ ................... ... ..... ...... ......... 127
Using Fluroecent Quenchers as Energy Acceptors .................................... 127
C o n c lus io n s ............. ......... .. .. ......... .. .. ......... ................................ 12 8

6 ENGINEERING A LIGHT-SWITCHING MOLECULAR BEACON (MB) FOR
RIBONUCLEASE H KINETIC STUDY............................... 147

Intro d u ctio n .............. ................. ............................................. ............... 14 7
Experim mental S section ........... ........ ......... .......... ............... .............. 149
Materials ................ ......... .................. 149
Instrum e nts ......... ............... ................ ............... ...... ...... ....... 14 9
Pyrene-MB Synthesis and Purifications........ ........ .................... ............... 149
Pyrene Beacon Assays for RNase H Studies............................................ 150
Results and Discussions............................. ............... 151
Design of Light-Switching MB .............................. ............... 152
Optim ization of cDNA lengths ......................................... ......................... 153
Light-Switching MB Assay for Ribonulcease H Kinetic Study..................... 154
Sequence Dependence on Ribonuclease H Cleavage Kinetics.................... 155
Conclusions .............. ......... ........... ......... ............... .......... 156

7 SUMMARY AND FUTURE DIRECTIONS ................................ ................. 162

Instrumentation Development and Nanomaterial Engineering for Live Cell
M apping and B ioanalysis ........................................ ............... .............. 162
Future D directions .......................... ...... ..... ........... ... ... ................ 164
Exploration of FCS/FCCS Applications for Molecular Interaction Studies
inside Living C ells ................................. ........... .. ............ ........... 165









INSTRUMENTATION DEVELOPMENTS AND NANOMATERIAL ENGINEERING FOR
LIVE CELL MAPPING AND BIOANALYSIS




















By

YAN CHEN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010









efficiency and different anti-PTK7 dye labeling efficiencies was determined from
fluorescence intensity in Figure 8A using equation4:
SI -I I = (AntiPTK7 + NP sgc8) -Cell A linear relationship was
0=- = 1- = 1- A linear relationship was


I1 I1


AntiPTK7 Cell


obtained between the fluorescence quenching efficiency and gold nanoparticle
diameters (d=5.4nm-18.4nm) for different labeling efficiencies. C) Plot of
S 1 4
-1 vs. NP diameter gave linear relationships between d=5.4nm-18.4nm.

The linear fits for different labeling efficiencies resulted in similar distance R
results (red: R=(13.382.59) nm; grey: R=(13.061.96) nm; green:
R=(13.121.40) nm).


* A -- Alex488-AB 200nM B --Alex488-AB 200nM
-- BHQ-sgc8







i .. m.. .-.. .. .. ..

-- Alex488-AB 200nM -- Alex488-AB 200nM
C O -- Dabcyl-10T-sgc8



IA



me -


-CEM cell only


- Alex488-AB 200nM
- BHQ-sgc8



- Dabcyl-10OT-sgc8


Figure 5-16 Flow cytometry analysis using different fluorescence quenchers as energy
acceptors. (A) Aptamer sgc8 was labeled with a BHQ functional group on its 5'
end and used as an energy acceptor on the aptamer binding site (dark blue
curve). (B) Same as (A) with a 10-polyT linker between the aptamer sgc8
sequence and the BHQ functional group (light blue curve). (C) Aptamer sgc8
was labeled with a Dabcyl functional group on its 5' end and used as an energy
acceptor on the aptamer binding site (green curve). (D) Same as (C) with a 10-
polyT linker added between the aptamer sgc8 sequence and the Dabcyl
functional group (purple curve). Experimental procedure was similar to Figure 1.
The fluorescence was determined with a FACScan cytometer (BD


144









monitor bound ligands from high concentrations of free ligands in order to determine

receptor densities on cell membrane. KK1H08 was selected for K-562 cells and

showed low binding affinity with Kd=29641nM112. It requires a high concentration of

total aptamers (0.5pM) to saturate all the aptamer binding sites. Our FCS studies prove

that it can differentiate bound aptamers from high background of free ones. A

Kd=27129 nM was obtained and used to determine the receptor density on the cell

membrane. (Figure 3-7). Further experiments using excess labeled aptamer sgc8

(250nM) incubated with HeLa cells tested by FCS also proves the capability of FCS to

differentiate bound aptamers from the high background of free aptamers (bound

fraction=161%). In addition, the receptor density determined under this condition

(45929 receptor/pm2) is comparable to the one determined previously under low

aptamer concentrations (55090 receptor/pm2), which further demonstrates the

robustness of this receptor density determination using FCS.

Competition Studies and Trypsin Experiments

In order to make certain that the method developed here is effective, control

experiments were conducted to further confirm the calculated results. To verify that the

receptor density is obtained through the specific labeling of receptors using fluorophore-

labeled aptamers, a competitive displacement with non-labeled aptamer was examined

(Figure 3-8A, B). In this experiment, cells were first incubated with 3.0nM of labeled

aptamers, and after 40 min, a 1000-fold molar excess of non-labeled aptamer was

added to compete against the labeled aptamer in binding to the target receptor. Since

the number of receptors on the surface is limited, the majority should be bound by the

non-labeled aptamer given the huge excess of the non-labeled population. If the labeled

aptamers are displaced by unlabeled aptamers, which signifies competition between









3 MAPPING RECEPTOR DENSITY ON LIVE CELL MEMBRANE USING FCS....... 59

Introduction ............... .................. ........ .................. 59
Experimental Section ...... ................... .................. 63
Cell Lines ................. ........ ........ ...... ............... 63
Aptamers Synthesis ............... ..... ......................... 64
A antibody ........................................................................................ 64
Trypsin Treatment of Cells............................ .................... 65
FCS Instrum entaion Set-Up............................................... .................... 65
Results and Discussions ........... .. ........ .................... 65
FCS Measurement and Analysis ................................... .. ............... 65
Binding of Aptamers to Membrane Receptors .............................. ......... 67
Specificity of Receptor Binding by the Aptamers ............................ 69
Binding Affinity Determ nation (Kd)............................................... ............... 69
D density Study.......................................................................... 71
Competition Studies and Trypsin Experiments ............................................. 74
Control Cell Line for Method Evaluation.......................................... 76
C o n c lu s io n s .......... ......... ......... ....................... ............... 7 6

4. UPGRADING FCS TO DUAL COLOR-FLUORESCENCE CROSS-
CORRELATION SPECTROSCOPY (FCCS) FOR WIDER APPLICATION TO
MOLECULAR INTERACTION STUDIES ................................. 89

Limitaions of FCS Autocorrelation Analysis for Molecular Interaction Studies........ 89
Theories of Dual-Color FCCS for Molecular Interaction Studies ........................... 91
Lab-Built FCCS Intrum entation Set-Up.............................................. ............... 93
Scheme of Lab-Built FCCS Set-Up.......................................... 93
Components of Lab-Built FCCS Set-Up............................................ 94
Excitation ........... a ......... .......... ..... ........................... 94
Beam com biners ......... ........................................... ............... 94
Objective ........... ............... ............ ............... 95
Dichroic mirrors ....................... ......... ................ 95
Filters..................... ............................ ............... 96
Pinholes ......................................... ............... 96
Correlators..................................................... .... .......... 97
Construction of the Lab Built FCCS Set-Up ............................. ............... 97
Apply FCCS for Intracelluar mRNA Detection.................................................... 98
S u m m a ry ................................................... .................. ............................... 9 9

5 ENGINEERING A SURFACE ENERGY TRANSFER (SET) NANORULER FOR
MEASURING PROTEIN BINDING SITE DISTANCES ON LIVE CELL
M E M B R A N E ............. ......... .. .. ......... ................................ ............... 1 0 4

Intro d u ctio n ............. ............ .. .. ........................... .................. ............... 10 4
Experimental Section .................. ..... ... ..... .. ......... ............. 107
Preparation of Gold Nanoparticles of Different Sizes................................... 107
A ptam er S ynthesis ................................. .......................... ............ 107









using 5'-amino-modifier-C6 linker phosphoramidite. The column then was flushed slowly

with dimethylformamide (DMF) (15 ml), piperidine (20%) in DMF (15 ml), trichloroacetic

acid (3%) in dichloromethane (15 ml), and then another DMF (15 ml). The CPG

contained within the column was released into DMF solution (1 ml) containing pyrene

butyric acid (57.7 mg, 200 pmol), dicyclocarbodiimide (41.3 mg, 200 pmol), and

dimethylaminopyridine (24.4 pg, 200 pmol). After stirring for 3 h, the solution was

centrifuged, and the supernatant was discarded. The pellet was washed three times

with DMF, methanol, and water, respectively, before incubated in a solution of

methylamine (50%) in ammonia at 65C for ~10 min. The resulting clear and colorless

supernatant was collected. Under UV radiation, an intense green fluorescence was

observed from the collected solution. The beacon solution was desalted with a

Sephadex G-25 column (NAP-5, Amersham Pharmacia) and dried in a SpeedVac. The

dried product was purified by HPLC using a C18 column with a linear elution gradient

with buffer B changing from 25% to 75% in 25 min at a flow rate of 1 ml/min. The

second peak in chromatography that absorbed at 260 and 350 nm, and emitted at 485

nm with 350 nm excitation, was collected as the product. The collected product then

was vacuum-dried, desalted with a G-25 column, and stored at -20C for future use.

Pyrene Beacon Assays for RNase H Studies: Assays were carried out in 100pl of

USB RNase H buffer (TRIS-HCI (20mM, pH7.5), KCI (20mM), MgCI2 (10mM), EDTA

(0.1mM), and DTT (0.1mM)) 154, containing RNase H (6 nM, 50units/ml) and DNA

beacon: RNA hybrid (100nM) at 1:1 ratio. An increase in fluorescence emission at

485nm, upon excitation at 340nm, indicates the hydrolysis progress of the hybrids. The

maximum fluorescence emission was determined by incubating the hybrids with excess


150









(51) Knox, S. J.; Meredith, R. F. Semin Radiat Oncol. 2000, 10, 73-93.

(52) Amstad, E.; Gillich, T.; Bilecka, I.; Textor, M.; Reimhult, E. Nano Letters 2009, 9,
4042-4048.

(53) Bunz, U. H. F.; Rotello, V. M. Angewandte Chemie International Edition, 49,
3268-3279.

(54) Schultz, D. A. Curr. Opin. Biotechnol. 2003, 14, 13.

(55) Chah, S.; Hammond, M.; Zare, R. Chem. Biol. 2005, 12, 323.

(56) Glynou, K.; loannou, P. C.; Christopoulos, T. K.; Syriopoulou, V. Anal. Chem.
2003, 75, 4155-4160.

(57) Oh, J. S.; Ha, G. W.; Cho, Y. S.; Kim M.J.; An, D. J.; D.S., S. Clin. Vaccine
Immunol. 2006, 13, 520-524.

(58) Tan, W.; Wang, K.; He, X.; Zhao, X. J.; Drake, T.; Wang, L.; Bagwe, R. P. Med.
Res. Rev. 2004, 24, 621.

(59) Kim, H. K.; Kang, S. J.; Choi, S. K.; Min, Y. H.; Yoon, C. S. Chem. Mater. 1999,
11,779.

(60) Smith, J. E.; Wang, L.; Tan, W. Trends Anal. Chem. 2006, 25, 848.

(61) Bagwe, R. P.; Yang, C. Y.; Hilliard, L.; Tan, W. Langmuir, 2004, 20, 8336-8342.

(62) Wang, J. Small 2005, 1.

(63) Aldaye, F. A.; Palmer, A. L.; Sleiman, H. F. Science 2008, 321, 1795-1799.

(64) Ingle, J. D.; Crouch, S. R. Spectrochemical Analysis: Prentice Hall, New Jersey,
1988.

(65) Tyagi, S.; Bratu, D. P.; Kramer, F. R. Nat.Biotechnol. 1998, 16.

(66) Yun, C. S.; Javier, A.; Jennings, T.; Fisher, M.; Hira, S.; Peterson, S.; Hopkins, B.;
Reich, N. O.; Strouse, G. F. Journal of the American Chemical Society 2005, 127,
3115-3119.

(67) Darbha, G. K.; Ray, A.; Ray, P. C. ACS Nano 2007, 1, 208-214.

(68) Jennings, T. L.; Schlatterer, J. C.; Singh, M. P.; Greenbaum, N. L.; Strouse, G. F.
Nano Letters 2006, 6, 1318-1324.

(69) Ray, P. C.; Darbha, G. K.; Ray, A.; Hardy, W.; Walker, J. Nanotechnology 2007,
18, 375504.


172









P = Scoered A ==D AtA = (100 cm2 /s)x(5 x 10-6 )x(1300 / m2)= 6.5x 10 (
10000

Therefore, the contribution of inter-receptor interaction is less than 1/10000 to the

total SET interaction. In other words, the contribution of the inter-receptor interaction

effects is negligible for the distance determinations.


168










1.2 fL with half axes wxy=0.235 pm and wz=3.93 pm. The sample was exposed to
laser for 30 seconds for obtaining the autocorrelation curve. Stable fluorescence
intensity was observed during the 30s detection time which indicates no
photobleaching. Different numbers of bound aptamers in the confocal volume at
different aptamer incubation concentrations (50nM, 150nM, 250nM, 300nM, 400nM
and 500nM) were calculated from the fitting of autocorrelation curve. The
equilibrium dissociation constants (KD) of the aptamer-cell interaction were obtained
by fitting the dependence of fluorescence intensity of specific binding on the
concentration of the aptamers to the equation Y=BmaxX/(Kd + X), using Microcal
Origin 6.0. From the binding curve, a 500nM aptamer concentration was enough to
saturate receptor binding sites on K562 cells.


80.

N
60.


40.

0
U '2n


1.0-

% 0.8-

o 0.6-



o
cc 0.4-

z 0.2-

0.0-


Competition 1000X non-labeled sgc8
Competition 1000X non-labeled Library


6 i i" 'o '" 1o
0 20 40 60 80 100
Time (min)

SCompetition 1000X non-labeled sgc8
o Competition 1000X non-labeled Library


0 20 40 60
Time (min)


80 100









detectors (avalanche photodiode (APD)). The tube lens of the microscope focuses the

emission beam onto pinholes at the front-ends of the detectors to reduce the collection

of fluorescence light from above and below the focal plane. The resulting optically

delimited detection volume is typically less than one femtoliter. At last, the correlator

collects the fluorescent signal from each channel and conducts the auto correlation and

cross-correlation calculations. Results are read out by the computer.

Components of the Lab-Built FCCS Set-Up

Excitation: There are three lasers as the excitation sources for the FCCS set-up.

They are one Ar-laser and two He-Ne lasers.

The Ar-laser (Innova 190C, Coherent Inc., Santa Clara, CA) is a coherent laser

that has multiple laser lines (458, 477, 488, 514 nm), in which 488nm has the strongest

laser intensity and is the one used for the FCCS measurements. Neutral density filters

are used to attenuate the beam intensity from 20 mW down to 0.2 mW to focus on the

sample species.

The He-Ne lasers emit at 543nm (maximum 0.5 mW) and 632.8nm (maximum

2.0 mW) (Thorlabs, Newton, NJ). Neutral density filters are also applied to attenuate the

beam intensity to a final 0.2 mW intensity.

Beam Combiner: Two beam combiners are used to overlap the three different

lasers lines to the same focus volume for the FCCS set-up. Referring to Figure 4-4, the

first beam combiner combines the laser beam of 543nm with 632.8nm by transmitting

543nm and reflecting 632.8nm (Chroma, Bellows Falls, VT). The second beam

combiner combines the three laser beams of 488nm, 543nm and 632.8nm by









Another aspect of this research was the development of molecular probes based

on nanomaterial engineering Two different types of molecular beacon (MB) probes were

designed for enzymatic activity studies3 and protein inhibition studies4.

In summary, this research mainly focuses on instrumentation development and

nanomaterial engineering for bioanalysis and biomedical applications, especially for cell

membrane receptor studies and intracellular measurements. A successful outcome from

these studies will lead to a better understanding of biological events and process.









straightforward preparation of the aptamer-nanoparticle conjugates by direct incubation

of the components.

Experimental Section

Preparation of gold nanoparticles of different sizes. Various sizes of gold NPs were

synthesized according to the method developed by Frens145. Gold NPs were adopted in

the system because of the simplicity and reproducibility of the synthetic and

bioconjugation techniques, as well as their unique fluorescence quenching property.

Briefly, 0.5 ml of 1% chloroauric acid (Sigma-Aldrich, St. Louis, MO) was added to 50 ml

of double-distilled water, and the solution was heated to boiling. Next, different amounts

of 1% sodium citrate (Riedel-deHaen, Seelze, Germany) were added to the solution to

synthesize different sizes of gold NPs. The solution was refluxed until a color change

from dark blue to red was observed. The sizes and absorption spectra of various gold

NPs were verified with the Hitachi H-7100 transmission electron microscope (Tokyo,

Japan) (Figure 5-1) and the Cary 100 UV-Vis spectrophotometer (Varian, Palo Alto, CA)

(Figure 5-2). The concentration of gold NPs in each aliquot was also determined by UV-

Vis spectroscopic measurements via the Beer's law (A = Ebc). The 5.4nm gold NPs

were purchased from Sigma-Aldrich (St. Louis, MO). Characteristic parameters for the

preparation and characterization of different sets of gold NPs are summarized in Table

5-1.

Aptamer Synthesis. All DNA oligomers were synthesized on an AB13400 DNA/RNA

synthesizer (Applied Biosystems, Foster City, CA) in our laboratory. All DNA

oligonuclease bases and 5'-modifiers were purchased from Glen Research, Sterling, VA.

The aptamer sgc8 was labeled with various 5'-modifiers. A DNA library containing a

randomized sequence of 41 nucleotides was labeled with a 5'-FITC, -TAMAR or -Cy5


107









modifier and used as a control for each detection channel. Sequence TD05, a control

aptamer for CEM cells, was labeled with 5'-Thiol-Modifier [DMTO(CH2)6-S-S-(CH2)6-

OP(N(iPr2))(O(CH2)2CN)] or 5'-Amino-Modifier [MMT-NH-(CH2)6-OP(N(iPr2))(OCNEt)]

for particle conjugations. For the deprotection procedures, sequences labeled with 5'-

TAMAR-Modifier were deprotected in TAMRA deprotection solution (0.05M potassium

carbonate in methanol) at 65C for 3-4 hours. Sequences labeled with 5'-Cy5-Modifier

were deprotected in ammonium hydroxide at room temperature for 24-36 hours.

Sequences labeled with 5'-Amino-Modifier were deprotected in ammonium hydroxide at

40C for 17 hours. All other sequences were deprotected in AMA (ammonium

hydroxide/40% aqueous methylamine 1:1) at 65C for 20 minutes. After deprotection, all

sequences were further purified by reverse phase HPLC (ProStar, Varian, Walnut Creek,

CA) on a C-18 column. A Cary Bio-300 UV spectrometer (Varian, Walnut Creek, CA)

was used to measure absorbance to quantify the manufactured sequences. DNA

sequences with different labeling are summarized in Table 5-2.

Preparation of aptamer-functionalized gold nanoparticles. The strategy for

functionalizing gold nanoparticles with thiol-modified oligonucleotides followed a

published procedure145. The thiolyated aptamers (sgc8 or TD05) were added to gold

NP stock solutions for surface modification. The amount of aptamer needed for the

different gold nanoparticle sizes (50 mL) was calculated from estimations of the surface

area for each gold NP to ensure saturated surface coverage, as listed in Table 5-3.

After 12 h, 2.5 ml 10x Phosphate Buffered Saline (PBS) (pH 7.4, Fisher Scientific,

Pittsburgh, PA) was added to the reaction mixture. After an additional 12 h, another 2.5

mL 10x PBS solution (pH 7.4) was added to make a final concentration of 1 x PBS


108










A B
t5
S +0
140.

t time
Average Fluorescence

"lO ti t t+z

0.4- G(0) oc 1/N
As time (tau) approaches 0
D 0.3- k C
SDiffusion
0.2


SG()= (F(t))
0.0-
S.... .. ... Auto Correlation Function
10 10 10 10 10
T (ms)

Figure 2-3 The working principle to obtain molecular diffusion in FCS study.









surface, thus demonstrating the validity of this "SET nanoruler" for mapping distance in

a real biological system.66 These "SET nanorulers" have the potential to become an

alternative to FRET for molecular interaction and distance measurement in cellular

systems, especially for applications demanding long observation times or large

distances. It represents the next leap forward in the use of optical probes to monitor

structural components within a cell membrane, and will open a new pathway for cellular

imaging.

Excited State Dimer-Excimer Formation

Pyrene is a polycyclic aromatic hydrocarbon (PAH) consisting of four fused

benzene rings, resulting in a flat aromatic system (left in Figure 1-5). It can form excited

state dimers (excimers, *Py+Py) upon close encounter of an excited state molecule (*Py)

with another ground state molecule (Py) 15, 77, 78 (right in Figure 1-5). When a pyrene

molecule is excited from the ground state (Py) to the excited state, the excited pyrene

(*Py) can relax to the ground state through internal conversion. During this process,

when such an excited pyrene (*Py) encounters with a second pyrene in ground

electronic state (Py), a complex (*Py+Py) with lower energy will form, which is named

excimer. The pyrene excimer emission has a broad, featureless emission centered at

480 to 500 nm, while the pyrene monomer emits in the 370 to 400 nm wavelength range

with two intensive peaks.

Similar to FRET, the formation of excimer is stringently distance-dependent, and

thus can be used as a unique signal transduction in the construction of molecular

probes. It can be adopted as the signaling moieties in the structure of MB to report the

conformational change of molecular probes upon target binding.78 The formation and









(111) Bice, A. N.; Zeeberg, B. R. Medical Imaging, IEEE Transactions on 1987, 6, 244-
249.

(112) Huang, Y.-F.; Chang, H.-T.; Tan, W. Analytical Chemistry 2008, 80, 567-572.

(113) Thompson, N. In Topics in Fluorescence Spectroscopy, 2002, pp 337-378.

(114) Zipfel, W. R.; Webb, W. W. In Methods in Cellular Imaging; Oxford University
Press: Oxford, UK, 2001, pp 216-235.

(115) Rigler, R.; Pramanik, A.; Jonasson, P.; Kratz, G.; Jansson, O. T.; Nygren, P.-A.;
Stahl, S.; Ekberg, K.; Johansson, B. L.; Uhlen, S.; Uhlen, M.; Jornvall, H.;
Wahren, J. Proceedings of the National Academy of Sciences of the United
States of America 1999, 96, 13318-13323.

(116) Pramanik, A.; Jureus, A.; Langel, U.; Bartfai, T.; Rigler, R. Biomedical
Chromatography 1999, 13, 119-120.

(117) Boonen, G.; Pramanik, A.; Rigler, R.; Haberlein, H. Planta Med 2000, 66, 7-10.

(118) Liu, J.; Lu, Y. Nat. Protocols 2006, 1, 246-252.

(119) Schurer, H.; Buchynskyy, A.; Korn, K.; Famulok, M.; Welzel, P.; Hahn, U.
Biological Chemistry 2005, 382, 479-481.

(120) Ikebukuro, K.; Okumura, Y.; Sumikura, K.; Karube, I. Nucl. Acids Res. 2005, 33,
e108-.

(121) Berezovski, M. V.; Lechmann, M.; Musheev, M. U.; Mak, T. W.; Krylov, S. N.
Journal of the American Chemical Society 2008, 130, 9137-9143.

(122) Raddatz, M.-S. L.; Dolf, A.; Endl, E.; Knolle, P.; Famulok, M.; Mayer, G.
Angewandte Chemie International Edition 2008, 47, 5190-5193.

(123) Osborne, S. E.; Ellington, A. D. Chemical Reviews 1997, 97, 349-370.

(124) Mallikaratchy, P.; Chen, H.; Tang, Z.; Meng, L.; Shangguan, D.; Parekh, P.; Kim,
Y.; Sefah, K.; Tan, W. American Pharmaceutical Review 2007, 10, 134-141.

(125) Shangguan, D.; Cao, Z.; Meng, L.; Mallikaratchy, P.; Sefah, K.; Wang, H.; Li, Y.;
Tan, W. Journal of Proteome Research 2008, 7, 2133-2139.

(126) MUller-Tidow, C.; Schwable, J.; Steffen, B.; Tidow, N.; Brandt, B.; Becker, K.;
Schulze-Bahr, E.; Halfter, H.; Vogt, U.; Metzger, R.; Schneider, P. M.; BCchner,
T.; Brandts, C.; Berdel, W. E.; Serve, H. Clin. Cancer. Res. 2004, 10, 1241-1249.

(127) Zeyu, X.; Dihua, S.; Zehui, C.; Xiaohong, F.; Weihong, T. Chemistry-A
European Journal 2008, 14, 1769-1775.


175









CHAPTER 7
SUMMARY AND FUTURE DIRECTIONS

Instrumentation Development and Nanomaterial Engineering for Live Cell
Mapping and Bioanalysis

Mapping living cells with good spatial and temporal resolution offers significant

potential for the understanding of important biological phenomena. Therefore, the

overall direction of this doctoral research has focused on the development of efficient

molecular probes and the construction of sensitive instruments for live cell receptor

mapping and bioanalysis.

Fluorescence Correlation Spectroscopy (FCS) is a sensitive technique for

observation of molecular interactions down to the molecular level. By extracting

information from molecular diffusion, FCS gives detailed information on molecular

interactions. In addition, it is also a non-invasive single-molecule-detection technique

that can be applied to the intracellular environment with low detection limits. Therefore,

this highly sensitive technique was chosen for this research project, and was applied for

direct measurements of membrane receptor density in its natural physiological

environment on the cell surface. A cellular model was constructed using a DNA ligand

aptamer for specific receptor targeting and labeling, and the receptor densities and

distribution profile on the cell surface were obtained for different types of cancer cells.

This successful outcome has proved the advantages of the FCS technique and its

applications for cell membrane receptor mapping.2

The previous membrane receptor density study using FCS has proven the

capability of using fluorescence auto-correlation for molecular interaction studies.

However, as discussed in Chapter 4, the application of FCS to binding analysis is

limited to the events in which binding significantly reduces the diffusion rate of the


162














I 10 nMfreesgc8

:A 10 nM sgc8 + HeLa Cells
0.04- ..


T D free = 0.235ms

S0.02- TD bound = 0.827ms




0.00-



1E-3 0.01 0.1 1 10 100 1000 10000
t (msec)


Detection Time (s)


- 10nM free Library
- 10nM Library + HeLa


-0..01 ......... .......... ... .. ...11 .. ...1... ....... 1
1E-3 0 0.1 .1 1 10 100 1000 10000 100000
t (msec)


0..05-


0.04-


0..03-

0..02-


0.01-


0.00-









12, while the 15-nm diameter gold nanoparticles quenched around half of the

fluorescence from the donor, their counterpart silica nanoparticles with the same

diameter showed no significant fluorescence shiftback compared to the maximum

fluorescence from the donors only. Therefore, we can therefore deduce that differences

in fluorescence quenching resulted from the specific energy transfer from the donor on

the antibody binding sites to the gold nanoparticles on the aptamer binding sites and

were not caused by the nonspecific steric hindrance effects in the donor binding.

Effects of Different Excitation Intensity and Labeling Efficiency on Distance

Determination

The binding site distance was obtained by studying the relationships between the

fluorescence quenching efficiencies and the change in NP diameters. However, some

systematic effects, such as the dye-antibody labeling efficiency effects and different

laser excitation intensity variations, could also change the quenching efficiency and

result in a different answer of binding site distance. Here, the influences of different

laser excitation intensities and different antibody/dye labeling efficiencies on the

calculations of binding site distance were investigated in a series of fluorescence

quenching experiments with gold NPs in different sizes. The laser excitation voltage

was varied from 650V to 700V and then to 750V in the flow cytometer, and the

fluorescence quenching was monitored as shown in Figure 5-14. It was observed that

higher laser intensities gave higher fluorescence signals. However, similar slopes were

obtained in the fluorescence decay for different laser intensities for nanoparticle

diameters ranging from 5.4nm to 18.4nm, while a plateau was achieved with

nanoparticle diameters larger than 18.4nm. The similar decay rates resulted in similar


125









shown in Figure 4-7, two DNA oligonucleotides have been designed, one labeled with a

green fluorophore and the other with a red fluorophore, that hybridize to adjacent

regions on the same mRNA target, resulting in an increase of cross-correlation signal.

The co-localization detection of the two different labeled DNA probes will significantly

reduce false positives, leading to the sensitive real-time monitoring of mRNAs in live

cells. Furthermore, the femtoliter-sized detection volume also ensures a low detection

limit in a spatially and temporally ordered manner and eventually improves the detection

sensitivity. This research, which is currently in progress, will provide a novel technique

for sensitive RNA detection and quantification in living cells. In the future, we will also

extend the applications for intracellular measurements and apply this sensitive

technique for protein-protein interaction studies in live cells.

Summary

A custom FCCS instrument has been successfully constructed for molecular

interaction studies. The technique not only inherits the merits of FCS as being a

sensitive tool for probing molecular diffusions, but also expands the application ranges

for molecular interaction studies. This lab-built instrument is being applied to

intracellular gene expression level detection.










sensitive biophysical tool on live cells, especially for studying weak molecular

interactions. In sum, our FCS density estimation approach can effectively improve

detection sensitivity and may have wide application for molecular recognition and

interaction studies, as well as density estimations for particles and other membrane

receptors.






8 Focus Laser Mior 1

*--Cell Membrane
i ......Mir / a
Cell



\ 1 I II I|l |"W lens
iMirror 2

Neutral
Dich roi / Density
....Dchroc or4 Filter

SFITC Trans- Computer
labeled membrane Filter
Aptamer receptor

SM ,,. Receptor Pinhole Correlator
Membrane Extracellular A




Figure 3-1 Scheme of FCS instrumental set-up and the illustration of aptamer/receptor
binding events on the cell membrane inside the focus. The magnified diagram
illustrates the geometry of the confocal volume with half-axis in length (w,) and
width (wxy).









sequences of synthetic DNA or RNA by repetitive binding of these oligonucleotides to

the target molecules.49 Through this iterative in vitro selection process, aptamers with

high specificity and affinity to their targets can be obtained. Most of the aptamers

reported so far have been selected using pure molecules, such as purified proteins, as

the targets. Aptamer selection against complex targets (such as red blood cells or a

single protein on live trypanosomes) was also demonstrated and interesting aptamers

have been generated.50 Aptamers have some advantages in molecular recognition and

imaging- low molecular weight, easy and reproducible synthesis, easy modification, fast

tissue penetration, low toxicity or immunogenicity, easy storage, high binding affinity

and specificity that are very comparable with antibodies. 51 Aptamers have shown great

promise in molecular recognition, diagnosis and therapy.

Cell-SELEX: A Molecular Evolution that Generates A Panel of Aptamers

To produce probes for molecular analysis of tumor cells, the Tan research

group has developed a novel method, the cell-based aptamer selection process (cell-

SELEX), for aptamer selection.9 Instead of using a single type of molecule as the target,

the cell-SELEX process uses whole cells as targets to select single-stranded DNA

aptamers that can distinguish target cancer cells from control cells (Figure 1-2). In

addition to those mentioned above for aptamers, the greatest advantage of cell-SELEX

technology is that there is no need of prior knowledge about the potential cancer

biomarkers for cancer on these cells. A group of cell-specific aptamers can be selected

in a relatively short period (4 to 10 weeks) without knowing which target molecules are

present on the cell surface after selection, the target molecules, which might be

important cancer biomarkers, maybe characterized, because the aptamer binding may









membranes. The resulting data, in turn, provide us with the aptamer concentration

required for saturating all the receptor binding sites in the density study. For this

purpose, we carried out binding experiments on individual cells with different

concentrations of aptamers. While keeping the same amount of HeLa cells in different

wells in the culture dish, we varied the concentration of FITC-labeled sgc8 aptamers

from 0.1 nM to 3.0 nM. We focused on the cell membrane with the 0.4fL ellipsoid

shaped confocal volume from a 1.2mW laser beam, the geometry of which had been

determined from the free dye calibration with half axes wxy=0.22 pm and wz=1.56 pm

(Figure 3-2). As a result of the difference in diffusion time of bound versus free

aptamers, we were able to assess the percentage of the bound aptamers (r) by fitting

the autocorrelation by equation (3-1). The fitting of the autocorrelation curve also

yielded the absolute number of total aptamers (N) inside the confocal volume, which

was interpreted as the reciprocal of the autocorrelation amplitude (in equation 4, when T

approaches 0, G(0)=1/N). As the total aptamer concentration increases, more and more

aptamers bind to the membrane receptors, so the number of bound aptamers inside the

confocal volume (total aptamer numberxbound fraction = Nxr) that was obtained

correspondingly increases before it reaches saturation. We plotted each bound aptamer

number (Nxr) versus the corresponding total aptamer concentrations used. As can be

seen in Figure 3-4 and Table 3-1, increasing concentrations of FITC-sgc8 clearly leads

to an increased bound aptamer number (table on the right). The femtoliter-sized (0.4fL)

observation volume created by FCS also allowed the detection of fluorescent aptamers

down to about two, while a low concentration (0.1 nM) of aptamers was used to incubate

with cells, which substantially validates FCS as a highly sensitive biophysical tool for









insight into particular molecular interactions, and dual-color Fluorescence Cross-

Correlation Spectroscopy (FCCS) is needed.

Theories of Dual-Color FCCS for Molecular Interaction Studies

The molecular weight limitation problem mentioned above can be solved by

labeling the two components with two spectrally distinct fluorophores and observing their

interactions, as shown in Figure 4-1 bottom. For this purpose, excitation needs to be

performed by two different lasers, and the two individual fluorescence emissions must

be detected in individual channels. Referring to Figure 4-1, the red and green signals

can be measured simultaneously and their cross-correlation function can be used as an

indication. The scheme of an FCCS instrumentation set-up is illustrated in Figure 4-2.

The experimental realization of a dual-color cross-correlation set-up is demanding,

because it also requires exact spatial superposition of the two laser beams, so that the

focal volumes overlap. Also, an additional dichroic mirror is needed in the emission

pathway between the first dichroic and the pinholes to split the fluorescence signals so

that each reaches its respective detector.

By labeling these two different kinds of molecules with two different fluorescent

tags, the cross-correlation between these two channels can be measured. From Figure

4-3, if there is no molecular interaction, the two different labeled molecular diffuse

independently, and there is low fluorescence cross-correlation from the two channels.

Once the two kinds of molecules interact with each other, either bound with each other

or bound to the same third molecule, they diffuse through the focus volume

simultaneously, inducing simultaneous fluctuations of the fluorescence signals in the

two color channels and thus a positive cross-correlation readout is obtained, and the









BIOGRAPHICAL SKETCH

Yan Chen was born in Guangzhou, China, in 1982. She attended the Zhixin

Middle School for her secondary education. She obtained a high ranking in the National

University Entrance Examination and attended University of Science and Technology of

China to study Chemical Physics. After obtaining her Bachelor of Science degree, Yan

came to the United States in the fall of 2005 to pursue her Ph.D. degree under the

supervision of Dr. Weihong Tan. She received her Doctor of Philosophy degree in

Analytical Chemistry from the University of Florida in June of 2010.


179









cell surface, but also free aptamer diffusing above the cell surface, will be seen. When a

portion of the aptamers in the confocal volume binds to the membrane receptors, the

bound aptamers are restricted to the cell surface and diffuse slower. Thus, the diffusion

time of the bound aptamers increases compared to the free ones. Diffusions of free

aptamers above the cell surface are described by the 3-D diffusion function, while

diffusions of bound aptamers on the cell membrane are represented by the 2-D diffusion

function. Thus, if the aptamer/receptor complex is stable on the time scale during its

transit through the detection region, then the overall autocorrelation function describing

the activities that occur in the confocal volume will represent a linear combination of the

autocorrelation functions of free and bound aptamers:





1 1 1 1
G(= (1-)- (3-1) 104
1+ 1 1+
free 2 bound
0 OZ Dfree



where N is the total absolute number of fluorescent molecules inside the focus,

TDfree is the diffusion time for the unbound labeled aptamer, and TDbound is the diffusion

time for the bound labeled aptamer. (1-r) is the fraction of the unbound aptamer

diffusing with TDfree, and r is the fraction of the bound aptamer diffusing with TDbound

Autocorrelation functions of FITC-labeled aptamers in solution and bound to the

cell membranes are respectively shown in Figure 3-2. Here we see that the binding of

aptamers to receptor results in an increase of the diffusion time, where TD = 0.827ms

(A), when compared to the free ones, where TD= 0.235ms (m), thus enabling FCS to









Control Cell Line for Method Evaluation

To further confirm the reliability of the approach for density estimation, a well

studied system of IgE receptors on the surface of RBL-2H3 cell line128 was also

conducted using FCS. A dissociation constant of 0.810.21 nM was obtained, and a 2.5

nM IgE concentration was indicated to be sufficient for saturating IgE receptor binding

sites on the FCS (Figure 3-9). Thirty RBL-2H3 cells were investigated to determine the

IgE receptor density on the cell surface per unit area by applying equation (3-2). Similar

to our aptamer sgc8 binding studies, a Gauss-shaped distribution of receptor density

was obtained (Figure 3-10). A mean density of around 1200 receptor/pm2 was obtained

with a variation of about 60 receptor/pm2. We determined the surface area of the RBL-

2H3 cells from confocal images of 80 individual cells (240 20 pm2/cell), and the IgE

receptor density could be estimated to be (2.88x105 1200) receptor/cell.

As shown in Table 3-2, the Kd value of IgE/receptor complexes obtained in FCS

gave results comparable to the Fluorescence Quenching method, as reported in the

literature 128, and a density with the same order of magnitude was also obtained for the

IgE receptor, which greatly supports the validity of the FCS approach for density

estimation.

Conclusion

In conclusion, we have reported the use of a highly sensitive technique,

Fluorescence Correlation Spectroscopy, for mapping receptor densities on live cell

membranes by introducing the fluorescently marked aptamer molecules, which target

specific membrane receptors with high affinity and selectivity. Full saturation of aptamer

binding to the cell surface is obtained at picomolar concentrations, which indicates the

high-affinity binding of the aptamer/receptor complexes (Kd=790150pM). The binding









147 I -I I I (AntiPTK7+NP sgc8)-Cell
equation : = 0 -=- 1 1. Inside the red
0I I0 AntiPTK7 Cell
rectangle, a linear relationship was obtained between the fluorescence
quenching efficiency and gold nanoparticle diameter (d=5.4nm-18.4nm) for
different laser intensities. Slopes were equal within experimental error (650V, (-

0.0770.0072); 700V, (-0.0720.0085); 750V, (-0.0730.013)). C) Plot -1

vs. NP diameter gave linear relationships between d=5.4nm-18.4nm (red
rectangle region in (B)). The linear fits for different laser intensities resulted in
similar distance R results (red: R=(13.431.40) nm; green: R=(13.741.82) nm;
blue: R=(13.562.73) nm).



A
Alex488 (6) 700V
1.0 I Alex488 (4) 700V
IZ I AAlex488 (2) 700V



o
0.8-


0.6-

0
FL 0.4-


S0.2-
0
z
0.0
SX.. X X_ X- & X, X" ,* X, .%P
0P O\vX%%S
^ ^^ 6 6^6^ 6 6^


142









the colocalization of both aptamer and antibody ligands on the receptor was confirmed

by competition studies (see Supplementary Materials). Later, a poly-T linker was added

between the aptamer and the fluorescent quencher to reduce the distance from the

quencher to the fluorescent donor (Figure 5-16B, D). However, similar results were

obtained, and no significant energy transfer was observed, indicating that no significant

energy transfer occurred from the fluorescent donor to the quencher on the two binding

sites.

These results once again proved that using gold nanoparticles as a surface

energy acceptor increased the probability of energy transfer and accounted for the

enhanced efficiency compared to the dipole-dipole interactions (FRET). Since

nanoparticle surface energy transfer (SET) has a surface and an isotropic distribution of

dipole vectors to accept energy from the donor, the energy transfer probability is greatly

increased.

Conclusion

In this paper, we have demonstrated, for the first time, the successful

construction of a "SET nanorular" on a live cell membrane to measure the protein

binding site distances. The distance between the aptamer and antibody binding sites in

the membrane protein PTK7 was obtained from the surface of leukemia T-cells (CEM)

in the natural physiological environment as (13.41.4) nm, with an error within 10%. No

protein separations or purifications were needed. The result also shows that this cell

membrane SET nanoruler can measure separation distances well beyond the detection

distance of FRET. Plus, since the energy acceptor nanoparticle in SET has a surface

and an isotropic distribution of dipole vectors to accept energy from the donor, it has a


128









CHAPTER 3
MAPPING RECEPTOR DENSITY ON LIVE CELLS USING FLUORESCENCE
CORRELATION SPECTROSCOPY

Introduction

Biological membranes are the sites where different elements of the cellular

machinery are brought together; therefore, they are central to the very phenomenon of

life108. Significantly different from bulk water, but integrated with it, biological

membranes create an environment in which many complex enzymatic reactions and

bioelectrical and biochemical signaling processes occur. Examples include the

conversion of metabolic energies into osmotic, electrical, and mechanical work,

transportation of materials between cellular compartments, and the processing of

information. In a broad sense, many cellular activities involve membrane-based ligand-

receptor interactions109, which are mediated by membrane-associated proteins that are

incorporated into the structures of the lipid bilayers. Consequently, biological

membranes are the primary target receptors for many drugs representing different

therapeutic categories. Therefore, while knowledge of molecular mechanisms

underlying ligand-receptor interaction has theoretical significance, there are also

practical implications for the discovery, design, and screening of novel therapeutic

agents. The binding of extracellular ligands to receptors also allows living cells to

constantly monitor and respond to changes in their environment. Therefore, the control

of receptor distribution and trafficking in a spatially and temporally ordered manner is

required to modulate cell behaviors, which range from cell division to differentiation.

