POLY(PHENYLENE ETHYNYLENE)S IN BIOSENSOR APPLICATIONS
By
JUAN ZHENG
B. Sc., Biological Chemistry
University of Toronto, 2000
Submitted to the Department of Chemistry
In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
MASSACHUSES INSTiU
At the
OF TECHNOLOGY
Massachusetts Institute of Technology
JUN 21 2005
[ju4 32 3
LIBRARIES
May, 2005
© Massachusetts Institute of Technology, 2005. All rights reserved.
of Author:
Signature
v
Ikpartmenlof Cl(mistry
May 24, 2005
7.
/7
Certified by:
Timtthy Swager
Thesis Supervisor
Accepted by:
Robert W. Field
Chairman, Departmental Committee on Graduate Studies
M~rjl~v6
w
This doctoral thesis has been examined by a Committee of the Department of
Chemistry as follows:
(N\
Professor Timothy F. Jamison:
vi
Chairman
Professor Timothy M. Swager:
Adz ~
u
Thesis Advisor
Professor Daniel S. Kemp:
2
Dedicated to my parents
3
Poly(phenylene ethynylene)s in biosensor applications
By
Juan Zheng
Submitted to the Department of Chemistry on May 24, 2005
In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy in Chemistry
ABSTRACT
Conjugated polymers have shown remarkable sensitivity for the detection of a
variety of analytes, ranging from explosives to biological molecules such as DNA.
This thesis presents three new applications of poly(phenylene ethynylene)s (PPEs)
in biosensing applications.
Biotinylated PPEs were synthesized for the detection of dye-labeled streptavidin
using energy transfer, in the aqueous phase and in the solid phase. These polymers
served as a model for multivalent biosensing. Energy transfer was enhanced for dyes
which have better orbital overlap with the polymer, indicating an electron exchange
energy transfer contribution to the overall signal.
In collaboration with Prof. Peter Seeberger's group, mannose-substituted PPEs were
synthesized. These polymers used the multivalent presentation of the sugar moieties
for the agglutination of Escherichia coli, and offer a rapid method for their detection.
The formation of brightly fluorescent bacterial clusters was extended to energy
transfer schemes.
Amphiphilic biotinylated PPEs were synthesized and used to probe interactions at
the air-water interface. Subtle changes in the polymer structure could lead to great
differences in protein-ligand interactions. The Langmuir technique offers a sensitive
method for understanding the fundamental properties of PPEs.
Thesis Supervisor: Timothy Swager
Title: Professor of Chemistry
4
Contents
List of abbreviations
CHAPTER
6
1
Biosensors using conjugated polymers - an introduction
7
CHAPTER 2
A model biosensor - biotin-functionalized PPEs for the detection of streptavidin
27
Introduction
Results and Discussion
28
33
Experimental
46
References
53
CHAPTER 3
Mannose-functionalized polymers for the detection of Escherichia coli
55
Introduction
56
Results and Discussion
59
Experimental
References
68
74
CHAPTER 4
Amphiphilic polymers at the air-water interface
77
Introduction
78
Results and Discussion
85
Experimental
References
98
101
Curriculum Vitae
105
Acknowledgments
107
Appendix
109
5
List of Abbreviations
Con A
concanavalin A
CP
conjugated polymer
EDAC
1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide
ET
energy transfer
DMF
N, N-dimethylformamide
FRET
fluorescence resonance energy transfer
Gal
galactose
HOMO
highest occupied molecular orbital
LB
Langmuir-Blodgett
LUMO
lowest unoccupied molecular orbital
Man
mannose
MPS-PPV
poly[lithium 5-methoxy-2-(4-sulfopropyloxy)- 1,4phenylenevinylene]
MV 2+
methyl viologen
PBS
phosphate buffered saline
PNA
peptide nucleic acid
PPE
poly(p-phenylene ethynylene)
RhB-strept
rhodamine B-streptavidin
ROMP
ring opening metathesis polymerization
ROX
carboxy-X-rhodamine
SPR
surface plasmon resonance
TFA
trifluoroacetic acid
T-red-strept
Texas Red X-streptavidin
6
Chapter
1
Biosensors Using Conjugated Polymers - An
Introduction
Partially adapted from: Zheng, J.; Swager, T. M. Adv. Polym. Sci. 2005, 177, 151179
7
A brief survey of current biosensor technology
A sensor provides information on our physical, chemical, and biological
environment. In nature, living organisms are outfitted with a myriad of sensors which
inform them about their surroundings, these in turn allow them to respond to various
environmental stimuli. The human sense of smell, for instance, can recognize and
discriminate volatile compounds with high sensitivity and accuracy. Specifically,
between 10 and 100 million receptors can exist in the human nasal area, detecting some
odors at the parts per trillion levels and sometimes even distinguishing stereoisomers.1
Around 10,000 chemicals can be differentiated depending on various combinations of
olfactory receptors.
As with nature's sensors, the successful artificial sensor must readily bind to its
analyte with great specificity, and this binding event should be reported in an easily
measurable manner. Sensors are composed of two components: a recognition element and
a transduction element. In the case of a biosensor, a bioreceptor serves as the recognition
component, and a transducer allows the visualization of the binding process. In addition,
the ideal biosensor would satisfy the following requirements: accuracy, real-time
response, sensitivity, specificity, reproducibility, robustness, and ease of use.2
Biosensors may be classified with respect to the element of biorecognition or with
respect to the method of signal transduction. The recognizing biomolecule may be
antibodies (monoclonal or polyclonal), protein receptors, whole cells, nuclei acids or
enzymes. Each of these may be coupled with one of four major transduction schemes:
electrochemical, mass sensitive, calorimetric or optical.3 Here, a brief overview of
commercial biosensors will be given according to the transducing method.
8
Electrochemical biosensors are best illustrated by glucose monitors for diabetes
patients. In these sensors, glucose is oxidized to gluconic acid according to Equation 1.
Glucose + 02 + glucose oxidase (cat) - gluconic acid + H 2 0 2 (Equation 1.1)
The first biosensor was demonstrated by Clark and Lyons in 1962, where the enzyme
glucose oxidase was coupled to an oxygen electrode, which monitors the consumption of
02. 4 Instead of using 02, the oxidation of H2 02 may be monitored.5 A mediator, which
transports electrons from the glucose oxidase to the anode surface and is reoxidized, is
also used in some cases. The mediators may be quinones, conductive organic salts,
ferricyanide or ferrocene species.6 Glucose sensors are some of the most successful and
well-researched commercial sensors to date. In addition to these amperometric
electrochemical biosensors, other methods of detection such as potentiometric and
conductimetric sensors are also possible.
Acoustic wave-based biosensors are based on the detection of mechanical
acoustic waves. They are mass sensitive biosensors and often use oscillating piezoelectric
crystals as the transduction elements. The crystal is coated with the biorecogntion
element. Upon binding of the analyte, a change in the resonance frequency occurs
according to the change in mass of the crystal.7 Acoustic wave biosensors offer label-free
detection and are very sensitive, cost effective, and easy to use. However, there is often a
long incubation period and problems
associated with surface contamination.