Furthermore, the ability to obtain quantitative information about ligand-receptor

interactions and receptor distribution over the cell surface will be of broad significance in

our understanding of cell membrane receptor characteristics and expression level,










(5.4nm-42.2nm) were used in these energy transfer experiments. "R" represents
the distance between the Alexa Fluor 488 dye on the anti-PTK7 heavy chain to
the center of the gold NP, and "d" is the diameter of the gold NPs; therefore, "d/2"
is the distance from the center of the gold NP to its surface, and "r" represents
the distance from the Alexa Fluor 488 dye on anti-PTK7 to the surface of the gold
NP.



No Gold NP-sgc8 5.4nm Gold NP-sgc8 10.1nrm Gold NP-sgc8 13.3nm Gold NP-sgc8 -CEM cell only

S-- Alex488antlPTK7 200nM
,, -- Alex488-antlPTK7 200nM
I + Gold NP-sgS84nM

100 10' 102 10 10 10 10 102 10 104 10o 10' 102 1 10 10 10 102 10 1 *.............................
Alexa Flour 488 Alexa Flour 488 Alexa Flour 488 Alexa Flour 488

S18Anm Gold NP-sgc8 20.4nm GoldNP-sgc8 25.5nm Gold NP-sgc8 31.2nm Gold NP-sgc8 42.2nm Gold NP-sgc8

Ui I SS l

; ,,. -, ,\ \ /', .
100 10' 10 1010 10 10' 10 103 104 10 10' 102 10F 10- 10 10F 102 10A 10 100 101 1 0 10F 10I
Alexa Flour 488 Alexa Flour 488 Alexa Flour 488 Alexa Flour 488 Alexa Flour 488


Figure 5-8 (Top) Flow cytometry assay to monitor the fluorescence intensities on live
cells. A) Fluorescence intensity from the Alexa488-labeled anti-PTK7 in the
presence of varying sizes of gold NP-aptamer conjugates on live cell membrane
was monitored using a Flow cytometer. On each frame, the red curve indicates


135









Specifically, for a two-component diffusion system, Equation 2-9 is simplified to:




11 1 1 1
G3 (r)= + rB
e& 1+- 1+ 1+- 1+
TA D B 2 C
Z Z
S\ (2-10)

FCS Instrumentation Set-Up

As explained in the theoretical discussion, the FCS measurement requires a

small laser focus volume to ensure detection of a very small number of molecules to

obtain a correlation function with high amplitude. The method also requires high photon

detection efficiency and discrimination from background fluorescence.

The schematic diagram of the FCS experimental setup is shown in Figure 2-4.

Light from a laser passes through the optical lenses (i.e., beam expanders) and is

reflected by a dichroic mirror into the back of an objective. The objective helps to focus

the laser beam onto the sample to form a small volume element. After excitation of the

fluorescent molecules inside the sample, the emitted fluorescent light is collected by the

same objective and transmitted through the dichroic mirror. Due to the difference in the

excitation and emission wavelengths, the dichroic mirror separates the emission light

from the excitation source. Then, the emission light passes through some emission

filters, and a pinhole to reach a photon detector. The dimensions of the laser beam

focus and the pinhole together define the confocal volume element. The detector signal

is fed into a digital signal correlator, which calculates the autocorrelation function of the

detected intensity fluctuations.










* B
- Linear fit of Data B from MB226 assay


0.7



" 0.5

0.4


=0.25s1


-OJ01 O00 0.01 0.02
[I/Substrate, nM]


Figure 6-6 The Linear weaver-Burk plot of reciprocals of initial rates versus substrate
concentration for the determination of kinetic parameters Km, kcat and Vmax of
RNase H in MB226 assay. From the plot, we got Km =0.019 pM, Vmax =1.65
nM/s and kcat =0.25 s-1. Reaction conditions: [RNA-DNA hybrid] =25 nM, 50 nM,
75 nM, 100 nM, 200 nM, 300 nM, 400 nM and 500nM. [RNase H]=6 nM (50
units/ml). Excitation at 340 nm; emission at 485 nm.



Table 6-1. Probes and oligonucleotides used in Ribonuclease H kinetic study


Name
MB226
MB226 shared stem cDNA-A
MB226 shared stem cDNA-G
MB226 shared stem RNA
MBS1
MBS1 shared stem cDNA-A
MBS1 shared stem cDNA-G
MBS1 shared stem RNA


Sequence
5' Pyr-CCT AGC TCT AAA TCA CTA TGG TCG CGC TAG G-Pyr 3'
5' CCT AGC GCG ACC ATA GTG ATT TAG A 3'
5' GCG ACC ATA GTG ATT TAG AGC TAG G 3'
5' GCG ACC AUA GUG AUU UAG AGC UAG G 3'
5' Pyr-CGC ACC TCT GGT CTG AAG GTT TAT TGG TGC G-Pyr 3'
5' CGC ACC AAT AAA CCT TCA G AC CAG A 3'
5' AAT AAA CCT TCA GAC CAG AGG TGC G 3'
5' AAU AAA CCU UCA GAC CAG AGG UGC G 3'


Boldface type indicates the stem sequences in the molecular beacon structures.



Table 6-2. Kinetic parameters of pyrene beacon assays for E. coli RNase H

Substrate Km (pM) kcat ( s1)
MB226 (25 mer) 0.019 0.25
MBS1 (25 mer) 0.031 0.33
24 mer 165 0.02
22 mer 151 0.07 0.1
14 mer 154 0.08 1.12


161


Km=-[S]o
=0.019pM


-0.02









curve marks the fluorescence background with cells only, and the blue curve and
purple curve show the binding of FITC-labeled control antibody FITC-labeled
isotype Mouse IgG2a and TMR-labeled unselected DNA library, respectively.
The table on the bottom shows the fluorescence intensities in both channel 1 (at
520nm) and channel 2 (at 580nm). Channel 1: excitation 488, emission: 520nm;
Channel 2: excitation 488, emission: 580nm; Channel 3: excitation 488, emission:
680nm. FITC: fluorescein isothiocyanate; TMR: Tetramethylrhodamine; Cy5:
Cyanine 5.


Surface Energy Transfer


R
r d/2


Alexa488-antiPTK7







PTK7
,I I Receptor


Gold NP-Aptamer
Conjugate


Figure 5-7 Scheme of "SET nanoruler" for measuring the distance between two binding
sites in receptor PTK7 on a live cell membrane. The yellow object represents a
PTK7 receptor in the lipid bilayer of the cell membrane, with two binding sites on
its extracellular domain. The blue moiety represents one of the receptor ligands,
anti-PTK7. The Alexa Fluor 488 dye is labeled on its heavy chain through its
primary amino groups. On the other side, the red sphere represents a gold
nanoparticle. Multiple sgc8 aptamers (black) with thiol labeling are used to modify
the surface of the gold nanoparticle. The aptamer-gold NP conjugate is brought
to the aptamer binding site on the receptor through aptamer-receptor binding.
The colocalization of both ligands on the receptor brings the Alexa Fluor 488 dye
on the antibody close to the gold NP on the aptamer binding site. When the
donor dye molecule and acceptor NP surface reach close proximity, quenching of
fluorescence from the cell surface results. A series gold NPs of different sizes


134









spatial distribution, as well as clustering and molecular changes, on the molecular level

in living biological specimens. These data would also, in turn, provide an important

database for drug discovery.

However, so far, only a limited number of approaches have been established for

estimating membrane receptor densities. Liquid scintillation counting, a standard

laboratory method in life science, measures radiation from beta-emitting nuclides, and

thus requires the use of radioactively labeled ligand10' 111. Fluorescence subtraction,

the most conventional fluorescence approach for density estimation, involves several

washing steps to remove unbound ligands, and the receptor density is determined by

fluorescence measurement of supernatant containing free ligands. Concentration of free

ligands is estimated by interpolation from a standard linear calibration curve 112

However, the half-life of the receptor-ligand complex is often shorter than, or equal to,

the time required for the separation of free and bound ligands. Specific interactions

between certain ligands (e.g., peptides, hormones, natural products) and their different

receptor subtypes are, therefore, often overlooked by the conventional fluorescence

subtraction method. In addition, the analysis may also be compromised by high

background levels of other membrane proteins that are expressed endogenously on the

membrane. Especially, in certain cases, the receptor numbers per cell are few;

therefore, no specific binding is detected because of high background. Alternatively, the

receptor of interest could be separated and then purified from over-expressing cells or

tissue. This biochemical purification, however, typically requires exchange of the

physiological lipid/lipid-protein environment by a detergent micelle, which may modify

the binding properties of the receptor. In an ideal receptor preparation format, a high









nanoparticle acceptor surface (smaller r), hence increasing energy transfer efficiency.

This, in turn, explains why the larger gold nanoparticles in the ruler resulted in better

fluorescence quenching. The plot of quenching efficiency (0) versus different gold

nanoparticle diameters (d) (Figure 5-10) shows a plateau after the particle diameter

reached ~18nm, indicating that the surface of the particle had reached the antibody

binding site and had thus achieved maximum fluorescence quenching. Consequently,

particles with diameters larger than d=18nm would not reach a higher degree of

quenching.

The distance-dependent quenching data were fit to the Nanoparticle Surface

Energy Transfer model employed by Jennings et. a/.70:

1
+-1 (1-4)
l+ rl


In equation (1-4), 0 is the energy transfer efficiency; ro is a constant value for a specific

dye-metal system, corresponding to the distance at which a dye will display equal

probabilities for energy transfer and spontaneous emission66; and r is the distance from

the dye molecule to the gold nanoparticle surface.

Substitution of r from equation (5-1) and rearranging gives:


1 =- + (5-2)
(1 J 2r, r,


4
By letting Y= -1 and X=d, (3) is simplified to:


X R
Y=- +-R (5-3)
2ro ro


122









R is the radius of the spherical particle.

From Equations (4-1) and (4-2), it is clear that there is a linear relationship

between the diffusion time and the molecular radius:

rD X R (4-3)

If it is assumed that the analyte is a spherical molecule, then the relationship

between its molecular weight and it radius can be described as:


R=3M 3 RccM3
4;r (4-4)

in which R is the radius of the spherical particle;

v is the volume of the particle;

M is the molecular weight of the particle.

Therefore, for particles with similar sizes/volume, from Equation (4-3) and (4-4), it

can be concluded that:

]
D ocM 3 :M o D3 (4-5)

Hence, in FCS experiments, the molecular weight of the analyte is proportional to

the cube of its diffusion time. Therefore, an eight-fold molecular weight difference

results in only a two-fold change in the diffusion time difference in the autocorrelation

curve, a severe limitation for multi-component systems. Thus, FCS is severely limited

for molecular interaction studies in common cases, where two molecules with similar

molecular weights interact with each other, because the bound complex does not have

an 8-fold molecular weight difference. In this case, single-color FCS does not provide










1001


KD=790 150 pM


0.0 0.5 1.0 1.5 2.0 2.5 3.0


Total Aptamer (nM)


Total Bound aptamer
aptamer number in confocal
(nM) volume (Nxr)
0.1 1.91.5
0.3 16.84.6
0.5 34.14.4
0.7 46.65.3
1.0 59.54.5
1.2 67.92.7
1.4 74.43.7
1.6 82.13.0
2.0 85.51.3
2.5 85.80.9
3.0 86.00.7


Figure 3-4 Binding of FITC-sgc8 to cell membrane on human cervical HeLa cells.
Number of the membrane-bound labeled aptamer (Nxr) was obtained as a
function of the total aptamer concentrations in the binding buffer. HeLa cells were
incubated with buffer containing different concentrations of labeled aptamers for
40 min at 4C. Each data point represents the mean of five separate
measurements. For each measurement, 1.2mW laser intensity was used at the
objective outlet. The confocal volume was determined by free dye calibration to
be 0.4 fL with half axes wxy=0.22 pm and wz=1.56 pm. Sample was exposed to
laser for 30 seconds for obtaining the autocorrelation curve. Stable fluorescence
intensity was observed during the 30s detection time which indicates no
photobleaching. Different numbers of bound aptamers in the confocal volume at









detection of rare messenger RNAs, and the absence of cytotoxic effects, allows

continuous monitoring of the physiological changes in live cells.

It is precisely the ability of molecular beacons to detect without separation that

makes them ideally suited as reporters in ultra small volumes and inside living cells.

Thus, of the many potential probes tested for sensitive detection, MBs have emerged

promising probes, which gain steady growth in application, such as in mechanism

studies of biological functions and the sensitive detection of several diseases' biological

species. In particular, with their high sensitivity, excellent specificity, flexible design, and

modifications, MBs are also vastly promising molecular tools for quantitative intracellular

studies.

Aptamers: molecules for bio-recognition

The application of MBs for molecular recognition sometimes is limited by the

range of targets and the nucleic acid sequence design. However, the introduction of

aptamers greatly extends the application of nucleic acid probes to a large range of

targets.

Molecular aptamers are single-stranded DNAs (ssDNAs) and RNAs that can

recognize target proteins, peptides and other small molecules. The dissociation

constants of aptamers to targets can range from 10-12 M-10-8 M. Aptamers recognize

their targets with high specificity, and can typically discriminate between protein targets

that are highly homologous or differ by only a few amino acids.47 The tertiary structures

formed by the ssDNAs are the basis for target protein recognition.48 These aptamers

are selected by a process called SELEX (Systematic Evolution of Ligands by

Exponential enrichment), where the aptamers are selected from libraries of random









CHAPTER 5
ENGINEERING A SURFACE ENERGY TRANSFER (SET) NANORULER FOR
MEASURING PROTEIN BINDING SITE DISTANCES ON LIVE CELL MEMBRANE

Introduction

Application of optical molecular rulers to address questions in biochemistry,

biodiagnostics, and bimolecular imaging allows routine measurement of molecular and

dynamic distance changes. Up to now, such measurements have typically been

addressed by optical methods based on Forster Resonance Energy Transfer (FRET)
136-139 between molecular donors and acceptors. However, the nature of the

dipole-dipole mechanism effectively constrains the length scales in FRET-based

methods to distances on the order of <10 nm (Ro = 6 nm).15'75, 76 Optical methods that

do not alter biomolecular function, but which enable investigation of both long-range

static and dynamic distances, would therefore facilitate the study of many

multicomponent complexes that are presently difficult to measure.

Recently developed localized surface plasmon resonance (LSPR) sensors have

been able to meet these requirements 140-142. Liu et. al. 140 have constructed a

nanoplasmonic molecular ruler, in which double-stranded DNA (dsDNA) is attached to a

gold nanoparticle (NP), for measuring nuclease activity and DNA footprinting. While this

plasmonic ruler performs very well in bulk solution experiments and show a longer

detection range than FRET, significant challenges exist when applied to cellular

systems because of the high Raman scattering background coming from the cells

themselves. This high background significantly reduces the signal-to-background ratio

of the plasmonic sensor and greatly hinders molecular distance measurements when

applied on the cell surface. 90 light scattering of gold nanoparticles has been used by

Bene et. al.143 to measure the distances of receptors on the cell surface. However, light


104









ACKNOWLEDGMENTS

First, I wish to express my gratitude to my exceptional research advisor, Dr.

Weihong Tan, for providing me the tremendous support, guidance, help and kindness

for my research and life during the five year PhD studies at University of Florida. His

constant advice, inspiration and encouragement have made me into a more mature and

confident person and a better scientist. I also appreciate the valuable opportunities that

he gave me to train my leadership and management skills, which will definitely have a

great impact throughout my life and career.

I would also like to express my sincere thanks to my committee members, Dr.

Nicole Horenstein, Dr. David Powell, Dr. Yunwei Cao and Dr. Donn Dennis for the

helpful discussion, advice and assistance, as well as the recommendations for the

fellowship applications. I also want to thank Dr. Ben Smith for all the help and support.

In particular, I would also like to thank Dr. Nicole Omenetto and his student, Jonathan

Merten, for the critical comments and help during the construction of the lab-built FCCS

instrument. I appreciate Dr. Gail Fanucci and Angelo Marcelo Veloro for the helpful

discussion and suggestions for the protein binding site measurement experiment. I also

want to thank Dr. Kathryn R. Williams for her kindness and help with the manuscripts.

This dissertation is a result of successful collaboration with many great scientists

in different areas. I especially would like to thank Dr. Chaoyong Yang and Ms. Hui Lin

for the guidance, tremendous help and friendship during me first two years of research.

I thank Dr. Alina C. Munteanu for her patient guidance to lead me into the research area

of instrumentation development. I am grateful to Dr. Yu-Fen Huang for the many helpful

discussions on different projects as a knowledgeable college and a valuable friend. I

really appreciate Dr. Huaizhi Kang as being a supportive friend and college, and for his









3-10 IgE receptor density distribution for RBL-2H3 cells. ............... ................... 87

4-1 Illustrations of molecular interaction studies by the autocorrelation function in
FCS and by the cross-correlation function in FCCS. ................................ .. 100

4-2 Scheme of FCCS instrumentation set-up. ........ ....... ..................... ......... 100

4-3 Molecular interactions characterized by autocorrelation curves and cross-
correlation curves. ............... ...... ..... ......... .. ................ ............... 101

4-4 Scheme of the lab-built FCCS set-up ............... ......... .................... 102

4-5 Picture of the overlap of the three laser beams. ........ ... ... .......... ... 102

4-6 Picture of the lab-built FCCS instrumentation set-up...................................... 103

4-7 Scheme of applying FCCS technique for intracelluar mRNA detection ........... 103

5-1 Characterizations of different sized gold nanoparticles ............... ............... 130

5-2 TEM images of different sizes of gold nanoparticles. .................................. 130

5-3 Ligand binding saturation concentration determinations. ............. ............... 131

5-4 Competition studies using Cy5-sgc8 to confirm the saturation binding of gold
NP-aptamer conjugates on the cell surface............. ..... .................. 131

5-5 Competition studies between aptamer sgc8 and antibody anti-PTK7 on
receptor PTK7 ........ ......... ......... .... .................. ............... 132

5-6 Measure the distance between two binding sites using FRET. ...................... 133

5-7 Scheme of "SET nanoruler" for measuring the distance between two binding
sites in receptor PTK7 on a live cell membrane.......................................... 134

5-8 Flow cytometry assay to monitor the fluorescence intensities on live cells and
histogram of the mean fluorescence intensity for the fluorescence quenching
assay determined from the flow cytometry results ................ ........ ....... 135

5-9 Confocal imaging assay for monitoring the fluorescence quenching on cell
surface with different sizes of gold nanoparticles ....... ............. ............. .. 136

5-10 Relationship between fluorescence quenching efficiency and gold
nanoparticle diameter ........................................ .............. 137

5-11 Binding site distance determination ................ ....... ......... ............ 138

5-12 Flow cytometry analysis to monitor the fluorescence quenching effect with
15-nm silica nanoparticles. .......................................................................... 139










B Alexa488-antiPTK7 + TMR-sgc8


-- Cell only
-- FITC-IgG2a
-- Cy5-Lib


- Cell only
-- FITC-IgG2a
-- TMR-Lib


w L
-- Alexa488-antiPTK7 + Cy5-sgc8 -- Alexa488-anliPTK7 + TMR-sgc8



o100 101 1 1 104 100 101 102 103 104
FL3 FL2


Figure 5-6 Measure the distance between two binding sites using FRET. (A) The figure
on the top indicates the Flow Cytometry assay to monitor the fluorescence
intensities at 680nm (channel 3) from the cell membrane with an excitation
source at 488nm. The green curve indicates the fluorescence intensity from the
cell membrane with saturated concentrations of Alexa488-labeled anti-PTK7
(200nM). The yellow curve shows the fluorescence intensity with saturated
concentrations of Cy5-labeled sgc8 (200nM). The red curve indicates the
fluorescence intensity in the presence of both Alexa488-antiPTK7 (200nM) and
Cy5-sgc8 (200nM) on the cell membrane. The black curve marks the
fluorescence background with cells only, and the blue curve and purple curve
show the binding of FITC-labeled control antibody FITC-labeled isotype Mouse
IgG2a and Cy5-labeled unselected DNA library, respectively. The table on the
bottom shows the fluorescence intensities in both channel 1 (at 520nm) and
channel 3 (at 680nm). (B) The figure on the top indicates the Flow Cytometry
assay to monitor the fluorescence intensities at 580nm (channel 2) from the cell
membrane with an excitation source at 488nm. The green curve indicates the
fluorescence intensity from the cell membrane with saturated concentrations of
Alexa488-labeled anti-PTK7 (200nM). The yellow curve shows the fluorescence
intensity with saturated concentrations of TMR-labeled sgc8 (200nM). The red
curve indicates the fluorescence intensity in the presence of both Alexa488-
antiPTK7 (200nM) and TMR-sgc8 (200nM) on the cell membrane. The black


133


FL1 FL3
Cell only 3.23 3.58
FITC-IgGZa 5.80 3.84
CyS-Lib 3.11 3.59
26.99 4.30
3.11 3.71
Alexa488-antiPTKT 27.93 4.53
+ Cy5-Sgc8


FL1 FL2
Cell only 3.23 3.24
FITC-IgG2a 5.80 4.08
TMR-Lib 3.04 3.19
26.99 8.31
3.05 4.77
Alexa488-antiPTK7 27.05 10.28
+ TMR-Sgc8


A Alexa488-antiPTK7 + Cy5-sgc8









CHAPTER 2
THEORIES AND INSTRUMENTATIONS OF FLUORESCENCE CORRELATION
SPECTROSCOPY

Fluorescence Correlation Spectroscopy (FCS) is based on the statistical analysis

of fluorescence intensity fluctuations and is used to investigate the dynamic properties

of single molecules in solution.82 83 It was first introduced in the 1970s as a method for

measuring molecular diffusion, reaction kinetics, and flow of fluorescent particles.8487

This method has recently found widespread application in the study of processes such

as diffusion in the cellular environment.88 Because of the substantial increase in

sensitivity of FCS89, allowing single-molecule analysis,90'91 The small volume elements

(<1 fL) in which the measurements are performed make it possible to evaluate

molecular processes in the cellular environment with rapid temporal and high spatial

resolution. FCS is also noninvasive, making it ideal for live cell measurements. Coupled

with imaging, FCS can be used on the subcellular level to dissect biological processes

in different compartments morphologically; e.g., dendritic spines, which have been

previously intractable to biochemical experiments. FCS measurements also allow the

determination of both cell-to-cell and position-to-position (within the same cell) variability.

Many techniques measure average values of a seemingly 'homogeneous' ensemble, in

which the heterogeneity of a process in a cell population or within the same cell is lost.

However, all this information, which can be biologically important to understand the

extent of variation in cellular responses, will be obtained by FCS. This research takes

advantage of the high detection sensitivity and non-invasive properties of FCS

technique, in particular the instrumentation developments and applications of FCS for

live cell mapping and intracellular measurements.









and was used with FCS to determine receptor densities and distributions on the cell

surface. With its intrinsic advantages of direct measurement, high sensitivity, rapid

analysis and single-cell measurement, this FCS density estimation approach holds

great potential for future applications in molecular interaction studies and density

estimations for subcellular structures and membrane receptors.2

To further understand the structure of PTK7 beyond its spatial distribution and

ligand-receptor interactions, a nanoparticle was used as a molecular ruler to measure

the distance between two binding sites on the receptor on live cells. Measuring

distances at molecular length scales in living systems is a significant challenge.

Methods like FRET (fluorescence resonance energy transfer) have significant limitations

due to short detection distances and strict orientations. To overcome these limitations

and construct a practical nanoruler for measuring distances on live cells, an SET-based

nanoruler, using aptamer-gold-nanoparticle conjugates with different diameters, was

developed to measure separation distances well beyond the detection limit of FRET.

Since application of fluorescence auto-correlation (FCS) to binding analysis is

limited to applications in which the binding event significantly reduces the diffusion of

the labeled species, the original FCS setup was upgraded to a novel three-channel

Fluorescence Cross-Correlation Spectroscopy (FCCS) setup. This lab-built FCCS not

only inherits the single-molecule detection capability from FCS, but also further extends

its applications for molecular interaction studies by labeling two species with two

spectrally distinct fluorophores. This technique has been establied and is now being

adapted for real-time monitoring of intracellular mRNA.









Figure 3-2 Autocorrelation functions of aptamer sgc8 and control free in solution and on
a cell membrane. (A) Top: Autocorrelation functions of aptamer sgc8 (10nM) free
in solution (m) and aptamer sgc8 (10nM) incubated with HeLa cells and bound to
membrane receptor PTK7 on the cell surface (A). The diffusion times (TD) of
aptamer sgc8 is increased from 0.235ms (free) to 0.827ms (bound). Bottom:
Fluorescence intensity (or count rate) fluctuation curves during the detection time
(30 seconds) for aptamer sgc8 (10nM) diffusion free in solution (blue line) and
bound to membrane surface (red line). Stable fluorescence fluctuations show no
photobleaching in the detection volume during the entire detection time. (B)
Control binding experiments with a randomized sequence of 41 nucleotides
(Library) were conducted under the same conditions. Both the free Library and
Library incubated with HeLa cells show similar diffusion times (0.435ms (free)
and 0.457ms (bound)), which indicates no binding interactions between Library
and cell membrane receptors.









(162) Fujimoto, K.; Shimizu, H.; Inouye, M. The Journal of Organic Chemistry 2004, 69,
3271-3275.

(163) Birks, J. B. Photophysics of Aromatic Molecules; Wiley-lnterscience: London,
1970.

(164) Dahim, M.; Mizuno, N. K.; Li, X.-M.; Momsen, W. E.; Momsen, M. M.; Brockman,
H. L. Biophysical Journal 2002, 83, 1511-1524.

(165) Haruki, M.; Noguchi, E.; Kanaya, S.; Crouch, R. J. J. Biol. Chem. 1997, 272,
22015-22022.

(166) Kanaya, E.; Uchiyama, Y.; Ohtsuka, E.; Ueno, Y.; Ikehara, M.; Kanaya, S. FEBS
Letters 1994, 354, 227-231.

(167) Chen, Y.; O'Donoghue, M. B.; Huang, Y.-F.; Kang, H.; Phillips, J. A.; Chen, X.;
Estevez, M.-C.; Tan, W. submitted 2010.

(168) Schlessinger, J.; Shechter, Y.; Cuatrecasas, P.; Willingham, M. C.; Pastan, I.
Proceeding of the National Academy of Sciences of the United States of America
1978, 75, 5353-5357.


178









problem could be minimized with silica NPs. The flexible silica chemistry also provides

versatile routes for surface modification. Different types of functional groups can be

easily introduced onto the NP surface for conjugation with biomolecules. In addition, the

silica surface makes these NPs chemically inert and physically stable.60 All these

properties make silica NPs excellent labeling reagents for bioanalysis and bioimaging.61

Combination of Nucleic Acid Probes and Nanoparticles for Biological

Applications

Thanks to their intrinsic molecular properties, the combination of functional

nucleic acid molecules and nanoparticles results in an unprecedented improvement in

molecular recognition. First, due to the high surface area to volume ratio, nanoparticles

can be modified to generate many types of DNA probes to enable more effective

molecular recognition. In addition, the phenomenon of cooperative interaction, also

called the synergistic effect, makes the recognition much easier towards targets with

more binding sites. When suitable nanoparticles serve as reporters, the signal of

recognition events can be amplified by thousands due to the several reasons.62 First,

because each NP carries multiple recognition molecules, there is a higher possibility of

target binding, which can improve the detection selectivity. Second, unlike other

biomolecules, DNA probes are stable after modification. The easy handling of DNA

strands and different modification strategies involving nanoparticles provide a vast

platform to achieve molecular recognition. Third, the unusual interactions between

nanoparticles and living systems make the application of functional DNA more practical

for molecular recognition and medical diagnostics. With the help of nanoparticles,

nucleic acids can escape digestion by, for example, cellular nucleases, and can be











B 0.9-
0.8
0.7 I
T
0.6- T
0.5-
11

-1 0.4
0.3- Alex488 (6) 700V
S* Alex488 (4) 700V
0.2 Alex488 (2) 700V
0.1
0.0-

0 5 10 15 20 25 30 35 40 45
NP Diameter (nm)


2.0-
C U Alex488 (6) 700V
Alex488 (4) 700V
1.8- Alex488 (2) 700V

1.6- Red: Y=(-0.0800.014)X+(2.140.18)
Grey: Y=(-0.083 0.011)X+(2.17 0.15)
1.4-

1.2-

1.0-
-L x
0.8-

4 6 8 10 12 14 16 18 20
NP Diameter (nm)


Figure 5-15 Effects of different antibody labeling efficiencies on fluorescence quenching.
A) Histogram of the fluorescence quenching assay with different antibody
labeling efficiencies (red: 6 Alexa Fluor 488 dyes on each anti-PTK7; grey: 4
dyes per anti-PTK7; green: 2 dyes per anti-PTK7) from the flow cytometry
analysis. The different labeling efficiencies for anti-PTK7 were achieved by
incubating different amounts of Alexa Fluor 488 dyes with anti-PTK7 during the
labeling procedure and were calculated and determined according to the UV
quantification (see Methods). Fluorescence quenching experiment procedures
were the same as described in Figure 1. Laser excitation at 488nm was constant
at 700V. All of the experiments for the fluorescence quenching assay were
repeated three times, and the average value was determined as the mean
fluorescence intensity. B) The relationship between fluorescence quenching


143









antibody anti-PTK7 and aptamer sgc8 have been identified as the molecular ligands for

the two binding sites on PTK7 respectively125.

Competition Studies between Aptamer Sgc8 and Antibody Anti-PTK7 on PTK7

Before constructing the prototype "SET nanoruler" on the live cell surface, the

colocalization of the two binding sites on the model receptor PTK7 was determined.

Monoclonal antibody anti-PTK7 and aptamer sgc8 have been identified as the

molecular ligands for the two binding sites on PTK7 respectively125.The specific

interactions of protein receptor PTK7 with these two ligands were confirmed by PTK7

gene silencing2 and gene transfection experiments'. The interactions between the two

ligands in the receptor PTK7 were first evaluated by competition studies to validate the

colocalization. Competition studies between the two ligands were conducted and

monitored using Flow Cytometer. Excess unlabeled sgc8 (100x) was used to compete

with Alexa488-labeled anti-PTK7 (200 nM) for CEM cell binding (Figure 5-5A). To

further investigate the possibility of co-binding of sgc8 and the antibody on PTK7, a

contrasting experiment was conducted by first labeling the aptamer with a FITC

fluorophore and incubating with CEM cells. Afterwards, excess non-labeled anti-PTK7

(100x) was added to compete with the aptamer binding (Figure 5-5B). Flow cytometry

results showed no obvious change in the Alexa488-anti-PTK7 binding, indicating that

the aptamer sgc8 and the antibody anti-PTK7 simultaneously bind to two different sites

of the extracellular domain of PTK7. This colocalization phenomenon served as the

basis for constructing the molecular ruler to measure the two binding sites on the cell

surface.


115









much lower requirement for interaction orientations than FRET and thus leads to a

wider application for distance measurements for various in vitro and cellular systems.

On the other hand, the SET system also shows its own advantages for cellular

measurements over its counterpart plasmonic rulers which also process long-distance

detection capability. The cell surface SET nanoruler adopted a series of different sized

gold nanoparticles (5nm-42nm) for the ruler construction, however, only small sized

particles can be used to build plasmonic rulers, because the scattering effect only

dominates in smaller sized nanoparticles (<2nm) due to the insufficient overlap for

fluorescence energy transfer. In contrast, particles with greater than a 2nm diameter

have higher probability and incidence of fluorescence quenching (like SET)144

Therefore, SET is a more suitable choice for our nanoruler construction with various

large sized (6nm-42nm) of gold nanoparticles. Moreover, high scattering background

from the cell surface also significantly reduces the detection signal-to-background ratio

and therefore prohibits the applications of plasmonic rulers to the cell surface.

In summary, "SET nanorulers" have the potential to become an alternative to

FRET for molecular interaction and distance measurement in cellular systems,

especially for applications demanding long observation times or large distances. It

represents the next leap forward in use of optical probes to monitor structural

components within a cell membrane and will open a new pathway for cellular imaging.


129









Therefore, during the last decade, there has been a dramatic rise of interest in the FCS-

related studies, and a number of books and review articles on FCS theories and

applications have been published.17' 101,102

Theories of FCS

In this section, the principles of FCS for molecular diffusion studies will be

discussed. Specifically, the fluorescence autocorrelation function will be used to explain

the geometry of the focus volume and the physics of molecular diffusion within this

focus volume.

Molecular Diffusion Obtained by the Autocorrelation Function in FCS

As shown in Figure 1, the signal obtained from conventional fluorescence

spectrometry measurements can be generally described in terms of two parameters, the

constant mean intensity < I > and a fluctuating contribution 61 (t). Although the mean

fluorescent intensity < I > is generally the information of interest, the fluorescence

fluctuations 61 (t) actually carry the information that pertains to molecular diffusion study

by FCS. Specifically, FCS is an analytical method that can measure the dynamics of

molecular processes from the observations of spontaneous fluorescence fluctuations in

a solution at thermal equilibrium. By extracting from the fluorescence fluctuation signal,

it gives the information about dynamic molecular events, such as diffusion or

conformational changes. This method also finds useful applications in the study of

thermodynamic and kinetic features of molecular interactions.

FCS measurement is accomplished by focusing the laser beam onto the sample

to create a fL-size focus volume and then monitoring the fluorescence signal within the

volume. The geometry of the focus volume is illustrated in Figure 2-2.103 The dimension









(70) Jennings, T. L.; Singh, M. P.; Strouse, G. F. Journal of the American Chemical
Society 2006, 128, 5462-5467.

(71) Griffin, J.; Ray, P. C. The Journal of Physical Chemistry B 2008, 112, 11198-
11201.

(72) Griffin, J.; Singh, A. K.; Senapati, D.; Rhodes, P.; Mitchell, K.; Robinson, B.; Yu,
E.; Ray, P. C. Chemistry -A European Journal 2009, 15, 342-351.

(73) Sen, T.; Haldar, K. K.; Patra, A. The Journal of Physical Chemistry C 2008, 112,
17945-17951.

(74) Sen, T.; Sadhu, S.; Patra, A. Applied Physics Letters 2007, 91, 043104-043104-
043103.

(75) Chance, R. R.; Prock, A.; Silbey, R. In Advances in Chemical Physics; I.
Prigogine, S. A. R., Ed., 2007, pp 1-65.

(76) Persson, B. N. J.; Lang, N. D. Physical Review B 1982, 26, 5409.

(77) Winnik, F. M. Chem.Rev. 1993, 93, 587-614.

(78) Fujimoto, K.; Shimizu, H.; Inouye, M. J.Org.Chem. 2004, 69.

(79) Sokol, D. L.; Zhang, X. L.; Lu, P. Z.; Gewitz, A. M. Proc.Natl.Acad.Sci.U.S.A.
1998, 95, 11538-11543.

(80) Kim, S. A.; Heinze, K. G.; Schwille, P. Nat Meth 2007, 4, 963-973.

(81) Lee, S.-J. R.; Escobedo-Lozoya, Y.; Szatmari, E. M.; Yasuda, R. Nature 2009,
458, 299-304.

(82) Magde, D.; Elson, E.; Webb, W. W. Physical Review Letters 1972, 29, 705.

(83) Douglas, M.; Elliot, L. E.; Watt, W. W. Biopolymers 1974, 13, 29-61.

(84) Aragon, S. R.; Pecora, R. Biopolymers 1975, 14, 119-137.

(85) Ehrenberg, M.; Rigler, R. Quarterly Reviews of Biophysics 1976, 9, 69-81.

(86) Ehrenberg, M.; Rigler, R. Chemical Physics 1974, 4, 390-401.

(87) Douglas, M.; Watt, W. W.; Elliot, L. E. Biopolymers 1978, 17, 361-376.

(88) Preuss, A. K.; Mayer, M.; Wohland, T.; Hovius, R.; Vogel, H. Biochemistry 2003,
42, 877-884.

(89) Rigler, R.; Mets, U.; Widengren, J.; Kask, P. European Biophysics Journal 1993,
22, 169-175.


173









ABSTRACT OF DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


INSTRUMENTATION DEVELOPMENTS AND NANOMATERIAL ENGINEERING FOR
LIVE CELL MAPPING AND BIOANALYSIS

By

Yan Chen

August 2010

Chair: Weihong Tan
Major: Chemistry

Intracellular molecular interaction studies have attracted great interest among cell

biologists. However, the intrinsic resolution of conventional optical microscopy only

allows the visualization of subcellular localizations. Therefore, a sensitive technique that

can observe molecular interactions down to the molecular level is in great need. By

extracting information from molecular diffusion, Fluorescence Correlation Spectroscopy

(FCS) gives detailed information on molecular interactions, and has become a non-

invasive single-molecule-detection technique that can be applied to the intracellular

environment with low detection limits.1 Therefore, the overall direction of this doctoral

research was the development of a self-built FCS setup. Combined with nanomaterial

engineering, this project has enabled membrane receptor studies and intracellular

measurements in live cells.

The Fluorescence Correlation Spectroscopy instrumentation was to direct

measurement of membrane receptor density of the PTK7 protein, an important cancer

marker, in its natural physiological environment on the cell surface. A cellular model

using a DNA ligand aptamer was designed for specific receptor targeting and labeling,









distance R results, which indicated the negligible effects of different laser intensities on

the distance determinations of the "SET nanoruler".

Different antibody/dye labeling efficiencies were also tested. With multiple

primary amino groups on the heavy chain, the anti-PTK7 could react with different

amounts of Alexa Fluor 488 dyes during the conjugation. By controlling the amount of

dye added, three different conjugates were prepared, having 2, 4 or 6 dyes, respectively,

on each antibody. As shown in Figure 5-15, results of the fluorescence quenching

experiments were similar to those of the laser intensity experiments.

Both of these studies indicated that the variations in laser source intensity and

antibody labeling efficiency have limited effects on the determination of the binding site

distances. Here it also showed the benefit of using gold NPs with different sizes instead

of a single size for the nanoruler construction. As seen from the above measurements,

the laser intensity and labeling efficiency effects could change the fluorescence

quenching efficiency for individual size of gold NPs. So if only a single sized NP was

used in the distance measurement, these effects could result in a different binding

distance. However, while a series of different sized NPs were applied for the

measurement, these effects were rule out and a similar decay slope was obtained,

which leaded to a similar results for the binding site measurements. Therefore, the use

of different sizes of nanoparticles in the ruler construction also introduces a feature to

enhance the precision for the distance calculations and determinations.


126




Full Text

PAGE 1

1 INSTRUMENTATION DEVELOPMENTS AND NANOMATERIAL ENGINEERING FOR LIVE CELL MAPPING AND BIOANALYSIS By YAN CHEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Yan Chen

PAGE 3

3 To My Beloved Grandma and Parents

PAGE 4

4 ACKNOWLEDGMENTS First, I wish to express my gratitude to my exceptional research advisor, Dr. Weihong Tan, for providing me the tremen dous support, guidance, help and kindness for my research and life during the five year PhD studies at Universi ty of Florida. His constant advice, inspirati on and encouragement have made me into a more mature and confident person and a better scientist. I also appreciate the valuable opportunities that he gave me to train my leadership and m anagement skills, which will definitely have a great impact throughout my life and career. I would also like to express my sincer e thanks to my committee members, Dr. Nicole Horenstein, Dr. David Powell, Dr. Yunwei Cao and Dr. Donn Dennis for the helpful discussion, advice and assistance, as well as the recommendations for the fellowship applications. I also want to than k Dr. Ben Smith for all the help and support. In particular, I would also like to thank Dr. Nicole Omenetto and his student, Jonathan Merten, for the critical comments and help duri ng the construction of the lab-built FCCS instrument. I appreciate Dr. Gail Fanucci a nd Angelo Marcelo Veloro for the helpful discussion and suggestions for the protein bindi ng site measurement experiment. I also want to thank Dr. Kathryn R. Williams fo r her kindness and help with the manuscripts. This dissertation is a result of successf ul collaboration with many great scientists in different areas. I especia lly would like to thank Dr. Chaoyong Yang and Ms. Hui Lin for the guidance, tremendous help and friendship during me first two years of research. I thank Dr. Alina C. Munteanu for her patient guidance to lead me into the research area of instrumentation development. I am grateful to Dr. Yu-Fen Huang for the many helpful discussions on different projects as a kno wledgeable college and a valuable friend. I really appreciate Dr. Huaizhi Kang as being a supportive friend and college, and for his

PAGE 5

5 encouragement and help during my difficult times. I apprecia te Micheal Marvos and Lu Peng for spending many difficult hours with me on the FCCS constructions. I am thankful to Meghan B. OÂ’Donoghue for being there whenever I need any help or advice. I thank Dr. Xiaolan Ch en for her help in particle synthes is and functionalization. I thank Dr. Yanrong Wu and Patrick Conlon for assist ing me with the pyrene molecular beacon synthesis. Also I would like to thank Zhi Zhu for the all the helpful discussions. It has been a great privilege to spend seve ral years in the Tan research group, and its members will always remain dear to me. I would like to thank Dr. Xiaoling Zhang, Dr. Ronghua Yang, Dr. Dihua Sh angguan, Dr. Zhiwen Tang, Dr. Marie Carmen Estevez, Dr. Wenjun Zhao, Dr. Zehui Cao, Dr. Josh E. Smith, Dr. Colin Medley, Dr. Prabodhika Mallikaratchy, Dr. Kwame Sefah, Dr. Liu Yang, Dr. Joseph Phillips, Dr. Youngmi Kim, Dr. Dosung Sohn, Dr. Karen Martinez, Dr. Hui C hen, Dr. Jilin Yan, Dr. Zeyu Xiao, Dr. Kelong Wang, Ling Meng, Jennife r Martin, Pinpin Sheng, Da lia Lopez Colon, Hui Wang, Suwussa Bamrungsap, Xiangling Xiong, and other s for their friendship and help. Each of them has made this journey very enjoyable and pleasant. Also, I am deeply grateful to my boyfri end, Fei Huang, for being a wonderful companion, friend, and lover. I thank him for being with me to persevere during the bad times, and celebrating the good times. Finally, I owe a huge debt of gratitude to my grandm a and my parents for their unfailing love, encouragement, support and compani on. Their great personality, endless love and constant guidance make me who I am. I really appreciate their unselfish dedication all this time, especially for my beloved grandma. She will live in my heart, every moment, forever and ever.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDG MENTS..................................................................................................4 LIST OF TABLES..........................................................................................................10 LIST OF FI GURES ........................................................................................................11 ABSTRACT ...................................................................................................................14 CHAPTER 1 INTRODUC TION....................................................................................................17 Molecular Probes for Li ve Cell M apping ................................................................17 Molecular Engineering of Nanomat erials ................................................................20 Molecular Engineering of Nucleic Acid Probes.................................................21 Molecular beacons (MB)s: molecular s witches for signal transduction......23 Aptamer: Molecules fo r Bio-Recogni tion....................................................24 Molecular Engineering of Nanoparti cles...........................................................26 Combination of Nucleic Acid Prob es and Nanopaticles for Biological Applicatio ns...................................................................................................29 Fluorescence Mechanisms fo r Signal Trans duction...............................................30 Jablonski Diagram ............................................................................................31 Fluoresecence Quenching ................................................................................32 Fluoresecence Resonance E nergy Transfe r (FRET).......................................35 Surface Energy Tr ansfer ( SET)........................................................................36 Excited State Dim er-Excim er............................................................................37 Challenges of Live Cell Anal ysis using Nanom aterials ...........................................38 Fluorescence Correlation Spectroscopy (FCS) for Live Ce ll Analysis....................38 2 THEORIES AND INSTRUMENTA TIONS OF FLUORESCENCE CORRELATION SPEC TROSCO PY.......................................................................43 History of FCS........................................................................................................44 Theories of FCS......................................................................................................46 Molecular Diffusion Obtained by Au tocorrelation Func tion in FCS..................46 Geometry of Focus Volume Obta ined by Free Dye Calibrat ion.......................50 Single-Component Diffusion and Mu lti-Component Diffusi on..........................51 FCS Instrumentat ion SetUp...................................................................................52 FCS for Molecular Diffusion Studies in Soluti on.....................................................53 Quantitative Studies of Molecular Numbers in the Fo cus Volume...................53 Molecular Diffu sion Ti me.................................................................................54 FCS for Live Ce ll Analysi s......................................................................................54

PAGE 7

7 3 MAPPING RECEPTOR DENSITY ON LIVE CELL MEMBRANE USING FCS.......59 Introducti on.............................................................................................................59 Experimental Section..............................................................................................63 Cell Li nes......................................................................................................... 63 Aptamers Sy nthesis .........................................................................................64 Antibody ...........................................................................................................64 Trypsin Treatment of Ce lls...............................................................................65 FCS Instrument aion SetUp.............................................................................65 Results and Di scussions.........................................................................................65 FCS Measurement and Analysi s.....................................................................65 Binding of Aptamers to Membrane Re ceptors.................................................67 Specificity of Receptor Binding by the Aptamers.............................................69 Binding Affinity Determination (Kd)...................................................................69 Density St udy...................................................................................................71 Competition Studies and Trypsin Exper iments................................................74 Control Cell Line for Method Eval uation...........................................................76 Conclusion s............................................................................................................76 4. UPGRADING FCS TO DUAL COLOR-FLUORESCENCE CROSSCORRELATION SPECTROSCOPY (FCCS) FOR WIDER APPLICATION TO MOLECULAR INTERA CTION ST UDIES................................................................89 Limitaions of FCS Autocorre lation Analysis for Molecu lar Interaction Studies........89 Theories of Dual-Color FCCS for Molecular Intera ction St udies .............................91 Lab-Built FCCS Intrum entation Se t-Up...................................................................93 Scheme of Lab-Built FCCS Se t-Up..................................................................93 Components of Lab-Built FCCS Se t-Up...........................................................94 Excitati on....................................................................................................94 Beam combi ners ........................................................................................94 Objectiv e ...................................................................................................95 Dichroic mi rrors .........................................................................................95 Filter s..........................................................................................................96 Pinholes ......................................................................................................96 Correlato rs..................................................................................................97 Construction of the Lab Built FCCS Set-Up .....................................................97 Apply FCCS for Intracellu ar mRNA Dete ction.........................................................98 Summary ................................................................................................................99 5 ENGINEERING A SURFACE ENERGY TRANSFER (SET) NANORULER FOR MEASURING PROTEIN BINDING SITE DISTANCES ON LIVE CELL MEMBRANE .........................................................................................................104 Introducti on...........................................................................................................104 Experimental Se ction............................................................................................107 Preparation of Gold Nanoparticles of Different Sizes.....................................107 Aptamer Synt hesis......................................................................................... 107

PAGE 8

8 Preparation of Aptamer-Functiona lized Gold N anoparticles ..........................108 Preparation of Aptamer-Functionaliz ed Silica N anoparticles .........................109 Antibody Label ing..........................................................................................110 Cell Cult ure....................................................................................................110 Flow Cytometric Analysi s...............................................................................111 Confocal Imaging Analysis .............................................................................112 Saturation Binding Concentra tion Determi nation...........................................112 Competition Studies to Confirm the Saturation Binding of Aptamer-NP Conjugates on Cell Membrane .....................................................................114 Results and Di scussions .......................................................................................115 Competition Studies between Aptamer Sgc8 and Antibody Anti-PTK7 on PTK7.............................................................................................................. 115 Measure the Distance between Two Binding Sites us ing FRET ....................116 SET Nanoruler C onstructi on..........................................................................117 Binding Site Distanc e Determinat ion..............................................................121 Potential Effects of Ster ic Hindrance on the Fluorescence Quenching..........124 Effects of Different Ex citation Intensity and Labeling Efficiency on Distance Determinatio n................................................................................................125 Effects of Different Antibody/Dye Label ing Efficiencies on the Calculations of Binding Site Distanc es...............................................................................127 Using Fluroecent Quenchers as Energy Acc eptors.......................................127 Conclusion s..........................................................................................................128 6 ENGINEERING A LIGHT-SWITCHI NG MOLECULAR BEACON (MB) FOR RIBONUCLEASE H KI NETIC ST UDY..................................................................147 Introducti on...........................................................................................................147 Experimental Se ction............................................................................................149 Material s........................................................................................................149 Instruments....................................................................................................149 Pyrene-MB Synthesis and Purifica tions.........................................................149 Pyrene Beacon Assays for RNase H St udies................................................150 Results and Di scussions .......................................................................................151 Design of Light-S witching MB........................................................................152 Optimization of c DNA lengt hes......................................................................153 Light-Switching MB Assay for Ri bonulcease H Kine tic Study........................154 Sequence Dependence on Ribonucl ease H Cleavage Ki netics.....................155 Conclusion s..........................................................................................................156 7 SUMMARY AND FUTURE DIRECTIONS............................................................162 Instrumentation Developm ent and Nanomaterial Engineering for Live Cell Mapping and Bioa nalysis ......................................................................................162 Future Direc tions.................................................................................................. 164 Exploration of FCS/FCCS Applications for Molecular Interaction Studies inside Living Cells ..........................................................................................165 Engineering Nanomaterials as Natura l Circuit Mimics for Bioanalysis...........166

PAGE 9

9 APPENDIX PROBABILITY CALCULATION OF IN TER-RECEPTOR SET INTERACTION CONTRIBUT ION .................................................................................................162 LIST OF REFE RENCES.............................................................................................162 BIOGRAPHICAL SKETCH ..........................................................................................179

PAGE 10

10 LIST OF TABLES Table page 3-1 Comparison of Kd Value of IgE/Receptor Complexes and IgE Receptor Density on RBL-2H3 Cells Us ing Different Met hods...........................................88 5-1 Summary of Properties for Gold N anoparticles of Diffe rent Sizes....................145 5-2 Summary of DNA sequenc es........................................................................... 145 5-3 Summary of aptamer-functionaliz ation for gold nanop articles ..........................146 6-1 Probes and oligonucleot ides used in Ribonucl ease H kineti c study.................161 6-2 Kinetic parameters of pyrene beacon assays for E. coli RNase H....................161

PAGE 11

11 LIST OF FIGURES Figure page 1-1 Schematic representati on of the structure of Mo lecular Beacons (MB) and spectral characteristics in a “closed” (no target no-fluorescece) vs “open” configuration (target + fluore scence) ..................................................................40 1-2 Fluorescent signal obtained from conv entional fluorescence spectrometry........40 1-3 Jablonski diagram...............................................................................................41 1-4 Schematic representati on of the FRET spectral overlap in tegral........................42 1-5 The Structure of pyrene and the sc hematic for the formation of pyrene excime r...............................................................................................................42 2-1 Fluorescent signal obtained from conv entional fluorescence spectrometer.......56 2-2 Gaussian laser excitation profile and focus volume of observati on....................56 2-3 The working principle to obtain molecular diffusion in FCS st udy.......................57 2-4 FCS experiment al setup....................................................................................58 2-5 FCS for molecular diffusi on studies in solution...................................................58 3-1 Scheme of FCS instrumental set-up and the illustration of aptamer/receptor binding events on the cell me mbrane inside t he focu s.......................................78 3-2 Autocorrelation functions of aptamer sgc8 and control free in solution and on cell memb rane....................................................................................................79 3-3 Selective binding of aptamers sgc8 to positive cells rather than negative contro ls...............................................................................................................81 3-4 Binding of FITC-sgc8 to cell me mbrane on human cervic al HeLa cells..............82 3-5 PTK7 receptor density dist ribution for H eLa cells. ..............................................83 3-6 Comparisons of PTK7 recept or density for HeLa and CE M...............................84 3-7 Binding of FITC-KK1H08 to cell membrane on human chronic myelogenous leukemia cells .....................................................................................................84 3-8 Competition studies and Trypsin treat ment........................................................85 3-9 Binding curve for FITC-labeled IgE ant ibody to IgE receptors on RBL-2H3 rat basophilic leukaem ia ce lls..................................................................................87

PAGE 12

12 3-10 IgE receptor density distri bution for RBL-2H 3 cells. ...........................................87 4-1 Illustrations of molecular interaction studies by the autocorrelation function in FCS and by the cross-correlati on function in FCCS.........................................100 4-2 Scheme of FCCS inst rumentaion se t-up..........................................................100 4-3 Molecular interactions characteriz ed by autocorrelation curves and crosscorrelation cu rves. ...........................................................................................101 4-4 Scheme of the lab-bui lt FCCS se t-up...............................................................102 4-5 Picture of the overlap of the three lase r beams. ..............................................102 4-6 Picture of the lab-built F CCS instrumentai on set-up.........................................103 4-7 Scheme of applying FCCS technique fo r intracelluar m RNA detecti on............103 5-1 Characterizations of differ ent sized gold nan oparticles ....................................130 5-2 TEM images of different size s of gold nanopar ticles........................................130 5-3 Ligand binding saturation conc entration determi nations. .................................131 5-4 Competition studies using Cy5-sgc8 to confirm the satu ration binding of gold NP-aptamer conjugates on the cell surf ace......................................................131 5-5 Competition studies between aptamer sgc8 and antibody anti-PTK7 on receptor PTK7.................................................................................................. 132 5-6 Measure the distance between tw o binding sites us ing FRET .........................133 5-7 Scheme of “SET nanoruler” for meas uring the distance between two binding sites in receptor PTK7 on a live cell me mbrane................................................134 5-8 Flow cytometry assay to monitor the fluorescence intensities on live cells and histogram of the mean fluorescence intensity for the fluorescence quenching assay determined from the flow cytometry re sults ..........................................135 5-9 Confocal imaging assay for monito ring the fluorescence quenching on cell surface with different size s of gold nanopar ticles.............................................137 5-10 Relationship between fluorescence quenching efficiency and gold nanoparticle diam eter.......................................................................................138 5-11 Binding site dist ance determinat ion..................................................................138 5-12 Flow cytometry analysis to monitor the fluorescence quenching effect with 15-nm silica nanopar ticles................................................................................139

PAGE 13

13 5-13 Competition studies using Cy5-sgc8 to confirm the saturation binding of silica NP-aptamer conjugates on the cell surf ace......................................................140 5-14 Effects of different laser intensitie s on fluorescence quenching efficiency.......140 5-15 Effects of different antibody labeling efficiencies on fluorescence quenching..142 5-16 Flow cytometry analysis using diffe rent fluorescence quenchers as energy acceptor s..........................................................................................................144 6-1 Schematic representat ion of the fluorescence mechanism using lightswitching pyrene assay for RNa se H activity study..........................................157 6-2 Steady-state fluorescence spectra of MB226 pyrene beacon assay for RNase H activity study. ....................................................................................158 6-3 Time-base fluorescence monitoring of RNase H cleavage activity in MB226 assay ................................................................................................................159 6-4 Steady-state fluorescence spectra of pyrene beacon (20 nM) with loop-cDNA (1:100) for optimizat ion of RNA sequence in MB226 assa y.............................159 6-5 Curves of cleavage of RNA strand from MB226 assay by E. coli RNase H at different enzyme concentration at 37 ............................................................160 6-6 The Linear weaver-Burk plot of recipr ocals of initial rates versus substrate concentration for the determinat ion of kinetic parameters Km, kcat and Vmax of RNase H in MB 226 assay ................................................................................161

PAGE 14

14 ABSTRACT OF DISSERTATION PRESEN TED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY INSTRUMENTATION DEVELOPMENTS AND NANOMATERIAL ENGINEERING FOR LIVE CELL MAPPING AND BIOANALYSIS By Yan Chen August 2010 Chair: Weihong Tan Major: Chemistry Intracellular molecular interaction studies have attracted great interest among cell biologists. However, the intrinsic resoluti on of conventional optical microscopy only allows the visualization of subcellular locali zations. Therefore, a sensitive technique that can observe molecular interactions down to the molecular level is in great need. By extracting information from mo lecular diffusion, Fluorescence Correlation Spectroscopy (FCS) gives detailed information on mole cular interactions, and has become a noninvasive single-molecule-detection techniq ue that can be applied to the intracellular environment with low detection limits.1 Therefore, the overall direction of this doctoral research was the development of a self-built FCS setup. Combined with nanomaterial engineering, this project has enabled memb rane receptor studies and intracellular measurements in live cells. The Fluorescence Correlation Spectrosc opy instrumentation was to direct measurement of membrane rec eptor density of the PTK7 pr otein, an important cancer marker, in its natural physi ological environment on the cell surface. A cellular model using a DNA ligand aptamer was designed for specific recept or targeting and labeling,

PAGE 15

15 and was used with FCS to determine receptor densities and distributions on the cell surface. With its intrinsic advantages of direct measurement, high sensitivity, rapid analysis and single-cell measurement, th is FCS density estimation approach holds great potential for future applications in molecular interaction studies and density estimations for subcellular st ructures and membrane receptors.2 To further understand the st ructure of PTK7 beyond its spatial distribution and ligand-receptor interactions, a nanoparticle wa s used as a molecular ruler to measure the distance between two binding sites on the receptor on live cells. Measuring distances at molecular length scales in living systems is a significant challenge. Methods like FRET ( fluorescence resonance energy transfer ) have significant limitations due to short detection distances and strict or ientations. To overcome these limitations and construct a practical nanoruler for meas uring distances on live cells, an SET-based nanoruler, using aptamer-gold-nanoparticle conj ugates with different diameters, was developed to measure s eparation distances well beyond the detection limit of FRET. Since application of fluorescence autocorrelation (FCS) to binding analysis is limited to applications in which the binding ev ent significantly reduces the diffusion of the labeled species, the original FCS se tup was upgraded to a novel three-channel Fluorescence Cross-Correlation Spectroscopy (FCCS) setup. This lab-built FCCS not only inherits the single-molecule detection c apability from FCS, but also further extends its applications for molecular interaction studies by labeling two species with two spectrally distinct fluorophores. This tec hnique has been estab lied and is now being adapted for real-time monitori ng of intracellular mRNA.

PAGE 16

16 Another aspect of this research was t he development of molecular probes based on nanomaterial engineering Two different types of molecu lar beacon (MB) probes were designed for enzymatic activity studies3 and protein inhibition studies4. In summary, this research mainly fo cuses on instrumentation development and nanomaterial engineering for bioanalysis and biom edical applications, especially for cell membrane receptor studies and intracellular measurements. A successful outcome from these studies will lead to a better understanding of biological events and processe.

PAGE 17

17 CHAPTER 1 INTRODUCTION Molecular Probes for Live Cell Mapping Mapping living cells with good spatial and tem poral resolution offers significant potential for the understanding of impor tant biological phenomena. Probing biomolecules in living cells not only allows us to precise characterization of molecular properties, but also helps to define the role of the molecule in cellular processes. Molecular probes emerge as a class of sens itive tools for probing these biological events down to the molecular level, pr oviding detailed information on molecular interactions in living cells. In particular, us e of nucleic acids has been a key factor in probe design due to their capabili ty for molecular recognition.5 The Watson-Crick type hydrogen-bonds, in addition to electrostatic forces, -stackings and hydrophobic interactions, not only make DNA duplication and transcription perfectly cyclical, but also bring the natural and specific inte ractions of nucleotide base pairs into the field of molecular recognition.5 By careful engineering of the st ructures of the nucleic acid probes, the Tan research group has developed various types of molecular probes via different recognition mechanism s, such as a molecular s witch, molecular beacon (MB), 6-8 and a specific molecular probe for bio-recognition, aptamer.9 To gain an understanding of molecular bi ological processes with accurate quantitation, advanced tools wit h controlled size and shape ar e in great demand. Over the past decade, significant advances have led to a large va riety of emerging nanomaterials.10-12 These materials, notable for their extremely small feature sizes and well-defined molecular shapes, have been conf irmed as reliable and effective tools,

PAGE 18

18 enabling scientists to observe, understand and manipulate the phenom ena of biological processes at the molecular level. O ne of the most exciting aspects of bionanotechnology is the use of nanomaterials to carry out target-specific functions.13 In particular, the introduction of nucleic acids into this “bio-matrix” results in nanomaterials with the added capability fo r molecular recognition.13 Therefore, to vi sualize biochemical reactions and events in living cells quantit atively calls for the combination of nanomaterials and nucleic acids to provide novel molecular tools. A typical molecular probe consists of a targeting moiety and a signaling component.14 The targeting moiety utilizes select ive molecular recognition to allow discriminative interaction with the target mo lecule in complex cellular environments. Depending on the signal transduction used for the probe design, the signaling component generates observable response w hen the probe and target binds. Because of its nondestructive nature, high sensitivity, flexible signaling schemes and multiplexing capability, fluorescence is the first choice in constructing the si gnaling components. The fluorescence signal transduction elements ava ilable for probe design involve different fluorescence energy transfer mechanisms,15 such as fluorescence quenching, fluorescence resonance energy transfer (FRE T), surface energy transfer (SET), and excited state dimer-excimer formation. In th is research, we have introduced different signal transduction mechanisms to the desi gn and engineering of molecular probes for various applications in molecular inte raction studies and live cell mapping. There is no doubt that molecular probes have played a fundamental role in live cell studies. At the same time, however, t he development of sens itive and non-invasive instrumentation is also important in accele rating the pace of intrac ellular measurements.

PAGE 19

19 Cell biologists strive to charac terize molecular interactions directly in the intracellular environment. To meet this demand, advanced fluorescence techniques, including highspeed sensitive CCD cameras, fast computer s, powerful lasers and high-resolution microscopy, have been developed. Most of these techniques depend on interpreting molecular interactions by tracking fluor escent signals. Recently, a new type of fluorescent technique, named Fluorescenc e Correlation Spectroscopy, has been developed and employed to observe molecular interactions by measuring the dynamics of biochemical reactions in the intact cell.16, 17 All these different techniques constitute a powerful tool set to interrogate molecular interaction information inside a living cell. This dissertation focuses on the design of selective molecular probes, which combine with sensitive fluorescence techni que developments for molecular interaction probing with membrane protein mapping on live cell surfaces. The following sections will discuss the molecular engineering of nucleic acid probes and nanoparticles and the basics of fluorescence signal transduction me chanisms. Specifically, the molecular recognition principles and applications of molecular beacons (MB) and aptamers will be reviewed. In addition, different types of nanoparticles will also be introduced for molecular probing, due to their controlla ble feature sizes and well-defined molecular shapes. Finally, the challenges of using mo lecular probes for complex biological systems will be discussed and a sensitive techniqu e for molecular interaction studies, named Fluorescence Correlation Spectroscopy (FCS), will be introduced to enhance the detection sensitivity.

PAGE 20

20 Molecular Engineering of Nanomaterials In the post-genomic era, gaining knowl edge about the biological world at the molecular level requires advanced bioanalytical tools. By developing novel techniques in nanoscience and nanotechnology, the essent ial relationships between structure and function of biomolecules at the nanoscale can be determined. Since most biological systems operate at the nanosca le, the field of biology has greatly benefited from advances in nanoscience, leading to new levels of sensitivity, prec ision, and resolution in biomolecular analysis. Over the past decade, significant advances have led to a large variety of emerging nanomaterials. 5, 10-12 These materials, notabl e for their extremely small feature sizes and well-defined molecu lar shapes, have been confirmed as reliable and effective tools, enabling scientists to observe, understand and manipulate the phenomena of biological processe s at the molecular level. Nanomaterials can be composed from metals, ceramics, polymers, or composites. Their molecular shapes can vary from spheres, cylinder s, to single-walled tubes or any well-defined struct ures. The defining characterist ic of nanomaterials is a very small size in the range of 1-100 nanometers (nm). One nanometer spans 3-5 atoms lined up in a row. By comparison, t he diameter of a human hair is about 5 orders of magnitude larger than a nanos cale particle. The nanoworld lies midway between the scale of atomic/quantum phenomena, and t he scale of bulk materials. Thus, nanomaterials do not simply comprise another step in miniaturizat ion; they form an entirely different arena. At t he level of nanoscale, some material properties are affected by the laws of atomic physics; thus, nanom aterials do not behave like traditional bulk materials. If so, what makes these nanomater ials so intriguing for biologists? Their

PAGE 21

21 extremely small-featured size which is on t he same scale as the critical size for biological phenomena. For example, since proteins are 10-1000 nm in size and cell walls are 1-100 nm thick, their interacti ons with nanomaterials may be quite different from those observed with larger-scale mate rials. Fundamental electronic, magnetic, optical, chemical, and biological properties are also different at this level. Nanomaterials in diverse shapes and mate rials, such as DNA and RNA probes, polymers, metallic (Au and Ag) NPs (nanoparticles)10, 12, 18, 19, silica NPs20, 21, magnetic NPs22-24, quantum dots (QDs)11, 25, 26, nanorods27, 28, and carbon nanotubes29, 30, have been widely applied to bioanalysi s. Different types of na nomaterials have their own unique properties and they have been adapted to different applications in the bioanalysis field, such as biosensing19, 31, 32, drug delivery33, 34, gene therapy35-37, and medical diagnostics13, 27, 38. This dissertation focuses on the devel opment and engineering of nucleic acid probes and nanoparticles, as well as their conjugates, for molecular recognition and live cell mapping studies. The molecular des igns, synthesis, functionalization, and applications in different bioanalysis pl atforms are described and investigated. Molecular Engineering of Nucleic Acid Probes Nucleic acids are ideal building blocks fo r the construction of molecular probes for living cell studies and bioanalysis. Sci entists have put tremendous effort into the engineering of nucleic acid pr obes with different structures, shapes and functions. There are several reasons why nucleic acid probes are regarded as one of the powerful tools for recognition.

PAGE 22

22 First of all, nucleic acid base pairing is one of the strongest and most specific biomolecular recognition mechanisms, allowing molecular targeting and recognition with high sensitivity and specificity.5 Second, nucleic acid sequences can be easily designed according to different needs, and they are also easy to synthesize and purify. Also, there are various types of modifications available for labeling nucleic acid sequences with different dyes and radioactive isotopes, as well as modified nucleic acid bases.39 Third, by designing nucleic acid probes into different secondary structures, different signal transduction moieties can be in tercalated into specific positions of the structure, to act as reporters of structures. This allows construction of molecular probes with different structures and signal transduction mechanis ms for target recognition reporting.7 Finally, an in vitro selection technique named SELEX40-42 make it possible to obtain nucleic acid sequences (a ptamers) that are capable of binding to a large range of target molecules, including ions, organic mo lecules, peptides, proteins, cells and tissues, with high affinity and selectivity. By intr oducing aptamer technol ogy, the targets of nucleic acid probes have been greatly expanded from traditional nucleic acid sequences to any type of target or component. Therefore, nucleic acid probes have been designed with different structures and shapes and applied in various areas of biolog y, medical science and chemistry. Today, nucleic acid probes, especially DNA probes, ar e instrumental and ubiquitous tools in exploring biological processe s and in medical diagnostics.

PAGE 23

23 Molecular beacon (MB): molecular switch for signal transduction Among the various nucleic acid designs, a hairpin shaped molecular beacon (MB) is one of specific recognition units wit h an excellent built-in signal transduction mechanism for detection without separations.14 The hairpin-shaped structures in MBs are constructed from a stem and a loop (Figure 1-1). One end of the MB stem contains a fluorophore with a quencher moiety at the ot her end. The MBs act as primarily closed switches, bringing the fluorophor e/quencher pair together to turn the fluorescence "off". When hybridized to its target, the MB undergoes a conformational change that opens the hairpin structure; as a result, t he fluorophore and the quenc her are separated and the fluorescence is turned "on". However, di gestion of the oligonucleotides enzymes will also lead to the separation between the fluorophore and quencher, which also makes the MB switch from the “off” to the “on” state. This problem can be avoided by introducing artificial nucleic acid bases.39 The MB’s signal transduction is based on a fluorescence resonance energy transfer (FRET) 6, 43, described further below in “Fl uorescence Methods for Signal Transduction”. The fluorophore and quencher pai r can also be customized for any experimental setup or conditi ons. These features have led to a wide applicability and new utility for MBs in areas where convent ional nucleic acid probes cannot function well.44 Since MBs’ development, they have pr ovided many exciting opportunities in DNA/RNA/protein studies.6, 45, 46 Furthermore, MBs can be designed for (i) selectivity with single base mismatch identification c apability, and (ii) detection without separation (real-time monitoring in homogeneous soluti on or living samples), (iii) sensitive

PAGE 24

24 detection of rare messenger RNAs, and the absence of cytotoxic effects, allows continuous monitoring of the physi ological changes in live cells. It is precisely the ability of molecular beacons to detect without separation that makes them ideally suited as reporters in ultra small volumes and inside living cells. Thus, of the many potential probes test ed for sensitive detec tion, MBs have emerged promising probes, which gain steady growth in application, such as in mechanism studies of biological functions and the sensit ive detection of several diseases’ biological species. In particular, with their high sensitivit y, excellent specificity, flexible design, and modifications, MBs are also vastly promising molecular tools for quantitative intracellular studies. Aptamers: molecules for bio-recognition The application of MBs for molecular re cognition sometimes is limited by the range of targets and the nucleic acid sequen ce design. However, the introduction of aptamers greatly extends the application of nucleic acid probes to a large range of targets. Molecular aptamers are single-stranded DNAs (ss DNAs) and RNAs that can recognize target proteins, peptides and ot her small molecules. The dissociation constants of aptamers to targets can range from 10–12 M-10–8 M. Aptamers recognize their targets with high specificity, and can ty pically discriminate between protein targets that are highly homologous or di ffer by only a few amino acids.47 The tertiary structures formed by the ssDNAs are the basi s for target protein recognition.48 These aptamers are selected by a process called SELEX (Systematic Evolution of Ligands by Exponential enrichment), where the aptamers are selected fr om libraries of random

PAGE 25

25 sequences of synthetic DNA or RNA by repetitive binding of these oligonucleotides to the target molecules.49 Through this iterative in vitro selection process, aptamers with high specificity and affinity to their tar gets can be obtained. Mo st of the aptamers reported so far have been selected using pure molecules, such as purified proteins, as the targets. Aptamer selection against comp lex targets (such as red blood cells or a single protein on live trypanosomes) was al so demonstrated and interesting aptamers have been generated.50 Aptamers have some advantages in molecular recognition and imaginglow molecular weight, easy and reproduc ible synthesis, easy modification, fast tissue penetration, low toxicity or immunogen icity, easy storage, hi gh binding affinity and specificity that are very comparable with antibodies. 51 Aptamers have shown great promise in molecular recogniti on, diagnosis and therapy. Cell-SELEX: A Molecular Evolution that Generates A Panel of Aptamers To produce probes for molecular analysis of tumor cells, the Tan research group has developed a novel me thod, the cell-based aptamer selection process (cellSELEX), for aptamer selection.9 Instead of using a single type of molecule as the target, the cell-SELEX process uses whole cells as targets to sele ct single-stranded DNA aptamers that can distinguish target cancer cells from c ontrol cells (Figure 1-2). In addition to those mentioned above for aptam ers, the greatest adv antage of cell-SELEX technology is that there is no need of pr ior knowledge about t he potential cancer biomarkers for cancer on these cells. A group of cell-specific aptamers can be selected in a relatively short period (4 to 10 weeks) without knowing which target molecules are present on the cell surface after selection, the target molecules, which might be important cancer biomarkers maybe characterized, because the aptamer binding may

PAGE 26

26 provide information about the molecular structur e of cell-surface rec eptors. There is no easy way to produce a similar panel of monoc lonal antibodies in such a short time without access purified samples of the possible antigens. Compared to 2-D gel electrophoresis and mass spectrometry for pr oteomic studies aim ed at identifying proteins, cell-SELEX first produces molecula r probes, which can then be to identify the target proteins. Thus, not only can the se lected aptamers be used as molecular probes for molecular analysis of cancer, but also t hey can be used as tools for identifying new biomarkers expressed by tumor cells or other cells in disease status. Molecular Engineering of Nanoparticles Nanoparticles (NP)s are defined as particulate dispersions or solid particles with sizes in the r ange of 1-100nm. Nanoparti cles are of great scientific interest, as they effectively form a bridge between bu lk materials and atomic or molecular structures. Bulk materials have constant int ensive properties regardless of their sizes. However, this is not the case for nanoparti cles, whose properties are size-dependent. In addition, the properties of materials change as their si ze approaches the nanoscale, which makes some specific properties of nanoparticles quite different from bulk materials, even though they ar e constructed with the same ty pes of atoms. For example, it now appears clear that nanoparticles will over come many of the significant chemical and spectral limitations of molecular fluorophor es. In most cases, these interesting and unique properties of nanoparticles are main ly due to the large surface area of the material, which dominates the contributions made by the small bul k of the material. Thanks to state-of-art synthetic techniqu es, methods for preparation and handling of

PAGE 27

27 sizeand shape-tunable nanoparticles are proliferating, and various types of nanoparticles having different sizes and shapes are becoming commercially available. Nanoparticles in diverse shapes and materials have been widely applied for bioanalysis. Nanoparticles can be used as quantit ation tags, such as optical detection of fluorescent nanoparticles and the electrochemic al detection of metallic nanoparticles. Encoded nanoparticles, such as striped metall ic nanoparticles, can also be applied as substrates for multiplexed bioassays. In addition, nanoparticles are also used to leverage signal transduction, for example in colloidal gold-bas ed aggregation assays. Moreover, functionalized nanoparticles can ex ploit specific physical or chemical properties to carry out novel processes, such as catalysis of biological reactions. Finally, nanoparticles are widely utilized in the life sciences for biosensing19, 31, 32, drug delivery33, 34, gene therapy35-37, and medical diagnostics13, 27, 38. There are different types of nanomaterials composed of different materials, such as metallic (Au and Ag) NPs (nanoparticles)10, 12, 18, 19, silica NPs20, 21, magnetic NPs22-24 and iron oxide NPs52. This dissertation will focus on appl ications of gold NPs and silica NPs for live cell mapping and bioanalysis. Gold NPs (commonly known as Au colloid or colloidal Au) 10, 12, 18, 19 are key materials in nanoscience and nanotechnol ogy and have been studied more than other types of NPs. an important property of gol d NPs is their precisely controllable size achieved by conventional synthesis met hods. Also, the broad absorption bands around 520nm of gold NPs make them good candidate s for use as fluorescent quenchers for multiple fluorophores. Therefore, gold NPs have been widely used for the signal transduction design in molecular probes.53 Furthermore, the presence of a plasmon

PAGE 28

28 absorption band and shapeand si ze-dependent optical prope rties make gold NPs suitable as colorimetric probes.54 The plasmon-resonance s pectra of free single particles differ significantly from those of aggregated NPs. Gold NPs have been used to develop highly sensitive detection schemes for many targets.55 Ultrasensitive analysis of oligonucleotides, proteins, and other biomol ecules has been achieved using gold NPs as biomarkers12. Gold NPs have been approved for in vivo use by the Food & Drug Administration and are already in commercia l products. One well-known example is the lateral flow strip developed for fast pathogen detection and point-of-care diagnosis56, 57. Silica NPs are currently used in many areas of bioanalysis. While polymer-based NPs have been used widely in bioanalysis and labeling, silica NPs, by comparison, show less aggregation and little dye leakage.58 Using appropriate synt hetic conditions, a large number of dye molecule s (either organic or inorganic) can be incorporated inside a single silica particle ( potentially tens of thous ands of dye molecules).20 When a large amount of dye is incorporated in such a small volume, some fluorescence-quenching phenomena occur within the NP. Nonetheless, the goal of obtaining a particle with brighter luminescence has been largel y successful. Dye-doped NPs produce a highly amplified optical signal compared to a si ngle-dye molecule. If applied appropriately in bioanalysis, silica NPs can provide a great improvement in analytical sensitivity. Moreover, since the dye is trapped inside the silica matrix, which provides an effective barrier against leakage, both photobleaching and photodegradation that often affect conventional dyes can be minimized.59 The excellent photostability makes these NPs suitable for applications where high intensit y or prolonged excitations are required. For example, intracellular optical imaging suffers severely from photobleaching, but this

PAGE 29

29 problem could be minimized with silica NPs. T he flexible silica chemistry also provides versatile routes for surface modification. Different types of functional groups can be easily introduced onto the NP surface for conj ugation with biomolecules. In addition, the silica surface makes these NPs chemically inert and physically stable.60 All these properties make silica NPs excellent label ing reagents for bioanalysis and bioimaging.61 Combination of Nucleic Acid Prob es and Nanoparticles for Biological Applications Thanks to their intrinsic molecular proper ties, the combination of functional nucleic acid molecules and nanoparticles re sults in an unprecedented improvement in molecular recognition. First, due to the high su rface area to volume ratio, nanoparticles can be modified to generate many types of DNA probes to enable more effective molecular recognition. In addition, the phenomenon of cooperative in teraction, also called the synergistic effect, makes the re cognition much easier towards targets with more binding sites. When suitable nanoparticles serve as reporters, the signal of recognition events can be amplified by thousands due to the several reasons.62 First, because each NP carries multiple recognition molecules, there is a higher possibility of target binding, which can improve the det ection selectivity. Second, unlike other biomolecules, DNA probes are stable afte r modification. The easy handling of DNA strands and different modification strategi es involving nanoparticles provide a vast platform to achieve molecula r recognition. Third, the unusual interactions between nanoparticles and living systems make the application of functional DNA more practical for molecular recognition and medical diagn ostics. With the help of nanoparticles, nucleic acids can escape digestion by, for example, cellular nucleases, and can be

PAGE 30

30 transported across the cell membranes to recogni ze bioactive substances. This offers a chance to recover the real recognition event in vivo Owing to these advantages, the combination of DNA molecular design a nd with a variety of nanoparticles has been widely applied in the interdisciplinary fiel ds of chemistry, biology and medicine.63 In summary, the combination of f unctional nucleic acid molecules and nanoparticles offers significant advantages for bioanalysis and molecular recognition studies. The high surface-area/volume ra tio of nanoparticles not only provides an effective platform for nucleic acid probes for molecular recognition, but can also increase the loading efficiencies of mo lecular probes to enhance the detection sensitivity. In addition, nanoparticles can not only work as molecular probe carriers, but can also act as reporters to simplify the design of molecular probes. In addition, the easy handling of DNA strands and different m odification strategies of nanoparticles provide a vast platform to achieve mole cular recognition. Finally, with the help of nanoparticles, nucleic acids can escape digest ion by, for example, cellular nucleases, and be transported across the cell membranes to recognize bioactive substances. This offers a better chance for applying nucleic acid probes for in vivo applications. Therefore, the combination of DNA molecular desi gn and different nanoparticles has been widely applied in chemistry, physics and medical research. Fluorescence Methods for Signal Transduction Fluorescence is a widely used tool for a va riety of investigations in biochemical, medical, and chemical research due to its high sensitivity, nondestructive nature, and multiplexing capabilities.15 This section presents diffe rent types of fluorescence mechanisms for signal transduction for use in the design of molecular probes.

PAGE 31

31 Jablonski Diagram15 A Jablonski diagram is used to illustrate the mechanism of fluorescence from the photophysical processes that occur during the absorption and subs equent reemission of light. Figure 1-3A shows a typical Jablonski diagram where S0 and S1 stand for the ground electronic state and the first singlet ex ited electronic states, respectively, while T1 stands for the triplet state. In each el ectronic state, there are different discrete vibrational levels. When a molecule is irr adiated by photons, it can be excited to upper vibrational levels in the singlet excited energy level S1 or higher singlet levels, depending on the magnitude of the absorbed energy. The exci tation process is very fast and usually occurs in the 10-15 s range. Through vibrational relaxation, molecules in higher vibrat ional energy levels of a specific excited state, S1 for example, rapidly relax to the lo west vibrational level of S1, in the next 10-12 s. In this process, energy is transfe rred as heat via collisions with the surrounding solvent molecules. If conditions are favorable, the excited molecule can then relax from the lowest ex cited state S1 to an upper vi brational level in S0 with emission of photons. This relaxation is referr ed as fluorescence. The average time for a molecule to stay in its excited state is called the fluorescence lifetime. Sometimes, a molecule in the excited state undergoes a proc ess called intersystem crossing, in which an electron is converting the molecule to a tr iplet state T1. In this case, the relaxation from the triplet state T1 to the ground state S0 with emi ssion of photons is referred as phosphorescence. Since fluorescence emission typically takes 10-10 to 10-6 s to occur, but vibrational relaxation genera lly occurs much faster (10-12 s) and therefore is completed prior to emission. Thus, fluore scence emission generally occurs from the

PAGE 32

32 lowest-energy vibrational state of S1. As a result, fluorescence emission energy is usually independent of the ex citation energy. In addition, because a small fraction of energy absorbed is lost during vibrational relaxation, there is an energy difference between the energy absorbed and emitted during the fluoresc ence process. Therefore, emission of a specific molecule usually appears at longer wavelengths than the absorption wavelength, which results in a “S toke’s shift” between the two peaks (Figure 1-3B). More importantly, this “Stoke’s shi ft” allows the spectral separation of the excitation photon from the emi ssion photon for sensitive studies. The signal transduction techniques used in fluorescence-based design consider how to associate the target recognition event with changes in the relative rates of fluorescence, which result in change in the in ternal conversion, external conversion, and inter-system conversion. In addition, if t he target binding event changes the electronic structure of the fluorophore, changes in fl uorescence excitation/emission wavelengths will be observed and can be used to signal the target binding event. The specific mechanisms of four differ ent signal transduction approaches used in the Tan laboratory are described below. They are fluorescence quenching, fluorescence resonance energy transfer (FRET), surface e nergy transfer (SET) and excited-state dimer formation. Fluorescence Quenching In addition to internal and external conv ersions, there are ot her non-radiative processes by which excited state molecule s relax to the ground state. Because these occur without photon emission, they are te rmed fluorescence quenching processes. Quenching and dequenching upon interaction with a specific molecular biological target

PAGE 33

33 form the basis for signal transduction design in molecular probes, such as molecular beacons. Fluorescence quenching occurs through two major mechanisms:15, 64 collisional quenching (also called dynamic quenching), and static quenching. For collisional quenching, an excited fluorophore collides with other molecules in the solution during its fluorescence lifetime, causing energy lo ss from the fluorophore. Therefore, the fluorophore returns to the ground state wit hout emitting photons. The decrease in fluorescence intensity due to the collis ional quenching can be described using the Stern-Volmer equation: 0 01[]1[]qF KQkQ F (1-1) in which K is the Stern-Volmer quenching constant, which is the product of kq (in M-1s-1), and the fluorescence lifetime in the absence of the quencher ( 0 in seconds), and [Q] is the quencher concentration. Collisional quenchers can be heavy ions, oxygen, halogens, amines, acrylamide and other materials. The collision rate fo r molecules in aqueous solutions at room temperature is about 1010 Lmol-1s-1. If all these collisional events result in fluorescence quenching, it can be estimated that the maximum value for kq is about 1010 Lmol-1s-1. For a fluorophore with a lifetime 0=1 ns, the Stern-Volmer quenching constant K is approximately 10 Lmol-1. Therefore, these estimations show that when the quencher concentration is below 10-3 M, the dynamic quenching of fluorescence is usually negligible. For two molecules that are brought together by linkers in many molecular probes, the collision rate is not diffusion rate controlled, and the dynamic quenching may be more prominent.