Functionalization of the crystal surfaces may also prove challenging.2
In the case of calorimetric biosensors, the presence of an analyte is detected by
measuring the evolution or absorption of heat in biological reactions. Isothermal titration
calorimetry and differential scanning calorimetry are two of the most common
commercial sensors employing this method. They have been used to study many different
kinds of interactions: protein-protein, protein-membrane and drug-receptor interactions.7
9
Optical biosensors occupy a prominent place in commercial biosensors today,
second only to electrochemical biosensors. They can employ a variety of methods when
transducing a recognition event: light absorption, fluorescence/phosphorescence,
bio/chemi-luminescence, reflectance, raman scattering, and refractive index. In addition
to their flexibility, optical sensors provide rapid response and are sensitive and robust.
Optical sensors are also suitable to miniaturization,
remote sensing and multi-analyte
sensing. Among the more successful commercial biosensors, surface plasmon resonance
(SPR) has emerged as a viable method for analyte detection. In SPR, light is coupled to a
gold surface by a prism or a grating. This results in the propagation of charged density
waves (surface plasmons) along the metal surface, creating an electromagnetic field. This
evanescent wave is highly sensitive to the dielectric constant changes in the medium,
where a change in the index of refraction at the surface of the sensor results upon binding
of analyte to receptors immobilized at the gold surface. This in turn causes a shift in the
SPR angle, the critical angle at which total internal reflection of light produces a
reduction in the reflected light intensity.3 A measurable shift in the resonance angle when
using monochromatic light occurs when there is analyte binding. When using
polychromatic light, the wavelength of the reflected light at a fixed angle can be
monitored. In this case, a binding event will result in a change in the color of the reflected
light.7 While SPR-based biosensors often use very small sample volumes, they may
require a preconcentration step to increase the numbers of pathogens. They are also
sensitive to ambient temperature drift.2
10
Poly(p-phenylene ethynylene)s for signal amplification
Poly(p-phenylene ethynylene)s (PPEs) are made of repetitive sequences of alkyne
and phenyl groups. Because of their extended conjugated structure, their highest occupied
(HOMO) and lowest unoccupied (LUMO) molecular orbitals give rise to conduction (*)
and valence 1r) bands. PPEs are typically insulating in their native neutral state but can
be made
conductive by oxidation (p-doping) or reduction (n-doping). Their
semiconductive nature has generated interest in developing electroluminescent polymers
for device applications. However, it is their photophysical characteristics that make PPEs
good candidates for use as transducers and they are now one of the most important
classes of conducting polymers for sensing purposes.
In the Swager group, PPEs are conjugated polymers of choice for sensing
applications for a variety of reasons. Their ease of synthesis by palladium-catalyzed
Sonogashira reaction using aryl diiodides and aryl diacetylenes is amenable to wide range
of polymers with specially tailored characteristics. For example, by varying the
substituents
on the aromatic rings, the polymers'
bandgaps, secondary structure, and
solubility can be tuned. PPEs also behave as rigid rods over their persistence lengths and
can maintain conjugation even when the aromatic rings are twisted with respect to one
another. This extended conjugated electronic structure leads to a narrower range of
energy levels with sharp spectral features.
In PPEs, excited states consisting of electron-hole pairs also referred to as
excitons, can be generated upon photoexcitation and these travel through the polymer
energy bands by F6rster or Dexter mechanisms. Recombination of the electron-hole pairs
can occur via radiative and non-radiative pathways. The emissive properties of
conjugated polymers are dominated by energy migration to recombination of the excitons
at the local minima of their band structures. Hence perturbations to the PPEs will be
11
reflected in its collective property. This has important implications for sensing
applications.
There are two mechanisms by which conjugated polymers effect signal
amplification in sensing events (Figure 1.1). In a "turn-off' sensor, an electron accepting
analyte provides a low energy trap, enabling electron transfer from the polymer to the
analyte. This favors non-radiative exciton recombination and results in fluorescence
quenching. In a "turn-on" sensor, the analyte provides a local narrowing of the polymer
bandgap, which facilitates exciton recombination and results in a red-shifted new
emission.
A
Conduction
Band
B
I
I
"
®-~
hv\
Vhv
E
hv
EBand
Figure1.1. Schematic representation of exciton migration in conjugated polymer and sensing mechanisms.
(A) Turn-off sensor: fluorescence quenching due to electron transfer from the polymer to the analyte. (B)
Turn-on sensor: local narrowing of the bandgap, facilitating radiative recombination of the exciton at a new
red-shifted wavelength.
The sensitivity of PPEs to perturbations in its band structure was illustrated by
endcapping a PPE with anthracene units (Structure 1). The polymers acted as an antenna
for harvesting optical energy and this was transferred to the anthracene units due to an
induced localized narrowing of the band gap (Figure 1.2). Radiative recombination of the
electron-hole pair resulted in greater than 95% of the emission occurring at the states
localized at the anthracene endgroups.8 The presence of anthracene was effectively
amplified.
Conduction Band
\-~------~;
'v
.AAWMMMMI
hv'
Valence Band
Molecular Axis
Figure 1.2. Energy migration in a semiconductive molecular wire with a decrease in band gap at the
terminus induced by anthracene units. (Reprinted with permission from reference 8, copyright 1995
American Chemical Society)
This amplification phenomenon can be applied to sensor design and was first
demonstrated. by our group in 1995.9 10 In these studies, the sensing ability for a methyl
viologen salt (MV2 + or N, N'-dimethyl-4-4'bipyridinium
bis(hexafluorophosphate)
was
measured for a single receptor fluorescent molecule 2 and a PPE 3 where many receptors
are "wired in series". In this sensing scheme, excitons were generated by photoexcitation
of the polymer. When they encounter a cyclophane-bound MV2+, a highly efficient
electron transfer reaction occurred to the analyte and the initially fluorescent polymer was
returned to the ground state without the emission of a photon (Figure 1.3). A 67-fold
increase in quenching enhancement was obtained by comparing the Stem-Volmer
quenching constants (Ksv) for 3 and 2, corresponding to an average of 134 phenylene
units sampled by the excitons. The quenching efficiency of MV2 + increased steadily with
increasing polymer molecular weight, reaching a maximum at 65,000 and plateaued
thereafter. The molecular weight dependence of quenching indicated that the diffusion
length of the exciton is less than the length of the polymer.
13
2+
MV
, N-
-
R = C10H25
R' = CON(C 8H1 7)2
3
2
The apparent binding constant Ksv obtained by Stem-Volmer quenching studies
was the product of the number of receptors visited by the exciton and the binding
constant of MV 2+ to the cyclophane receptor. For this reason polymer 3 and its
monoreceptor model 2 were designed so that the binding constant for methyl viologen to
the receptor was known for both systems. This allowed the calculation of the true
amplification factor of 67.
INOL
n ,
itted
Fluorescnt ('h eniorwn
liv
'+PQ'
-PQ+
0JVJ
't i
I
T
i!', rcuciuilor
Itltnr,,c' ..
ll
1...
cn
ý,iIiI
".%i i
Wn
k
l ('Clh'llol
Ii
'"
t
,
l,
in crini~.icii
Ic,'
_
__
_
i
l (010
N
redltLIon in CTii',10io1
Figure 1.3. Conceptual illustration of signal amplification by wiring receptors in series. (Reprinted with
permission from reference 10, copyright 1995 American Chemical Society)
The 67-fold amplification obtained for polymer 3 was restricted by an inherent
limitation of the "wired in series" design. As the exciton traveled in a one dimensional
random walk process down the polymer chain, it had equal opportunity to visit a
preceding or an ensuing receptor. This represented 1342 random stepwise movements for
134 phenylene ethynylene units and so much of the receptor sampling by the exciton was
redundant. To increase the efficiency of receptor sampling, maximization of the number
of different receptors that an exciton could visit throughout its lifetime was required. To
achieve this end we extended the polymer sensor into two dimensions by use of a thin
film and thereby increased the sensitivity.