PAGE 34

34 In static quenching, the quencher fo rms a non-fluorescent complex ( i.e, dark complex) with the fluor ophore in the ground state.15, 64 The decrease in the fluorescence intensity for static quenching is described by Equation (1-2): 0[] 1[]1 ([][]) F FQ KQ FFQ (1-2) in which K is the formation constant of the dark complex, and [F Q], [F], and [Q] are the concentrations of the dark complex the fluorophore, and the quencher, respectively. There are two simple ways to disti nguish static quenching from dynamic quenching, though both of them result in a decrease of em ission intensity.15, 64 The first involves lifetime measurements. In stat ic quenching, the lifetime does not change, because the only observed fluorescence is from the uncomplexed fluorophore, which has the same lifetime as before quenching. In contrast in a dynamic quenching mechanism the lifetime shows the same order of decrease as the intensity. Second, temperature plays different roles in the tw o processes. In static quenching, higher temperature dissociates weakly bound comple xes and alleviates static quenching. For dynamic quenching, higher temperature c auses faster diffusion and more quenching. But in many cases, both static and dynam ic quenching processes occur in the same system. Static quenching plays an impor tant role in molecular probes. For example, it is involved in the fluorescence que nching of fluorophores in MBs.65 Many fluorophore/ quencher pairs, including tetramethylr hodamine(TMR)/DABCYL, EDANS/DABCYL, eosine/DABCYL, fluorescein/TMR and TMR/TM R display absorption spectral changes when they are brought close together in t he hair-pin conformation, indicating the formation of non-fluorescent co mplexes in closed-stem MBs.65 The static quenching that

PAGE 35

35 occurs in MBs leads to higher quenching efficiency than other process, such as fluorescence resonance energy transfer (FRET).65 Fluorescence Resonance Energy Transfer (FRET) Fluorescence resonance energy transfer (FRET) is an interaction between a donor chromophore and an acceptor chromophore.15 The donor chromophore, initially in its electronic excited state, may transfe r energy to an acceptor chromophore through nonradiative dipole–dipole coup ling. FRET requires an ov erlap between the emission spectrum of the donor wit h the absorption spectrum of the acceptor, as shown in Figure 1-4. Such an overlap allows the excitation en ergy to be transferred from the donor to the acceptor if the two molecules are coupled by a dipole-dipole interaction within a distance of 10 nm. FRET efficiency defined as Equation (1-3), depends strongly on the distance between the donor and the acceptor molecules as described in the following equation: 6 01 1() E r R (1-3) where the Forster radius R0 is the distance at which energy transfer is 50% efficient, and r is the distance between the donor and the acceptor. The efficiency of FRET is dependent on the inverse sixth power of the intermolecular separation, making it usef ul over distances comparable to the dimensions of biological macromolecules. Thus, FRET is an important technique for investigating a variety of biological phenomena that prod uce changes in molecular proximity. When FRET is used as a contrast mechanism, colocalization of proteins and

PAGE 36

36 other molecules can be imaged with spatial re solution beyond the limits of conventional optical microscopy. Surface Energy Transfer (SET) Recently, several groups66-74 have reported the phenomenon of nanoparticle surface energy transfer (SET), in which energy transfer flows from a donor molecule to a nanoparticle surface at a much slower decay rate than the dipole-dipole energy transfer in FRET, the 1/d4 distance dependence of SET is described in the Equation (14): 66 4 01 1() E r R (1-4) where the Forster radius R0 is the distance at which energy transfer is 50% efficient, and r is the distance between the donor and the surface of t he metal particle. While SET is similar to FRET, in that the interaction is dipole–dipole in nature, it is geometrically different from FRET bec ause an acceptor nanoparticle has a surface and an isotropic distribution of dipole vectors to accept energy fr om the donor, leading to a dipole-surface resonance mechanism.67-70 This arrangement effectively breaks the inherent detection barriers of FRET (~10 nm ), thereby increasing the probability of energy transfer and ultimately enhancing the efficiency of SET over FRET. The intensity quenching mechanism via c oupling of the oscillating electronic dipole of a dye to a metal surface with lo ss of energy was developed by Chance, Prock and Silbey,75 and by Persson and Lang76 for metals. Aside from bulk systems, recent attempts have conjugated different lengths of dsDNA onto the metal nanoparticle

PAGE 37

37 surface, thus demonstrating t he validity of this “SET nanor uler” for mapping distance in a real biological system.66 These “SET nanorulers” have the potential to become an alternative to FRET for molecular intera ction and distance measurement in cellular systems, especially for applications demand ing long observation times or large distances. It represents the next leap forward in the use of optical probes to monitor structural components within a cell membr ane, and will open a new pathway for cellular imaging. Excited State Dimer-Excimer Formation Pyrene is a polycyclic aromat ic hydrocarbon (PAH) consisting of four fused benzene rings, resulting in a flat aromatic system (left in Fi gure 1-5). It can form excited state dimers (excimers, *Py+Py) upon close encounter of an excited state molecule (*Py) with another ground state molecule (Py) 15, 77, 78 (right in Figure 1-5). When a pyrene molecule is excited from the ground state (Py) to the excited state, the excited pyrene (*Py) can relax to the ground state through internal conversion. During this process, when such an excited pyrene (*Py) enc ounters with a second pyrene in ground electronic state (Py), a complex (*Py+Py) with lower energy will form, which is named excimer. The pyrene excimer emission has a broad, featureless emission centered at 480 to 500 nm, while the pyrene monomer emit s in the 370 to 400 nm wavelength range with two intensive peaks. Similar to FRET, the forma tion of excimer is stri ngently distance-dependent, and thus can be used as a unique signal transduct ion in the construction of molecular probes. It can be adopted as the signaling moieties in the stru cture of MB to report the conformational change of molecular probes upon target binding.78 The formation and

PAGE 38

38 dissociation of pyrene excimer switch from green light to blue light, which also finds a wide application for signal transduction for ot her nucleic acid probe s, such as aptamer switches.7 Challenges of Using Molecular Pr obes for Intracellular Analysis Many of the applications discussed above exhibit limitations and challenges to intracellular applications of molecular pr obes. One of the major challenges is low sensitivity. For example, MBs have been repor ted to be able to detect as low as 10 copies of mRNA sequences,79 but most MB applications so far are limited to detecting highly expressed or stimulat ed genes. Another challenge lies in the restrictions in the molecular probe design and engineer ing. It is very difficult to generate a universal and conventional approach for molecular probe desi gn to intercalate the signal transduction moieties for target recogniti on reporting. For example, al though aptamers with different sequences can recognize different target, it is always a difficult task to design an aptamer switch with a signal change to r eport the binding. T herefore, generating solutions to these problems is crucial for t he better performance and wider application of molecular probes, especially fo r intracellular analysis. Fluorescence Correlation Spectroscopy (FCS) for Live Cell Analysis The development of sensitive instrum entation with low det ection limits can definitely facilitate the design of molecular probes for improved performance in live cell analysis. Fluorescence correlation spectroscopy (FCS), which belongs to a class of single molecule detection techniques, has recently been applied to intracellular environments and has great potentia l for addressing these problems.80

PAGE 39

39 FCS is exquisitely sensitive, provides access to a multitude of measurement parameters in real time, is noninvasive and offers rapid tempor al and high spatial resolution. Through analysis of minute s pontaneous fluorescence fluctuations, the autocorrelation function provides an ideal method for studying diffusion and dynamics of proteins at nanomolar concentra tions in living cells. Coupled with imaging, FCS can be used on the subcellular level to morphologically dissect biological processes in different compartments, such as dendritic spines,81 that have been previously intractable to biochemical experiments. FCS measurements also allow the determination of both cellto-cell and position-to-position (within the same cell) variability. Many techniques measure average values of a seemingl y ‘homogeneous’ ensemble, in which the heterogeneity of a process in a cell population or within t he same cell is lost. This information may be biologically important to understand the extent of variation in cellular responses. Besides the sensitivity improv ements, FCS also changes the mode of observing molecular interactions from t he change of fluorescence intensity to the difference in molecular diffusi on. This different approach c an greatly simplify molecular probe design for signaling molecular in teractions. The detailed theories and instrumentations of FCS are discussed in Chapter 2.

PAGE 40

40 Figure 1-1. Schematic repres entation of the structure of Molecular Beacons (MB) and spectral characteristics in the “closed” (no target no-fluorescece) vs “open” configuration (target + fluorescence) states. Figure 1-2. Schematic of cell-bas ed aptamer selection (cell-SELEX).9 Briefly, a ssDNA pool is incubated with the target cells to conduct the positive selection. After washing, the bound DNAs are eluted by heating in binding buffer. Then the eluted DNAs are incubated wit h the control cells (negative cells) for a counterselection. After centrifugation, t he unbounded ssDNAs in the supernatant are

PAGE 41

41 collected, and then amplifi ed by PCR. The amplified DNAs are used for next selection round. The selection process is monitored using fluorescent imaging by confocal microscopy or fluorescent analysi s by flow cytometry. After around 1520 rounds of selection, the final po ol is cloned and sequenced. Aptamer candidates are identified and tested. Figure 1-3. (A) A typical Jablonski diagram. (B) The excitation and emission spectrum for a widely used dye molecule, Fluorescein.

PAGE 42

42 Figure 1-4. Schematic repr esentation of the FRET spectral overlap integral. Figure 1-5. The Structure of pyrene (left) and the schematic for the formation of a pyrene excimer (right).

PAGE 43

43 CHAPTER 2 THEORIES AND INSTRUMENTATIONS OF FLUORESCENCE CORRELATION SPECTROSCOPY Fluorescence Correlation Spectroscopy (F CS) is based on the statistical analysis of fluorescence intensity fluctuations and is used to investigate the dynamic properties of single molecules in solution.82, 83 It was first introduced in the 1970s as a method for measuring molecular diffusion, reaction kine tics, and flow of fluorescent particles.84-87 This method has recently found widespread app lication in the study of processes such as diffusion in the cellular environment.88 Because of the substantial increase in sensitivity of FCS89, allowing single-molecule analysis,90, 91 The small volume elements (<1 fL) in which the measurements are performed make it possible to evaluate molecular processes in the cellular environm ent with rapid temporal and high spatial resolution. FCS is also noninvasive, making it ideal for live cell measurements. Coupled with imaging, FCS can be used on the subcellula r level to dissect biological processes in different compartments morphologica lly; e.g., dendritic spines, which have been previously intractable to biochemical ex periments. FCS measurem ents also allow the determination of both cell-to-ce ll and position-to-posit ion (within the same cell) variability. Many techniques measure average values of a seemingly ‘homogeneous’ ensemble, in which the heterogeneity of a process in a ce ll population or within the same cell is lost. However, all this information, which can be biologically important to understand the extent of variation in cellular responses, wil l be obtained by FCS. This research takes advantage of the high detection sensitivity and non-invasi ve properties of FCS technique, in particular the instrumentati on developments and applications of FCS for live cell mapping and intracellular measurements.

PAGE 44

44 History of FCS The theory of FCS is based on molecula r diffusion, and the origin of the technique can be traced back to the discovery of the Brownian motion of molecules. The first person to describe the mathemati cs behind Brownian motion was Thorvald N. Thiele in 1880 in a paper92 on the method of least squares. Then it was followed independently by Louis Bachelier in his PhD thesis "The theory of speculation" in 1900,93 in which he proposed a stochastic analysi s of the stock and option markets. However, it was Albert Einstein94 (1905) and Marian Smoluchowski95 (1906) who independently brought the soluti on of the problem to the a ttention of physicists, and presented it as a way to indirectly confirm the existence of atoms and molecules. Albert Einstein explained that the movement of parti cles suspended in liquids, which is called Brownian motion, is due to the random thermal motion of the solvent molecules. He also predicted that the displacement of a particle should be, on average, proportional to the square root of its diffusion time. This predict ion was later been proved experimentally by Jean Perrin in 1909, whose studies provided the most compelling evidence for molecular diffusion.96 This was also the first time that scientists analyzed a fluctuating signal to study molecular-scaled properties, t he first example of “molecular fluctuation analysis”. Between the late 1960’s and early 1970’s, t he invention of lasers led to the development of an important technique, named dynamic light scattering (DLS), for molecular-size characterization. This advance opened the door for molecular-scaled analysis.

PAGE 45

45 The concept of Fluorescence Correlation Spectroscopy (FCS) was first introduced by Elson, Magde, and Webb97 as a fluorescence-based analogue of DLS at the beginning of 1972. Compar ed to DLS, there are several key advantages using fluorescence measurements for the analysis of molecular motion. Fi rst, fluorescence is a very sensitive tool, and it allows the spec ific analytes under study to be specifically labeled for differentiation from other molecu les in solution. Second, fluorescence has a larger detection range for molecular sizes compared to DLS, es pecially for small molecules. Finally, fluorescence analysis allows characterization of chemically reacting systems, which generally do not create a large enough index of refraction change for study by DLS. However, spontaneous chemical reactions can create fluctuations in the molecular diffusion properties and/or other fluorescence characteristics that can analyzed via fluorescence detection with much lower LODs. In 1992, Rigler et al.98 first discovered that by combining FCS with single molecule confocal microscopy, it was possi ble to overcome many of the challenges originally encountered with t he earlier versions of FCS, su ch as high detection limits, high scattering background and large excitation vo lumes. This idea was later confirmed by the experiments done by Rigler et al.99 and Zare et al.100 in 1994. They demonstrated that confocal microscopy could be used for directly detection of the fluorescence emitted by individual molecules as they di ffused through the microscopic focal volume of the confocal microscope. This discovery was an important step forward for molecular diffusion studies using FCS, as it allowed analysis of molecular diffusion based on the fluorescent signal from a sma ll number of molecu les, or even a single molecule. This advance helps scientist to observe molecular properties at the single molecular level.

PAGE 46

46 Therefore, during the la st decade, there has been a dramatic rise of interest in the FCSrelated studies, and a number of books and review articles on FCS theories and applications have been published.17, 101, 102 Theories of FCS In this section, the principles of FCS for molecular diffusion studies will be discussed. Specifically, the fl uorescence autocorrelation function will be used to explain the geometry of the focus volume and the physics of molecular diffusion within this focus volume. Molecular Diffusion Obtained by the Autocorrelation Function in FCS As shown in Figure 1, the signal ob tained from conventional fluorescence spectrometry measurements c an be generally described in te rms of two parameters, the constant mean intensity < I > and a fluctuating contribution I (t). Although the mean fluorescent intensity < I > is generally t he information of interest, the fluorescence fluctuations I (t) actually carry the information that pertains to molecular diffusion study by FCS. Specifically, FCS is an analytical method that can meas ure the dynamics of molecular processes from the observations of spontaneous fluorescence fluctuations in a solution at thermal equilibrium. By extracti ng from the fluorescence fluctuation signal, it gives the information about dynamic mole cular events, such as diffusion or conformational changes. This method also fi nds useful applications in the study of thermodynamic and kinetic features of molecular interactions. FCS measurement is accomplished by fo cusing the laser beam onto the sample to create a ƒL-size focus volume and then m onitoring the fluorescence signal within the volume. The geometry of the focus vo lume is illustrated in Figure 2-2.103 The dimension

PAGE 47

47 of the ellipsoid-shaped volume element is defined by the half-axis length ( z) and halfaxis radius ( xy). The principle of use of fluorescence fl uctuations to obtain molecular diffusion information is illustrated in Figure 2-3. When a molecule (represented by a red square in Figure 2-3A) enters the excitation volume element (represented by the blue region), it emits photons which can be det ected by the photon detector in FCS. And depending on the random path of the molecu le, it may remain a while, emitting more photons (maybe even crossing the entire detection volume element ), or, alternately, it may quickly exit the volume element having emitted only a single photon to the detector. This randomized behavior can be recorded and analyzed by FCS. Diffusion of fluorophores into and out of the focus volume alters the lo cal concentration of the fluorophores inside the tiny focus volume, contributing to the spontaneous fluorescence intensity fluctuations (Figure 2-2B). After obtaining the fluorescence fluctuations, FCS compares the fluorescence intensity measured at time t () I t with that of a later time () t () It over the mean fluorescence intensity I yields the normalized intensity autocorrelation function () G : 2()() ()1 ItIt G I (2-1) 104 In order to use this equation for the evaluation of experimentally obtained autocorrelation functions, Equation (2-1 ) has been derived to a single-component solution. The resulting equation is consideri ng only diffusion along the axial dimensions of the laser beam. 91, 100, 105, 106

PAGE 48

48 3 2111 () 1 1D D Z D xyG N (2-2) 104 in which, 3()DG represents the three-dimensional diffusion of fluorescent molecules; N represents the absolute number of fluorescent molecules that occupy the focus volume; D represents the mean diffusi on time of the molecules; z and xy are the half-axis length and width of the focus volume. Figure 2-3D shows that G3D( ) decays with from a maximum at = 0 ms. The maximum amplitude of G( ) is the reciprocal of the av erage number of molecules, N, occupying the observation volume, which occurs when approaches 0. 31 ()DG N (when = 0ms) (2-3) Hence, a higher number of molecules in the observation volume (higher concentration) results in lower correlation am plitude. With a fixed focus volume, N is proportional to the molecular concentration. T herefore, there is a detection range of molecular concentrations for mo lecular diffusion studies by FCS. High concentrations of molecules (i.e. > 500nM) will alleviate the contribution of each moleculeÂ’s diffusion in and out of the focus volume to the final fluore scent fluctuation output, thus, resulting in a low antocorrelation. On the other hand, a suitable range of molecular concentration (pM to ~100 nM) allows a significant contribution of individual molecu lar diffusions in the focus volume to the overall fluorescence fluctuation signal. Hence, pM to 100 nM

PAGE 49

49 represents a suitable range of molecular conc entrations for molecula r diffusion studies in FCS. The mean diffusion time D describes the average time it takes for a molecule to diffuse through the observation volume. It represents a char acteristic fingerprint for specific molecules in specific diffusi on states, because molecules with different molecular weights take different times to diffuse through the same observation volume, and the same molecule also diffuses differently in different states (e .g., free in solution vs. bound to membrane surface). Specifically, the average time it take s for a molecule to diffuse through the observation volume, D depends on the width of the focus volume, xy. The relationship of D to the half-axis radius of the focus volume, xy, is given by: 24 x y DD (2-4) 104 where D is molecular diffusion coefficien t. Hence, measuring the decay rate of () G vs. gives the diffusion coefficient of the anal yte and its molecular size. Larger molecules have smaller diffusion coefficients and t herefore larger mean diffusion times In the case where molecular diffusion takes place only on a two-dimensional surface (i.e., Z >> x y ), 21 1Z D xy 1. Therefore, () G in (2-2) becomes: 211 () 1D D G N (2-5)104

PAGE 50

50 Therefore, Equations (2-2) and (2-5) represent the time-dependent correlation function for 3Dand 2D-diffusion, respectively. Geometry of Focus Volume Obta ined by Free Dye Calibration In order to obtain the mole cule diffusion information ( D and N) for unknown molecules (or molecules with unknown diffusion c oefficient D) using Equation (2-2), the geometry of the focus volume needs to be determined. The dimension of the observation volume element is defined by the half-axis in length ( z) and radius ( xy), as shown in the magnified diagram in Figure 2-2. These dimensions can be obtained by a calibration study using a dye molecule with known diffusion coefficient D. For a dye molecule with known diffusion c oefficient D and known concentration C, the half-axis width of the focus volume can be obtained: 24 4xy D xyDD D (2-6) Due to the relationship between the effe ctive volume of the focus volume (Veff) and the molecular concentration C: eff AN V NC (2-7) Therefore, the half-axis l ength of the focus volume, z, can be obtained: 3/22 3/22 eff effxyZZ x yV V (2-8) A femtoliter-sized ellipsoid shaped confocal volume with half axis radius xy and half axis length z in the m range is generally obtaine d for the geometry of the focus volume in FCS. This femtoliter-sized, opt ically defined observation volume is the

PAGE 51

51 engineering feature that make s FCS a highly sensitive biophysical tool for studying molecular interactions, especially in live cells. Single-Component Diffusion a nd Multi-Component Diffusion Single-component diffusion generally de scribes molecular diffusion of one specific type of molecule in its specific state or conformation in side the laser focus volume. The approach to obtain the molecular diffusion information is similar to that described above. The dimensions of the focu s volume are first obtained from a free dye calibration. Then with the fixed xy and z in Equation (2-2), the diffusion time D of the molecule in the single-com ponent system can be obtained. Besides single-component diffusion, the la rge diffusion time range accessible by FCS makes it possible to analyze the superposit ion of various diffusion (and/or blinking) processes that take place on different time scales in a single FCS measurement. Their respective time scales can be revealed by fi tting the autocorrelation curve to a multicomponent diffusion model. For instance, the model for multi-components XA, XB, XC, Â…, with diffusion times A B C Â…, reads 3 2221111111 () ... 111 111DABC eff ABC ZZZ ABC xyxyxyGrrr N (2-9) where rA, rB, rC, Â…, represent the fr actions of components XA, XB, XC, Â…, respectively, and are obtained from G( ) curve fitting. xy and z represent the geometry of the focus volume shared by components XA, XB, XC, Â….

PAGE 52

52 Specifically, for a two-component diffusi on system, Equation 2-9 is simplified to: 3 2211111 () 11 11DAB eff AB ZZ AB xyxyGrr N (2-10) FCS Instrumentation Set-Up As explained in the theoretical discu ssion, the FCS measurement requires a small laser focus volume to ensure detection of a very small number of molecules to obtain a correlation function with high ampl itude. The method also requires high photon detection efficiency and discrimination from background fluorescence. The schematic diagram of the FCS experimental setup is shown in Figure 2-4. Light from a laser passes through the optic al lenses (i.e., beam expanders) and is reflected by a dichroic mirror into the back of an objective. The objective helps to focus the laser beam onto the sample to form a sm all volume element. A fter excitation of the fluorescent molecules inside the sample, the emitted fluorescent light is collected by the same objective and transmitted through the dich roic mirror. Due to the difference in the excitation and emission wavele ngths, the dichroic mirror separates the emission light from the excitation source. Then, the em ission light passes through some emission filters, and a pinhole to r each a photon detector. The dim ensions of the laser beam focus and the pinhole together define the confocal volume element. The detector signal is fed into a digital signal correlator, which calculates the autocorrelation function of the detected intensity fluctuations.

PAGE 53

53 Specifically, the FCS m easurements in this research were performed on a custom apparatus, which has been devel oped from an Olympus IX70 confocal microscope. The excitation lig ht of a 488 nm line of an Ar+ laser is focused onto the sample through an Olympus 60x water-immersion objective (numerical aperture 1.2). The fluorescence is collected by the same obj ective, separated from the excitation light by a dichroic mirror, then sent onto an avalanche photodiode (APD) (SPCM-AQR-14, Perkin Elmer) by the means of an optical fiber with 50 m-inner diameter after passing through a 515–555 nm bandpass filter. The sample is illuminated with an excitation power of 2 mW at the back aperture of t he objective. The meas urements are performed in a volume element of 0.44 fL with halfaxes xy = 0.22 m and z = 1.56 m Autocorrelation is processed by a hardw are correlator (ALV 5000/EPP, ALV-GmBH, Langen, Germany). FCS for Molecular Diffusion Studies in Solution FCS is a highly sensitive tool for molecula r diffusion studies in solution. There are two direct read-outs from the autocorrelation curves, the absolute number of fluorescent molecules in the focus volume, N, and the mean diffusion time D. Quantitative Studies of Molecular Numbers in the Focus Volume Figure 2-5A shows the autocorrelati on function measured for dye Rhodamine 123 at different concentrations (0.01nM10nM). The increase in the number of molecules N is proportional to the increase in concentration, and the autocorrelation decreases as N increases, as given by Equation 2-3.

PAGE 54

54 Molecular Diffusion Time Figure 2-5B shows the molecular we ight dependence of the autocorrelation function of 5 M fluorescein-dextran complexes. In creased molecular weight makes the molecules diffuse more slowly, and is reflec ted in a shift of the characteristic decay times of the autocorrelation func tions to longer delay times. From these two direct r ead-outs of the autocorrelati on curve, other molecular parameters, such as the molecular concentration C (mo l/L), binding constant Kd (mol-1) and kinetic rate constants kon (mol-1s-1) and koff (s-1), can also be obtained. FCS for Live Cell Analysis FCS has proven to be a highly sensitive t ool for studying molecular diffusion in solution. Moreover, thanks to its sma ll focus volume, low LODÂ’s, and noninvasive detection properties, it has also found wide app lication in live cell analysis, including the study of processes such as diffusion in the cellular environment.88,89,90, 91 FCS has also been applied to measure ligand bi nding to cell surface re ceptors of both tyrosine kinases and G protein-coupled re ceptor (GPCR) super families,107 and also to observe some molecular events in cells.103 There are several advantages of using FCS for live cell analysis: (1) FCS measures biological molecules in the nat ural cell physiological environment. (2) Because there is no need for radioactive labelin g, molecules can be rapidly and directed detected inside live cells. (3) Beyond its capability to obtain molecule diffusion information, FCS also provides deta iled kinetics information about moleuclar interactions (4) Single-cell information and ce ll profiles can be obtai ned. This capability is not possible with other strategies, such as flow cytometry. (5 ) The femtoliter-sized

PAGE 55

55 observation volume created by FCS permits the detection down to a single fluorescent molecule, making this technique a highly sensitive biophysical tool for live cells, especially for studying weak molecular interactions. In sum, the FCS approach can effectiv ely improve detection sensitivity and may have wide application for molecular rec ognition and interaction studies. The next chapter describes the applicat ion of FCS for cell membr ane receptor density mapping on live cell surfaces.

PAGE 56

56 Figure 2-1 Fluorescent signal obtained from conventional fluorescence spectrometry. Figure 2-2103 Left: Gaussian laser excita tion profile. Right: Focus volume of observation. The dimension of the observation volume element is defined by the half-axis length ( z) and width ( xy= x= y).

PAGE 57

57 Figure 2-3 The working principle to obt ain molecular diffusion in FCS study. (ms)

PAGE 58

58 Figure 2-4 FCS experimental set-up.101 Figure 2-5 FCS for molecular diffusion studies in solution. (A) Autocorrelation curves with different concentrations of dye R hodamine 123. (B) Autocorrelation curves for fluorescein-dextran complexes of different molecular weights. Dichroic mirror Pinhole

PAGE 59

59 CHAPTER 3 MAPPING RECEPTOR DENSITY ON LIVE CELLS USING FLUORESCENCE CORRELATION SPECTROSCOPY Introduction Biological membranes are the sites wher e different element s of the cellular machinery are brought together; therefore, they are centra l to the very phenomenon of life108. Significantly different from bulk wate r, but integrated with it, biological membranes create an environment in which many complex enzymatic reactions and bioelectrical and biochemica l signaling processes occu r. Examples include the conversion of metabolic energies into os motic, electrical, and mechanical work, transportation of materials between cellu lar compartments, and the processing of information. In a broad sense, many cellular activities involve membrane-based ligandreceptor interactions109, which are mediated by membr ane-associated proteins that are incorporated into the structures of the lipid bilayers. Consequently, biological membranes are the primary target receptor s for many drugs representing different therapeutic categories. Ther efore, while knowledge of molecular mechanisms underlying ligand-receptor interaction has t heoretical significance, there are also practical implications for the discovery, design, and screening of novel therapeutic agents. The binding of extracellular ligands to receptors also allows living cells to constantly monitor and respond to changes in their environment. Therefore, the control of receptor distribution and trafficking in a spatially and temporally ordered manner is required to modulate cell behaviors, which r ange from cell division to differentiation. Furthermore, the ability to obtain quantit ative information about ligand-receptor interactions and receptor distribution over t he cell surface will be of broad significance in our understanding of cell membrane receptor characteristics and expression level,

PAGE 60

60 spatial distribution, as well as clusteri ng and molecular changes, on the molecular level in living biological specimens. These data would also, in turn, provide an important database for drug discovery. However, so far, only a limited number of approaches have been established for estimating membrane receptor densities. Li quid scintillation c ounting, a standard laboratory method in life science, measures radiation from beta-em itting nuclides, and thus requires the use of r adioactively labeled ligand110, 111. Fluorescence subtraction, the most conventional fluorescence approach for density estimation, involves several washing steps to remove unbound ligands, and the receptor density is determined by fluorescence measurement of supernatant c ontaining free ligands. Concentration of free ligands is estimated by interpolation from a standar d linear calibration curve 112. However, the half-life of the receptor-ligand complex is often shorter than, or equal to, the time required for the separation of free and bound ligands. Specific interactions between certain ligands (e.g., peptides, hormones, natural pro ducts) and their different receptor subtypes are, ther efore, often overlooked by the conventional fluorescence subtraction method. In addition, the analys is may also be compromised by high background levels of other membrane proteins that are expressed endogenously on the membrane. Especially, in certain cases, the receptor number s per cell are few; therefore, no specific bindi ng is detected because of high ba ckground. Alternatively, the receptor of interest could be separated and then purified from over-expressing cells or tissue. This biochemical purification, how ever, typically requires exchange of the physiological lipid/lipid-protein environment by a detergent micelle, which may modify the binding properties of the receptor. In an ideal recept or preparation format, a high

PAGE 61

61 enrichment of the receptor of interest would be combined with t he receptor as it is found, i.e., naturally integrated within its own lipid/lipid-protein environment. To address the above problems, we report here an effective approach for direct measurement of membrane rec eptor density in its natural physiological environment on the cell surface using a highly sensitive technique, Fluorescence Correlation Spectroscopy (FCS). FCS is an ideal biophy sical and bioanalytical tool for studying concentrations, propagation, interactions and internal dynamics of molecules at nanomolar concentrations in living cells 17, 102, 113, 114. It is capable of analyzing even minute fluorescence-intensity fluctuations about the equilibrium of a small ensemble of molecules. These specific fluctuations in frequency act like the “fingerprint” of a molecular species detected when entering and l eaving the instrument’s femtoliter-sized, optically defined observation volume, which is created by a focused laser beam. This small laser volume element (<1ƒL), about 250 nm in diameter, allows the detection of single molecules16 as well as the measurement of molecular properties at specific coordinates on the cell membrane or inside cells115. However, while this highly sensitive technique has been developed and adapted to cellular measurements ever since its breakthrough in the early 1990s, there has so far been no single report on the application of receptor densit y studies on live cell membranes. With its single-molecule detection sensitivity, FCS allows the detection of receptor binding sites at the molecular level in a cell membrane’s nativ e environment on the cell surface115-117. Thus, it permits the identification of target receptors as well as detailed ligand-receptor interaction kinetics. By measuring ligand-receptor intera ctions on individual cells, FCS can obtain binding affinity and receptor density information from just a small number of cells (less

PAGE 62

62 than 50), which was not possible before by c onventional binding techniques, such as flow cytometry. This also makes our approach crit ical in certain situations where it is not feasible to obtain millions of cells. Also, t he practicality of the FC S technique arises from the elimination of washing steps ot herwise needed to separ ate unbound from bound ligands16. Change in diffusion times for ligands upon binding to membrane receptors enables FCS to differentiate between the receptorÂ’s bound and free ligand fractions. In order to mark the memb rane receptors of interest for density study on FCS, fluorophore-labeled nucleic acid aptamers were chosen for receptor recognition. As probes for molecular recognition, the effi cacy of DNA aptamers has been successfully demonstrated in many unique applications9, 118-122. Aptamers are single-strand oligonucleotides derived from a selection process called SELEX (Systematic Evolution of Ligands by Exponential enrichment). In our study, the aptamers were selected from whole intact biological live cell s through a process called cell-SELEX9, 121, 122 and these aptamers are capable of binding to their targ et molecules on the cell membrane surface with high affinity and specificity123. Compared with molecular pr obes currently available for receptor recognition, such as monocl onal antibodies, aptamers offer significant advantages over existing antibody-based recogni tion procedures in that they offer higher binding affinity (higher retention/reduced dissociation) and specificity to the target (ability to determine variations on the protei n target down to single amino acid changes), higher selectivity against mutated pr otein epitopes, and potentially reduced immunogenicity and increased tumor penetration associated with their size. They also possess lower molecular weight and can be chemically synthesized and easily modified 9, 118-120. These features make aptamers promisi ng probes for recognizing target-specific

PAGE 63

63 receptors and identifying bind ing sites on the cell membrane surface, which could also be recognized as potential biomarkers124. Nowadays, more and more aptamers have been generated as specific probes for mole cular signatures on the cell surface. However, there are only limited ways to fully study and use them for biological studies. Here, we combine aptamer recognition with the highly sensitive FCS tool to perform a ligand-receptor interaction study on the ce ll membrane. This technique yields the information required to estimate receptor density and distribution, thus extending the potential applications of apt amers generated from cell-SELEX, as well as FCS as an effective biophysical tool fo r cell membrane surface study Using an in vitro cell-SELEX procedure9, 121, 122, we selected an aptamer, sgc8, towards T-cell ALL CCRF-CEM cell line. We also successfully elucidated its target protein, human protein tyrosine kinase-7 (PTK7) by using this newly selected aptamer125. PTK7 has been discovered to be highly expressed on the cell membrane in a series of leukemia cell lines124. It is recognized as a potential cancer biomarker, having a role characteristic of tumo rs, i.e., signal amp lifier or modulator126. In order to demonstrate FCS as an effective approach for m apping receptor dens ities on live cells, we chose two different cell types with differ ent expression levels of PTK7 on the cell membrane as proof of principle. This is the first study using FCS to estimate the density of membrane receptor PTK7 on different cell types through aptamer/receptor interactions. Experimental Section Cell Lines CCRF-CEM (human leukaemia cells), HeLa (cervix adenocarcinoma) K-562 cells (CCL-243, human chronic myelogenous leuk emia) and RBL-2H3 (Rat Basophilic

PAGE 64

64 Leukaemia cells) cell lines were obtained fr om American Type Cu lture Collection. CCRF-CEM and HeLa cells were cultured in RPMI 1640 medium (American Type Culture Collection), with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) and 0.5 mg/ml Penicillin-Streptomycin (American Ty pe Culture Collection, Manassas, VA) at 37oC under a 5% CO2 atmosphere. RBL-2H3 cells were cultured in Dulbecco's modified Eagle's medium (American Type Cu lture Collection), with 15% fetal bovine serum (Invitrogen, Carlsbad, CA) at 37oC under a 5% CO2 atmosphere. All the cells were grown in 8-well Nunc chambers (Nalge Nunc Inc., IL, USA) to a density of ~ 1000 cells/well. The cells were washed before and after aptamer incubation with DulbeccoÂ’s phosphate buffer (Sigma) with 5 mM MgCl2. Aptamer Synthesis Aptamer sgc8 (5Â’-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA-3Â’), aptamer KK1H08 (5Â’-ATC CA G AGT GAC GCA GCA GAT CAG TCT ATC TTC TCCTGA TGG GTT CCT AGT TAT A GG TGA AGC TGG ACA CGG TGG CTT AGT-3Â’) were synthesized on ABI3400 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA). The aptamer was labeled with 5Â’-FITC modifier. A DNA library containing a randomized sequence of 41 nucl eotides was used as a control. The completed sequences were then deprotect ed in AMA (ammonium hydroxide/40% aqueous methylamine 1:1) at 65C for 20 mi nutes and further purified with reverse phase HPLC (ProStar, Varian, Walnut Cr eek, CA) on a C-18 column. A Cary Bio-300 UV spectrometer (Varian, Walnut Creek, CA) was used to measure absorbance to quantify the manufactu red sequence. Antibody FITC-labeled IgE antibody was purchased from Miltenyi Biotec Company.