PPEs often vn-stack and form excimers in the solid state. To circumvent this
problem we designed PPE films that incorporate rigid 3-dimensional pentiptycene
scaffolds in the polymer backbone (Figure 1.4). 11, 12 These polymers formed porous films
and discriminately bound to various analytes of suitable size and electronic properties.
Strongly electron deficient analytes such as 2,4,6-trinitrotoluene (TNT) and 2,4dinitrotoluene (DNT) caused fluorescence quenching by electron transfer mechanism.
Films of polymer 12 were quenched by 50% within 30s of exposure to TNT and by 75%
within 60s, despite its low equilibrium vapor pressure of 7 ppb.
\77'"
RO
R = C14H29
4
Polymer Backbone
Pentiptycene Groups
Figure1.4. PPE containing rigid pentipticene monomers in the backbone, which provided a porous
structure for diffusion and docking of analyte and which also prevented vT-stacking.
As greater sensitivity was realized in 2-D compared to 1-D, to further maximize
the sensitivity of the polymer energy migration in three dimensions was studied. The
Langmuir-Blodgett (LB) technique was used to construct layers of aligned PPE in order
to facilitate dipolar Forster type processes for efficient intermolecular energy transfer
from the PPE to surface acridine orange acceptors. Increasing the number of polymer
layers steadily enhanced the acridine orange emission, with the energy transfer peaking at
16 layers." In analogy to the amplification observed for the two dimensional process, the
energy trapping efficiency was maximized in 3-D as the exciton did not retrace its steps.
These insights have been applied to directed energy transfer with PPEs by layering
polymer films with decreasing bandgaps on top of one another. Energy was preferentially
transferred to the surface of the thin film, where the bandgap was the smallest (Figure
1.5).14 This ability to control the exciton pathway has important implications for the
design of sensors.
h\h"
I"
0
\
RO
-
OC2H5
C12H25 0
I
RO
OR
OR
O
O
n
Ro
R'= CON(CaH,7) 2
3
Figure1.5. Energy transfer from polymers 5 to 3 to 6. The films had decreasing bandgaps from left to
right. Emission occurred primarily from the most red-shifted polymer. (Reprinted with permission from
reference 14, copyright 2001 American Chemical Society)
While aggregation is usually an undesired characteristic for PPEs, when properly
controlled and designed it can be used for sensing purposes. This was demonstrated by
the detection of potassium ions with use of a 15-crown-5 substituted PPE. K+ induced a
2:1 complex with 15-crown-5 while Li + and Na+ formed 1:1 complexes with the same
crown ether. As a consequence, aggregation between polymer chains occurred only with
K+ and this was observed at a polymer to ion ratio of 0.5:1 (Figure 1.6). 15 Aggregation
was manifested by a diminished emission and appearance of a bathochromic band in the
absorption spectrum upon addition of the ion to a solution of 7. No change in the
absorbance and emission spectra was observed even at 1500 fold excess of Na+ and Li +.
The comonomer's steric and electronic properties were also important in facilitating nstacking interactions between polymer chains. A selective sensor for potassium ion was
thus constructed.
C-0
6""-)
.
Cl~H,·,O;/
3R
-o
by
R~
Ký: 0 p
7 R=CH3
CloH2
Figure1.6. Schematic representation of K ÷ ion-induced aggregation. (Reprinted with permission from
reference 15, copyright 2000 Wiley-VCH Verlag GmbH)
PPEs for biological applications
The amplification ability of conjugated polymers can be used in biosensors and
this capability becomes especially relevant considering the often minute quantities of
biological analytes. Biological interactions such as those between proteins and their
ligands, DNA strands, carbohydrates and cells could all be potentially amplified using
conjugated polymers. To be compatible with biological systems the polymer should be
hydrophilic and this is usually accomplished by installing ionic groups onto the polymer
backbone. The transducer should also be highly selective in only amplifying the desired
signal while minimizing response due to non-specific interactions.
Numerous biosensing assays based on quenching schemes have been designed
with ionic poly(phenylene vinylene)s. 16-20 One of the earliest examples for such a
biosensor involved a biotin-functionalized viologen that quenches an anionic conjugated
polymer (poly[lithium 5-methoxy-2-(4-sulfopropyloxy)-1,4-phenylenevinylene],
MPS-
PPV) in water. The quencher could be removed by adding avidin, resulting in
fluorescence recovery (Figure 1.7). The authors claimed that limitations of this bioassay
lay in the requirements for the analyte. A protein that was too small may be unable to
fully remove and segregate the quencher from the polymer. Changes in the charge of the
quencher and in the protein may also affect the sensitivity of the ionic polymer.
.#-0
HN
+
Unquenched
MPS-PPV
Quenched MPS-PPV
=
+
Streptavidin
NH
H
--
,Si\/
I-
Figure1.7. Removal of a biotin-labeled quencher by avidin resulted in recovery of the MPS-PPV
fluorescence.
Some of the intricacies of biosensor design were illustrated by a later report,21
which refuted the results from this early paper on avidin detection. Following a
systematic evaluation of all the contributing factors, a subtle interplay between steric bulk,
hydrophobic content and ionic groups were determined to have contributed to the
misguided interpretations. In water, mixing avidin to the conjugated polymer resulted in
an increase in fluorescence intensity as the large and positively charged avidin elongated
and separated the aggregated polymer chains, thereby reducing self-quenching. When
avidin was added to a mixture of biotin-functionalized quencher and conjugated polymer
in water, the electrostatic binding of the quenchers to the anionic polymers was
overwhelmed by the interactions with avidin, so the previously observed increased
fluorescence was erroneously interpreted as a fluorescence recovery. Indeed, if an excess
of the stoichiometric amount of avidin was added to the solution, an intensity equivalent
to 150% of the unquenched polymer resulted. Any resemblance to fluorescence recovery
was purely coincidental. In buffered solutions designed to maximize biotin-avidin
association, the quenching effected by the biotin-functionalized quencher was diminished
by 2-3 orders of magnitude due to electrostatic screening. Contrary to the earlier report,
enhanced quenching was observed when avidin was added to the solution. The authors
attributed this to a number of reasons: as the protein was cationic, addition of avidin
created a biotin-avidin complex that had a larger net positive charge, which resulted in a
larger association constant with the anionic polymer. Hydrophobic interactions between
the polymer and the protein could also contribute to the enhanced association.
Furthermore, encapsulation of the biotin-functionalized quencher into avidin was not
complete. This culminated in increased association of the exposed quencher to the
polymer resulting in enhanced overall quenching.
Underlining the importance of electrostatic interactions in biological systems,
electron transfer quenching analogous to that between MV2+ and conjugated polymers
was observed with proteins. Fluorescence quenching of poly[lithium 5-methoxy-2-(4sulfobutoxy)- 1,4-phenylenevinylene]
was carried out with cytochrome
c, a cationic
protein at neutral pH that can undergo rapid electron transfer. 8 Binding of the protein to
the polymer was sensitive to the pH of the solution and dropped by up to 6 orders of
magnitude when the pH was increased to where the protein was slightly anionic. Control
experiments were conducted with myoglobin and lysozyme. No electron transfer could
occur in the case of myoglobin, however at a pH where it had the same surface charge as
cytochrome c at neutral pH, significant quenching could still be obtained. Quenching of
the polymer was also observed upon addition of the cationic lysozyme. In this case, there
existed certainly some dependence of quenching efficiency on the electron transfer ability
19
of the protein analyte. However, the charge of the protein also played a significant role in
causing nonspecific quenching of the polymer, possibly by inducing aggregation and
subsequent self-quenching.
To create a biosensor that was specific toward a particular substrate, a watersoluble PPE containing carboxylic acid terminated oligo(ethylene glycol) side chains was
synthesized. The polymer was amenable to functionalization with biomolecules, such as a
peptide cleavable by the enzyme trypsin. The peptide sequence was also functionalized
with an electron accepting 2, 4-dinitrophenylamino group that could associate with the
fluorescent polymer and serve as a quencher. Upon exposure to trypsin, an increase in
fluorescence of approximately 1 order of magnitude occured as a result of hydrolysis and
subsequent diffusion of the quencher away from the polymer. Control experiments
showed that fluorescence recovery is significantly slowed down by a trypsin inhibitor.
Fluorescence increase did not occur with a non-peptidic substrate or in the presence of
inhibitor alone.22
GLGGGGN
AcHN
N
O
GPLGRGGGG
COOH
AcHN
02N
HN
h
02
NH
NO2
hv
hv2N
Figure1.8. Schematic representation of a conjugated polymer turn-on sensor based on quencher removal
upon exposure to the protease trypsin. Covalent attachment of the quencher caused the polymer to stay in a
non-fluorescent state. Upon removal of the quencher, the polymer fluorescence recovered.
While quenching/unquenching experiments are useful and provide sensitive
response to analytes, a turn on sensor that generates an emission at a new wavelength
offers advantages such as improved sensitivity and selectivity, as well as diminished
response to nonspecific interactions.
An obvious method is to use fluorescence
resonance energy transfer (FRET) to transducer the recognition event. For this purpose
our group demonstrated the efficient energy transfer from a PPE to a fluorescent pH-
sensitive dye. Films of cationic PPE 8 and the anionic fluoresceinamine appended
polyacrylate 9 were coated onto a glass substrate using layer by layer deposition. The
absorption cross section, energy migration efficiency and emission efficiency of the
pendant fluoresceinamine dye could change as a function of pH. At high pH, the dye was
highly absorptive and fluorescent, acting as a shunt and withdrawing energy from the
light harvesting conjugated polymer. At low pH, the absorbance decreased and there was
no fluorescence. Excitation of the PPE at 420 nm resulted in a 10 fold increase in
emission of the dye relative to its emission obtained by direct excitation. At pH 11, -90%
of the PPE's emission was transferred to the dye.23
\ /
Ci-
6
0
.7
A/ it NaO
3S0
;CI-
C
0
8
Cr
Ht
H
.HO
450
500
550
600
650
Figure 1.9. A film composed of 9 sandwiched between two layers of 8. The fluorescence spectra spanning
from 435 to 650 nm and the spectra beginning at 515nm were excited at 420 and 500nm, respectively.
Inset: the emission maximum of the fluoresceinamine band after excitation at 420nm plotted against pH.
(Reprinted with permission from reference 23, copyright 2000 American Chemical Society)
Energy transfer platforms were used in the design of sensors that detect negatively
charged peptide nucleic acid (PNA)-DNA,24 2-
7
DNA-DNA,82
-3
, RNA-peptide 3 1 duplexes,
and RNA assemblies32 . Typically these assays use cationic polymers based on
poly(fluorene)s,
a fluorophore-labeled
probe DNA/PNA/peptide
strand, and a non-
labeled single-stranded DNA/RNA target. Upon formation of the negatively charged
recognition duplex, the fluorophore was brought to close proximity with the polymer by
electrostatic interactions between the anionic duplex and cationic polymer and emission
from the dye occurred due to energy transfer. A three-tiered energy transfer assay was
also constructed where energy transfer occurred from the conjugated polymer to a
21
fluorescein labeled DNA, which in turn transferred energy to an intercalated ethydium
bromide in the double-stranded DNA duplex 29. This could potentially improve selectivity
and optical resolution of the biosensor.
8)
* 0
c.
CI
S-
A
SB
8
0E
0
G
Figure 1.10. Cationic poly(fluorene) with a fluorescently labeled peptide nucleic acid probe sequence. Path
A) Recognition between the probe sequence and the anionic complementary DNA target strand resulted in
an anionic complex which associates with the cationic polymer. Energy transfer occured as the
chromophore was in close proximity to the polymer. Path B) No recognition between the probe sequence
and a mismatched DNA strand. The chromophore was not brought to close proximity with respect to the
conjugated polymer. No energy transfer occurred. (Reprinted with permission from reference 28, copyright
2003 American Chemical Society)
Affinitychromic sensors using sensitive cationic water-soluble polythiophenes
could detect a variety of recognition events between biotinylated polythiophene-avidin, 33
DNA-DNA 34 36 and aptamer-protein.37 Electrostatic and conformational differences that
occur upon recognition disrupt the planarization and aggregation of the polymer
backbone, leading to visible absorption and fluorescence changes. For example, in the
detection of DNA, the native polymer in buffer solution was a yellow solution
corresponding to a random coil conformation. When a single stranded capture DNA is
added, the polymer-DNA solution became red as the polymer adopted to a highly
conjugated, planar conformation. Upon addition of the complementary DNA strand and
hybridization, the solution turned yellow due to the polymer adopting a less conjugated,
non-planar form. Single base mismatch detection was possible by monitoring the
colorimetric changes, as the hybridization was imperfect and therefore had different
complexation kinetics when compared to the perfect complement. Remarkably, analyte
detection levels down to 10-15 could in some cases be achieved.35
e
Hybridization
zooL
WE-r
Positively charged
Polythiophene
eee
8
e
-e( e
8 '
Single-stranded
DNA probe
I;e
"Duplex"
"TrIplex"
Figure1.11. A native random coil poly(thiophene) became more conjugated and planar upon complexation
with a single stranded DNA. When the complementary DNA strand bound to form a double helix, the
poly(thiophene) became more distorted and less conjugated. (Reprinted with permission from reference 36,
copyright 2002 Wiley-VCH Verlag GmbH)
Biosensors based on poly(diacetylene)s were also investigated by other groups.3843 These proved successful in detecting biologically relevant agents such as the influenza
virus, cholera toxin and catalytic activities of phospholipases. Analogous to colorimetric
poly(thiophene)s, poly(diacetylene)s could undergo changes in their conjugation length
upon binding of an analyte. Whether in a film, a liposome, or immobilized into a sol-gel
matrix, these polymers proved valuable in providing a rudimentary model for cell
membrane recognition processes.
References
(1)
Breer, H. Handbookof Biosensorsand ElectronicNoses: Medicine,Food and
Environment; CRC Press: Boca Raton, 1997.
(2)
Leonard, P.; Hearty, S.; Brennan, J.; Dunne, L.; Quinn, J.; Chakraborty, T.;
O'Kennedy, R. Enzyme Microb. Tech. 2003, 32, 3-13.
(3)
Rodriguez-Mozaz,
S.; Marco, M.-P.; Lopez de Alda, M. J.; Barcel6, D. Anal.
Bioanal. Chem. 2004, 378, 588-598.
(4)
Clark, L. C.; Lyons, M. Ann. NYAcad. Sci. 1962, 102, 29-45.
(5)
Updike, S. J.; Hicks, G. P. Nature 1967, 214, 986-988.