PAGE 65

65 Trypsin Treatment of Cells Cells were first washed with washing buffe r (500mL) to remove the FBS (Fetal Bovine Serum) in the medium or the binding buffer, which might quench the function of trypsin, then incubated with 10 diluted trypsin (500mL, 0.005%)/EDTA (0.53 mm) in HBSS at 37C. Fluorescence intensity and autocorrelation of bound aptamers on a single cell membrane were monito red for 90 minutes using FCS. FCS Instrumental Set-Up As shown in Figure 3-1, a light beam of 20 mW coming from an Ar+ laser is routed through two mirrors and expanded by a beam expander, which is made up of a planconcave lens and a plan-convex lens. The intens ity of the excitation light is attenuated by a neutral density filter. The expanded beam is reflected by two other mirrors and enters the back port of an invert ed microscope, where the laser beam is reflected by a dichroic mirror and then focused into the sample through a microscope objective lens. The fluorescence from the sample is collect ed by the same microscope objective lens. The filtered fluorescence is focused by the tube lens of the microscope and exits through the side port with a focus close to t he body of the microscope. The signal is then filtered with a bandpass emi ssion filter and focused onto a multimode fiber, which works as a pinhole. The fiber is coupled to a single photon counting module (SPCM) which detects the fluorescent signal. The dete ctor sends the signal to the input channel of the hardware correlator. We detected the laser beam intensity at the objective outlet to be 1.2 mW, sufficient excitation power for single-photon experiments17, 113 Results and Discussion FCS Measurement and Analysis As we discussed in Chapter 2, Fluorescence Correlation Spectroscopy

PAGE 66

66 measurement is accomplished by focusing an excitation laser beam onto the sample and then monitoring the fluorescence fluctuations derived from the fo cal region of the laser beam (Figure 3-1). Diffusion of fluorophores into and out of the focal volume alters the local concentration of the fluorophores, thus giving rise to spontaneous fluorescence intensity fluctuations. Correlating the fluores cence intensity fluctuations at time t, I(t), with that of a later time (t+ ), I(t+ ), over the mean fluorescence intensity , yields the normalized intensity autocorrelation function G ( ): 2()() ()1 ItIt G I (2-1) 104 Then, we are able to monitor the real-tim e aptamer/receptor interactions on the cell membrane by tracking the diffusion of t he fluorescent-labeled aptamers into and out of the focal volume. More spec ifically, a derivative of G ( ), which assumes a single component solution and only considers diffu sion along the 3D axial dimensions ( xy and z) of the laser beam, yields equation (2-2): 3 2111 () 1 1D D Z D xyG N (2-2) 104 The amplitude of G ( ) depends on the absolute numbe r of molecules, N, occupying the observation volume. A higher number of molecules in the observation volume (higher concentration) results in a lower correlation amplitude. Mean diffusion time D describes the average time it takes for a molecule to diffuse through the observation volume. It works as a characteri stic fingerprint for specific molecules in specific diffusion states, as molecules with different molecular we ights take different

PAGE 67

67 times to diffuse through the same observa tion volume, and the same molecule also diffuses differently in different states, e. g., free in solution vs. bound to membrane surface. The dimension of the observation volume element is defined by the half-axis in length ( z) and width ( xy), as shown in the magnified di agram in Figure 3-1. A 0.4-ƒL ellipsoid shaped confocal volume with half axes xy=0.22 m and z=1.56 m was obtained from free dye calibration of R hodamine 123, a photostable dye with known diffusion coefficient (D=310-10m2/s). This femtoliter-sized, optically defined observation volume is the engineering feat ure that makes FCS a highly s ensitive biophysical tool for studying molecular interactions on cell membranes. Binding of Aptamer to Membrane Receptor For the aptamer/receptor binding study, light coming from a laser beam is focused on the cell membrane. If we cons ider that the thickness of t he lipid bilayer (~ 4 nm) is 3 orders of magnitude smaller t han the typical axial length ( z=1.56 m) of the FCS confocal volume in its geometry, for the fluorophore-labeled aptamer binding to a cell membrane, the diffusion at the membrane takes place only at a 2D surface which will be that surface section which is equal to the confocal volume. Then, the diffusion of bound aptamers on the membrane c an be treated as two-dimens ional diffusions and G2D ( ) becomes: 211 () 1D D G N (2-5) 104 Equations (2-2) and (2-5) represent the time-dependent correlation function for 3D and 2-D diffusion of aptamers, respectively. When the vo lume element with half axes xy and z is projected onto the cell surface, not only bound aptamer diffusing at the

PAGE 68

68 cell surface, but also free aptamer diffusi ng above the cell surface, will be seen. When a portion of the aptamers in the confocal volu me binds to the membrane receptors, the bound aptamers are restricted to the cell surface and diffuse sl ower. Thus, the diffusion time of the bound aptamers in creases compared to the fr ee ones. Diffusions of free aptamers above the cell surface are descri bed by the 3-D diffusion function, while diffusions of bound aptamers on the cell memb rane are represented by the 2-D diffusion function. Thus, if the aptamer /receptor complex is stable on the time scale during its transit through the detection r egion, then the overall autocorrelation function describing the activities that occur in the confocal vo lume will represent a linear combination of the autocorrelation functions of free and bound aptamers: 21111 ()1 11 1freebound DD free Z D xyGrr N (3-1) 104 where N is the total absolute number of fl uorescent molecules inside the focus, D free is the diffusion time for the unbound labeled aptamer, and D bound is the diffusion time for the bound labeled aptamer. (1-r) is the fraction of the unbound aptamer diffusing with D free, and r is the fraction of the bound aptamer diffusing with D bound. Autocorrelation functions of FITC-label ed aptamers in solution and bound to the cell membranes are respectively shown in Figur e 3-2. Here we s ee that the binding of aptamers to receptor results in an increase of the diffusion time, where D = 0.827ms ( ), when compared to the free ones, where D= 0.235ms ( ), thus enabling FCS to

PAGE 69

69 differentiate bound aptamers from free ones. The corresponding fluorescence intensity fluctuations are shown on the bottom in Fi gure 3-2. Stable fluorescence fluctuations show that no photobleaching effects had occurr ed during the entire detection time in the detection volume. Control experiments with a randomized sequence of 41 nucleotides (Library, Figure 3-2B), which have been shown as the negative binding control to receptor PTK7 in cell-SELEX9, show similar diffusion times fo r aptamers free in solution and on cell membranes. It indicates only spec ific interactions between aptamers and membrane receptors will give the response of increased diffusion time on cell surface. Specificity of Binding by the Aptamers The specificity of aptamer/receptor intera ctions, as demonstrated by the specific labeling of target receptors by aptamers, provides the basis for studying receptor density using FCS. Results shown in Figur e 3-3 demonstrate the s pecific binding of aptamer sgc8 in different concentrations to its membrane-bound receptors in different cell types. The formation of complexes betw een the receptor and aptamer is identified by the change in the autocorrelation function in the positive target cells, which is not shown in the negative control cells that lack PTK7 expressions on the cell membrane9, 127. Binding Affinity Studying the density and distri bution of membrane receptor s at the molecular level in living biological specimens has been challenging. The above experiments have shown the efficacy of FCS in terms of ligand binding specificity to membrane receptors. However, since FCS is also capable of measuring detailed molecular interaction between ligands and receptors, this property wa s applied to study the binding affinities of aptamer-receptor interacti ons in their native physiolog ical environment on the cell

PAGE 70

70 membranes. The resulting data, in turn, pr ovide us with the apt amer concentration required for saturating all the receptor bi nding sites in the density study. For this purpose, we carried out binding experiment s on individual cells with different concentrations of aptamers. While keeping t he same amount of HeLa cells in different wells in the culture dish, we varied the concentration of FITC-labeled sgc8 aptamers from 0.1 nM to 3.0 nM. We focused on the cell membrane with the 0.4ƒL ellipsoid shaped confocal volume from a 1.2mW laser beam, the geometry of which had been determined from the free dye calibration with half axes xy=0.22 m and z=1.56 m (Figure 3-2). As a result of the difference in diffusion time of bound versus free aptamers, we were able to assess the per centage of the bound aptame rs (r) by fitting the autocorrelation by equation (3-1). The fitting of the autocorrelation curve also yielded the absolute number of total aptamers (N) inside the confocal volume, which was interpreted as the reciprocal of t he autocorrelation amplitude (in equation 4, when approaches 0, G(0)=1/N). As the total aptamer concentrati on increases, more and more aptamers bind to the membr ane receptors, so the number of bound aptamers inside the confocal volume (total aptamer numberbound fraction = Nr) that was obtained correspondingly increases befor e it reaches saturation. We plotted each bound aptamer number (Nr) versus the corresponding tota l aptamer concentrations used. As can be seen in Figure 3-4 and Table 3-1, increasing concentrations of FITC-sgc8 clearly leads to an increased bound aptamer num ber (table on the right). The femtoliter-sized (0.4ƒL) observation volume created by FCS also a llowed the detection of fluorescent aptamers down to about two, while a low concentration (0.1nM) of aptamers wa s used to incubate with cells, which substantially validates FCS as a highly sensitive biophysical tool for

PAGE 71

71 studying molecular interactions on live cell s. In the aptamer sgc8 binding curve on human cervical cancer HeLa cells, a pico molar range dissociation constant (790150 pM) was determined via non-linear regression. This Kd result is in good agreement with the flow cytometry result, where a Kd value of 810 pM was obtained9, which further confirms the validity of using FCS to det ect and quantify aptamer/rec eptor interactions. The binding curve also indicates that a 3. 0 nM aptamer concentra tion of FITC-sgc8 is sufficient for saturating receptor PTK7 binding sites on HeLa cells using FCS. However, only a small amount of cells (less than 60 in total) was needed here to obtain the binding curve for aptamer-receptor interacti ons using FCS, while millions of cells are required for flow cytometry to obtain a si milar binding curve. This makes FCS an invaluable technique for obtaining binding information under conditions where only a limited amount of ce lls can be obtained. Density Study Receptor density was obtained from each individual cell surface through aptamer/receptor binding. Ex periments were performed at 4 C to inhibit the receptormediated endocytosis of aptamers [17] and thereby ensure the efficient saturation of receptor binding with aptamers on the membrane. When the volume element is projected onto a single cell surface, in t he presence of excess FITC-labeled aptamers, both the bound FITC-labeled aptamer diffusi ng at the cell surface and the unbound FITC-labeled aptamer diffusing above the ce ll surface are observed. Thus, the parameter N, which could be obt ained from the fitting of autoco rrelation function, as we discussed earlier, characterizes the total absolute number of apt amers inside the confocal volume, which is the sum of bound aptamers on t he membrane and free aptamers above the cell surface. While r gives the bound fraction of aptamers, (Nr)

PAGE 72

72 stands for the total number of bound aptamers. Assume that the waist of the confocal volume ellipsoid is situated on the cell me mbrane such that t he highest fluorescence intensity is given. In this case, the surfac e area on the membrane covered by the focus, where all the bound aptamers are located, will be a circular area with a radius equal to xy (shown in the magnified diagram of Figure 3-1). Ther efore, the receptor density estimated in the confocal volume can be expressed as the total number of bound aptamers divided by the area covered: 2()xyNr Density (3-2) Fifty HeLa cells were investigated us ing aptamer sgc8 to carry out the experimental measurement s and determine the PTK7 re ceptor density on the cell surface per unit area by applying the above formula. The 0.4ƒL ellipsoid shaped confocal volume from a 1. 2mW laser beam with half axes xy=0.22 m and z=1.56 m was projected onto each cell membrane. The ci rcular area covered by the focus volume could be estimated to be ( xy)2 = (0.22m)2 = 0.15m2. Within this tiny covered area, the number of specific receptors that occupied the area could be interpreted by the amount of specifically tagged aptamers at the saturati on condition. As determined from the earlier binding affinity study, a 3. 0nM aptamer concentrati on of FITC-sgc8 was used to saturate all the PTK7 binding sites on HeLa cells. Autocorrelation curves were obtained from the membranes of fifty indi vidual cells. An average of Nr=84 bound aptamer sgc8 was obtained with a variation of 14, which indicate d that there were an average of 84 bound aptamers that occupied the 0.15m2 covered area in the confocal volume. By applying these parameters into fo rmula (3-2), receptor density is obtained. A Gauss-shaped distribution of the receptor d ensity is shown in Figure 3-5. A mean

PAGE 73

73 density of around 550 receptors/m2 is obtained with a variation of about 90 receptors/m2. A key advantage of the FCS density approach is s uggested by this result. That is, the method gives not only the mean density, but, because of the ease of using a single-cell approach to measure mu ltiple cells, it also give s the statistical distribution among different cells. In order to compare the expr ession levels of the same receptors on the membrane of different cell types, aptamer sgc8 is appl ied to label the receptor PTK7 on both human cervical cancer HeLa cells and CCRF-CE M leukemia cells. A higher receptor density of about 1300190 receptor/m2 is found on CEM cells than on HeLa cells (55090 receptor/m2), which indicates the higher receptor PTK7 expression level on the CEM cell surface. Alternatively, a hi gher receptor density per unit area also indicates that the distance between two adj acent receptors is smaller for CEM cells (~28nm) than for HeLa cells (~43nm). However, in contrast to HeLa cells, a broader distribution of density was obtai ned for CEM, as shown in Figure 3-6. By these results, this expression level study demonstrates that the use of FCS provides a potential tool for drug delivery studies because it offers a m eans of interpreting the loading ability for drugs onto the cell surface through binding to membrane proteins. Specifically and simply, the loading factor im pacts the quantity of a given dr ug that can be internalized by receptor-mediated endocytosis. Obviously, t hen, different receptor expression levels on different cell types will yield distinct l oading efficiency for the same drug into cells. In order to demonstrate the robustness of using FCS to measure receptor density, especially in the case of pr esence of a high background fr om free labeled ligands, we also chose a weaker binding li gand, aptamer KK1H08, to prove the capability of FCS to

PAGE 74

74 monitor bound ligands from high concentrations of free ligands in order to determine receptor densities on cell membrane. KK1 HO8 was selected for K-562 cells and showed low binding affi nity with Kd=29641nM112. It requires a high concentration of total aptamers (0.5 M) to saturate all the aptamer bi nding sites. Our FCS studies prove that it can differentiate bound aptamers from high backgr ound of free ones. A Kd=27129 nM was obtained and used to dete rmine the receptor density on the cell membrane. (Figure 3-7). Fu rther experiments using ex cess labeled aptamer sgc8 (250nM) incubated with HeLa cells tested by FCS also proves the capability of FCS to differentiate bound aptamers from the high background of free aptamers (bound fraction=161%). In addition, the recept or density determined under this condition (45929 receptor/m2) is comparable to the one determined previously under low aptamer concentration s (55090 receptor/m2), which further demonstrates the robustness of this receptor density determination using FCS. Competition Studies and Trypsin Experiments In order to make certain that the method dev eloped here is effective, control experiments were conducted to further confirm the calculated results. To verify that the receptor density is obtained through the specif ic labeling of receptors using fluorophorelabeled aptamers, a competit ive displacement with non-l abeled aptamer was examined (Figure 3-8A, B). In this experiment, cells were first incubated with 3.0nM of labeled aptamers, and after 40 min, a 1000-fold molar excess of non-labeled aptamer was added to compete against the labeled aptamer in binding to the target receptor. Since the number of receptors on the surface is limited, the majority should be bound by the non-labeled aptamer given the huge excess of the non-labe led population. If the labeled aptamers are displaced by unlabeled aptamers, which si gnifies competition between

PAGE 75

75 them, it would indicate that the aptamers are in teracting with their target receptor specifically. From Figure 38, it can be observed that the addition of non-labeled aptamers did result in reductions of both fluorescence count ra te (fluorescence intensity) on cell membrane (Figure 3-8A) and membr ane-bound aptamer densit y (Figure 3-8B). After about 90 minutes, the aptamer binding was almost entirely displaced. This proves that the labeled aptamers recognized the target receptor specifically and did bind to them on the cell membrane. Control exper iments (blue circles) with a randomized sequence of 41 nucleotides, which have been s hown as the negative binding control to receptor PTK7 in cell-SELEX9, were conducted under t he same conditions. No significant decrease of density and count rates were observed within the first 60 minutes compared with competition studi es using non-labled sgc8. The small decrease after 60 minutes is believed coming from koff from the labeled sgc8 itse lf with time increased, but not from specific binding co mpetition. Further experimen ts with proteinase trypsin treatment (Figure 3-8C) also indicated tha t, by removing the tar get protein receptors from the cell surface wit h proteinase and reducing their densit y artificially, the binding of aptamers on the membrane was decreased. This in turn, resulted in the decrease of membrane-bound aptamer density obtained from FCS detection. Since the aptamer must have been binding to its targe t, and not simply undergoing a nonspecific interaction with the cell membr ane surface, this evidence se rves to further validate the results obtained from the above density study. Overall, thes e results support the use of FCS as a comprehensive tool for detailed rec eptor/ligand interaction studies, such as the determination of binding affinity (Kd), dissociation rate (koff) and other kinetic parameters, which other density st udy approaches cannot accomplish.

PAGE 76

76 Control Cell Line for Method Evaluation To further confirm the reliability of the approach for density estimation, a well studied system of IgE receptors on the surface of RBL-2H3 cell line128 was also conducted using FCS. A dissociation constant of 0.810.21 nM was obtained, and a 2.5 nM IgE concentration was indicated to be suffi cient for saturating IgE receptor binding sites on the FCS (Figure 3-9). Thirty RBL-2H 3 cells were investigated to determine the IgE receptor density on the cell surface per unit area by applying equation (3-2). Similar to our aptamer sgc8 binding studies, a G auss-shaped distribution of receptor density was obtained (Figure 3-10). A m ean density of around 1200 receptor/m2 was obtained with a variation of about 60 receptor/m2. We determined the surface area of the RBL2H3 cells from confocal images of 80 individual cells (240 20 m2/cell), and the IgE receptor density could be estimated to be (2.88105 1200) receptor/cell. As shown in Table 3-2, the Kd value of IgE/receptor complexes obtained in FCS gave results comparable to the Fluorescenc e Quenching method, as reported in the literature 128, and a density with the same order of magnitude was also obtained for the IgE receptor, which greatly supports the validity of the FCS approach for density estimation. Conclusion In conclusion, we have reported the us e of a highly sensitive technique, Fluorescence Correlation Spectroscopy, fo r mapping receptor densities on live cell membranes by introducing the fluorescently marked aptamer molecules, which target specific membrane receptors with high affinity and selectivity. Full saturation of aptamer binding to the cell surface is obtained at pi comolar concentrations, which indicates the high-affinity binding of the aptamer/receptor complexes (Kd=790150pM). The binding

PAGE 77

77 properties of aptamer/receptor complexes we re investigated and fu rther applied for the determination of receptor densities. Human protein tyrosine kinase-7 was found to have a larger density in CCRF-CEM human l eukemia cells (1300190 receptor/m2) than HeLa cervical cancer cells (55090 receptor/m2). Competition studies and control cell line experiments proved the va lidity of the density estimati on approach. This strongly indicates the intrinsic advantages of the FCS approach for recept or density studies when compared with the conventional methods: (1) FCS measures receptor density in the natural cell surface physiological environ ment. (2) There is no need for radioactively labeled ligands or washing steps to remove unbound ligands, which results in a fast and direct detection for receptor density. (3) Beyond its application for density estimation, FCS also provides detailed kinetics info rmation about ligand-receptor interactions, which may not be available from other density study approaches. (4) Single-cell information and density distribution pattern s can be obtained, which is almost impossible by other strategies. (5) Only a small number of cells are needed for FCS to perform ligand-receptor interaction studies, as well as density estimation, compared to other methods, which makes FCS a critical t ool in the situations where only a limited amount of cells can be obtained. (6) Rec eptor expression leve ls and distribution patterns on different cell types can be easily obt ained from FCS detection. This provides basic information for drug loading effici ency on cell membranes and serves as a potential tool for drug delivery studies. (7) For receptor density estimation, FCS has an advantage over conventional approaches by eliminating the need for standard curve calibration. (8) The femtoliter-sized obser vation volume created by FCS permits the detection of fluorescent molecules down to two and therefore appear s to be a highly

PAGE 78

78 sensitive biophysical tool on live cells, es pecially for studying weak molecular interactions. In sum, our FCS density es timation approach can effectively improve detection sensitivity and may have wide ap plication for molecular recognition and interaction studies, as well as density estimations for particl es and other membrane receptors. Objective lens Dichroic Filter PinholeAPD Correlator Computer Mirror 2 Mirror 1 Mirror 3 Mirror 4 Neutral Density Filter Laser Cell Cell Membrane Focus • FITC labeled Aptamer • Transmembrane receptor • Membrane • Receptor Extracellular Domain z z xy xy Figure 3-1 Scheme of FCS inst rumental set-up and the illustr ation of aptamer/receptor binding events on the cell membrane insi de the focus. The magnified diagram illustrates the geometry of the confocal volume with half-axis in length ( z) and width ( xy).

PAGE 79

79 1E-30.010.1110100100010000 0.00 0.02 0.04 G(t)t (msec)Dfree= 0.235msDbound = 0.827ms 10 nMfree sgc8 10 nMsgc8+ HeLaCells 051015202530 52 54 56 58 60 051015202530 52 54 56 58 60 Count RateDetection Time (s) A 1E-30.010.1110100100010000100000 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 10nM Library + HeLa1E-30.010.1110100100010000100000 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 G(t)t ( msec ) 10nM free LibraryBDfree= 0.435msDbound = 0.457ms

PAGE 80

80 Figure 3-2 Autocorrelation functions of aptamer sgc8 and control free in solution and on a cell membrane. (A) Top: Autocorrelation functions of aptamer sgc8 (10nM) free in solution ( ) and aptamer sgc8 (10nM) incubated with HeLa cells and bound to membrane receptor PTK7 on the cell surface ( ). The diffusion times ( D) of aptamer sgc8 is increased from 0.235m s (free) to 0.827ms (bound). Bottom: Fluorescence intensity (or count rate) fluc tuation curves during the detection time (30 seconds) for aptamer sgc8 (10nM) di ffusion free in solution (blue line) and bound to membrane surface (red line). Stable fluorescence fluctuations show no photobleaching in the detection volume during the entire detection time. (B) Control binding experiments with a r andomized sequence of 41 nucleotides (Library) were conducted under the same conditions. Both the free Library and Library incubated with HeLa cells show si milar diffusion time s (0.435ms (free) and 0.457ms (bound)), which indicates no binding interactions between Library and cell membrane receptors.

PAGE 81

81 1E-30.010.11101001000100000.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 G(t)t (msec) 25nM aptamer with CEM cells 30nM aptamer with CEM cells 60nM aptamer with CEM cells 100nM aptamer with CEM cells 200nM aptamer with CEM cellsPositive Cells 1E-30.010.1110100100010000 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 G(t)t (msec) 25nM aptamer with Ramos cells 30nM aptamer with Ramos cells 60nM aptamer with Ramos cells 100nM aptamer with Ramos cells 200nM aptamer with Ramos cellsNegative Cells Figure 3-3 Selective binding of aptamers sgc8 to positive cells rather than negative controls. Different concentrations of aptamers were incubated with positive and negative cells for 40 min at 4C before FCS measurements. Flow cytometry results have proved Ramos cells as the negative control cell line for sgc8 binding as it lacks PTK7 expression on the membrane. For each FCS measurement, 1.2mW laser intensity was used at the obj ective outlet. Sample was exposed to laser for 30 seconds to obtain the aut ocorrelation curve. Stable fluorescence intensity was observed during the 30s detection time which indicates no photobleaching. For positive cells, increased concentrations of aptamers lead to decreased autocorrelation amplitudes.

PAGE 82

82 KD=790 150pM0.00.51.01.52.02.53.0 0 20 40 60 80 100 Bound AptamerNumberTotal Aptamer(nM) Total aptamer (nM) Bound aptamer number in confocal volume (Nr) 0.1 0.3 0.5 0.7 1.0 1.2 1.4 1.6 2.0 2.5 3.0 1.91.5 16.84.6 34.14.4 46.65.3 59.54.5 67.92.7 74.43.7 82.13.0 85.51.3 85.80.9 86.00.7 Figure 3-4 Binding of FITC-sgc8 to ce ll membrane on human cervical HeLa cells. Number of the membr ane-bound labeled aptamer (N r) was obtained as a function of the total aptamer concentrations in the bindi ng buffer. HeLa cells were incubated with buffer containi ng different concentrations of labeled aptamers for 40 min at 4C. Each data point r epresents the mean of five separate measurements. For each measurement, 1. 2mW laser intensity was used at the objective outlet. The confocal volume was determined by free dye calibration to be 0.4 ƒL with half axes xy=0.22 m and z=1.56 m. Sample was exposed to laser for 30 seconds for obtaining the aut ocorrelation curve. Stable fluorescence intensity was observed during the 30s detection time which indicates no photobleaching. Different numbers of bound apt amers in the confocal volume at

PAGE 83

83 different aptamer incubation concentrations were calculated from the fitting of autocorrelation curve and listed in the table on the right. The equilibrium dissociation constants (Kd) of the aptamer-cell interact ion were obtained by fitting the dependence of fluorescence intensity of specific binding on the concentration of the aptamers to the equation Y=BmaxX/(Kd + X), using Microcal Origin 6.0. Rece p tor Densit y ( Number/ m2 ) Number of Cells Figure 3-5 PTK7 receptor densit y distribution for HeLa cells. HeLa cells were incubated with 3.0nM sgc8 aptamers for 40 min at 4 C to saturate all the receptor binding sites before FCS measurement. For each FCS measurement, 1.2mW laser intensity was used at the objective out let. The confocal volume was determined by free dye calibration to be 0.4 ƒL with half axes xy=0.22 m and z=1.56 m. Sample was exposed to laser for 30 seconds to obtain the autocorrelation curve. Stable fluorescence intensity was observed during the 30s detection time which indicates no photobleaching. Receptor dens ity was obtained from individual cell detection using formula (5). 50 cells we re studied and the density distribution is shown above.

PAGE 84

84 HeLaCEM0 200 400 600 800 1000 1200 1400 1600 Receptor Density ( / um2)Cell Line Figure 3-6 Comparisons of PTK7 receptor density for HeLa and CEM. PTK7 receptor densities were obtained from the FCS m easurements on 50 individual cells for each cell line and calculated using formula (3-5). A mean PTK7 receptor density on CEM cells is found to be about 1300 receptor/m2 with a variation of about 190 receptor/m2. HeLa cells have a mean receptor density around 550 receptor/m2 with a variation of about 90 receptor/m2. 0100200300400500 0 100 200 300 400 500 600 700 Aptamer concentration (nM)Bound Aptamer NumberKD= 271 29nM Figure 3-7 Binding of FITC -KK1H08 to cell membrane on human chronic myelogenous leukemia cells (K562). Number of the membrane-bound labeled aptamer (Nr) was obtained as a function of t he total aptamer concentrations in the binding buffer. K562 cells were incubated with buffer contai ning different concentrations of labeled aptamers for 40 min at 4C. Each data point represents the mean of three separate measurements. For each measurement, 0. 12mW laser intensity was used at the objective outlet. The confocal volume wa s determined by free dye calibration to be

PAGE 85

85 1.2 ƒL with half axes xy=0.235 m and z=3.93 m. The sample was exposed to laser for 30 seconds for obtaining the aut ocorrelation curve. Stable fluorescence intensity was observed during the 30s detection time which indicates no photobleaching. Different numbers of bound apt amers in the confocal volume at different aptamer incubation concent rations (50nM, 150nM, 250nM, 300nM, 400nM and 500nM) were calculated from the fitt ing of autocorrelation curve. The equilibrium dissociat ion constants (KD) of the aptamer-cell interaction were obtained by fitting the dependence of fluorescence in tensity of specific binding on the concentration of the apta mers to the equation Y=BmaxX/(Kd + X), using Microcal Origin 6.0. From the binding curve, a 500nM aptamer concentration was enough to saturate receptor binding sites on K562 cells. 020406080100 20 40 60 80 Count Rate (kHz)Time (min) Competition 1000X non-labeled sgc8020406080100 20 40 60 80 Count Rate (kHz)Time (min) Competition 1000X non-labeled LibraryA 020406080100 0.0 0.2 0.4 0.6 0.8 1.0 Normalized DensityTime (min) Competition 1000X non-labeled sgc8020406080100 0.0 0.2 0.4 0.6 0.8 1.0 Normalized DensityTime (min) Competition 1000X non-labeled LibraryB

PAGE 86

86 051015202530354 0 0.0 0.2 0.4 0.6 0.8 1.0 Normalized DensityTime (min) Tripsin TreatmentC Figure 3-8 Competition studies and Trypsin treatment. (A) Ti me course of fluorescence count rate for the displacement of FITC-labeled sgc8 binding by non-labeled sgc8 on the membrane of HeLa cells. Cell s were first incubated with 3.0nM of FITC-sgc8 for 40 min at 4 A 1000-fold molar excess of non-labeled sgc8 was then added to compete against t he labeled sgc8 in binding to the target receptor PKT7. Fluorescence count rates were m onitored for 90min. Blue squares are competition studies with a control sequenc e, which is a non-la beled DNA library containing a randomized sequence of 41 nucleotides. HeLa cells were first incubated with 3.0nM of FI TC-sgc8 for 40 min at 4 A 1000-fold molar excess of non-labeled control sequence (library ) was then added to compete the binding of labeled sgc8 to the target receptor PKT7. (B) Time course of normalized membrane-bound aptamer density change of di splacement of FITC-labeled sgc8 binding by non-labeled sgc8 on HeLa ce ll membrane. Same reaction condition was used as (A). Changes in membrane-bound FITC-sgc8 density were monitored for 90min. Blue circles are competition with the control sequence (library). (C) Time course of normaliz ed density change with trypsin treatment on HeLa cell membrane. Cells were first in cubated with 3.0nM of FITC-sgc8 for 40 min at 4 to saturate all the receptor binding sites. Then, a 10 diluted trypsin solution was added. Changes in memb rane-bound FITC-sgc8 density were monitored for 90min at 37

PAGE 87

87 0.00.51.01.52.02.53.0 0 20 40 60 80 100 120 140 160 180 200 220 Bound Antibody NumberTotal Antibody (nM)KD=0.81 0.21nM Figure 3-9 Binding curve for FITC-labeled Ig E antibody to IgE receptors on RBL-2H3 rat basophilic leukaemia cells. Number of the membrane-bound FITC-labeled IgE antibodies (Nr) was obtained as a function of the total antibody concentration in the binding buffer. RBL-2H3 cells were in cubated with different concentrations of antibodies for 40 min at 4C. Each data poi nt represents the mean of five separate measurements. The equilibriu m dissociation constants (Kd) of the aptamer-cell interaction were obtained by fitting t he dependence of fluorescence intensity of specific binding on the concentration of the aptamers to the equation Y=BmaxX/(Kd + X), using Microcal Origin 6.0. Number of CellsReceptor Density (number/ m2)B B

PAGE 88

88 Figure 3-10 IgE receptor density distributi on for RBL-2H3 cells. Cells were incubated with 2.5nM FITC-IgE antibodies at 4C for 40 min to saturate a ll the IgE receptor binding sites. Thirty indi vidual cells were under FC S measurement. The 0.4 ƒL ellipsoid shaped focus volume with half axes xy=0.22 m and z=1.56 m formed a 0.15m2 covered area on the membrane. A mean density of IgE receptors of around 1200 receptor/m2 was obtained with a variation of about 60 receptor/m2. Table 1 Comparison of Kd Value of IgE/Receptor Comple xes and IgE Receptor Density on RBL-2H3 Cells Using Different Methods. Methods for Kd Determination Dissociation Constant KD (nM) IgE Receptor Density (receptor/cell) Fluorescence Quenching128 0.71 5105 FCS 0.810.21 2.88105 1200

PAGE 89

89 CHAPTER 4 UPGRADING FCS TO DUAL COLOR-FL UORESCENCE CROSS CORRELATION SPECTROSCOPY (FCCS) FOR WIDER APPLICATION TO MOLECULAR INTERACTION STUDIES Limitations of FCS Autocorrelation Anal ysis for Molecular Interaction Studies Although the autocorrelation analysis in FCS appears to be a sensitive tool for molecular interaction studies, these can be achieved only if binding results in a measurable change in mobility (or other quant ity accessible from single-color data). Examples of applicable systems include so me membrane studies where the ligand undergoes a significant reduction in mobility when bound to the plasma membrane of a cell, as discussed in Chapter 3, or to an ar tificial membrane that is (quasi-) planar. A number of binding experiment s have also been performed in vitro, based on the change in diffusion time.129-133 As discussed in Chapter 2, fr om Equation (2-4), it is clear that the mean diffusion time D is related to the reverse reciprocal of the diffusion coefficient D: 21 4xy DD D D (4-1) According to the Einstein-Stokes Equation, the diffusion coefficient D is inversely proportional to the radius of the molecule: 1 6 kT DD R R (4-2) in which k is BoltzmannÂ’s constant; T is the absolute temperature; is viscosity;

PAGE 90

90 R is the radius of the spherical particle. From Equations (4-1) and (4-2), it is cl ear that there is a linear relationship between the diffusion time and the molecular radius: D R (4-3) If it is assumed that the analyte is a spherical molecule, then the relationship between its molecular weight and it radius can be described as: 1 1 3 33 4 M R RM (4-4) in which R is the radius of the spherical particle; is the volume of the particle; M is the molecular wei ght of the particle. Therefore, for particles with similar sizes/ volume, from Equation (4-3) and (4-4), it can be concluded that: 1 3 3 D DMM (4-5) Hence, in FCS experiments, t he molecular weight of the analyte is proportional to the cube of its diffusion time. Therefore, an eight-fold molecular weight difference results in only a two-fold change in the diffu sion time difference in the autocorrelation curve, a severe limitation for multi-componen t systems. Thus, FCS is severely limited for molecular interaction studies in common cases, where two molecules with similar molecular weights interact with each ot her, because the bound comp lex does not have an 8-fold molecular weight difference. In this case, single-color FCS does not provide

PAGE 91

91 insight into particular mo lecular interactions, and dual-color Fluorescence CrossCorrelation Spectroscopy (FCCS) is needed. Theories of Dual-Color FCCS fo r Molecular Interaction Studies The molecular weight limitation probl em mentioned above can be solved by labeling the two components with two spectrally distinct fluorophores and observing their interactions, as shown in Figure 4-1 bottom. For this purpose, excitation needs to be performed by two different lasers, and the tw o individual fluorescence emissions must be detected in individual chann els. Referring to Figure 41, the red and green signals can be measured simultaneously and their crosscorrelation function can be used as an indication. The scheme of an FCCS instrument ation set-up is illustrated in Figure 4-2. The experimental realization of a dual-col or cross-correlation set-up is demanding, because it also requires exact spatial superpos ition of the two lase r beams, so that the focal volumes overlap. Also, an additional di chroic mirror is needed in the emission pathway between the first dichroic and the pin holes to split the fluorescence signals so that each reaches its respective detector. By labeling these two different kinds of molecules with two di fferent fluorescent tags, the cross-correlation between these two channels can be measured. From Figure 4-3, if there is no molecular interaction, the two different labeled molecular diffuse independently, and there is low fluorescence cross-correlation fr om the two channels. Once the two kinds of molecules interact with each other, either bound with each other or bound to the same third molecule, t hey diffuse through the focus volume simultaneously, inducing simultaneous fluctuat ions of the fluorescence signals in the two color channels and thus a positive cr oss-correlation readout is obtained, and the

PAGE 92

92 molecular interaction is detected. More im portantly, there is no limitation on molecular weight differences for the detection of the molecular interactions. In FCCS measurements, quantitative info rmation can also be obtained. Both an autocorrelation curve for each channel and t he cross-correlation curve can be obtained from the FCCS measurement, as shown in Figure 4-1 bottom. For the autocorrelation curves (solid lines), the green and the red autocorrelation curves describe the diffusion times ( diff, red and diff, green) and amplitudes ( Go,red and Go,green) for each channel. For the cross-correlation curve (dashed line), the am plitude of the cross-correlation curve is determined ( Go,x), in which x represents t he complex. In the ideal case, if binding of the redand the green-labeled molecules is 1:1, and chromophore brightness is not influenced by the binding and cross-talk arti facts can be neglected, the degree of binding of green with respect to all red molecules ( ) is given by the ratio of the crosscorrelation amplitude to the green autocorrelation amplitude, as follows: 0, ,0,x x totalredgreenG C CG (4-6)1 Therefore, FCCS inherits the quantitative capability from FCS measurements, and it is an extension of FCS for molecula r interaction analysis. Thanks to these improvements, FCCS has found wi der applications for molecula r interaction studies. It has been applied to monitor molecular intera ctions and enzymatic reactions, as well as dynamic colocalization, for example, of cargo in small vesicles.1 Detection of these interactions quantitatively and selectively in t he full complexity of living cells allows these processes to be characterized in t heir original context. The data provides numerical values for use in computer simu lations in systems biology and offers the possibility of drug screening.

PAGE 93

93 Lab-built FCCS Instrumentation Set-Up Chapter 3 demonstrated the application of FCS to study the ligand-receptor interactions on live cell surface through monitoring the molecular diffusions of fluorescent-labeled ligands, in which the lig ands undergo a significant reduction in mobility when bound to the plasma membrane of a cell.2 However, in order to expand the range for molecular interaction studies, es pecially for the cases where a significant change of mobility is not obtained, the FC S instrument has been upgraded into a multichannel dual-color FCCS set-up. Scheme of Lab-Built FCCS Set-Up The lab-built FCCS uses a confocal type of microscope setup (Figure 4-4), which includes three different exci tation channels at different wavelengths: 488nm, 543nm and 632.8nm. The focus volumes from the three s eparate laser lines are aligned to overlap with each other in order to co-excite molecule s in the same region. This is achieved by the employment of two beam combiners, which are wavelength selective to transmit/reflect specific wavelengths to obtai n the overlap of the three laser lines. Two of the three channels can form a dual-color cross-correlation: 488nm/632.8nm (Figure 4-4A) or 488nm/543nm (Figure 4-4B), which depends on the choice of dye pairs in the FCCS measurement. The overlapped laser beam enters the objective from the back and is focused at the same region in t he sample and excites their respective fluorophores. The fluorescence emissions from the sample are collected through the same objective and are separated from the excitation light by the primary dichroic mirror. The emission beam from the two different c hannels reaches a secondary dichroic mirror to be separated from each other prior to r eaching their respective emission filters and

PAGE 94

94 detectors (avalanche photodiode (APD)). The tube lens of the microscope focuses the emission beam onto pinholes at the front-ends of the detect ors to reduce the collection of fluorescence light from above and below the focal plane. The resulting optically delimited detection volume is typically less than one femtoliter. At last, the correlator collects the fluorescent signal from each c hannel and conducts the auto correlation and cross-correlation calculations. Result s are read out by the computer. Components of the La b-Built FCCS Set-Up Excitation: There are three lasers as the exci tation sources for the FCCS set-up. They are one Ar+-laser and two He-Ne lasers. The Ar+-laser (Innova I90C, Coherent Inc., Santa Clara, CA) is a coherent laser that has multiple laser lines (458, 477, 488, 514 nm ), in which 488nm has the strongest laser intensity and is the one used for the FCCS measurements. N eutral density filters are used to attenuate the beam intensity from 20 mW down to 0.2 mW to focus on the sample species. The He-Ne lasers emit at 543nm (max imum 0.5 mW) and 632.8nm (maximum 2.0 mW) (Thorlabs, Newton, NJ). Neutral density f ilters are also app lied to attenuate the beam intensity to a final 0.2 mW intensity. Beam Combiner: Two beam combiners are used to overlap the three different lasers lines to the same focu s volume for the FCCS set-up. Referring to Figure 4-4, the first beam combiner combines the laser beam of 543nm with 632.8nm by transmitting 543nm and reflecting 632.8nm (Chroma, Bellows Falls, VT ). The second beam combiner combines the three lase r beams of 488nm, 543nm and 632.8nm by

PAGE 95

95 transmitting the shorter wavelength 488nm and reflecting the longer wavelength 543nm and 632.8nm beams (Chroma, Bellows Falls, VT ). Objective: The fluorescent light from the samp le from each channel is collected using the same microscope objective lens that is used for excitation. For improved focus of the excitation volume and collection of em issions, high power objectives are used. Although oil-immersion objectives have t he highest numerical aperture (NA ~ 1 4) and are used for good focusing, they are designe d to focus and collect light in a high refractive index environment (tha t of the immersion oil or of the cover glass). However, this optical quality is reliable only when the fo cal plane is right at the surface of the cover glass. For situations where the focal plane is at a certain di stance from the cover glass, i.e., to focus the laser excitation deep er inside the aqueous sample, use of an oilimmersion objective will generate optical aberrations as well as sub-optimal fluorescence.17 This problem can be solved through the use of water-immersion objectives, as in the design of the conf ocal microscopy. Although having a smaller numerical aperture (NA ~ 1 2), the water-immersion objective have a clear advantage of focusing the excitation light and collecting the emission efficiently, especially for experiments with aqueous samples. In t he FCCS set-up, a 60x water immersion objective with N.A. = 1.2 was used. Dichroic Mirrors: Two different dichroic mirrors are used in the FCCS set-up. The function of the first di chroic mirror is to separat e the emission beam from the excitation wavelengths for all three channe ls. It is a triple line beamsplitter 488/543/632.8 (Chroma, Bellows Falls, VT ), which selectively reflects the excitation wavelength (488nm, 543nm and 632.8nm) to the back of the object for excitation. The

PAGE 96

96 emission beams from the thr ee channels are transmitted through the first dichroic mirror to reach the second dichroic. The second dichroic mirror separates the emission wavelength from each channel to reach its respective APD detector. It is a dichroic long pass mirror (550 DCXR) that reflects any wavelengths below 550nm and transmits wavelengths above 550nm (Chroma, Bellows Falls, VT ). Filters: A bandpass interference filter is pl aced before the pinhole for each detector. For channel 488nm, a bandpass filter with a center wavelength (CWL) at 520nm and a full width at half maximum (FWHM) of 40nm (52020nm) is used. For channel 543nm, a bandpass filter (59020nm) is applied. And for channel 632.8nm, a bandpass filter (68030nm) is used. These bandpa ss filter have special coatings on the surface and are designed to transmit light onl y within a small wavelength window. They effectively cut off the interference light and reduce the nonspecific background, while maximizing the transmission of the emissions at specific wa velengths to the detector. Pinholes: The fluorescence from the emission f ilter is focused onto a pinhole in the image plane. Light originating from t he focus passes through the pinhole aperture, while light from other regions is preferentially blocked. T he pinhole was introduced into the instrument by focusing the fluorescence onto a multimode fiber with a 50 m internal diameter, where the fiber face acts as the pinhole. Therefor e, changing the fiber allows easy alteration of the pinhole aperture size without the need for subs tantial realignment of the instrument.