(6)
D'Orazio, P. Clin. Chim. Acta 2003, 334, 41-69.
(7)
Cooper, M. Anal. Bioanal. Chem. 2003, 377, 834-842.
(8)
Swager, T. M.; Gil, C. J.; Wrighton, M. S. J. Phys. Chem. 1995, 99, 4886-4893.
(9)
Zhou., Q.; Swager, T. M. J. Amer. Chem. Soc. 1995, 117, 7017-7018.
(10)
Zhou, Q.; Swager, T. M. J. Amer. Chem. Soc. 1995, 117, 12593-12602.
(11)
Yang, J.-S.; Swager, T. M. J. Amer. Chem. Soc. 1998, 120, 5321-5322.
(12)
Yang.,,J.-S.; Swager, T. M. J. Amer. Chem. Soc. 1998, 120, 11864-11873.
(13)
Levitsky, I. A.; Kim, J.; Swager, T. M. J. Amer. Chem. Soc. 1999, 121, 1466-1472.
(14)
Kim, .J.;McQuade, D. T.; Rose, A.; Zhu, Z.; Swager, T. M. J. Amer. Chem. Soc.
2001, 123, 11488-11489.
(15)
Kim, J.; McQuade, D. T.; McHugh, S. K.; Swager, T. M. Angew. Chem. Int. Ed.
2000, 39, 3868-3872.
(16)
Chen, L.; McBranch, D. W.; Wang, H.-L.; Helgeson, R.; Wudl, F.; Whitten, D. G.
P. Nal. Acad. Sci. USA 1999, 96, 12287-12292.
(17)
Wang, D.; Gong, X.; Heeger, P. S.; Rininsland, F.; Bazan, G. C.; Heeger, A. J. P.
Natl. Acad. Sci. USA 2002, 99, 49-53.
24
(18)
Fan, C.; Plaxco, K. W.; Heeger, A. J. J. Amer. Chem. Soc. 2002, 124, 5642-5643.
(19)
Kushon, S. A.; Ley, K. D.; Bradford, K.; Jones, R. M.; McBranch, D. W.; Whitten,
D. G. Langmuir 2002, 18, 7245-7249.
(20)
Kushon, S. A.; Bradford, K.; Marin, V.; Suhrada, C.; Armitage, B. A.; McBranch,
D. W.; Whitten, D. G. Langmuir 2003, 2003, 6456-6464.
(21)
Dwight, S. J.; Gaylord, B. S.; Hong, J. W.; Bazan, G. C. J. Amer. Chem. Soc.
2004, 126, 16850-16859.
(22)
Wosnick, J. H.; Mello, C. M.; Swager, T. M. J. Amer. Chem. Soc. 2005, 127,
3400-3405.
(23)
McQuade, D. T.; Hegedus, A. H.; Swager, T. M. J. Amer. Chem. Soc. 2000, 122,
12389-12390.
(24)
Xu, Q.-H.; Gaylord, B. S.; Wang, S.; Bazan, G. C.; Moses, D.; Heeger, A. J. P.
Natl. Acad. Sci. USA 2004, 101, 11634-11639.
(25)
Liu, B.; Bazan, G. C. J. Amer. Chem. Soc. 2004, 126, 1942-1943.
(26)
Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. P. Natl. Acad. Sci. USA 2002, 99,
10954-10957.
(27)
Gaylord, B. S.; Massie, M. R.; Feinstein, S. C.; Bazan, G. C. P. Natl. Acad. Sci.
USA 2005, 102, 34-39.
(28)
Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Amer. Chem. Soc. 2003, 125, 896900.
(29)
Wang, S.; Gaylord, B. S.; Bazan, G. C. J. Amer. Chem. Soc. 2004, 126, 5446-
5451.
(30)
Liu, B.; Bazan, G. C. Chem. Mater. 2004, 16, 4467-4476.
(31)
Wang, S.; Bazan, G. C. Adv. Mater. 2003, 15, 1425-1428.
(32)
Liu, B.; Baudrey, S.; Jaeger, L.; Bazan, G. C. J. Amer. Chem. Soc. 2004, 126,
4076-4077.
25
(33)
Bemier, S.; Garreau, S.; Bera-Ab6rem, M.; Gravel, C.; Leclerc, M. J. Amer. Chem.
Soc. 2002, 124, 12463-12468.
(34)
Leclerc, M.; Ho, H.-A. Synlett 2004, 2, 380-387.
(35)
Dor6, K.; Dubus, S.; Ho, H.-A.; Lvesque, I.; Brunnette, M.; Corbeil, G.;
Boissinot, M.; Boivin, G.; Bergeron, M. G.; Boudreau, D.; Leclerc, M. J. Amer.
Chem. Soc. 2004, 126, 4240-4244.
(36)
Ho, H.-A.; Boissinot, M.; Bergeron, M. G.; Corbeil, G.; Dor6, K.; Boudreau, D.;
Leclerc, M. Angew. Chem. Int. Edit. 2002, 41, 1548-1551.
(37)
Ho, H.-A.; Leclerc, M. J. Amer. Chem. Soc. 2004, 126, 1384-1387.
(38)
Charych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D. Science 1993, 261,
585-588.
(39)
Reichert, A.; Nagy, J. O.; Spevak, W.; Charych, D. H. J. Amer. Chem. Soc. 1995,
1995, 829-830.
(40)
Yamanaka, S. A.; Charych, D. H.; Loy, D. A.; Sasaki, D. Y. Langmuir 1997, 13,
5049-5053.
(41)
Spevak, W.; Foxall, C.; Charych, D. H.; Dasgupta, F.; Nagy, J. O. J. Med. Chem.
1996, 39, 1018-1020.
(42)
Pan, J. J.; Charych, D. H. Langmuir 1997, 1997, 1365-1367.
(43)
Jelinek, R.; Okada, S.; Norvez, S.; Charych, D. H. Chem. Biol. 1998, 5, 619-629.
26
Chapter 2
A Model Biosensor - Biotin-Functionalized PPEs for
the Detection of Streptavidin
Partially adapted from: Zheng, J.; Swager, T.M. Chem. Commun. 2004, 2798-2799.
27
Introduction
The cell surface is a complex and constantly fluctuating environment. Comprised
mostly of phospholipids which form the lipid bilayer, other molecules such as proteins
and saccharides also protrude from the surface. These serve as reporters, relaying
information regarding nutrients, toxins, extracellular matrix, attractants or foreign
invaders to the cell. Recognition events for cell and pathogen adhesions also occur at the
cell interface and are mediated by specific ligand-receptor interactions, which are often
multivalent. The valency corresponds to the number of separate identical connections that
one particle can have with another. In a multivalent
recognition
scheme, there is
simultaneous association of two or more ligands with receptors.
For a chemical system, the receptor is a protein with a pocket on its surface; and
the ligand is a molecule that can fit into this pocket. In cell surface biology, however,
both the receptor and the ligand can be called the receptor, with the convention that the
receptor is the species on the target cell surface. There are many mechanisms by which
multivalent interactions can operate. The chelate effect, where many binding sites are
occupied simultaneously, can occur when multivalent ligands bind to oligomeric
receptors. In this case, the first binding event accounts for the translational entropy cost,
and subsequent bindings only proceeds through conformational entropy cost. As
receptors can diffuse within the lipid membrane, ligands can cluster receptors and lead to
cellular signaling pathways. More than one binding site on the same receptor may also be
bound by the ligand. Additionally, higher affinities may be observed, as the ligand
displays a higher local concentration of binding elements (Figure 2.1).1
28
(a)
(b)(..
(d)
("T6
(C)a'
Currnt Opnon n Cheml
Biology
Figure2.1. Different mechanisms in multivalent sensing. (a) Chelate effect by multivalent ligands. (b)
Receptor clustering. (c) Recognition occurring at different binding sites on the same receptor. (d) A
multivalent ligand has higher apparent local concentration when binding to a receptor. (Reprinted with
permission from reference 1, copyright 2000 Elsevier)
Multivalent interactions present many advantages in biology. The strength of
multivalent interactions can be much stronger than a single interaction. It is possible to
generate a graded response depending on the number of interactions. There is also an
evolutionary advantage associated with multivalency, as one uses already existing
receptors. The possibility of binding to more than one type of receptor or ligand at the
same time can generate finer specificities. Conformational changes may also be effected
by having large surface areas involved in the recognition process, these can lead to
important consequences in signaling pathways.2
Energy transfer in multivalent sensing
Energy transfer is a phenomenon that is often used in biology to study
biomolecular structure and dynamics. 3' 4 Fluorescence resonance energy transfer (FRET)
occurs from the excited donor to an acceptor via coulombic interactions and is most
commonly observed. It is a dipole-dopole interaction and not the emission and
reabsorption of a photon. The rate of energy transfer depends on the spectral overlap
between the emission spectrum of the donor and the absorption spectrum of the acceptor,
the quantum yield of the donor, the transition dipole orientation of the donor and the
acceptor, and the distance between the two species.4 FRET is often used as a spectral
ruler to measure distances in biological applications. The F6rster distance 5 is defined as
the distance at which energy transfer is 50% efficient, and is typically in the range of 2090A. The rate of energy transfer is inversely proportional to r6, where r is the distance
between the donor and the acceptor.
In addition to fluorescence resonance energy transfer, also referred to as F6rster
energy transfer, a second mechanism of energy transfer can occur via electron exchange
between the excited donor and the ground-state acceptor. In this case, it is a collisional
mechanism, where the electron clouds of the two species overlap significantly in space
(Figure 2.2). The rate of energy transfer is directly related to the spectral overlap between
the donor and the acceptor, but it is independent of the absorption characteristics of the
acceptor. The rate of electron exchange energy transfer decreases as e-2R/L and becomes
negligible beyond 5-10A. It is independent of the oscillator strength of the D*-+D and
A--A* transitions.
(b) Electron Exchange Energy transfer
(a) Fluorescence Resonance Energy Transfer
LUMO
.
-
C ouombicl
Interae4ion
A
h
Joo
HOMO0 -0
D*
hV
D
A*
0
00
•
D*
A
D
A*
Figure2.2. Schematic representation of energy transfer mechanisms. (a) Coulombic interaction between
the excited donor and ground state acceptor results in energy transfer to the acceptor. (b) Electron exchange
energy transfer occurs via orbital overlap between the donor and the acceptor.
FRET has been used to study multivalent interactions and in biosensors. Swanson
et al. created biosensors that detect cholera toxin, a ligand with five identical binding sites
for the ganglioside GM1 receptor on the cell surface. 6 -9 A two-tiered energy transfer
scheme was designed to minimize the simultaneous excitation of the acceptor when the
donor was excited. In this case, the GM1 molecules were labeled with a donor, an
intermediate fluorophore with an absorption spectrum overlapping with the fluorescence
spectrum of the donor and a fluorescence spectrum overlapping with the acceptor, and an
acceptor whose absorption did not overlap with the emission of the donor. The labeled
GM1 were incorporated into a glass microsphere supported lipid membrane. Upon
exposure to cholera toxin, the three fluorophores were brought to close proximity, and
energy transfer occurred from the donor to the acceptor via the intermediate.6
Strong donor
& no acceptor
fluorescence
lfUflBilayer
U
Donor GM
4 CT
IntermediateGMI
Excitation
/
Emission
AcceptorGMI
Weak donor
& strong
acceptor
fluorescence
nSli\\ IS
U
UL ILUIL
Figure 2.3. Clustering of the fluorophore-labeled GM1 receptors by cholera toxin resulted in energy
transfer from the donor-labeled GM 1 to the acceptor-labeled GM1 via the intermediate chromophorelabeled GM1. (Reprinted with permission from reference 6, copyright 2001 Elsevier)
Biotinylated poly(p-phenylene ethynylene)s as model biosensor
In recent years, the fluorescence properties of conjugated polymers (CPs) have
been actively investigated in the design of highly sensitive chemical and biological
sensors, the majority of which have been based upon the amplification of fluorescence
quenching.0 -13 In contrast to these turn-off sensors, a turn-on sensor using FRET with
CPs as light-harvesting donors1 4 - 16 has the advantage of being more sensitive and
31
selective. Although FRET is a tool widely used in biology to study biomolecular structure
and dynamics,3 4 its use with CPs as a method of transduction for sensing biological
molecules is not common ' 71 8 In this chapter, we detail a model biosensor based on the
multivalent interactions between biotinylated poly(p-phenylene ethynylene)s and
fluorophore-labeled streptavidin.
Streptavidin is a 53 kDa tetrameric protein produced by the bacterium
Streptomyces avidinii. It derives its name from avidin, a homologous protein found in
chicken egg white, with which it shares one of the strongest associations for a single
interaction between a receptor and a small organic ligand, biotin. Each subunit of the
protein binds one molecule of the vitamin biotin with high affinity, with the association
constant at 1)4-105.19,20 The similarity also extends to other properties. Both proteins are
tetramers and share an identical 33% primary sequence at the homologous core. These
proteins may function as antibiotics that deplete the environment of the essential vitamin
biotin.
Unlike avidin,
streptavidin
lacks carbohydrate
side chains.
It also has an
isoelectric point that is close to neutrality at physiological conditions (pI = 5-6 compared
to 10 for avidin). As a consequence, streptavidin frequently exhibits lower non-specific
binding than avidin. It is a homotetramer with 24-32 lysine residues per unit and does not
contain cycteine residues, carbohydrate side chains or associated cofactors. Streptavidin
is stable over a wide pH range and is very heat stable, requiring up to 20 minutes at
1000 C in 0.2(!/oSDS to dissociate the subunits. 2 1 To separate the protein into dimers, 6M
urea can be used. Streptavidin
connected, aritiparallel
subunits are organized as eight-stranded,
sequentially
sheets. Pairs of streptavidin barrels hydrogen bond to form the
tetramer. The crystal structures of streptavidin and avidin with bound biotin have been
solved at increasingly finer resolution over the years. It seemed logical to use biotin-
streptavidin as a model system for creating a multivalent biosensor, as the interactions
were very well understood and could be considered to be essentially irreversible. While
other groups have used the streptavidin-biotin recognition system to model biosensor
32
design in conjunction with conjugated polymers in affinitychromic22' 23 and agglutination
assays,24 no examples using energy transfer has been demonstrated.
Results and Discussion
Synthesis of biotinylated PPEs
There are two main routes by which a biotinylated PPE can be synthesized. While
there is an obvious advantage to modifying a universal polymer once it has been
polymerized to generate a library of polymers, making a well-characterized biotinylated
monomer for polymerization leads to a polymer that is well-defined. This becomes
important when one is making a model system, where variables need to be limited so that
the results could be interpreted unambiguously.