PAGE 97

97 Correlators: The set-up contains a hardware multiple tau digital correlator (ALV 5000/EPP, ALV-GmBH, Langen, Ge rmany) with 100% real time correlation from the initial 125 ns on. Construction of the Lab-Built FCCS Set-Up The commercially available lasers and opti cs were assembled with the right sized optic holders into a rigid framework (as s hown in Figure 4-4) on an optical bench. The overlap of the three laser lines are shown in the pictur e in Figure 4-5. As shown in Figure 4-6A, a continuous-wave Ar+ laser (III) and two He-Ne lasers (I and II) are used as the light sources. A mirror (1) and two beam combiners (2 and 3) are used to overlap the three laser lines. The laser beam is routed through a neutral density filter (4) so that the intensity of the exci tation light can be easily adjusted. Then the laser beam is expanded to a diameter of ~10 mm using a beam expander (5). The beam expander is made up of a plan-concave lens and a plan-convex lens of focal lengths -5 and 10 cm, respectively. The di stance between the lenses adds up to the sum of their focal lengths to ensure proper beam collimation. The expanded beam is reflected by two other mirrors and enters the back port of an Olympus IX70 inverted microscope (7). Here, the lase r beam is reflected by the fi rst dichroic mirror held inside the slider box of the microscope (not shown in the picture), and is then focused into the sample through a microscope objective lens The fluorescence from the sample is collected by the same microscope objective lens used for excitation and is separated from the excitation light by the same dichroic mirror, which now passes the fluorescence. After going through the first dichroic mi rror, the detection of the fluorescence emission is changed by the prism on the botto m of the confocal microscope and is

PAGE 98

98 reflected out of the microscope. As shown in Figure 4-6B, it then reaches the second dichroic mirror (1) to separate the emission wavelengths for each channel. Wavelengths below 550 nm are reflected to reach the emi ssion filter (2) and the pinhole (5), and then APD detector (7). Wavelengths higher t han 550 nm, on the other hand, are transmitted through the second dichroic mirror (1) to arrive at emission filter (3 ) and pinhole (4), and then APD detector (8). For each detection c hannel, the fiber front face works as the pinhole and the fiber holder al lows translation along the radi al (x and y) directions and axial (z) direction. At last, the detectors (7 and 8) send the signal to the input channel of the hardware correlator (8). The autocorre lation and cross-correlation curves are read out from the computer. Apply FCCS for Intracelluar mRNA Detection This lab-built FCCS not only inherits t he single-molecule detection capability from FCS, but also further extends its applicat ions for molecular interaction studies by labeling the two species with two spectrally distinct fluorophores. By taking advantage of FCCS co-localization capabi lities and the high sensitivity and low detection limit properties, this technique was further modified for real-time monitoring of intracellular mRNA. The ability to detect, localize, quantif y and monitor the expression of specific genes in living cells in realtime will o ffer unprecedented opportunities for advancement in molecular biology, disease pathophysiolog y, drug discovery and medical diagnostics. This innovative FCCS strategy stands in contrast to current methods for quantifying gene expression which are not able to pr ovide real-time monitoring (e.g., PCR134 and microarray135). Moreover, those conventional methods often result in false positives and suffer from high background from the intracellular environm ent. In this research, as

PAGE 99

99 shown in Figure 4-7, two DNA oligonucleot ides have been designed, one labeled with a green fluorophore and the other with a red fluorophore, that hybridize to adjacent regions on the same mRNA target, resulting in an increase of cross-correlation signal. The co-localization detection of the two diffe rent labeled DNA probes will significantly reduce false positives, leading to the sensit ive real-time monitori ng of mRNAs in live cells. Furthermore, the femtoliter-sized detec tion volume also ensures a low detection limit in a spatially and te mporally ordered manner and event ually improves the detection sensitivity. This research, which is current ly in progress, will provide a novel technique for sensitive RNA detection and quantification in living cells. In the future, we will also extend the applications for intracellular measurement s and apply this sensitive technique for protein-protein inte raction studies in live cells. Summary A custom FCCS instrument has been su ccessfully constructed for molecular interaction studies. The technique not only inherits the merits of FCS as being a sensitive tool for probing molecular diffusions but also expands t he application ranges for molecular interaction studies. This lab-built instrument is being applied to intracellular gene expression level detection.

PAGE 100

100 Figure 4-1 Illustrations of molecular interact ion studies by the autocorrelation function in FCS and by the cross-correlation function in FCCS. Figure 4-2 Scheme of FCCS instrumentation set-up.101

PAGE 101

101 Figure 4-3 Molecular interactions characte rized by autocorrelation curves and crosscorrelation curves. () G () G () G () G () G () G () G () G () G

PAGE 102

102 Figure 4-4 Scheme of the lab-built FCCS set-up. (A) Using channels 488nm/633nm for dual-color cross-correlation measurement s. (B) Using channels 488nm/543nm for dual-color cross-correlation measurements. Figure 4-5 Picture of the over lap of the three laser beams.

PAGE 103

103 Figure 4-6 Pictures of the lab-built FCCS instrumentat ion set-up. (A) Lasers and beamcombiners. (B) Emission filters and detectors. Figure 4-7 Scheme of applying FCCS technique for intracellular mRNA detection. 1 2 4 5 3 6 7 III II I A B 1 2 3 4 5 6 7 8

PAGE 104

104 CHAPTER 5 ENGINEERING A SURFACE ENERGY TRANSFER (SET) NANORULER FOR MEASURING PROTEIN BINDI NG SITE DISTANCES ON LIVE CELL MEMBRANE Introduction Application of optical molecular rulers to address questions in biochemistry, biodiagnostics, and bimolecular imaging allows routine measurement of molecular and dynamic distance changes. Up to now, such measurements have typically been addressed by optical methods based on Frs ter Resonance Energy Transfer (FRET) 136-139 between molecular donors and acceptor s. However, the nature of the dipole dipole mechanism effectively constrai ns the length scales in FRET-based methods to distances on the order of <10 nm (R0 6 nm).15, 75, 76 Optical methods that do not alter biomolecular function, but wh ich enable investigation of both long-range static and dynamic distances, would ther efore facilitate t he study of many multicomponent complexes that are pr esently difficult to measure. Recently developed localized surface plasmon resonance (LSPR) sensors have been able to meet these requirements 140-142. Liu et. al. 140 have constructed a nanoplasmonic molecular ruler, in which doubl e-stranded DNA (dsDNA) is attached to a gold nanoparticle (NP), for meas uring nuclease activity and DNA footprinting. While this plasmonic ruler performs very well in bu lk solution experiments and show a longer detection range than FRET, significant c hallenges exist when applied to cellular systems because of the high Raman scatte ring background coming from the cells themselves. This high background significant ly reduces the signal-to-background ratio of the plasmonic sensor and greatly hinder s molecular distance measurements when applied on the cell surface. 90 light scattering of gold nanoparticles has been used by Bene et. al.143 to measure the distances of receptor s on the cell surface. However, light

PAGE 105

105 scattering system normally requires the dye and t he particle surface to locate in close proximity (<10nm) in order to have the scatt ering occurred; in longer distance range, fluorescence quenching effect will be the dominant factor in the interactions.143 In addition, it was reported that light scattering effect only dominates in smaller sized nanoparticles (<2nm) due to the insufficient over lap for fluorescence energy transfer. In contrast, particles with greater than a 2nm diameter have higher probability and incidence of fluorescence quenching.144 Therefore, there’s a great limitation on the choice of nanoparticle sizes for the ruler cons tructions. All these restrictions make the detection of interactions between large co mplexes on live cells extremely difficult. Therefore, alternative met hods are needed for constructing molecular rulers with a still larger detection range fo r these cellular systems. One such option involves fluore scence measurements which have wide applications in cellular systems as well as in vivo systems. Recently, several groups66-74 have reported the nanoparticle surface energy transfer (SET) where energy transfer flows from a donor molecule to a nanoparticle su rface at a much slower decay rate than the dipole-dipole energy transfer in FRET, with a 1/d4 distance dependence.66 While SET is similar to FRET, in that the inte raction is dipole–dipole in nature, it is geometrically different from FRET becaus e an acceptor nanoparticl e has a surface and an isotropic distribution of dipole vectors to accept energy from the donor, leading to a dipole-surface resonance mechanism.67-70 This arrangement effectively breaks the inherent detection barriers of FRET, thereby increasing the pr obability of energy transfer and ultimately enhancing the efficiency of SET over FRET. The intensity quenching mechanism via coupling of the oscillating elec tronic dipole of a dye to a metal surface

PAGE 106

106 with loss of energy was developed by Chance, Prock and Silbey,75 and by Persson and Lang76 for bulk metals. Aside from bulk system s, recent attempts have conjugated different lengths of dsDNA onto the metal nanoparticle surface, th us demonstrating the validity of this “SET nanoruler” for mappi ng distance in a real biological system.66 However, in all these systems, the gold nan oparticles are only used as carriers. Instead of functionalized as a real nanoruler to measure distance between two moieties, these systems can only work as the artificial rule rs for characterizing the SET phenomenon. In addition, for these models to be effective, high conjugation requirements must be established to place only one dsDNA onto t he surface. Therefore, although ample literatures have described the applications of SET for distance measurements in bulk system, there is no single application so far to apply these SET models to the cellular systems. To overcome these limitations and construct a practical SET nanoruler for measuring distances on live cells, we are the first to propose to 1) use aptamer-gold nanoparticle conjugates to constr uct a “SET nanoruler” in a r ange of different sizes and, 2) for the first time, apply it to the surface of a live cell to monitor the distance between two binding sites on a membr ane receptor. Specifically, by taking advantage of the fixed binding site distance on a receptor, we us e different ligands to bring the organic fluorophore and the metal nanoparticles into clos e proximity of fixe d length. Moreover, by varying the size of particles, the distance from the dye molecule to the metal particle surface can be manipulated. In this way, we are able to obtain the distance between the two binding sites on the live cell me mbrane. An added advantage is the easy and

PAGE 107

107 straightforward preparation of the aptamernanoparticle conjugates by direct incubation of the components. Experimental Section Preparation of gold nanopartic les of different sizes. Various sizes of gold NPs were synthesized according to the method developed by Frens145. Gold NPs were adopted in the system because of the simplicity and reproducibilit y of the synthetic and bioconjugation techniques, as well as thei r unique fluorescence quenching property. Briefly, 0.5 ml of 1% chloroauric acid (S igma-Aldrich, St. Louis, MO) was added to 50 ml of double-distilled water, and t he solution was heated to boiling. Next, different amounts of 1% sodium citrate (Rie del-deHan, Seelze, Germany) were added to the solution to synthesize different sizes of gold NPs. T he solution was refluxed until a color change from dark blue to red was obs erved. The sizes and absorption spectra of various gold NPs were verified with the Hitachi H-7100 transmission electron microscope (Tokyo, Japan) (Figure 5-1) and the Cary 100 UV-Vis spectrophotometer (Varian, Palo Alto, CA) (Figure 5-2). The concentration of gold NPs in each aliquot was also determined by UVVis spectroscopic measurement s via the BeerÂ’s law (A = bc). The 5.4nm gold NPs were purchased from Sigma-Al drich (St. Louis, MO). Charac teristic parameters for the preparation and characterization of different sets of gold NPs are summarized in Table 5-1. Aptamer Synthesis. All DNA oligomers were syn thesized on an ABI3400 DNA/RNA synthesizer (Applied Biosyst ems, Foster City, CA) in our laboratory. All DNA oligonuclease bases and 5Â’-modifiers were pur chased from Glen Research, Sterling, VA. The aptamer sgc8 was labeled with various 5Â’-modifiers. A DNA library containing a randomized sequence of 41 nucleotides was label ed with a 5Â’-FITC, -TAMAR or -Cy5

PAGE 108

108 modifier and used as a control for each det ection channel. Sequence TDO5, a control aptamer for CEM cells, was labeled with 5Â’-Thiol-Modifier [DMTO(CH2)6-S-S-(CH2)6OP(N( i Pr2))(O(CH2)2CN)] or 5Â’-Amino-Modifier [MMT-NH-(CH2)6-OP(N( i Pr2))(OCNEt)] for particle conjugations. Fo r the deprotection procedur es, sequences labeled with 5Â’TAMAR-Modifier were deprotected in TAMR A deprotection solution (0.05M potassium carbonate in methanol) at 65C for 3-4 hours. Sequences labeled with 5Â’-Cy5-Modifier were deprotected in ammonium hydroxide at room temperatur e for 24-36 hours. Sequences labeled with 5Â’-Amino-Modifier we re deprotected in ammonium hydroxide at 40C for 17 hours. All other sequences were deprotected in AMA (ammonium hydroxide/40% aqueous methylamine 1:1) at 65C for 20 minutes. After deprotection, all sequences were further purified by reverse phase HPLC (ProStar, Varian, Walnut Creek, CA) on a C-18 column. A Cary Bio-300 UV s pectrometer (Varian, Walnut Creek, CA) was used to measure absorbance to quant ify the manufactured sequences. DNA sequences with different labeling are summarized in Table 5-2. Preparation of aptamer-functi onalized gold nanoparticles. The strategy for functionalizing gold nanoparticl es with thiol-modified o ligonucleotides followed a published procedure145. The thiolyated aptamers (sgc8 or TDO5) were added to gold NP stock solutions for surface modifica tion. The amount of aptamer needed for the different gold nanoparticle sizes (50 mL) was ca lculated from estimations of the surface area for each gold NP to ensure saturated su rface coverage, as listed in Table 5-3. After 12 h, 2.5 ml 10 Phosphate Buffered Saline (PBS) (pH 7.4, Fisher Scientific, Pittsburgh, PA) was added to the reaction mixt ure. After an additional 12 h, another 2.5 mL 10 PBS solution (pH 7.4) was added to make a final concentration of 1 PBS

PAGE 109

109 solution (pH 7.4 0.1) with 11.9 mM PO4 3 137mM NaCl and 2.7 mM KCl. The reaction mixture was then allowed to react fo r 12 h. Unbound oligonucleotides were subsequently removed by centrifugation an d resuspension of the pellet (14,000 rpm, twice). DNA-coated gold NPs were resu spended, and the concentrations were determined by measuring the absorptions using a Cary Bio-300 UV spectrometer (Varian, Walnut Creek, CA). Condensed to a concentration of 10 nM, the DNA-gold NP conjugates were stored in a 1 PBS solution (pH 7.4 0.1, 11.9 mM PO4 3 137mM NaCl and 2.7 mM KCl). Preparation of aptamer-functi onalized silica nanoparticles. 3-Aminopropyl(3oxobutanoic acid)-functionalized silica NPs (~15 nm (DLS), 2.5 % (w/v) in DMF) were purchased from Sigma-Aldrich (St. Louis, MO) and used as the control NPs for fluorescence quenching experiments. The par ticles were first washed with doubledistilled water three times (centrifuge rate : 14,000 rpm) to remove dimethylformamide (DMF) solvent and resuspended in DI water to a final concentration of 200 nM. Before DNA conjugation, the carboxy lic acid groups of silica NPs were activated with 1-ethyl-3(3-dimethylaminopropyl) carbodiimide (EDC) and N -Hydroxysuccinimide (NHS) by adding 1.0 ml of carboxylic acid-functiona lized silica NPs to 2.0 mL of 1 PBS buffer (pH 7.4) containing 6.5 mg of EDC and 5.8 mg of NHS. After 1.0 h of stirring at 25 C, the particles were centrifuged and redispers ed in 1 PBS buffer (pH 7.4) to give the NHS ester-terminated silica NPs. Ex cess amino-labeled aptamers (5Â’-NH2-sgc8 and 5Â’NH2-TDO5) (200 l, 1 mM) were added and incubat ed with the silica NPs under gentle shaking overnight at 4 C. (Aptamer:NP = 1:1000 molar ratio) The conjugates were

PAGE 110

110 washed three times with PBS buffer and finally reconstituted in binding buffer and stored at 4C until use. Antibody labeling. The pure anti-PTK7 (100 mg/L) wa s purchased from Miltenyi Biotec, Inc. (Auburn, CA). Before dye conjugation, the antibody was concentrated 10 to 1 mg/mL using a 30 KDa cut-off membrane (Fisher Scientific, Pittsburgh, PA) and centrifuged at 12,000 rpm. The Alexa Fluor 488 Monoclonal Antibody Labeling Kit was purchased from Invitrogen (C arlsbad, CA). The Alexa-488 reactive dye has a tetrafluorophenyl (TFP) ester moiety that reac ts efficiently with primary amines on the anti-PTK7 to form a stable dye-protein c onjugate. The antibody was labeled with an Alexa-488 reactive dye in a sodium bicar bonate buffer (pH 8.3) fo r 1 hour at room temperature. Dye-labeled ant i-PTK7 was separated from the unlabeled anti-PTK7 using a spin column (1100 g, 5 min) with a 30,000 MW size-exclusion puri fication resin in PBS (pH 7.2, plus 2 mM sodium azide). After centrifugation, the labeled anti-PTK7 was collected in the collecti on tube in approximately 100 L of PBS (pH 7.2), with 2 mM sodium azide; free dye remained in the co lumn bed. The concentration and degree of labeling were determined using the formulas given in the prot ocol. The conjugates were stored at 4 C, protected from light. For the isotype-negativ e control antibody, the FITClabeled Mouse IgG2a was purchased directly from Miltenyi Biotec, Inc. (Auburn, CA) and also stored at 4 C, protected from light. Cell culture. CCRF-CEM (CCL-119, T cell line, human ALL) and Ramos (CRL-1596, B cell line, human Burkitt's lymphoma) we re obtained from American Type Culture Collection. Both cell types were cultured in RPMI 1640 medium (American Type Culture Collection), with 10% fetal bovine serum (I nvitrogen, Carlsbad, CA) and 0.5 mg/ml

PAGE 111

111 Penicillin-Streptomycin (American Type Cu lture Collection, Manassas, VA) at 37oC under a 5% CO2 atmosphere. Cells were washed bef ore and after incubation with wash buffer [4.5 g/L gl ucose and 5 mM MgCl2 in Dulbecco's PBS with calcium chloride and magnesium chloride (Sigma-Aldrich, St. Louis, MO)]. Binding buffer used for selection was prepared by adding yeast tRNA (0.1 mg/m l; Sigma-Aldrich, St. Louis, MO) and BSA (1 mg/ml; Fisher Scientific, Pittsburgh, PA) to wash buffer to reduce background binding. Flow Cytometric Analysis. Saturation concentration of Alexa488-labeled anti-PTK7 (200 nM, Figure 5-3A) was incubated with CCRF-CEM cells (1 106) at 4C to prevent receptor internalizations146 for 20 min in the dark in a 200L volume of binding buffer containing 20% FBS. Cells were then washed once with 700 L of the binding buffer, then incubated with saturation concentrations of different sizes of gold NPs (5.4nm: 8nM; 10.1nM-42.2nm: 4nM, saturation concentration det ermination shown in Figure 5-3B) or with 4nM 15nm silica NP aptam er conjugates for another 20 min. Then cells were washed twice with 0.1% sodi um azide, suspended in 200 L of binding buffer with 0.1% sodium azide, and subjected to flow cytometric analysis wit hin 15 min. The fluorescence was determined with a FACScan cytometer (BD Immunocytometry Systems) by counting 10,000 events. A green laser at 488nm with different excitation voltages (650V, 700V, 750V) was used as the excitation sour ce. The FITC-labeled isotype Mouse IgG2a antibody was used as a negative control to determine nonspecific binding. The cell fluorescence background was determined fr om samples with cells only as the fluorescence background for later calculations. Confocal Imaging Analysis. For confocal imaging, trea tment steps for fluorescence quenching experiments were the same as descr ibed in Flow Cytometric Analysis. 30 L

PAGE 112

112 of cell suspension bound with Alexa488-ant i-PTK7 and aptamer-gol d NP conjugates was dropped on a thin glass slide placed above a 20 objective on the confocal microscope and then covered with a covers lip. Imaging of the cells was performed on an Olympus FV500-IX81 confocal microscope. A 5-mW, 488-nm Argon laser was the excitation source for the Al exa Fluor 488 dyes throughout the fluorescence quenching experiments. The objective used for imaging was a XLUMPLFL20XW 20 waterimmersion objective with a numerical aperture of 0.95 (Olympus). Saturation Binding Concentr ation Determination. In order to monitor the fluore scence quenching phenomenon between the two binding sites on live cell membrane, both Alexa488-labeled anit-PTK7 and aptamer-gold NP conjugates (aptamer sgc8 or TDO5) we re added to the cell surface at saturation concentrations to saturate all the aptam er and antibody binding sites on the cell membrane. The saturation concentrations fo r Alexa488-antiPTK7 were determined as following: varying concentrations of Alexa48 8-labeled anti-PTK7 (5 nM 300 nM) were incubated with CCRF-CEM cells (1 106) at 4C to prevent receptor internalizations3 for 20 min in the dark in a 200l volume of binding buffer containing 20% fetal bovine serum (FBS). Cells were then washed twice with 700 L of the binding buffer with 0.1% sodium azide, suspended in 200 L of binding buffer with 0.1% sodium azide, and subjected to flow cytometric analysis withi n 15 min. The fluorescence was determined with a FACScan cytometer (B D Immunocytometry Systems) by counting 10,000 events. A green laser at 488nm with an excitation voltage 750V was used as the excitation source. The FITC-labeled isotype Mous e IgG2a antibody was used as a negative control to determine nonspecific binding. All of the experiments fo r binding assay were

PAGE 113

113 repeated four times. The mean fluorescence intensity of target cells labeled by Alexa488-anti-PTK7 was used to calculate specific binding by subtracting the mean fluorescence background intensity of cells only. The equilibrium dissociation constants ( Kd) of the antibody–cell in teraction were obtained by fitting the dependence of fluorescence intensity of specific bindi ng on the concentration of the Alexa488-antiPTK7 to the equation Y = B max X /( KD+ X ), using SigmaPlot (Jand el, San Rafael, CA). Binding curves were also determined for diff erent sizes of gold NP -aptamer conjugates on the cell surface. The binding curve of Alexa488-antiPTK7, shown in Figure 5-3A, clearly indicates that concentrations above 200 nM assure saturation binding. Fluorescence quenching were then obtained by incubating 1 106 CCRF-CEM cells with 200 nM Alexa488-antiPTK7 and different am ounts (0 nM-15 nM) of specific sizes of gold NP-aptamer conjugates. For a given si ze, increasing concentrations of gold NPaptamer conjugates lead to hi gher degree of fluorescence quenc hing from the Alexa488 dyes on the antibody until the maximum fl uorescence quenching is reached, where all the aptamer binding sites have been saturated, as shown in Figure 5-3B. The saturation concentrations were determined for different sizes of gold-aptamer conjugates (5.4 nm, 10.1 nm, 18.4 nm) on the cell surface. The bindi ng curves indicated that for 5.4 nm gold NP-aptamer conjugates, 8 nM was sufficient to saturate the aptamer binding sites and reached maximal fluorescence quenching. For larger size conjugat es (10.1 nm and 18.4 nm), 4 nM was determined as the satura tion concentration. Therefore, in the fluorescence quenching experiments, we used 8 nM of the 5.4 nm gold NP-aptamer conjugates and 4 nM of larger sized conjuga tes (10.1 nm – 42.2 nm) to saturate the aptamer binding sites.

PAGE 114

114 Competition Studies to Confirm the Saturation Binding of Aptamer-NP Conjugates on Cell Membrane. Competition studies were carried out to confirm the saturation binding of gold NP-aptamer conjugates on the ce ll membrane. Cells were first incubated with saturation concentrations of Alexa488-antiPTK7 ( 200nM) and different sized gold NP-sgc8 conjugates (8nM for 5.4nm; 4nM for 10.1-42.2nm) for 20 min on ice. After the cells were washed twice with 700 L of binding buffer (with 0.1% NaN3), an additional 25 nM Cy5labeled aptamer sgc8 was added to the ce ll surface and incubated for another 20 min on ice. Then, the cells were wa shed twice and suspended in 200 L of binding buffer (with 0.1% NaN3). Images of the cells stained wit h Cy5-labeled aptamers were taken using a 2-mW, 633 nm He-Ne laser as the ex citation source and collected by the same 20 objective on the confocal microscope at 680nm. As shown in Figure 5-4, no staining of Cy5-sgc8 was observed from the cell membrane for cells incubated with Alexa488antiPTK7 and gold NP-sgc8 beforehand (c), indi cating all the aptamer binding sites on the cell membrane were saturated with the gold NP-sgc8 conjugates already. However, for cells incubated with Alexa488-antiPTK7 and the conjugates with control aptamers, gold NP-TDO5 (d), cells show similar fl uorescence staining of Cy5-sgc8 on the membrane as in cases with no gold NP-aptam er conjugates (a, b). The competition studies using Cy5-sgc8 confirmed the satura tion binding of different sized gold NP-sgc8 conjugates on the aptamer binding sites in the fluorescence quenching experiments. Results and Discussion The cell membrane receptor Protein Tyro sine Kinase 7 (PTK7 or CCK-4), an important biomarker recept or for T-cell acute lymphobl astic leukemia (T-ALL) 9, 125, was chosen as the target molecule for constr uction of the “SET nanoruler”. Monoclonal

PAGE 115

115 antibody anti-PTK7 and aptamer sgc8 have been identified as the molecular ligands for the two binding sites on PTK7 respectively125. Competition Studies between Aptamer Sg c8 and Antibody Anti-PTK7 on PTK7 Before constructing the prototype “SET nanoruler” on the live cell surface, the colocalization of the two bi nding sites on the model rec eptor PTK7 was determined. Monoclonal antibody anti-PTK7 and aptam er sgc8 have been identified as the molecular ligands for the two binding sites on PTK7 respectively125.The specific interactions of protein receptor PTK7 with these two ligands were confirmed by PTK7 gene silencing2 and gene transfection experiments1. The interactions between the two ligands in the receptor PTK7 were first ev aluated by competition studies to validate the colocalization. Competiti on studies between the two ligands were conducted and monitored using Flow Cytometer. Excess un labeled sgc8 (100) was used to compete with Alexa488-labeled anti-PTK7 (200 nM) for CEM ce ll binding (Figure 5-5A). To further investigate the possi bility of co-binding of sgc8 and the antibody on PTK7, a contrasting experiment was conducted by first labeling the apt amer with a FITC fluorophore and incubating with CEM cells. A fterwards, excess non-labeled anti-PTK7 (100) was added to compete with the aptamer binding (Figure 5-5B). Flow cytometry results showed no obvious change in the Alex a488-anti-PTK7 binding, indicating that the aptamer sgc8 and the antibod y anti-PTK7 simultaneously bi nd to two different sites of the extracellular domain of PTK7. Th is colocalization phenomenon served as the basis for constructing the molecular ruler to measure the two bi nding sites on the cell surface.

PAGE 116

116 Measure the Distance between Two Binding Sites using FRET. Preliminary trials to meas ure the two binding site di stance were conducted using the FRET design as shown in Figure 5-6. By labeling the two ligands with different dye pairs (Alexa488-antiPTK7/Cy5-sgc8 or Al exa488-antiPTK7/TMR-sgc8), we monitored the energy transfer from the donor dye (Alexa Fluor 488 on antiPTK7) to the acceptor (TMR/Cy5 on sgc8) in Flow Cytometry. In the FRET experiment in the presence of Alexa488-antiPTK7 and Cy5-sgc8, no signifi cant fluorescence shift was observed in channel 3 at 680nm with an exci tation source at 488nm compared to the Alexa488antiPTK7 or Cy5-sgc8 only (A lexa488-antiPTK7 only: 4.30; Cy5-sgc8 only: 3.71; both: 4.53). In the FRET experiment with Alexa48 8-antiPTK7 and TMR-sgc8 in channel 2 at 580nm excited at 488nm, severe channel le akage was observed from channel 1 to channel 2 for TMR-labeled sgc8 due to the spectr al overlap, which resulted in significant fluorescence shift in channel 2 for the a cceptor only (TMR-sgc8) even without the presence of fluorescence donor (Alexa488-an tiPTK7). Therefore, no significant fluorescence difference was observed betw een the FRET signal and TMR-sgc8 only (TMR-sgc8 only: 8.31; Alexa488/TMR-sgc8: 10.28). All these results have showed the weak fluorescent signals between the two ligands using the FRET design. Since no competition was observed between the two ligands on PTK7, which supported the simultaneous bi nding of two ligands to t he PTK7 receptor on the membrane. This colocalizati on phenomenon proved PTK7 as an ideal model system for the development of a nanoruler to measure binding site distances on live cells. And preliminary tests with a FRET design by la beling two ligands with different dye pairs

PAGE 117

117 showed weak FRET signals, which may result from unsuitable dipole-dipole orientations on the cell surface or a dist ance between binding sites that exceeded the detection limit (<10nm) of FRET.15, 75, 76 Routine X-ray crystallography was also difficult to measure the binding site distances bec ause of problems wit h the separation and purification of the pr otein receptor from the cell membrane and prot ein complex. On the other hand, by applying our “SET nanoruler”, there is no need for protein separations and binding site distances c an be obtained in the natural cell surface physiological environment, without the concern of protein conformational changes after separations from the cell membrane. SET Nanoruler Construction In the effort to construct a “SET nanor uler” to measure the distance between the two binding sites, we employed the two ligan ds, sgc8 and anti-PTK7, to bring an organic fluorophore and a metal nanoparticle to each of the binding sites on the cell membrane. Gold NPs were chosen as the energy acc eptor in the SET design because of their unique fluorescence quenching pr operty, as well as their simplicity and reproducibility for synthetic and bioconjugation preparations (c ompared to other metal NPs, e.g., silver NPs). The gold nanoparticle was functional ized with excess thiol-modified sgc8 aptamers on its surface, and the antibody anti-PTK7 wa s modified with the organic fluorophore Alexa Fluor 488 through the primary amines on the heavy chain (preparations described in Experimental Section). As a consequence of the colocalization of both binding sites on receptor PTK7, the binding of ligand sgc8 and anti-PTK7 to their individual binding si te brought the fluorophore and gold nanoparticle into close proximity, effectively resulting in fluorescence energy transfer from one to the

PAGE 118

118 other. As shown in Figure 57, the distance, R, from the fluorophore on the antibody binding site to the center of the nanoparticle on the aptamer binding, is equivalent to the distance between the two bindi ng sites on the PTK7 receptor to be measured. In order to simplify the model, R was divided into two parts: the distance r, from the fluorophore to the surface of the nanoparticl e, and the distance from the surf ace of the particle to its center, which is the radius of t he particle, d/2, with R= r + (d/2 ). Therefore, as the size of the gold nanoparticle (d) is varied, the dist ance from the fluorophore to the particle surface (r) changes accordingly. Although some variations in position around the center point of the aptamer binding site can potentia lly occur, one million cells were counted each time to cancel out these variances. In addition, a series of gold nanoparticles of different sizes were adopted to fit in the bi nding pockets to avoid this steric effect. Therefore, by controlling t he sizes of the gold nanoparticl es, the distance from the fluorophore molecule to the surface of gol d nanoparticles could be manipulated, and the relationship between the size variations of the gold nanoparticles and the change in the energy transfer efficiency could be evaluated. First, gold nanoparticles of different size s were prepared with precisely controlled diameters from 5 nm, 10 nm 13 nm, 15 nm, 18 nm, 20 nm, 25 nm, and 31 nm to 42 nm, as described in Experimental Section. The sizes were controlled by using different amounts of reducing agent (sodi um citrate) and were moni tored afterwards by TEM imaging. The average diameter of each size was obtained by measuring the sizes of 100 particles from the TEM images with a narrow size distribution (shown in Table 5-1). In order to obtain the specif ic SET information from the two binding sites on the same receptor, the nanoparticle size range (d=5 .4nm-42.2nm) was chosen to avoid

PAGE 119

119 crosslinking between individual receptors or inter-receptor SET effects. According to our previous receptor density studies of PT K7 on the cell surface using Fluorescence Correlation Spectroscopy2, the distance between two adjacent receptors is around 40nm. Therefore, the sizes of nanoparticles c hosen for the SET ruler constructions are smaller than the average inter-receptor distanc e, leading to a low chance for receptor crosslinking. In addition, a probability calculation shown in Appendix A indicates that the contribution of inter-receptor SET due to re ceptor lateral diffusion is less than 1/10000 to the overall observed SET phenomenon. Mo reover, since the di stance from a dye molecule to the gold NP on the same receptor is much smaller compared to that to a gold NP on an adjacent receptor (at least 40 nm), the major contribution of the SET effect observed is, therefore, coming from interactions between the dye molecule and the gold NP on the same receptor. In order to eliminate the concentration effect on the SET measurement, quantitative binding experiments were conduct ed for both dye-labeled anti-PTK7 (Figure 5-3A) and each size of aptamer-gold NP conj ugates (Figure 5-3B) on their own binding sites on the cell membrane. For dyelabeled anti-PTK7, a 200nM saturation concentration was obtained at the antibody bind ing sites; and for each size of aptamergold conjugates, different concentrations of the conjugates (0nM-15nM) were incubated on the cell surface of same am ounts of cells (1 million) to obtain a quantitative binding curve, and a plateau was achieved to deter mine the saturation concentration on the surface for that specific si zed conjugate (8nM for 5.4nm, 4nM for 10.1nm and larger NPs). We used these saturation concentrations to conduct all the following SET experiments to keep the consistency in measurements and ensure that they are all

PAGE 120

120 under the same saturation labeling conditions for all sized particles conjugates on the cell surface. Plus, a competition study usi ng the same aptamer sgc8 with a different label (Cyanine 5, 650nm/670nm) was conducted after each SET experiment (Figure 5-4) to once again confirm the saturation bindi ng of the gold nanopartic le-atpamer sgc8 conjugates on the cell surface. No stain of Cy5-sgc8 on the surface of cell membrane indicated that all the aptamer binding si tes are saturated by the gold NP-sgc8 conjugates. Therefore, both binding site s were ensured under saturation binding conditions by their individual ligands in the SET nanoruler for the distance measurements. After the SET nanoruler were construct ed onto the cell membrane and saturated the two binding sites, the sizes of the gol d nanoparticles on the aptamer sites were varied from d=5.4nm to 42.2nm. Both flow cytometry (Figure 5-8) and confocal imaging (Figure 5-9) were used to monitor the c hanges in the fluorescence energy transfer efficiency from the Alexa Fluor 488 dyes to the surface of t he gold nanoparticle while the NP sizes were varied. In presence of the aptamer-gold NP conjugates on the aptamer binding sites, fluorescence quenc hing from the Alex a Fluor 488 donor was observed compared to the situation wit h no gold NP conjugates; and while the diameters of the NPs became la rger, the surface of gold NPs were getting closer to the dye molecules, and theref ore, more fluorescence quenching was obtained. A quantitative monitoring of the fluorescence quenching was shown in the histograms from the flow cytometry measurements (Figure 5-8B ). When the fluorescence quenching efficiencies was plotted against the change in diameters of the gold NPs on the aptamer binding sites, a sharp fluoresc ence decrease from the Alexa Fluor 488 on

PAGE 121

121 anti-PTK7 was observed as the sizes of aptamer-NP conjugates grew from d=5.4nm (92.8% fluorescence intensity compared to no NPs) to d=18.4nm (35.9% fluorescence intensity compared to no NPs) While the aptamer-NP conjugat es reached a diameter of 18nm, the fluorescence quenching reached a fluor escence plateau with similar intensity as the IgG2a control, and larger NP conj ugate sizes (18.4nm-42.2nm) remained similar fluorescence quencher efficiencies. This d= 18nm converging point not only indicated the situation of maximum fluorescence quenchi ng, but also provide the information of the minimum diameter of gold NPs on the apt amer site whose surface can reach the dye molecules on the antibody sites. Theref ore, the distance between the dye molecule and the gold NP reached minimum to produc e the maximum fluorescence quenching. Therefore, a fluorescence pl ateau was obtained and gold NP conjugates with diameters larger than 18nm gave a similar degree of fluorescence quenching. All these quantitative fluorescence quenching efficiency results against the change of gold NPs on the aptamer binding sites were adopted into the further calculations and discussions in the following part “Binding Site Distance De termination” to further obtain the distance between the two binding sites on the cell membrane. Binding Site Distance Determination As illustrated in Figure 5-7, the relationshi p of the binding site distance (R), the NP diameter (d) and the distance from the dy e molecule to the gold NP surface (r) is given by: 2 d rR (5-1) Since the binding site distance R is fix ed, increasing the diameter of the gold nanoparticle will result in closer proxim ity between the fluorescence donor and the

PAGE 122

122 nanoparticle acceptor surface (smaller r), henc e increasing energy transfer efficiency. This, in turn, explains why the larger gold nanoparticles in the ruler resulted in better fluorescence quenching. The plot of quenching efficiency ( ) versus different gold nanoparticle diameters (d) (Figur e 5-10) shows a plateau a fter the particle diameter reached ~18nm, indicating that the surfac e of the particle had reached the antibody binding site and had thus achieved maxi mum fluorescence quenching. Consequently, particles with diameters larger than d=18nm would not reac h a higher degree of quenching. The distance-dependent quenchi ng data were fit to t he Nanoparticle Surface Energy Transfer model employed by Jennings et. al.70: 4 01 1 r r (1-4) In equation (1-4), is the energy transfer efficiency; r0 is a constant value for a specific dye-metal system, corresponding to the dist ance at which a dye will display equal probabilities for energy transfe r and spontaneous emission66; and r is the distance from the dye molecule to the gold nanoparticle surface. Substitution of r from equat ion (5-1) and rearranging gives: 1 4 001 1 2 dR rr (5-2) By letting Y= 1 41 1 and X= d (3) is simplified to: 002 X R Y rr (5-3)

PAGE 123

123 According to the literature72, unlike large particle sizes (>20nm) in which r0 values vary a lot for different particle sizes, fo r gold particle sizes between 5-15nm, a very similar r0 value was obtained around 6-8nm. Therefore, the r0 values do not have significant variance for different sizes for the particle size ranges (5-15nm) that chosen for the distance calculations and it is r easonable during our calculation to assume r0 as a constant in this range. Since r0 is a constant, X and Y obey a linear relationship with slope= 01 2 r and Yaxis intercept= 0 R r. The plot of Y versus X in Figure 5-11 is linear between 5.4 and 18.4 nm with equation: 1 41 (1)(0.0770.0072)()(2.070.095) dnm (5-4) From the slope, 0 11 6.49 (0.077)2 rnm nm and from the intercept, 6.492.0713.4 R nmnm Thus, the obtained distance between the two binding sites is R= (13.41.4) nm, which is larger than the detection limit for FRET. This explains the weak FRET signal from these two binding sites obtained in previous FRET studies (Figure 5-7) and proves that nanoparticle dipole-surface energy transfer can detect larger interaction distances than dipole-dipole interactions.

PAGE 124

124 Potential Effects of Steric Hi ndrance on the Fluorescence Quenching The suspected fluorescence quenching fr om the gold nanoparticles themselves in the system was excluded by a control experiment with same sized gold NPs conjugated with a control apt amer sequence TDO5, which had been confirmed with no bindings to receptor PTK7146. The control experiments conduc ted in both flow cytometry (green curves in Figure 5-8) and confocal imaging (with white titles) showed no significant fluorescence quenching compared to the situation with no NPs, indicating that the energy transfer between the two binding sites was indeed resulted from the specific ligand-receptor interactions whic h brought the donors and acceptors into close proximity and that the metal par ticle in the system did not have nonspecific interactions with the dyes. So far, surface energy transfer has been the only plausible basis for the fluorescence quenching phenomenon. However, it is also possible that the binding of nanoparticles to the aptamer binding sites coul d hinder the binding of the fluorescence donors to the antibody binding sites, and ther efore lead to the decrease of fluorescent signals. To investigate the possible steric hindrance of the nanoparticles, control experiments were performed with silica nano particles (Figure 5-12). A 15-nm silica nanoparticle was chosen as the control syst em for two reasons: 1) 15-nm gold nanoparticles showed significant fluorescenc e energy transfer in the gold NP assays (~50%) and 2) there is no fluorescence absor ption in the visible range for silica nanoparticles; therefore, they should not accept any energy from the fluorescence donor. Saturation binding (Figure 5-13) indica ted that the SiNP-sgc8 conjugates indeed bound to the aptamer binding si tes on the cell surface. However, as shown in Figure 5-

PAGE 125

125 12, while the 15-nm diameter gold nanoparticles quenched arou nd half of the fluorescence from the donor, their counter part silica nanoparticles with the same diameter showed no significant fluoresc ence shiftback compared to the maximum fluorescence from the donors onl y. Therefore, we can ther efore deduce that differences in fluorescence quenching resulted from the specific energy transfer from the donor on the antibody binding sites to the gold nanopartic les on the aptamer binding sites and were not caused by the nonspecific steric hindrance effects in the donor binding. Effects of Different Exci tation Intensity and Labe ling Efficiency on Distance Determination The binding site distance was obtained by studying the relationships between the fluorescence quenching efficiencies and the c hange in NP diameters. However, some systematic effects, such as the dye-antibody labeling e fficiency effects and different laser excitation intensity variations, c ould also change the quenching efficiency and result in a different answer of binding site distance. Here, the influences of different laser excitation intensities and different ant ibody/dye labeling efficiencies on the calculations of binding site distance were investigated in a series of fluorescence quenching experiments with gold NPs in differ ent sizes. The laser excitation voltage was varied from 650V to 700V and then to 750V in the flow cytometer, and the fluorescence quenching was monitored as show n in Figure 5-14. It was observed that higher laser intensities gave higher fluoresc ence signals. However, similar slopes were obtained in the fluorescence decay for diff erent laser intensities for nanoparticle diameters ranging from 5.4nm to 18. 4nm, while a plateau was achieved with nanoparticle diameters larger th an 18.4nm. The similar decay rates resulted in similar

PAGE 126

126 distance R results, which indicated the negligib le effects of different laser intensities on the distance determinations of the “SET nanoruler”. Different antibody/dye labeling efficiencie s were also tested. With multiple primary amino groups on the heavy chain, the anti-PTK7 could react with different amounts of Alexa Fluor 488 dyes during the conjugation. By controlling the amount of dye added, three different conjugates were prepar ed, having 2, 4 or 6 dyes, respectively, on each antibody. As shown in Figure 5-15, results of the fluorescence quenching experiments were similar to those of the laser intensity experiments. Both of these studies indicated that the variations in laser source intensity and antibody labeling efficiency have limited effect s on the determination of the binding site distances. Here it also showed the benefit of using gold NPs with different sizes instead of a single size for the nanoruler construc tion. As seen from the above measurements, the laser intensity and labeling efficien cy effects could change the fluorescence quenching efficiency for i ndividual size of gold NPs. So if only a single sized NP was used in the distance measurem ent, these effects could resu lt in a different binding distance. However, while a series of different sized NPs were applied for the measurement, these effects were rule out and a similar decay slope was obtained, which leaded to a similar results for the bi nding site measurements. Therefore, the use of different sizes of nanoparticles in the rule r construction also introduces a feature to enhance the precision for the distanc e calculations and determinations.