We chose to synthesize a monomer that was substituted with only one biotin to
provide binding accessibility for streptavidin, while minimizing the divalent binding of
one streptavidin onto the same repeat unit (if it were symmetrically functionalized with
biotin).2 5 This provides a model for multivalent binding, as the protein could potentially
bind to multiple polymer strands. Biotinylated monomer 5 was synthesized according to
Scheme 2.1.. 1,4-diiodo-2,5-hydroquinone was reacted with tri(ethylene glycol)
monomethyl ether p-toluenesulfonate to generate 1. A second Williamson ether synthesis
using bromoethylacetate afforded 2, which was saponified to give the carboxylic acid 3.
Reaction with thionyl chloride yielded an acid chloride, which was then reacted with a
protected amine linker 10 to give 4. Deprotection using trifluoroacetic acid and reaction
with N-hydroxy-succinimido biotin afforded the final monomer 5.
In order to create a biosensor that was amenable to physiological conditions, it
was important that the polymer be water soluble. Using a carboxylic acid functionalized
comonomer 6, generously given by Jordan Wosnick of the Swager group, a water-soluble
33
biotinylated
PPE 8 and its non-biotinylated
relative 9 were synthesized for solution
energy transfer experiments via a Sonagashira-Hagihara cross-coupling reaction.26
OH
OH
i TsO,.__/O_o_
o0
°J4O0
Br_Qk
/
0
_
NaH, DMF, 900°C
K2CO3
2-butanone, reflux I
OH
O --
o
-O -,
o.-
O
o
O
o
L
I
H
KOH, MeOH
900C, reflux
NO
°
1. SOC12, reflux
0
H
I
2.H2N
100
ON
-- 010
CHC 3, NEt3
reflux
3
4
0
kNH
H
o0H
1. TFA
2. DMF, NEt3
HN
HN
"
5,f,
0
N
H
NH
O-O
S
0H
NH
eov
"'"%OH
5
Scheme 2.1. Synthesis of biotinylated monomer 5.
Polymer 8 was constructed from two diiodobenzene monomers at loading ratios
of 1:4 (biotinylated to non-biotinylated monomers) that were polymerized by a crosscoupling reaction with a diacetylene monomer. Morpholine was used as the solvent in
polymerization reactions because it solubilizes the polymer chains while also serving as
the amine base in the coupling reaction. Purification of the water soluble polymers was
by dialysis against water in 10,000-MWCO dialysis tubing. Polymer 9 did not contain
any biotinylated monomer and served as the control polymer.
34
~O o
H
0
HNNH
0
XIH
N
H
-oo°-NoH
H~~~~~o
I0
morpholine,
O'ofAo
o../'OO/
OH
O /OO
O_o/
0.8eq.
6
0.2 eq.
5
0
.
Cul,
60°C
Pd(PPh)O
n
oN°
1.0 eq.
7
O
O
oNH
o
O
x/y = 1/4
HNNH
Scheme 2.2. Synthesis of water-soluble biotinylated polymer 8.
Cul,
Pd(P
)4
3
/
OH
0
0
O-Oo-'O"VO
OH
,
morpholine,60°C
OH
1.0 eq.
6
o
0
-
-
\ /
/
O
0
1.0 eq.
7
0
0
HO
Scheme 2.3. Synthesis of water-soluble control polymer 9.
Solution energy transfer assays
As an initial assay, biotinylated polymers 8 and control polymer 9 were incubated
with fluorescein-labeled streptavidin (3.5 dyes/protein) at room temperature, in 50 mM
Tris buffer at pH 7.5 for five minutes. Fluorescein was selected as its absorbance
maximum at 490 nm overlaps well with the emission maximum of polymer 8 at 486 nm
(excitation at 440nm, Figure 2.4). This would favor FRET by the F6rster mechanism
between the polymer donor and dye acceptor upon binding of labeled streptavidin to
biotin. When 0.030 nmol of labeled streptavidin was added to 2.16 nmol of 8, an increase
in the fluorescein's
emission was observed (Figure 2.5). This was at first glance
promising. However, the data was complicated by several factors. Firstly, the emission of
35
the fluorescein overlapped with that of the polymer. Secondly, fluorescein absorbed at the
excitation wavelength, thus the observed emission was due to direct excitation and also
due to energy transfer. In order to separate these contributing factors, the data was subject
to a deconvolution process. Nevertheless, a control experiment was carried out with 9.
Minimal energy transfer was observed without biological recognition.
Wavelength (nm)
Figure 2.4. Polymer 8 emission and fluorescein-labeled streptavidin spectra.
450
500
550
600
650
700
Wavelength (nm)
Figure2.5. Addition of fluorescein-labeled streptavidin to polymer 8 in tris buffer at pH 7.5. Fluorescence
emission decrease of the polymer was accompanied by fluorescence emission increase of the fluorescein.
The overlapping fluorescence spectra were deconvoluted to separate fluorescein's
emission from that of the polymer. Varying amounts of fluorescein labeled streptavidin
were added to 4.32 nmol of 8. After data processing, much of the fluorescein emission
was due to direct excitation (Figure 2.6) and very little was actually due to energy
transfer. Although the degree of enhancement in the fluorescence emission was low,
these results indicated that biological recognition was necessary for ET from the polymer
to the dye-labeled streptavidin.
x 10
7
x 10
2.5
7
0.17 nmol
2
1.5
1
0.
450
500
550
600
650
700
0
x 10
-
450
500
550o
00
650
700
Actual data
Polymer fluorescence contribution
(fitted data)
Fluoresceil labeled streptavidm:
fluorescence contribution in
presence of FRET (fitted data)
Sum of fitted data
Fluorescein labeled streptavidin:
fluorescence emission in
absence of FRET
Figure2.6. Deconvoluted spectra showing contribution due to energy transfer to the fluorescein emission.
In order to better determine ET between the polymer donor and dye-acceptor, a
more red-shifted rhodamine B-labeled streptavidin (RhB-strept) was used in the solution
phase ET assays with 8 (Figure 2.7). To our surprise, higher ET was observed even
though RhB had a diminished spectral overlap with 8 (emission maximum 8: 486 nm,
absorption maximum RhB-strept: 574 nm, 4.6 dyes/protein). At this point we decided to
screen 8 with Texas red XTM-labeled streptavidin (T-red-strept) (absorption maximum
591 nm, 2.9 dyes/protein, Figure 2.9). Remarkable ET was observed. For both dyes the
emission due to ET was amplified compared to direct excitation of the dyes at their
absorbance maximum (Figures 2.8 and 2.10). This was consistent with the lightharvesting properties of conjugated polymers.
"(D
U
U,
CU
vC:
u,
U,
750
E
.0
.0
<Oj
U_
a)
0
ILl
400
450
500
550
600
650
Wavelength (nm)
Figure2.7. Polymer 8 emission and rhodamine-labeled streptavidin spectra.
0
to
E
w
IZ
(..
C
450
500
550
600
Wavelength (nm)
650
700
Figure 2.8. Addition of rhodamine-labeled streptavidin to polymer 8 in tris buffer at pH 7.5. Fluorescence
emission decrease of the polymer was accompanied by fluorescence emission increase of the rhodamine.
--
I
II
450
500
Texas Red Streptavidin
Polymer Emission
'
'
550
II
600
Wavelength (nm)
'
I
'
650
Figure2.9. Polymer 8 emission and Texas Red X-labeled streptavidin spectra.
I
700
0;
U
U
0
450
500
550
600
650
700
Wavelength (nm)
Figure 2.10. Addition of Texas Red X-labeled streptavidin to polymer 8 in tris buffer at pH 7.5.