PAGE 127

127 Effects of Different Anti body/Dye Labeling Efficienci es on the Calculations of Binding Site Distances Different antibody/dye labeling efficiencie s were also tested as shown in Figure 5-15. With multiple primary amino groups on the heavy chain, the anti-PTK7 could react with different amounts of Alex a Fluor 488 dyes during the conjugation. By controlling the amount of dye added, three diffe rent conjugates were prepared, having 2, 4 or 6 dyes, respectively, on each antibody. As shown in Figure 5-15, results of the fluorescence quenching experiments were similar to those of the laser intensity experiments. Both of these studies indicated that t he variations in laser source intensity and antibody labeling efficiency have limited effects on the determi nation of the binding site distances. Using Fluorescent Quencher as Energy Acceptors If the fluorescence energy transfer from the antibody donor to the surface of aptamer-gold nanoparticle conjugates was in fact acting as a molecular ruler capable of reporting the distance between the two binding sites, we asked whether it would be possible to replace the gold nanoparticles wi th a fluorescent quencher and still be able to detect the energy transfer. Two different fluorescent quenchers, Bla ck Hole Quencher I (BHQ) and Dabcyl, were chosen to replace the gold nanoparticles to investigate whether it is still be able to detect the energy transfer. As shown in Figure 5-16A, C, an Al exa488-antibody on the antibody binding site and a fluorescent quenc her-linked aptamer sgc8 on the aptamer binding side, respectively, the flow cytom etry results suggested that no significant fluorescence decrease was observed, indica ting that no significant energy transfer occurred from the fluorescent donor to the q uencher on the two binding sites. However,

PAGE 128

128 the colocalization of both aptamer and antibody ligands on the receptor was confirmed by competition studies (s ee Supplementary Materials). Later, a poly-T linker was added between the aptamer and the fl uorescent quencher to reduce the distance from the quencher to the fluorescent donor (Figure 516B, D). However, similar results were obtained, and no significant ener gy transfer was observed, i ndicating that no significant energy transfer occurred from the fluorescent donor to the quencher on the two binding sites. These results once again proved that us ing gold nanoparticles as a surface energy acceptor increased t he probability of energy transf er and accounted for the enhanced efficiency compared to the dipoledipole interactions (FRET). Since nanoparticle surface energy transfer (SET) has a surface and an isotropic distribution of dipole vectors to accept energy from the donor, the energy transfer probability is greatly increased. Conclusion In this paper, we have demonstrated, for the first time, the successful construction of a “SET nanorular” on a liv e cell membrane to measure the protein binding site distances. The distance between the aptamer and antibody binding sites in the membrane protein PTK7 was obtained from the surface of leukemia T-cells (CEM) in the natural physiological environment as (13.41.4) nm, with an error within 10%. No protein separations or purifications were needed. The result also shows that this cell membrane SET nanoruler can measure separa tion distances well beyond the detection distance of FRET. Plus, since the energy acceptor nanoparticle in SET has a surface and an isotropic distribution of dipole vectors to accept energy from the donor, it has a

PAGE 129

129 much lower requirement for interaction or ientations than FRET and thus leads to a wider application for distanc e measurements for various in vitro and cellular systems. On the other hand, the SET system also shows its own advantages for cellular measurements over its counterpart plasmonic rulers which also process long-distance detection capability. The cell surface SET nanor uler adopted a series of different sized gold nanoparticles (5nm-42nm) for the ruler construction, however, only small sized particles can be used to build plasmonic rulers, because the scattering effect only dominates in smaller sized nanoparticles (<2nm) due to the insufficient overlap for fluorescence energy transfer. In contrast, par ticles with greater than a 2nm diameter have higher probability and incidence of fluorescence quenching (like SET)144. Therefore, SET is a more suitable choice for our nanoruler construction with various large sized (6nm-42nm) of gold nanoparticl es. Moreover, high scattering background from the cell surface also si gnificantly reduces the detecti on signal-to-background ratio and therefore prohibits the applications of plasmonic rulers to the cell surface. In summary, “SET nanorulers” have the potential to become an alternative to FRET for molecular interaction and dist ance measurement in cellular systems, especially for applications dem anding long observation times or large distances. It represents the next leap forwar d in use of optical probes to monitor structural components within a cell membrane and will open a new pathway for cellular imaging.

PAGE 130

130 Figure 5-1 Characterizations of different sized gold nanoparticles. (A) UV absorption spectra of different sized gold nanoparticles. (B) The visual colors of different size gold nanoparticles.

PAGE 131

131 Figure 5-2 TEM images of different sizes of gold nanoparticles. The average diameter of NPs for each size was determined by m easuring the size of 100 particles from the TEM images. The red title in each box indicated the mean diameter. Figure 5-3 Ligand binding saturation concentration determinations. A) KD binding curve for ligand Alexa488-antiPTK7 on CEM cells. B) KD binding curves for 5.4, 10.1 and 18.4 nm gold NP-aptamer conjugates on CEM cells. R ed curve: 5.4 nm gold NP-aptamer conjugates; Blue curve: 10. 1 nm gold NP-aptamer conjugates; Purple curve: 18.4 nm gol d NP-aptamer conjugates. Figure 5-4 Competition studies using Cy5-sgc8 to confirm the saturation binding of gold NP-aptamer conjugates on the cell surfac e. (a) Plain cells incubated with 25nM 0510152025 15 20 25 30 35 0510152025 15 20 25 30 35 0510152025 15 20 25 30 35 NP Concentration (nM) Fluorescence Intensity 5.4nm 10.1nm 050100150200250300 30 40 50 60 70 80 Fluorescence IntensityAlex488_antiPTK7 (nM) B A a) Cell only + 25nM Cy5-sgc8 d) Alexa488-antiPTK7 and Gold NP-TDO5 + 25nM Cy5-sgc8 10m c) Alexa488-antiPTK7 and Gold NP-Sgc8 + 25nM Cy5-sgc8 10m 10m b) Alexa488-antiPTK7 + 25nM Cy5-sgc8 10m

PAGE 132

132 Cy5-sgc8 show fluorescence staining of Cy5 on the cell membrane. (b) Cells incubated with Alexa488-antiP TK7 can also be stained by Cy5-sgc8. (c) Cells incubated with Alexa488-ant iPTK7 and gold NP-sgc8 conjugates show minimal fluorescence. For different sized conjugat es, similar results are shown. And here we picked 15nm conjugates as an exampl e. (d) Cells incubated with Alexa488antiPTK7 and the conjugates with control apt amers, gold NP-TDO5, show staining of Cy5-sgc8 on the cell membrane. Figure 5-5 Competition st udies between aptamer sgc8 and antibody anti-PTK7 on receptor PTK7. (A) Flow Cytometry assay to monitor the fluorescence intensities from the Alexa488-labeled anti-PTK7 in the presence and absence of 100X nonlabeled sgc8. The red curve indicates the fluorescence intensity from cell membrane with saturated concentrations of Alexa488-labeled ant i-PTK7 (200 nM) in the absence of non-labeled sgc8. The bl ue curve shows the fluorescence shift in the presence of 100 non-labeled sgc8. The black curve marks the fluorescence background with cells only, and the green curve shows the binding of control antibody FITC-labeled isotype Mouse IgG2a. (B) Flow Cytometry assay to monitor the fluorescence intensitie s from the FITC-labeled sgc8 in the presence and absence of 100X non-labeled anti-PTK7. The red curve indicates the fluorescence intensity from cell me mbrane with saturated concentrations of FITC-sgc8 (200 nM) in the absence of non-labeled anti-PTK7. The blue curve shows the fluorescence shift in the pres ence of 100 non-labeled anti-PTK7. The black curve marks the fluorescence ba ckground with cells only, and the green curve shows the binding of control anti body FITC-labeled isotype Mouse IgG2a. FL1: fluorescence channel 1 in Flow Cytometer at 520 nm. FITC: fluorescein isothiocyanate.

PAGE 133

133 Figure 5-6 Measure the distance betw een two binding sites using FRET. (A) The figure on the top indicates the Flow Cytometr y assay to monitor the fluorescence intensities at 680nm (channel 3) from th e cell membrane with an excitation source at 488nm. The green curve indicate s the fluorescence intensity from the cell membrane with saturated concentrati ons of Alexa488-labeled anti-PTK7 (200nM). The yellow curve shows the fl uorescence intensity with saturated concentrations of Cy5-labeled sgc8 ( 200nM). The red curve indicates the fluorescence intensity in the presence of both Alexa488-antiPTK7 (200nM) and Cy5-sgc8 (200nM) on the cell memb rane. The black curve marks the fluorescence background with cells only, and the blue curve and purple curve show the binding of FITC-labeled contro l antibody FITC-labeled isotype Mouse IgG2a and Cy5-labeled unselected DNA librar y, respectively. The table on the bottom shows the fluorescence intensit ies in both channel 1 (at 520nm) and channel 3 (at 680nm). (B) The figure on t he top indicates the Flow Cytometry assay to monitor the fluorescence intens ities at 580nm (channel 2) from the cell membrane with an excitation source at 488nm. The green curve indicates the fluorescence intensity from the cell me mbrane with saturated concentrations of Alexa488-labeled anti-PTK7 (200nM). T he yellow curve shows the fluorescence intensity with saturated concentrations of TMR-labeled sgc8 (200nM). The red curve indicates the fluorescence intens ity in the presence of both Alexa488antiPTK7 (200nM) and TMR-sgc8 (200nM) on the cell membrane. The black

PAGE 134

134 curve marks the fluorescence background wi th cells only, and the blue curve and purple curve show the binding of FI TC-labeled control antibody FITC-labeled isotype Mouse IgG2a and TMR-labeled uns elected DNA library, respectively. The table on the bottom shows the fluoresc ence intensities in both channel 1 (at 520nm) and channel 2 (at 580nm). Channel 1: excitation 488, emission: 520nm; Channel 2: excitation 488, emission: 580nm; Channel 3: excita tion 488, emission: 680nm. FITC: fluorescein isothiocyanate; TMR: Tetramethylr hodamine; Cy5: Cyanine 5. Figure 5-7 Scheme of “SET nanoruler” for measuring the distance between two binding sites in receptor PTK7 on a live cell me mbrane. The yellow obj ect represents a PTK7 receptor in the lipid bilayer of the cell membrane, with two binding sites on its extracellular domain. The blue moiety represents one of th e receptor ligands, anti-PTK7. The Alexa Fluor 488 dye is labeled on its heavy chain through its primary amino groups. On t he other side, the red sphere represents a gold nanoparticle. Multiple sgc8 apt amers (black) with thiol l abeling are used to modify the surface of the gold nanoparticle. The aptamer-gold NP conjugate is brought to the aptamer binding site on the rec eptor through aptamer-receptor binding. The colocalization of both ligands on the receptor brings the Alexa Fluor 488 dye on the antibody close to the gold NP on the aptamer binding site. When the donor dye molecule and acceptor NP surf ace reach close proximity, quenching of fluorescence from the cell surface results. A series gold NPs of different sizes

PAGE 135

135 (5.4nm-42.2nm) were used in these energy transfer experiments. “R” represents the distance between the Al exa Fluor 488 dye on the anti-PTK7 heavy chain to the center of the gold NP, and “d” is the di ameter of the gold NPs; therefore, “d/2” is the distance from the c enter of the gold NP to it s surface, and “r” represents the distance from the Alexa Fluor 488 dye on anti-PTK7 to the su rface of the gold NP. Figure 5-8 (Top) Flow cytometry assay to monitor the fluorescence intensities on live cells. A) Fluorescence intensity from the Alexa488-labeled anti-PTK7 in the presence of varying sizes of gold NPaptamer conjugates on live cell membrane was monitored using a Flow cytometer. On each frame, the red curve indicates

PAGE 136

136 the fluorescence intensity from cell me mbrane with saturated concentrations of Alexa488-labeled anti-PTK7 with no gold NP-aptamer conjugate. The red arrow indicates the mean intensity. The blue cu rve shows the fluorescence shift in the presence of gold NP-sgc8 conjugates. The gap between the blue and red arrow indicates that the fluorescence intensity decreased with the increasing size of the gold NP conjugates. The green curve repres ents the fluorescence intensity in the presence of the same-sized gold NP, but the NP was conjugat ed with a control aptamer sequence, TDO5, which does not bi nd the receptor PTK7. No significant fluorescence shift was shown for the control. The black curve marks the fluorescence background with cells only, and the yellow curve shows the binding of control antibody FITC-labeled isotype Mouse IgG2a. (Botto m) Histogram of the mean fluorescence intensity for the fluorescence quenching assay determined from the flow cytometry re sults. All of the experiments for the fluorescence quenching assay were repeated three times, and the average value was determined as the mean fluorescence intensity. Cell only Alex488_antiPTK7 only a1) 5.4nm NPsgc8 a2) 10.1nm NPsgc8 a3) 13.3nm NPsgc8 a4) 15.4nm NPsgc8 b1) 5.4nm NPTDO5 b2) 10.1nm NPTDO5 b3) 13.3nm NPTDO5 b4) 15.4nm NPTDO5 a5) 18.4nm NPsgc8 a6) 20.4nm NPsgc8 a7) 25.5nm NPsgc8 a8) 31.2nm NPsgc8 b5) 18.4nm NPTDO5 b6) 20.4nm NPTDO5 b7) 25.5nm NPTDO5 b8) 32.1nm NPTDO5 a9) 42.2nm NPsgc8 b9) 42.2nm NPTDO5 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m 100 m

PAGE 137

137 Figure 5-9 Confocal imaging assay for mo nitoring the fluorescence quenching on cell surface with different sizes of gold nanoparticles. The two box es framed in red show the fluorescence images of CEM ce lls with saturated concentration of Alexa488-labeled anti-PTK7 only and fl uorescence background with cells only. Boxes (a1)-(a9) (with yellow titles) show the fluorescence of cells in the presence of different sizes of gold NP-sgc8 conj ugates. Boxes (b1)-(b9) (with white titles) indicate the fluorescence with the contro l aptamers TDO5 and the same sizes of gold NPs as those shown in boxes (a1)-(a9) Vertical comparisons of same sizes of gold NPs with aptamer sgc8 and cont rol sequence TDO5 clearly show the difference in the fluorescence quenching efficiency. Figure 5-10 Relationship between fluorescence quenching efficiency and gold nanoparticle diameter. The fluorescence quenching efficiency was determined from the quantitation of the fluorescence intensity from the Flow Cytometry Analysis (Fig. 1B). The fl uorescence quenching efficiency was determined from147: 0 00(7_8) 11 7 II I AntiPTKNPsgcCell IIAntiPTKCell in which I0=fluorescence intensity in the absence of the quencher (gold NP-aptamer) (i.e., cells with Alexa488-antiPTK7 only); I=fluorescence intens ity in the presence of both Alexa488-antiPTK7 and the quenc her (gold NP-aptamer).

PAGE 138

138 Figure 5-11 Binding site distance determination. A) By plotting 1 41 1 vs. nanoparticle diameter d, a linear relationship was obtained between d=5.4nM to d=18.4nM, and it reached a plateau with d larger than 18.4nM (d= 18.4nM-42.2nM). B) The expanded plot of linear region.

PAGE 139

139 --CEM cell only --FITC-IgG2a (Control) --Alex488-antiPTK7 200nM --Alex488-antiPTK7 200nM + 15nm SiNP-sgc8 4nM --Alex488-antiPTK7 200nM + 15nm SiNP-TDO5 4nM Events FL1 100 101 102 103 104 Figure 5-12. Flow cytometry analysis to mo nitor the fluorescence quenching effect with 15-nm silica nanoparticles. The red curve indicates the fluorescence intensity from cell membrane with saturated concent rations of Alexa488-labeled anti-PTK7 in the absence of the SiNP-aptamer conjugates. The blue curve shows the fluorescence shifts in the presence of the 15-nm SiNP-sgc8 conjugates, and the green curve represents the fluorescence in tensity in the pres ence of the 15-nm SiNP-TDO5 conjugates, the control aptamer sequence. No significant fluorescence shiftback was observed in the presence of either SiNP-sgc8 or SiNP-TDO5, compared to the red curv e with the Alexa488labeled anti-PTK7 only. The black curve marks the fluore scence background with cells only, and the purple curve showed the binding of cont rol antibody FITC-labeled isotype Mouse IgG2a.

PAGE 140

140 Figure 5-13. Competition studies using Cy5-sgc8 to confirm th e saturation binding of silica NP-aptamer conjugates on t he cell surface. (a) Cells incubated with Alexa488antiPTK7 and 15nm SilicaNP-sg c8 conjugates show minimal fluorescence. (b) Cells incubated with Alexa488-antiPTK7 and the conjugates with control aptamers, silica NP-TDO5, show staining of Cy5-sgc8 on the cell membrane. a) Alexa488-antiPTK7 and 15nm SiNP-sgc8 + 25nM Cy5-sgc8 50m 50mb) Alexa488-antiPTK7 and 15nm SiNP-TDO5 + 25nM Cy5-sgc8

PAGE 141

141 Figure 5-14 Effects of different laser intens ities on fluorescence quenching efficiency. A) Histogram of the fluorescence quenching assay with different laser excitation intensities (Red: 750V; Green: 700V; Blue: 650V) from flow cytometry analysis. Experimental procedures were the same as described in Fig. 1. In all these experiments, the labeling efficiencies of Alexa Fluor 488 dyes on the anti-PTK7 were kept the same, and there were four dyes on each antibody according to the labeling protocol. All of the experiments for fluorescence quenching assay were repeated three times, and the average va lue was determined as the mean fluorescence intensity. B) Relati onship between fluorescence quenching efficiency and gold nanoparticle diameter. Fluorescence quenching efficiencies were determined from fluorescence intens ity in Fig. 6A using the following

PAGE 142

142 equation147: 0 00(7_8) 11 7 II I AntiPTKNPsgcCell IIAntiPTKCell Inside the red rectangle, a linear relationship wa s obtained between the fluorescence quenching efficiency and gold nanoparticle diameter (d=5.4nm-18.4nm) for different laser intensities. Slopes were equal within experimental error (650V, (0.0770.0072); 700V, (-0.0720.0085); 750V (-0.0730.013)). C) Plot 1 41 1 vs. NP diameter gave linear relati onships between d=5.4nm-18.4nm (red rectangle region in (B)). The linear fits for different laser intensities resulted in similar distance R results (red: R=(13. 431.40) nm; green: R=(13.741.82) nm; blue: R=(13.562.73) nm). 0.0 0.2 0.4 0.6 0.8 1.0 Normalized Fluorescence Intensity Alex488 (6) 700V Alex488 (4) 700V Alex488 (2) 700VC e ll o n l y F I T C I g G 2 a A n t i P T K 7 o n l y + 5.4 n m N P + 10 1n m N P + 13 .3n m N P + 1 5 .4 n m N P + 18 4n m N P + 2 0. 4 n m N P + 2 5 .5 n m N P + 31. 2n m N P + 42 .2n m N PA

PAGE 143

143 051015202530354045 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Alex488 (6) 700V Alex488 (4) 700V Alex488 (2) 700VNP Diameter (nm) 1 41 1 B 468101214161820 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Alex488 (6) 700V Alex488 (4) 700V Alex488 (2) 700VNP Diameter (nm) Red: Y=(-0.080 0.014)X+(2.14 0.18) Grey: Y=(-0.083 0.011)X+(2.17 0.15) Green: Y=(-0.082 0.0077)X+(2.15 0.11) 1 41 1 C Figure 5-15 Effects of different antibody labeling efficiencies on fluorescence quenching. A) Histogram of the fl uorescence quenching assay with different antibody labeling efficiencies (red: 6 Alexa Fluor 488 dyes on each anti-PTK7; grey: 4 dyes per anti-PTK7; green: 2 dyes per anti-PTK7) from the flow cytometry analysis. The different labeling efficienc ies for anti-PTK7 were achieved by incubating different amounts of Alexa Fluor 488 dyes with anti-PTK7 during the labeling procedure and were calculated and determined according to the UV quantification (see Methods). Fluoresc ence quenching experiment procedures were the same as described in Figure 1. Laser excitation at 488nm was constant at 700V. All of the experiments for the fluorescence quenching assay were repeated three times, and the average va lue was determined as the mean fluorescence intensity. B) The rela tionship between fluorescence quenching

PAGE 144

144 efficiency and different anti-PTK7 dye l abeling efficiencies was determined from fluorescence intensity in Figure 8A using equation4: 0 00(7_8) 11 7 II I AntiPTKNPsgcCell IIAntiPTKCell A linear relationship was obtained between the fluorescence quenching efficiency and gold nanoparticle diameters (d=5.4nm-18.4nm) for different labeling efficiencies. C) Plot of 1 41 1 vs. NP diameter gave linear rela tionships between d=5.4nm-18.4nm. The linear fits for different labeling effi ciencies resulted in similar distance R results (red: R=(13.382.59) nm; grey: R=(13.061.96) nm; green: R=(13.121.40) nm). --CEM cell only --FITC-IgG2a --Alex488-AB 200nM --BHQ-sgc8 --BHQ-10T-sgc8 --Dabcyl-sgc8 --Dabcyl-10T-sgc8 --Alex488-AB 200nM --BHQ-sgc8 --Alex488-AB 200nM --BHQ-10T-sgc8 --Alex488-AB 200nM --Dabcyl-sgc8 --Alex488-AB 200nM --Dabcyl-10T-sgc8AB CD Figure 5-16 Flow cytometry analysis using different fluorescence quenchers as energy acceptors. (A) Aptamer sg c8 was labeled with a BHQ functional group on its 5Â’ end and used as an energy acceptor on t he aptamer binding site (dark blue curve). (B) Same as (A) with a 10-polyT linker between the aptamer sgc8 sequence and the BHQ functional group (light blue curve) (C) Apta mer sgc8 was labeled with a Dabcyl functional gr oup on its 5Â’ end and used as an energy acceptor on the aptamer binding site (g reen curve). (D) Same as (C) with a 10polyT linker added between the aptamer sgc8 sequence and the Dabcyl functional group (purple curve). Experimen tal procedure was similar to Figure 1. The fluorescence was determined with a FACScan cytometer (BD

PAGE 145

145 Immunocytometry Systems) by counti ng 10,000 events. Laser excitation at 488nm with 750V excitation intensity was used as the excitation source. The FITC-labeled isotype Mouse IgG2a antibod y was used as a negative control to determine nonspecific binding (yellow cu rve). Cell fluorescence background was determined from samples with cells only as the fluorescence background for later calculation (black curve). Table 5-1. Summary of Properties for Gold Nanoparticles of Different Sizes (50mL 0.01% HAuCl4 for each preparation). Volume of Sodium Citrate (1%, mL) Average diameter (nm)b Color max (nm) A N/Aa 5.4 1.0 Brownish Red 517 B 4.0 10.1 1.6 Red 519 C 3.5 13.3 1.5 Red 520 D 3.0 15.4 2.3 Red 519 E 2.0 18.4 4.6 Red 520 F 1.5 20.4 4.3 Red 520 G 1.0 25.4 5.9 Red 522 H 0.85 31.2 5.7 Pinkish Red 524 I 0.75 42.4 8.4 Pinkish Red 532 a. 5.4nm NPs were purchased from Sigma-Aldrich, St. Louis, MO. b. The average diameter of NPs for each size was determined by measuring the size of 100 particles from the TEM images. Table 5-2. Summary of DNA sequences. Name DNA Sequences sgc8 5Â’ -ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-FITC 5Â’ ( FITC )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-TAMAR 5Â’ ( TAMAR )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-Cy5 5Â’ ( Cy5 )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-S-S 5Â’ ( S-S-C6 linker )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-NH2 5Â’ ( NH2-C6 linker )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-BHQ 5Â’ ( BHQ )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-Dabcyl 5Â’ ( Dabcyl )-ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-10T-BHQ 5Â’ ( BHQ )TTTTTTTTTT ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ sgc8-10TDabcyl 5Â’ ( Dabcyl )-TTTTTTTTTT ATC TAA CTG CTG CGC CGC CGG GAA AAT ACT GTA CGG TTA GA3Â’ Lib-FITC 5Â’ ( FITC )-NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NN3Â’

PAGE 146

146 Lib-TAMAR 5Â’ ( TAMAR )NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NN 3Â’ Lib-Cy5 5Â’ ( Cy5 )NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NN 3Â’ TDO5-S-S 5Â’ ( S-S-C6 linker )ATC CAG AGT GAC GCA GCA GAT CAG TCT ATC TTC TCCTGA TGG GTT CCT AGT TAT AGG TGA AGC TGG ACA CGG TGG CTT AGT3Â’ TDO5-NH2 5Â’ ( NH2-C6 linker )-ATC CAG AGT GAC GCA GCA GAT CAG TCT ATC TTC TCCTGA TGG GTT CCT AGT TAT AGG TGA AGC TGG ACA CGG TGG CTT AGT3Â’ Table 5-3. Summary of aptamer-func tionalization for gold nanoparticles. Diameter of gold NPs (d, nm) Estimated surface area of one gold NP (S= d2) (nm2) Concentration of gold NP stock solutions from synthesis (nM, 50 ml) Volume of aptamer solution added to 50 ml gold NP stock solution (l, 1 mM) Total aptamer added : Gold NPb A 5.4 91.6 62.2a 134.0 431:1 B 10.1 320.3 9.5 71.7 1509:1 C 13.3 555.4 4.2 54.5 2595:1 D 15.4 744.7 2.7 47.0 3481:1 E 18.4 1063.1 1.6 39.5 4938:1 F 20.4 1306.8 1.2 35.5 5917:1 G 25.4 2025.8 0.6 28.5 9500:1 H 31.2 3056.6 0.3 23.0 15333:1 I 42.4 5645.0 0.1 17.0 34000:1 a The 5.4 nm gold NP stock solution was purchased from Sigma-Aldrich (St. Louis, MO). b The ratios of total aptamer added to gold NPs were calculated from the estimations of surface areas of different sizes of gold NPs to ensure saturated surface coverage.

PAGE 147

147 CHAPTER 6 ENGINEERING A LIGHT-SWITCHING MOLECULAR BEACON (MB) FOR RIBONUCLEASE H KINETIC STUDY Introduction RNase H is a ribonuclease that specia lly degrades the RNA strand in a RNADNA hybrid to produce 3'-hydroxyl and 5'phosphate terminated products. It is a nonspecific endonuclease and catalyzes the cleavage of RNA via an endonucleolytic mechanism 148, aided by an enzyme-bound divalent me tal ion. However, DNA strands or unhybridized RNA strands will not be degraded. The enzyme is involved in several important cellular processes including DNA r eplication, DNA repair and transcription. 149 Members of the RNase H family can be found in nearly all organism s, from archaea and prokaryota to eukaryota. RNase H also has wide applications in molecular biology and biotechnology in terms of it s unique cleaving property. Retrov iral RNase H, a part of the viral reverse transcriptase enzyme, is an im portant pharmaceutical target, as it is absolutely necessary for the proliferation of retr oviruses, such as HIV. Inhibitors of this enzyme could therefore provide new drugs against diseases like AIDS. E. coli RNase H usually requires at least 6 base pairs of RNA -DNA hybrids as substrates to bind and cleave effectively in solution, while the hy brid length required in living cells may be somewhat greater. 150 The complete digestion of poly (rA):poly(dT) with E. coli RNase H yields oligoribonucleotides with varying chain lengths, ranging from monomers to hexamers. 151 In order to understand more about t hese functions and processes, and more importantly, to screen new drugs against retroviruses, it is necessary to develop a fast, real-time, sensitive and isotope-label free system to assa y the cleavage activity of RNase H. A number of traditional met hods have been used to assay the enzyme

PAGE 148

148 activities and evaluate the kinet ic parameters, such as the acid soluble release of RNA fragment 152, gel electrophoresis 151 and HPLC 153. Acid soluble and gel electrophoresis techniques require radioi sotope-labeled substrates, and the HPLC method needs micromolar concentration for substrate. All of these methods are indirect, discontinuous and time-consuming. Recently, an RNA-DNA deplex is incorporated in a fluoresent probe for Rnase H real time study. However, only very short substrate sequences can be analysized. And the complexity in the RNA -DNA oligonucleotide synthesis may also result in a low yield and restrict its gener al utility in enzyme activity studies. 154 Here we describe a real-time fluor escence method in which the signal transduction was achieved by taking adv antage of the light-switching excimer mechanism inherent to molecular beacons (MBs). Molecular beacon is a singlestranded DNA that can form an intramolecular hairpin structure with a fluorophore and quencher at either end. 14 DNA MB assays have been described for a few enzyme studies, such as single-stranded specific DNases, endonuclease BamHI and small nonenzyme DNA cleavage agents 155-157. The enzyme activities have been detected and characterized by taking advantages of the si gnal transduction mechanism built into the MBs. The change in fluorescence signals reflec ts the conformational change of the MBs, which is a result of the cleavage by the enzymes. 156 Most of these beacon designs are based on fluorescence quenching 158 and fluorescence resonance energy transfer (FRET) 158. Though each technique has its own advantages, some limitations remain. For example, a quenching-based FRET mole cular probe always has incomplete quenching, resulting in a si gnificant probe background. 159

PAGE 149

149 Experimental Section Materials: The sequences of DNA and RNA oligon ucleotides prepared are listed in Table 6-1. DNA synthesis reagents were pur chased from Glen Research (Sterling, VA). Two difference sequences of DNA MBs (excim er probes both with 6 base pairs in stem, sequences shown in Table 6-1) were l abeled at both ends with pyrene. The RNA sequences were synthesized and purified by Sigma-Proligo (The W oodlands, TX, USA.). E. coli RNase H with (enzyme activity of 2000 units/ml (240 nM) was purchased from Sigma-aldrich, Inc, where one unit of RNa se H hydrolyzes 1.0 nanomole RNA in 3Hlabeled poly(dA):poly(dT) to acid soluble material in 20 min at 37 C. The enzyme specific activity is 419,972.4 units per mg of E. coli RNase H. 154 The RNase H inhibitor, EDTA was purchased from Fisher Scientific International Inc. (H ampton, NH, USA). The calibration dye SYBR-Green was purchased fr om Invitrogen (Carlsbad, CA, USA). Instruments: An ABI3400 DNA/RNA synthesizer (A pplied Biosystems) was used for DNA synthesis. Probe purification was perform ed with a ProStar HPLC (Varian) where a C18 column (Econosil, 5U, 250 4.6 mm) from Alltech Associates was used. UV-Vis measurements were performed with a Cary Bio-300UV spectr ometer (Varian) for probe quantitation. Steady-state fluorescence m easurements were performed on a FluorologTau-3 spectrofluorometer (Jobin Yvon, Edison NJ). For emission spectra, 340 nm was used for excitation. Pyrene-MB Synthesis and Purification: A solid-phase synthesis method was used to couple pyrene to MB sequences at both 3Â’ and 5Â’ ends. The synthesis started with a 3Â’amino-modifier C7 controlled pore glass (C PG) column at 1-mol scale. After the synthesis of the DNA beacon sequence, a 5Â’-amine was added to the sequence by

PAGE 150

150 using 5Â’-amino-modifier-C6 linker phosphoramid ite. The column then was flushed slowly with dimethylformamide (DMF) (15 ml), piperidine (20%) in DM F (15 ml), trichloroacetic acid (3%) in dichloromethane (15 ml), and then another DMF (15 ml). The CPG contained within the column was released into DMF solution (1 ml) containing pyrene butyric acid (57.7 mg, 200 mol), dicyclocarbodiimide ( 41.3 mg, 200 mol), and dimethylaminopyridine (24.4 g, 200 mol). After stirring for 3 h, the solution was centrifuged, and the supernatant was discar ded. The pellet was washed three times with DMF, methanol, and water, respective ly, before incubated in a solution of methylamine (50%) in ammonia at 65C for ~10 min. The resulting clear and colorless supernatant was collected. U nder UV radiation, an intense green fluorescence was observed from the collected solution. The beacon solution was desalted with a Sephadex G-25 column (NAP-5, Amersham Pharmacia) and dried in a SpeedVac. The dried product was purified by HPLC using a C18 column with a linear elution gradient with buffer B changing from 25% to 75% in 25 min at a flow rate of 1 ml/min. The second peak in chromatography that abs orbed at 260 and 350 nm, and emitted at 485 nm with 350 nm excitation, was collected as the product. The collected product then was vacuum-dried, desalted with a G-25 colu mn, and stored at -20C for future use. 7 Pyrene Beacon Assays for RNase H Studies: Assays were carried out in 100l of USB RNase H buffer (TRIS-HCl (20mM, pH7.5), KCl (20mM), MgCl2 (10mM), EDTA (0.1mM), and DTT (0.1mM)) 154, containing RNase H (6 nM, 50units/ml) and DNA beacon: RNA hybrid (100nM) at 1:1 ratio. An increase in fluorescence emission at 485nm, upon excitation at 340nm, indicates the hy drolysis progress of the hybrids. The maximum fluorescence emission was determi ned by incubating the hybrids with excess

PAGE 151

151 RNase H. For determination of the Michae lis-Menten kinetic parameters, the beacon assay concentration was varied from 25nM to 500nM, in the range of the previously reported Km values for RNase H. 160, 161 Dye calibration curve using SYBR GreenER™ is applied to determine the accu rate concentration of the hybrid. SYBR GreenER™ is a double-strand oligonucleotide binding dye that can quantitatively differentiate duplex from single strand oligonucl eotides after hybridization. In all kinetic experiments, initial rate measurements we re obtained in the first 30 second, with an enzyme concentration (6nM, 50units/ml). All experiments were done at 37C and repeated for 2-3 times. Average values were us ed for calculation. Data in Table 6-2 and Table 6-3 were obtained by performing curv e fitting to the Michaelis-Menten equation using OriginPro 7.0 (Microcal Software Inc.). Results and Discussion Here, we describe the molecular engineer ing of a light-switch ing excimer beacon probe for RNase H activity monitoring. Some spatially sensitive fluorescent dyes, such as pyrene 15, 160, 162, 163 and BODIPY Fl 161, 164 can form excimer upon proximity of an excited-state molecule with another ground-state molecule. The excimer results in an emission wavelength shift to a longer wa velength compared to the monomer. The formation of excimer between two pyrene mole cules that are connected by a flexible covalent chain, such as a DNA chain, is highly useful to probe spatial arrangement. Similar to FRET, the spatially dependent pr operty of excimer formation can be used as a signal transduction in the development of effective molecular probes. This unique technique is especially useful for the desi gn of MBs, which can undergo conformational change upon target binding. By attaching pyrene molecules 162 to both ends, an excimer

PAGE 152

152 switching MB probe has been developed. In t he absence of target molecules, the DNA beacon stays in its closed form and forms a l oop-stem hairpin structure, which brings the two pyrene moieties to close proximity and allows the formation of an excimer that emits at ~485nm. When binding to the comple mentary DNA or RNA, the MB opens up. Both pyrene molecules are spatially s eparated and only the monomer emission peaks (at 378nm and 397nm) are observed. The change in emission wavelength serves as a unique real-time method to trac e a series of conformational changes of the DNA probe based assays162. This emission wavelength switching solves the probe background signal problem that occurs with FRET mole cular probes. So it can help improve the precision of the kinetic param eter quantitatio n which also takes the background signal into account. Design of Light-Switching MB A 31mer light-switching pyrene beacon a ssay has been designed for the realtime study of RNase H activity with high se nsitivity. (Sequences shown in Table 6-1) The mechanism to monitor the activity is shown schematically in Figure 6-1. A dualpyrene-labeled DNA beacon is free in solu tion without the target binding. Both the pyrene moieties are brought to close proximit y by the beacon hairpin structure, allowing the formation of an excimer. The excimer emission at 485 nM has a fluorescence enhancement of about 37-fold, as shown in Fi gure 6-2. The bindin g of the DNA beacon to its RNA target leads to the opening up of the hairpin structure and thus spatially separates the pyrene moieties from each other. The fluorescence emission will have a blue shift and give the monomer emission at 378 nm and 397 nm. As we monitor the excimer signal in the whole process, this RNA-DNA hybrid, which serves as the

PAGE 153

153 substrate for the RNase H cleavage, has a low fluorescence background at 485nm (Figure 6-2). After the addition of the enzyme, only the RNA strand will be cleaved from the duplex 148, which sets free the DNA beacon. The re storation of the hairpin structure brings the pyrene moieties back toget her and gives a dramatic fluorescence enhancement at the excimer emission at 485nm, and the real-time fluorescence monitoring is shown in Figure 6-3. This crit ical step is the one we used for the RNase H kinetic parameters study. Comparing to normal fluorophore-quencher labeled MB design, whose fluorescence signal will be quenched at this step; the pyrene beacon assay results in a fluorescence increase and thus has a much higher sensitivity. It brings a lot more convenience to the calc ulation with the signal enhancement in the design. Also, Scheme 5-1 reveals another im portant advantage of the light-switching excimer signaling approach over traditional gel electrophoresis experiments: real-time detection without separation. Because only the cleaved duplex give s excimer emission, the uncleaved targets do not have to be separ ated from the solution for detection. In addition, since it is a real-time detection, the fluorescence assay gives us a much clearer picture for the enzym e cleavage activity and also minimizes the inconvenience of being discontinuous in experim ent manipulation. In order to confirm that the signal enhancement comes from the cleavage of the RNA strand by the RNase H, a complementary DNA is added to the soluti on after the enzyme cleavage. The free beacon hybridizes with the complementar y DNA and opens up its hairpin structure again as shown by the dramatic decrease in excimer emission. (Figure 6-3) Optimization of cDNA lengthes

PAGE 154

154 The oligonucleotide targets for norma l MBs have the complementary sequence to the loop sequence of the beacons. Howe ver, sometimes beacons can exhibit a substantial amount of intermolecular interactions as a result of sticky-end pairing of the beacon stems in the presence of target nucleic acids. 8 Two complementary sticky ends from two beacon hybrids can pair to form a s hort double helix, leading to association of the two hybrids at one end. With sticky-end pairing, two separated pyrene molecules are drawn back together again, causing high background in excimer emission and false negative results in the enzyme activity study, as shown in Figure 6-4. This problem is more severe when the probe concentrations are high in solution. In order to avoid this sticky-end pairing problem, shared stem oli gonucleotide targets ar e used in the pyrene beacon assays design. We have designed two di fferent sequence of shared stem cDNA to test the fluorescence signal for the optimization of the target RNA sequence. (Sequence shown in Table 6-1) We find that the sequence with a G base on the 5Â’ end gives a lower background at the excimer em ission than the one which has an A base. According to literature 165, this is because that the G base on the 5Â’ end of the DNA strand can partially quench t he fluorescence of the fluorophor e due to energy transfer. We will use this sequence for the design of the RNA stand in our kinetic studies. Moreover, using shared stem RNA to solve the sticky-end pairing problem also has the potential to increase the assay concentration from 20nM to 100nM in order to obtain a higher fluorescence signal. Light-Switching MB Assay fo r Ribonulcease H Ki netic Study Then the excimer beacon assay was used to obtain the kinetic parameters for E. coli RNase H monitoring the cleavage of t he RNA strand from t he RNA-DNA hybrid.

PAGE 155

155 Figure 6-5 shows that the cleavage of RNA is enzyme concentration dependent, indicating that the excimer beacon assay c an be used for RNase H detection over a range of enzyme concentration. The initial ra te of hydrolysis can be used for RNase H kinetic studies. The kinetic parameters of RNase H for the 25bp RNA-DNA hybrid are Km=0.019 M and kcat =0.25 s-1. The details of how thes e parameters are obtained are shown in Figure 6-6. The values are similar to those radioisotope labeled duplexes with similar length of base pairs (14 mer: Km=0.08 M and kcat =0.1 s-1154; 22 mer: Km=0.07 M and kcat =1.1 s-1151; 24 mer: Km=0.02 M) (Shown in Table 6-2) using the gel electrophoresis methods. The variation in Km and kcat is the result of varied assay conditions, substrates and methods. Prev ious reported methods such as gel electrophoresis and HPLC analysis resulted in a wide range for the RNase H kinetic parameters using RNA -DNA hybrids. Sequence Dependence on Ri bonuclease H Cleavage Kinetics In addition, early observation confirms that the cleavage sites of RNase H are not sequence dependent. 151 However, the enzyme does exhibit sequence dependence effect on the kinetic properties of the cleavage activity. 154 We synthesized a 25mer pyrene beacon, MBS1 (Sequence shown in Tabl e 6-1), which has the same GC content as the MB226 beacon in the whole hybrid bu t much higher GC content in stem. The beacon has also been applied to study the kinetics. While similar results have been obtained due to the same length of the hybrids, it suggests a slight increase in Michaelis constant (Km=0.031M) and turnover rate (kcat=0.33 s-1) for the MBS1 assay. (Comparison shown in Table 6-2) It suggests that the kinetic characteristics of the enzymeÂ’s cleavage activities show slight s equence preference. And it is most likely due

PAGE 156

156 to the only difference between these two assays – GC contents in beacon’s stem sequence. A larger Michaelis constant means a higher ratio between the velocities of enzyme-substrate complex de gradation and formation. Higher GC content in stem makes the product after cleavage more stable in the beacon’s hairpin structure, which leads to a higher complex degradation rate. The results indicate that E. coli RNase H does have slight sequence pref erence and slightly prefer t he substrate with higher GC contents in stem in the pyrene beacon assay. Conclusion In summary, with its unique properties, the dual-pyrene-label pyrene beacon is finding interesting applications in enzyme acti vity studies. Integrated with a novel signal transduction mechanism, the binding and clea ving elements can be used as sensitive probes for enzymatic monitoring and kinetic analysis. We have demonstrated that the light-switching excimer approach is an excellent signal transduction for MB development with specific detection purpose. The switching of the excimer signal indicates that the conformational change of the pyrene beacon can be used to monitor the enzymatic activity. This signaling approach has its meri ts for application in kinetic analysis for several reasons. First, the real-time det ection via monitoring the signal transduction corresponding to conformational change gives a real-time portray of what is happening in the reaction. Without the laborious work for stopping reaction and taking out samples every few minutes such as in gel electrophore sis methods, it can avoid any indirect and time-consuming problems. Also free from any need to stop r eactions in the middle by any inhibitors, it can give a more prec ise response only coming from the cleavage activity of the enzyme. Second, shorter detection time is needed. Real-time monitoring

PAGE 157

157 can avoid any other detection and analysis steps afterwards 166, so it shortens the detection time for each sample and minimizes the environmental effect to afford more precise detection. The real-t ime light-switching assays we have developed here show great advantages in these aspects for its simple, rapid method, easy construction and high sensitivity. These properti es will enable useful applications to construct specially designed light-switching pyrene beacons for differ ent types of enzyme activity studies in extracellular or even in tracellular environments. Figure 6-1 Schematic repres entation of the fluorescence mechanism using lightswitching pyrene assay for RNase H activity study. A DNA beacon (black) is endlabeled with pyrene molecules (green) that are brought close to each other by the beaconÂ’s loop-stem hairpin structure. The two pyrene molecules have an excimer emission in the green light range ~485 nm a). Upon binding to its target RNA (red), the DNA beacon forms a hybrid wit h the RNA strand and opens its hairpin structure. The two pyrene molecules are s patially separated, which results in the monomer emissions at ~378nm and 397nm in the blue light range. b). The addition of RNase H results in the specif ic cleavage of the RNA strand from the RNA-DNA hybrid. The restor ation of the hairpin st ructure in the DNA beacon brings the two pyrene molecules closed bac k together to form an excimer again. c). Complementary DNA (purple) is added to form a duplex with the DNA pyrene beacon to confirm that the enzyme just cleaved the RNA strand but did not

PAGE 158

158 degrade the DNA beacon. Monomer emi ssion for the pyrene molecules is detected again in this duplex structure. 360380400420440460480500520540560 0 5 10 15 20 25 30 35 40 Fluorescence IntensityWavelength (nm) Bufffer MB (100nM) MB: cDNA (1:100) MB: RNA after annealing (1:1) Figure 6-2 Steady-state fluorescence spectra of MB226 pyrene beacon assay for RNase H activity study. Excitation at 340nm was used for all spectra. The red line represents the fluorescence emissi on from the free pyrene beacons with 100nM concentration in solution before target binding. There is a ~37-fold signal to background fluorescence enhancement at 485nm emission. The blue line represents the fluorescence emission after the hybridization of beacon to target RNA by annealing at 1:1 ratio with an 8fold decrease compared to the free beacons. The green line is the emission spec trum of the beacon: cDNA duplex at 1:100 ratio. A dramatic decrease in ex cimer wavelength (~485 nm) happens after hybridization, and a corresponding increase in monomer signals (378 nm and 397 nm) is also shown.

PAGE 159

159 Figure 6-3 Time-base fluorescence monitori ng of RNase H cleavage activity in MB226 assay. While monitoring the excimer signal in the system, the addition of RNase H induces the fluorescence enhancem ent due to the cleavage of the RNA strands from the pyrene beacons. After reaching plateau, shared stem cDNA is added to hybridize with the free beacon, wh ich results in a dramatic decrease in excimer emission. This reaction is show n to confirm the cleavage mechanism of the pyrene beacon assays. Reaction condi tions: [RNA-DNA hybrid] =100 nM, [RNase H] =6 nM (50 units/ml), [cDNA] =10 M. Excitation wavelength: 340 nm; emission wavelength: 485 nm (excimer emission). 360380400420440460480500520540560580600620 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 Fluorescence IntensityWavelength (nm) Buffer Pyrene-MB (20nM) Pyrene-MB + loop-cDNA (1:100)

PAGE 160

160 Figure 6-4 Steady-state fluorescence spectra of pyrene beacon (20 nM) with loop-cDNA (1:100) for optimization of RNA sequence in MB226 assay. The duplex of pyrene beacon and loop-cDNA gives a high background signal after hybridization due to the sticky-end pairing problem. 02004006008001000120014001600 0 20000 40000 60000 80000 100000 120000 140000 5 units/ml 15 units/ml 22.5 units/ml 30 units/ml 50 units/ml 70 units/mlFluorescence IntensityTime (s) Figure 6-5 Time curves of cleavage of RNA strand from MB226 assay by E. coli RNase H at different enzyme concentration at 37 Reaction conditions were: [RNADNA hybrid]=100 nM; [RNase H]=0.6 nM (5 units/ml), 1.8 nM (15 units/ml), 2.7 nM (22.5 units/ml), 3.6 nM (30 units/ml), 6 nM (50 units/ml), 8.4 nM (70 units/ml); 1 unit of E. coli RNase H=1.210-13 mol.

PAGE 161

161 Figure 6-6 The Linear weaver-Burk plot of re ciprocals of initial rates versus substrate concentration for the determinat ion of kinetic parameters Km, kcat and Vmax of RNase H in MB226 assay. From the plot, we got Km =0.019 M, Vmax =1.65 nM/s and kcat =0.25 s-1. Reaction conditions: [RNA-DNA hybrid] =25 nM, 50 nM, 75 nM, 100 nM, 200 nM, 300 nM, 400 nM and 500nM. [RNase H]=6 nM (50 units/ml). Excitation at 340 nm; emission at 485 nm. Table 6-1. Probes and oli gonucleotides used in Ri bonuclease H kinetic study Name Sequence MB226 5Â’ Pyr CCT AGC TCT AAA TCA CTA TGG TCG C GC TAG G Pyr 3Â’ MB226 shared stem cDNA-A 5Â’ CCT AGC GCG ACC ATA GTG ATT TAG A 3Â’ MB226 shared stem cDNA-G 5Â’ GCG ACC ATA GTG ATT TAG A GC TAG G 3Â’ MB226 shared stem RNA 5Â’ GCG ACC AUA GUG AUU UAG A GC UAG G 3Â’ MBS1 5Â’ Pyr CGC ACC TCT GGT CTG AAG GTT TAT T GG TGC G Pyr 3Â’ MBS1 shared stem cDNA-A 5Â’ CGC ACC AAT AAA CCT TCA G AC CAG A 3Â’ MBS1 shared stem cDNA-G 5Â’ AAT AAA CCT TCA GAC CAG A GG TGC G 3Â’ MBS1 shared stem RNA 5Â’ AAU AAA CCU UCA GAC CAG A GG UGC G 3Â’ Boldface type indicates the stem sequence s in the molecular beacon structures. Table 6-2. Kinetic param eters of pyrene beacon assays for E. coli RNase H Substrate Km (M) kcat ( s-1) MB226 (25 mer) 0.019 0.25 MBS1 (25 mer) 0.031 0.33 24 mer 165 0.02 22 mer 151 0.07 0.1 14 mer 154 0.08 1.12

PAGE 162

162 CHAPTER 7 SUMMARY AND FUTURE DIRECTIONS Instrumentation Development and Nano material Engineering for Live Cell Mapping and Bioanalysis Mapping living cells with good spatial and tem poral resolution offers significant potential for the understanding of important bi ological phenomena. Therefore, the overall direction of this doctoral research has focused on the development of efficient molecular probes and the construction of sens itive instruments for live cell receptor mapping and bioanalysis. Fluorescence Correlation Spectroscopy (FCS) is a sensitive technique for observation of molecular interactions dow n to the molecular level. By extracting information from molecular diffusion, FCS gives detailed information on molecular interactions. In addition, it is also a non-invasive single -molecule-detection technique that can be applied to the intr acellular environment with low detection limits. Therefore, this highly sensitive technique was chosen fo r this research project, and was applied for direct measurements of membrane recept or density in its nat ural physiological environment on the cell surface. A cellular model was constructed using a DNA ligand aptamer for specific receptor targeting and labeling, and the receptor densities and distribution profile on the cell surface were obtained for different types of cancer cells. This successful outcome has proved the advantages of the FCS technique and its applications for cell membrane receptor mapping.2 The previous membrane receptor dens ity study using FCS has proven the capability of using fluorescence auto-correla tion for molecular interaction studies. However, as discussed in C hapter 4, the applic ation of FCS to binding analysis is limited to the events in which binding signifi cantly reduces the diffusion rate of the

PAGE 163

163 labeled species. Therefore, to overcome th is limitation, the original FCS setup was upgraded to a novel three-channel dual-col or Fluorescence Cross-Correlation Spectroscopy (FCCS) setup in our lab. The di fficulties of overlapping three femtolitersized observation volumes were overcome by careful alignments and calibrations during the setup. This lab-built F CCS not only inherits the single -molecule detection capability from FCS, but also further ex tends its applications for mole cular interaction studies by labeling two species with two spectrally dist inct fluorophores. By taking advantage of its co-localization capabilities and the high sensit ivity and low detection limit properties, we are now further adapting this technique for real-time monitoring of intracellular gene expression levels. Besides instrumentation developm ent, nanomaterials have been used to construct a nanoruler to study the detailed st ructures of individua l membrane receptor by mapping its binding site distances on live cell membrane. This SET-based nanoruler uses aptamer-gold-nanoparticle conjugates wit h different diameters to monitor the distance between two binding sites on a re ceptor in the natural physiological environment of the cell surf ace. This nanoruler has been proved to successfully measure separation distances we ll beyond the detection limit of FRET (~10 nm). This is a significant development because many membrane proteins will change their conformations if separated from the cell membranes for in-vitro studies, resulting in an inability to measure the real binding site dist ances in these proteins. Thus, for the first time, we have successfully developed an effectiv e SET nanoruler for live cells with long distance, simple construction, fast detection and low fluorescence background.167

PAGE 164

164 Nanomaterial engineering was also us ed to develope molecular probes for protein detection and enzymatic activity studi es. Due to the versatility, nucleic acids have extensively served as a novel class of recognition elements to detect a broad range of targets, such as ions, small mole cules, nucleic acids and proteins to living organisms. Two different types of molecu lar beacon (MB) probes have been engineered for enzymatic activity studies3 and protein inhibition studies4. In summary, this PhD research has focused on instrumentation development and nanomaterial engineering for bioanalysis and biom edical applications, especially for cell membrane receptor studies and intracellular measurements. A successful outcome from these studies will lead to a better understanding of biolog ical events and processes. Future Directions Exploration of FCS/FCCS Ap plications for Molecular Interaction Studies inside Living Cells In order to further extend the applications of FCS/FCCS for molecular interaction studies, especially for intracellular measur ements, we are adapting this technique for real-time monitoring of intracellular mRNA. The ability to detect, localize, quantify and monitor the expression of specific genes in living cells in real time will offer unprecedented opportunities for advancement in molecular biology, disease pathophysiology, drug discovery and medica l diagnostics. This innovative FCCS strategy stands in contrast to current methods for quantifying gene expression which are not able to provide real -time monitoring (e.g., PRC and microarray). Moreover, those conventional methods often result in false positives and suffer from high background from the intracellular environment. In this research, as shown in Figure 4-7,

PAGE 165

165 two DNA oligonucleotides were designed, one labeled with a gr een fluorophore and the other with a red fluorophore, t hat hybridize to adjacent regions on the same mRNA target, resulting in an increase of cross-corre lation signal. The co-localization detection of the two differently labeled DNA probes will significantly r educe false positives, leading to the sensitive real-time m onitoring of mRNAs in live cells. Furthermore, the femtolitersized detection volume also ensures a low de tection limit in a spatially and temporally ordered manner and eventually improves the det ection sensitivity. Ongoing research will provide a novel technique for sensitive RNA detection and quantificat ion in living cells. In the future, we will also further extend t he application for intracellular measurements and apply this sensitive technique for proteinprotein interaction studies in live cells. Engineering Nanomaterials as Natu ral Circuit Mimics for Bioanalysis The complex behavior of living systems is generated by molecular interactions among genes, proteins and metabolites, and the precise control of molecular interactions relies heavily on the existence of feedback circuits in the underlying intera ction networks. One typical regulatory mechanism is the negat ive feedback control circuit, in which an initiator of a reaction can be inhibited by its own products. In this way, the reactants and products of a reaction interact with each other to keep their concentrations at steadystate. Learning from the living system, we have applied nucleic acid probes to chemically mimic this type of regulation and have constructed a se lf-regulated protein inhibitor by introducing a feedback control circuit. By taking advantage of the unique property of the RNase H in cleaving only RN A strands from the RNA/DNA hybrid, an intramolecular RNA/DNA hy brid was engineered containing a DNA aptamer inhibitor selected for HIV-1 RNase H. Cleavage of the DNA aptamer from t he hybrid by HIV-1

PAGE 166

166 RNase H will lead to its binding of the aptamers to the HIV1 RNase H, thus deactivating by the cleavage function. In this way, the catalytic cleavage reaction can be terminated so a feedback control circuit. Th is feedback control circuit was further adapted to regulate the inhibitor function of thrombin to construct a self-regulated protein inhibitor. Due to this low molecu lar weight, ease of reproducible production, versatility in application, and most importantly, flexible sequence design of the nucleic acids, this feedback control holds great potential to become a universal method for making protein inhibitors.

PAGE 167

167 APPENDIX A PROBABILITY CALCULATION OF IN TER-RECEPTOR SET INTERACTION CONTRIBUTION According to the mean free path calcul ation in molecular collision study, we regard individual receptor as a circle with a diameter d of 13.4nm on t he cell surface. Previous Fluorescence Recovery After Photobleaching (FRAP) studies168 indicated that receptors are randomly diffusing on the surf ace with a lateral diffusion coefficient D of 10-10 cm2/s. Therefore, during the detection time the number of receptors that an individual receptor collides with during its late ral diffusion pathway actually indicates the probability of inter-receptor interactions. Assume the diffusion rate of a receptor is and the detection time is t so the surface area a receptor has trav eled during the detection time will be coveredSdt (1) While the lateral diffusion coefficient Dd so coveredSDt (2) Therefore, the number of re ceptors that an individual receptor collides with ( P ) during the detection time t will be coveredPSADtA (3) in which A is the density of receptor on the cell surface. According to our experiment conditions, the detection time for individual cells in the flow Cytometry will be 665sec/10510sec/ tcellscell. The receptor density of PTK7 on CEM cells were determined previously in the FCS study2 as 21300190/ Arecptorm So, by putting all these values in formula (3), the chance for inter-receptor collisions during the detection time will be

PAGE 168

168 102625 cov1 (10/)(510)(1300/)6.510 10000eredPSADtAcmssm Therefore, the contributi on of inter-receptor interaction is less than 1/10000 to the total SET interaction. In other words, the co ntribution of the inter-receptor interaction effects is negligible for the distance determinations.

PAGE 169

169 LIST OF REFERENCES (1) Bacia, K.; Kim, S. A.; Schwille, P. Nat Meth 2006, 3 83-89. (2) Chen, Y.; Munteanu, A.; Huang, Y.-F.; Phil lips, J.; Zhu, Z.; Mavros, M.; Tan, W. Chemistry A European Journal 2009, 15 5327-5336. (3) Chen, Y.; Yang, C. J.; Wu, Y.; Conl on, P.; Kim, Y.; Lin, H.; Tan, W. ChemBioChem 2008, 9 355-359. (4) Chen, Y.; Huang, J.; Wang, H.; Zhu, Z.; Tan, W. submitted 2010. (5) Bruce, A.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walters, P. Molecular Biology of the Cell ; Garland Science: New York and London, 2002. (6) Tan, W.; Wang, K.; Drake, T. J. Current Opinion in Chemical Biology 2004, 8 547-553. (7) Yang, C. J.; Jockusch, S.; Vic ens, M.; Turro, N. J.; Tan, W. Proceeding of the National Academy of Sciences of the United States of America 2005, 102 17278-17283. (8) Tan, L.; Li, Y.; Drake, T. J.; Moroz, L.; Wang, K.; Li, J.; Munteanu, A.; Yang, C. J.; Martinez, K.; Tan, W. The Analyst 2005, 130 1002-1005. (9) Shangguan, D.; Li, Y.; Tang, Z.; Cao, Z. C.; Chen, H. W.; Mallik aratchy, P.; Sefah, K.; Yang, C. J.; Tan, W. Proceedings of the National Academy of Sciences 2006, 103 11838-11843. (10) Daniel, M.-C.; Astruc, D. Chemical Reviews 2003, 104 293-346. (11) Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H. Nat Mater 2005, 4 435-446. (12) Mirkin, C. A.; Letsinger, R. L. ; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382 607609. (13) Omid C. Farokhzad; Ji anjun Cheng; Benjamin A. Tepl y; Ines Sherifi; Sangyong Jon; Philip W. Kantoff; Jero me P. Richie; Langer, R. Proceeding of the National Academy of Sciences of t he United States of America 2006, 103 6315-6320. (14) Tyagi, S.; Kramer, F. R. Nat Biotech 1996, 14 303-308. (15) Lakowicz, J. R. Principles of fluorescence spectroscopy 2nd ed.; Springer: Heidelberg, 1999. (16) Rigler, R. Journal of Biotechnology 1995, 41 177-186.

PAGE 170

170 (17) Rigler, R.; Elson, E. S. Fluorescence Correlation Spectroscopy: Theory and Applications ; Springer: Berlin, 2001. (18) Rosi, N. L.; Giljohann, D. A.; Thaxton, C. S.; Lytton-J ean, A. K. R.; Han, M. S.; Mirkin, C. A. Science 2006, 312 1027-1030. (19) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277 1078-1081. (20) Zhao, X.; Bagwe, R. P.; Tan, W. Advanced Materials 2004, 16 173-176. (21) Santra, S.; Zhang, P.; W ang, K.; Tapec, R.; Tan, W. Analytical Chemistry 2001, 73 4988-4993. (22) Jae-Hyun, L.; Young-wook, J.; S oo-In, Y.; Jeon-Soo, S.; Jinwoo, C. Angewandte Chemie International Edition 2006, 45 8160-8162. (23) Tatjana Atanasijevic; Maxim Sh usteff; Peter Fa m; Jasanoff, A. Proceeding of the National Academy of Sciences of the United States of America 2006, 103 14707-14712. (24) Lutz, J.-F.; Stiller, S.; Hoth, A .; Kaufner, L.; Pison, U.; Cartier, R. Biomacromolecules 2006, 7 3132-3138. (25) Alivisatos, A. P .; Gu, W.; Larabell, C. Annual Review of Biomedical Engineering 2005, 7 55-76. (26) Han, M.; Gao, X .; Su, J. Z.; Nie, S. Nat Biotech 2001, 19 631-635. (27) Huang, Y.-F.; Sefah, K.; Bamrung sap, S.; Chang, H.-T.; Tan, W. Langmuir 2008, 24 11860-11865. (28) Rex, M.; Hernandez, F. E.; Campiglia, A. D. Analytical Chemistry 2005, 78 445451. (29) Ajayan, P. M. Chemical Reviews 1999, 99 1787-1800. (30) Tasis, D.; Tagmatarchis N.; Bianco, A.; Prato, M. Chemical Reviews 2006, 106 1105-1136. (31) Park, J.-S.; Cho, M. K.; Lee, E. J.; A hn, K.-Y.; Lee, K. E.; Jung, J. H.; Cho, Y.; Han, S.-S.; Kim, Y. K.; Lee, J. Nat Nano 2009, 4 259-264. (32) Heller, D. A.; Jin, H.; Ma rtinez, B. M.; Patel, D.; Mille r, B. M.; Yeung, T.-K.; Jena, P. V.; Hobartner, C.; Ha, T.; Silv erman, S. K.; Strano, M. S. Nat Nano 2009, 4 114-120. (33) Younsoo, B.; Shigeto, F. ; Atsushi, H.; Kazunori, K. Angewandte Chemie International Edition 2003, 42 4640-4643.

PAGE 171

171 (34) Uri Raviv; Daniel J. N eedleman; Youli Li; Herbert P. M iller; Leslie Wilson; Safinya, C. R. Proceeding of the National Academy of Sciences of the United States of America 2005, 102 11167-11172. (35) Jae-Hyun, L.; Kyuri, L.; Seungo, M.; Yuhan, L.; Taewan, P.; Jinwoo, C. Angewandte Chemie International Edition 2009, 48 4174-4179. (36) Peisheng, X.; Shi-Yan, L.; Qun, L.; Ed ward, V.; Jun, R.; Williams, M.; Zhaojie, Z.; Maciej, R.; Youqing, S. Angewandte Chemie International Edition 2008, 47 1260-1264. (37) Davide, P.; Ravi, S.; David, M.; Mathi eu, E.; Jean-Paul, B.; Maurizio, P.; Kostas, K.; Alberto, B. Angewandte Chemie International Edition 2004, 43 5242-5246. (38) L. R. Hirsch; R. J. Staffo rd; J. A. Bankson; S. R. Sers hen; B. Rivera; R. E. Price; J. D. Hazle; N. J. Halas; West, J. L. Proceeding of the National Academy of Sciences of the United States of America 2003, 100 13549-13554. (39) Wang, L.; Yang, C. J.; Medley, C. D.; Benner, S. A.; Tan, W. Journal of the American Chemical Society 2005, 127 15664-15665. (40) Robertson, D. L.; Joyce, G. F. Nature 1990, 344 467-468. (41) C. Tuerk; L.Gold Science 1990, 249 505-510. (42) Ellington, A. D.; Szostak, J. W. Nature 1990, 346 818-822. (43) Fang, X.; Li, J. J.; Tan, W. Analytical Chemistry 2000, 72 3280-3285. (44) Piatek, A. S.; Tyagi, S.; Po l, A. C.; Telenti, A.; Miller, L. P.; Kramer, F. R.; Alland, D. Nat Biotech 1998, 16 359-363. (45) Drake, T. J.; Tan, W. Appl Spectrosc. 2004, 58 269-280. (46) Nitin, N.; Santangelo, P. J. ; Kim, G.; Nie, S.; Bao, G. Nucl. Acids Res. 2004, 32 e58-. (47) Conrad, R.; Keranen, L. M.; Ellington, A. D.; Newton, A. C. Journal of Biological Chemistry 1994, 269 32051-32054. (48) Macaya, R. F.; Schultze, P.; Smit h, F. W.; Roe, J. A.; Feigon, J. Proceedings of the National Academy of Sciences of the United Stat es of America 1993, 90 3745-3749. (49) Famulok, M.; Mayer, G. Curr Top Microbiol Immunol. 1999, 243 123-136. (50) Hicke, B. J.; Marion, C.; Chang, Y.-F.; Gould, T.; Lynott, C. K.; Parma, D.; Schmidt, P. G.; Warren, S. Journal of Biological Chemistry 2001, 276 4864448654.

PAGE 172

172 (51) Knox, S. J.; Meredith, R. F. Semin Radiat Oncol. 2000, 10 73-93. (52) Amstad, E.; Gillich, T.; Bile cka, I.; Textor, M.; Reimhult, E. Nano Letters 2009, 9 4042-4048. (53) Bunz, U. H. F.; Rotello, V. M. Angewandte Chemie International Edition 49 3268-3279. (54) Schultz, D. A. Curr. Opin. Biotechnol. 2003, 14 13. (55) Chah, S.; Hammond, M.; Zare, R. Chem. Biol. 2005, 12 323. (56) Glynou, K.; Ioannou, P. C.; Ch ristopoulos, T. K.; Syriopoulou, V. Anal. Chem. 2003, 75 4155-4160. (57) Oh, J. S.; Ha, G. W.; Cho, Y. S.; Kim M.J.; An, D. J.; D.S., S. Clin. Vaccine Immunol. 2006, 13 520-524. (58) Tan, W.; Wang, K.; He, X.; Zhao, X. J.; Dr ake, T.; Wang, L.; Bagwe, R. P. Med. Res. Rev. 2004, 24 621. (59) Kim, H. K.; Kang, S. J.; Choi, S. K.; Min, Y. H.; Yoon, C. S. Chem. Mater. 1999, 11 779. (60) Smith, J. E.; Wang, L.; Tan, W. Trends Anal. Chem. 2006, 25 848. (61) Bagwe, R. P.; Yang, C. Y.; Hilliard, L.; Tan, W. Langmuir, 2004, 20 8336-8342. (62) Wang, J. Small 2005, 1 (63) Aldaye, F. A.; Palmer A. L.; Sleiman, H. F. Science 2008, 321 1795-1799. (64) Ingle, J. D.; Crouch, S. R. Spectrochemical Analysis : Prentice Hall, New Jersey, 1988. (65) Tyagi, S.; Bratu, D. P.; Kramer, F. R. Nat.Biotechnol. 1998, 16 (66) Yun, C. S.; Javier, A.; Jennings, T.; Fisher, M.; Hira, S.; Peterson, S.; Hopkins, B.; Reich, N. O.; Strouse, G. F. Journal of the American Chemical Society 2005, 127 3115-3119. (67) Darbha, G. K.; Ray, A.; Ray, P. C. ACS Nano 2007, 1 208-214. (68) Jennings, T. L.; Schlatterer, J. C.; Singh, M. P.; Greenbaum, N. L.; Strouse, G. F. Nano Letters 2006, 6 1318-1324. (69) Ray, P. C.; Darbha, G. K.; Ray, A.; Hardy, W.; Walker, J. Nanotechnology 2007, 18 375504.

PAGE 173

173 (70) Jennings, T. L.; Singh, M. P.; Strouse, G. F. Journal of the American Chemical Society 2006, 128 5462-5467. (71) Griffin, J.; Ray, P. C. The Journal of Physical Chemistry B 2008, 112 1119811201. (72) Griffin, J.; Sing h, A. K.; Senapati, D.; Rhodes, P.; Mitchell, K.; Robinson, B.; Yu, E.; Ray, P. C. Chemistry A European Journal 2009, 15 342-351. (73) Sen, T.; Haldar K. K.; Patra, A. The Journal of Physical Chemistry C 2008, 112 17945-17951. (74) Sen, T.; Sadhu, S.; Patra, A. Applied Physics Letters 2007, 91 043104-043104043103. (75) Chance, R. R.; Prock, A.; Silbey, R. In Advances in Chemical Physics ; I. Prigogine, S. A. R. Ed., 2007, pp 1-65. (76) Persson, B. N. J.; Lang, N. D. Physical Review B 1982, 26 5409. (77) Winnik, F. M. Chem.Rev. 1993, 93 587-614. (78) Fujimoto, K.; Shimizu, H.; Inouye, M. J.Org.Chem. 2004, 69 (79) Sokol, D. L.; Zhang, X. L.; Lu, P. Z.; Gewitz, A. M. Proc.Natl.Acad.Sci.U.S.A. 1998, 95 11538-11543. (80) Kim, S. A.; Heinze K. G.; Schwille, P. Nat Meth 2007, 4 963-973. (81) Lee, S.-J. R.; Escobedo-Lozoya, Y.; Szatmari, E. M.; Yasuda, R. Nature 2009, 458 299-304. (82) Magde, D.; Elson, E.; Webb, W. W. Physical Review Letters 1972, 29 705. (83) Douglas, M.; Ellio t, L. E.; Watt, W. W. Biopolymers 1974, 13 29-61. (84) Aragon, S. R.; Pecora, R. Biopolymers 1975, 14 119-137. (85) Ehrenberg, M.; Rigler, R. Quarterly Reviews of Biophysics 1976, 9 69-81. (86) Ehrenberg, M.; Rigler, R. Chemical Physics 1974, 4 390-401. (87) Douglas, M.; Watt W. W.; Elliot, L. E. Biopolymers 1978, 17 361-376. (88) Preuss, A. K.; Mayer, M.; Wohland, T.; Hovius, R.; Vogel, H. Biochemistry 2003, 42 877-884. (89) Rigler, R.; Mets, ; Widengren, J.; Kask, P. European Biophysics Journal 1993, 22 169-175.

PAGE 174

174 (90) Rigler, R.; Mets, U. 1993; 239-248. (91) Eigen, M.; Rigler, R. Proceeding of the National Aca demy of Sciences of the United States of America 1994, 91 5740-5747. (92) Lauritzen, S. L. Thiele : pioneer in statistics Oxford, 2002. (93) Bachelier, L. Annales Scientifiques de lÂ’cole Normale Suprieure 1900, 3 21-86. (94) Einstein, A. Annalen der Physik 1905, 17 549-560. (95) Smoluchowski, M. v. Annalen der Physik 1906, 326 756-780. (96) Perrin, J. Ann. Chim. Phys. 1909, 18 5-114. (97) Magde, D.; Elson, E.; Webb, W. W. Phys. Rev. Lett. 1972, 29 705-708. (98) Rigler, R.; Widengren, J.; Mets, Interaction and kinetics of single molecules as observed by fluorescence correlation spectroscopy ; Springer: Berlin, 1992. (99) Eigen, M.; Rigler, R. Proc. Natl Acad. Sci. USA 1994, 91 (100) Nie, S. M.; Chiu, D. T.; Zare, R. N. Science 1994, 266 1018-1022. (101) Schwille, P.; Haustein, E. Fluorescence Correlation Spectroscopy: An introduction to its concepts and applications (102) Schwille, P. Cell Biochemistry and Biophysics 2001, 34 383-408. (103) Gosch, M.; Rigler, R. Advanced Drug Delivery Reviews 2005, 57 169-190. (104) Pramanik, A.; Rigler, R. Biological Chemistry 2005, 382 371-378. (105) Rigler, R.; Mets, U. ; Widengren, J.; Kask, P. Eur. J. Biophys. 1993, 22 169-175. (106) Krichevsky, O.; Bonnet, G. Rep. Prog. Phys. 2002, 65 251-297. (107) Briddon, S. J.; Middleton, R. J.; Cordeaux, Y.; Flavin F. M.; Weinstein, J. A.; George, M. W.; Kellam, B.; Hill, S. J. Proc. Natl. Acad. Sci. U. S. A. 2004, 101 (108) Watts, A. Nat Rev Drug Discov 2005, 4 555-568. (109) Takayama, S.; Shimosato, H.; Shiba, H.; Funato, M.; Che, F.-S.; Watanabe, M.; Iwano, M.; Isogai, A. Nature 2001, 413 534-538. (110) Colabufo, N. A.; Berardi, F.; Calo, R.; Leopoldo, M.; Perrone, R.; Tortorella, V. European Journal of Pharmacology 2001, 427 1-5.

PAGE 175

175 (111) Bice, A. N.; Zeeberg, B. R. Medical Imaging, IEEE Transactions on 1987, 6 244249. (112) Huang, Y.-F.; Chang, H.-T.; Tan, W. Analytical Chemistry 2008, 80 567-572. (113) Thompson, N. In Topics in Fluorescence Spectroscopy 2002, pp 337-378. (114) Zipfel, W. R .; Webb, W. W. In Methods in Cellular Imaging ; Oxford University Press: Oxford, UK, 2001, pp 216-235. (115) Rigler, R.; Pramanik, A.; Jonasson, P.; Kratz, G.; Jansson, O. T.; Nygren, P.-A.; Stahl, S.; Ekberg, K.; Johansson, B. L.; Uhlen, S.; Uhlen, M.; Jornvall, H.; Wahren, J. Proceedings of the Nati onal Academy of Sciences of the United States of America 1999, 96 13318-13323. (116) Pramanik, A.; Jureus, A.; Langel U.; Bartfai, T.; Rigler, R. Biomedical Chromatography 1999, 13 119-120. (117) Boonen, G.; Pramanik, A .; Rigler, R.; Haberlein, H. Planta Med 2000, 66 7-10. (118) Liu, J.; Lu, Y. Nat. Protocols 2006, 1 246-252. (119) Schurer, H.; Buchynskyy, A.; Korn, K. ; Famulok, M.; Welzel, P.; Hahn, U. Biological Chemistry 2005, 382 479-481. (120) Ikebukuro, K.; Okumura, Y.; Sumikura, K.; Karube, I. Nucl. Acids Res. 2005, 33 e108-. (121) Berezovski, M. V.; Lechmann, M.; Musheev M. U.; Mak, T. W.; Krylov, S. N. Journal of the American Chemical Society 2008, 130 9137-9143. (122) Raddatz, M.-S. L.; Dolf, A.; Endl, E .; Knolle, P.; Famulok, M.; Mayer, G. Angewandte Chemie International Edition 2008, 47 5190-5193. (123) Osborne, S. E. ; Ellington, A. D. Chemical Reviews 1997, 97 349-370. (124) Mallikaratchy, P.; Chen, H.; Tang, Z.; Meng, L.; S hangguan, D.; Parekh, P.; Kim, Y.; Sefah, K.; Tan, W. American Pharmaceutical Review 2007, 10 134-141. (125) Shangguan, D.; Cao, Z.; M eng, L.; Mallikaratchy, P.; Se fah, K.; Wang, H.; Li, Y.; Tan, W. Journal of Proteome Research 2008, 7 2133-2139. (126) Mller-Tidow, C.; Schwble, J.; Steff en, B.; Tidow, N.; Brandt, B.; Becker, K.; Schulze-Bahr, E.; Halfter, H.; Vogt, U.; Metzger, R.; Schneider, P. M.; Bchner, T.; Brandts, C.; Berdel W. E.; Serve, H. Clin. Cancer. Res. 2004, 10 1241-1249. (127) Zeyu, X.; Dihua, S.; Zehui C.; Xiaohong, F.; Weihong, T. Chemistry A European Journal 2008, 14 1769-1775.

PAGE 176

176 (128) Erickson, J.; Goldstein, B.; Holowka, D.; Baird, B. Biophysical Journal 1987, 52 657-662. (129) Meseth, U.; Wohland, T.; Rigler, R.; Vogel, H. Biophys. J. 1999, 76 (130) Kinjo, M. ; Rigler, R. Nucleic Acids Res. 1995, 23 1795-1799. (131) Rauer, B.; Neumann, E .; Widengren, J.; Rigler, R. Biophys. Chem. 1996, 58 (132) Schwille, P.; Oehlens chlager, F.; Walter, N. G. Biochemistry 1996, 35 1018210193. (133) Schuler, J.; Frank, J.; Trier, U. ; Schafer-Korting, M. ; Saenger, W. A. Biochemistry 1999, 38 8402-8408. (134) Chen, C.; Ridzon, D. A .; Broomer, A. J.; Zhou, Z.; Lee, D. H.; Nguyen, J. T.; Barbisin, M.; Xu, N. L.; Mahuvakar, V. R.; Andersen, M. R.; Lao, K. Q.; Livak, K. J.; Guegler, K. J. Nucl. Acids Res. 2005, 33 e179-. (135) Shingara, J.; Keiger, K. ; Shelton, J.; Laosinchai-Wolf, W.; Powers, P.; Conrad, R.; Brown, D.; Labourier, E. RNA 2005, 11 1461-1470. (136) Weiss, S. Science 1999, 283 1676-1683. (137) Zhuang, X.; Bartley, L. E.; Babcock, H. P.; Russell, R.; Ha T.; Herschlag, D.; Chu, S. Science 2000, 288 2048-2051. (138) Yildiz, A.; Forkey, J. N.; McKinney, S. A.; Ha, T.; Goldman, Y. E.; Selvin, P. R. Science 2003, 300 2061-2065. (139) Blanchard, S. C.; Kim, H. D.; Gonzalez, R. L.; Jr., P., Joseph D. ; Chu, S. Proceeding of the National Ac ademy of Sciences of the United States of America 2004, 101 12893-12898. (140) Liu, G. L.; Yin, Y.; Kunc hakarra, S.; Mukherj ee, B.; Gerion, D.; Jett, S. D.; Bear, D. G.; Gray, J. W.; A livisatos, A. P.; Lee, L. P.; Chen, F. F. Nat Nano 2006, 1 4752. (141) Anker, J. N.; Hall, W. P.; Lyandres, O. ; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Nat Mater 2008, 7 442-453. (142) Sonnichsen, C.; Re inhard, B. M.; Liphardt, J.; Alivisatos, A. P. Nat Biotech 2005, 23 741-745. (143) Bene, L.; Szentesi, G.; Matyus L.; Gaspar, R.; Damjanovich, S. Journal of Molecular Recognition 2005, 18 236-253.

PAGE 177

177 (144) Fan, C.; Wang, S.; Hong, J. W.; Bazan, G. C.; Plax co, K. W.; Heeger, A. J. Proceeding of the National Ac ademy of Sciences of the United States of America 2003, 100 6297-6301. (145) Frens, G. Nature Phys. Sci. 1973, 241 20-22. (146) Xiao, Z.; Shangguan, D.; C ao, Z.; Fang, X.; Tan, W. Chemistry A European Journal 2008, 14 1769-1775. (147) Ghosh, S. K.; Pal, A.; Kundu, S.; Nath, S.; Pal, T. Chemical Physics Letters 2004, 395 366-372. (148) Crouch, R. J.; Toulme, J. J. Ribonuclease H ; John Libbey: Paris, 1998. (149) Miller, H. I.; Riggs A. D.; Gill, G. N. J. Biol. Chem. 1973, 248 2621-2624. (150) Stein, C. A. The Journal of Clinical Investigation 2001, 108 641-644. (151) Kanaya, E.; Kanaya, S. European Journal of Biochemistry 1995, 231 557-562. (152) Dirksen, M. L. ; Crouch, R. J. J. Biol. Chem. 1981, 256 11569-11573. (153) Hogrefe, H. H.; Hogrefe, R. I.; Walder, R. Y. ; Walder, J. A. J. Biol. Chem. 1990, 265 5561-5566. (154) Rizzo, J.; Gifford, L. K.; Z hang, X.; Gewirtz, A. M.; Lu, P. Molecular and Cellular Probes 2002, 16 277-283. (155) Strouse, R. J.; Hakki, F. Z.; Wang, S.; DeFusco, A. ; Garrett, J.; Schenerman, M. A. BioPharm 2000, 13 40-47. (156) Li, J. J.; Ge yer, R.; Tan, W. Nucl. Acids Res. 2000, 28 e52-. (157) Biggins, J. B.; Prudent, J. R.; Marshall, D. J.; R uppen, M.; Thorson, J. S. Proceedings of the Na tional Academy of Sciences of the United States of America 2000, 97 13537-13542. (158) Li, J. J.; Fang, X.; Tan, W. Biochemical and Biophysical Research Communications 2002, 292 31-40. (159) Crooke, S. T.; Lemonidis, K. M.; Neilson, L.; Griffey, R.; Lesnik, E. A.; Monia, B. P. Biochem J. 1995, 312 599-608. (160) Winnik, F. M. Chemical Reviews 1993, 93 587-614. (161) Pagano, R. E.; Martin, O. C. ; Kang, H. C.; Haugland, R. P. J Cell Biol 1991, 113 1267-1279.

PAGE 178

178 (162) Fujimoto, K.; Shimizu, H.; Inouye, M. The Journal of Organic Chemistry 2004, 69 3271-3275. (163) Birks, J. B. Photophysics of Aromatic Molecules ; Wiley-Interscience: London, 1970. (164) Dahim, M.; Mizuno, N. K.; Li, X.-M.; Momsen, W. E.; Momsen, M. M.; Brockman, H. L. Biophysical Journal 2002, 83 1511-1524. (165) Haruki, M.; Noguchi, E.; Kanaya, S.; Crouch, R. J. J. Biol. Chem. 1997, 272 22015-22022. (166) Kanaya, E.; Uchiyama, Y.; Ohtsuka, E.; Ueno, Y.; Ikehar a, M.; Kanaya, S. FEBS Letters 1994, 354 227-231. (167) Chen, Y.; O'Donoghue, M. B.; Huang, Y.-F .; Kang, H.; Phillips J. A.; Chen, X.; Estevez, M.-C.; Tan, W. submitted 2010. (168) Schlessinger, J.; Shechter, Y.; Cuatre casas, P.; Willingham, M. C.; Pastan, I. Proceeding of the National Ac ademy of Sciences of the United States of America 1978, 75 5353-5357.

PAGE 179

179 BIOGRAPHICAL SKETCH Yan Chen was born in Guangzhou, Chi na, in 1982. She attended the Zhixin Middle School for her secondary education. She obtained a high ranking in the National University Entrance Examinat ion and attended University of Science and Technology of China to study Chemical Physics. After obtai ning her Bachelor of Science degree, Yan came to the United States in the fall of 2005 to pursue her Ph.D. degree under the supervision of Dr. Weihong Tan. She receiv ed her Doctor of Philosophy degree in Analytical Chemistry from the Univ ersity of Florida in June of 2010.