Fluorescence emission decrease of the polymer is accompanied by fluorescence emission increase of the
Texas Red.
Control experiments with 9 showed no ET upon addition of both dye-labeled
streptavidin derivatives (Figure 2.11). A control experiment with the addition of a biotin
pre-saturated solution of T-red-strept to biotinylated 8 was also carried out (Figure 2.12).
Again no decrease in fluorescence of the polymer and no ET to the dye were observed.
450
500
550
600
Wavelength (nm)
650
700
450
500
550
800
Wavelength (nm)
650
Figure 2.11. Control experiments showing no energy transfer. Left: addition of rhodamine B-labeled
streptavidin to polymer 9. Right: addition of Texas Red X-labeled streptavidin to polymer 9.
0
U
0
U
Ur.
450
500
600
550
Wavelength (nm)
650
700
Figure2.12. Addition of a biotin presaturated Texas Red X-labeled streptavidin to polymer 8. No energy
transfer was observed.
The quantum yields of the streptavidin-bound dyes varied upon binding to
polymer 8, presumably due to an aggregation or an environmental change in their vicinity.
This effect was observed by directly exciting the dyes at their maximum absorbance
(where the polymer does not absorb) using the same polymer concentration as in Figures
2.8 and 2.10. In the presence of 8, RhB-strept's quantum yield was diminished by 38%
while that of T-red decreased by 63%. Nevertheless, greater emission intensity was
observed for T-red-strept despite the greater decrease in its quantum yield as compared to
RhB-strept. The strong emission response from T-red-strept was therefore not due to a
simple improvement in its quantum efficiency.
To study the nature of the interactions between the free dyes and 8 we determined
the Stern-Volmer quenching constants from fluorescence emission and lifetime
measurements in 50 mM Tris buffer at pH 7.4. Upon addition of the non-protein
conjugated fluorescent dyes (fluorescein, RhB and sulforhodamine 101 (Texas redTM
parent dye)) to 8, the apparent Ksv values were determined to be 26,300 M-', 91,800 M-1
and 97,900 M-1 respectively. The bi-molecular quenching constant kq ranged from 1.25 x
1014 M-S-1 to 3.4 x 1014 M-IS- for the three dyes, which greatly exceeded the diffusion
constant and was indicative of static quenching. The dyes therefore had an inherent
affinity for the conjugated polymer backbone. A more planar conformation and greater
hydrophobic character for Texas redTMcompared to RhB and fluorescein may permit
better stacking and orbital interaction with the CP backbone, allowing for greater ET. In
the case of dye-labeled streptavidin, the biological recognition first brought the dyes into
closer proximity with the polymer. Conformational and hydrophobic characteristics of
the dyes then tailored the extent of orbital mixing with the polymer: the flatter Texas
RedTMinteracted most intimately with the planar conjugated polymer backbone. This may
contribute to the better ET even at decreased spectral overlap between the CP donor and
dye acceptor..
Thin film assays
The solution-based assays provided us with insights into the energy transfer
mechanism from PPEs to the fluorescently-labeled streptavidin. To complement these
studies, we synthesized organic solvent-soluble polymers 13 and 14.
These were
designed with a pentiptycene in the backbone to promote greater thin-film quantum
yield.27 They can be spin-coated onto glass coverslips and incubated in aqueous solutions
containing various analytes. Following incubation, the coverslips can be thoroughly
rinsed and dried.
1
H
(
oH
NH
0:Y1
|
+
A
Cul, Pd(PPh ) ,
3 4
toluene, DIPA, DM
1I>
60°C
0.2 eq.
5
1
0.8 eq.
11
I
1.0 eq.
12
Scheme 2.4. Synthesis of biotinylated polymer for thin film studies.
42
0~8\ ///Cul,
Pd(PP
)DMF
4,
3
toluen
DIPA
0
o-to
+
-(
1 0 eq
1.0 eq.
- o~oo
All
oW0
-
/
, /
1.0 eq.
.
Polymer 14
Scheme 2.5. Synthesis of control non-biotinylated polymer for thin film studies.
Thin film experiments have demonstrated to have superior sensitivity 7as the
excitons can sample a greater surface area compared to solution experiments. However,
when a polymer film is incubated with biological analytes, a commonly encounted
problem is non-specific binding. To increase specificity, incubation of polymers 13 and
14 with the dye-labeled streptavidin was screened against different detergents: sodium
dodecyl sulfate, Tween 20, Triton X-100 and octyl -glucoside. Non-specific binding was
least pronounced when samples were incubated in the presence of Triton X-100, a nonionic detergent. It was therefore included in all incubations. It was observed that RhB-
strept exhibited better ET than T-red-strept (Figure 2.13). However a small shoulder due
to non-specific binding was nonetheless observed in the case of polymer 14 incubated
with RhB-strept. This finding suggested that the smaller RhB dye was able to interact
more intimately with the sterically restrictive structure of polymers 13 and 14, leading to
greater ET.
43
0
C)
C)
0
450
500
550
600
Wavelength (mn)
650
700
Figure2.13. Fluorescence emission spectra of polymers 13 and 14. Non specific binding was observed
with RhB-strept. RhB-strept displayed greater energy transfer compared to T-red strept.
To verify the affinity of the dyes with the conjugated polymers, incubation of
polymers 13 and 14 was carried out with the streptavidin-free dyes (Figure 2.14). Indeed,
the free RhB dye associated with both 13 and 14, while free sulfo-rhodamine 101 (Texas
redTM) associated with neither. ET is therefore significantly dependent on factors that
influence the degree of interaction between the polymer and dye.
r-j
i'
¢IC)I
r
E
.
*
't)
r
4_
5O
4
50()0
550()
Wavelength
60()
650
7()(00
(Un)
Figure 2.14. Incubation of polymers 13 and 14 with small molecule dyes rhodamine and sulforhodamine
101, the Texas Red parent dye. Rhodamine displayed preferential association with the polymers.
Conclusions
Biotinylated water-soluble and organic solvent-soluble PPEs were synthesized for
use in energy transfer detection schemes. When dye-labeled streptavidin was added to the
water soluble polymers, energy transfer occurred from the polymer to the dye.
Surprisingly, with decreased spectral overlap, greater energy transfer was observed. This
was attributed to an electron exchange energy transfer mechanism component, where
there is orbital overlap between extended flat Texas red dye and polymer backbone. The
results were corroborated by the thin-film experiments, where the incorporation of
pentypticene units into the polymer facilitated association of the smaller rhodamine with
the concave hydrophobic pockets. The structure of the dye with relation to the polymer
was an important consideration in the design of sensors using energy transfer.
45
Experimental.
General. H and 13C NMR spectra for monomers and polymers were recorded on a
(Varian 300 MHz) or on a Varian VXR-500 (500 MHz) instrument. The chemical shift
data for each signal are given in units of 6 (ppm) relative to tetramethylsilane (TMS)
where 6 (TMS) = 0, and referenced to the solvent residual. High-resolution mass spectra
were obtained on a Finnigan MAT 8200 system using sector double focus and an electron
impact source with an ionizing voltage of 70V, and with a Bruker DALTONICS APEX II,
3 Tesla, FT-ICR-MS with ESI source or EI/CI source. UV-visible absorption spectra
were measured with a Cary 50 UV/visible spectrometer. Fluorescence spectra were
measured with a SPEX Fluorolog-2 fluorometer (model FL112, 450W xenon lamp). The
spectra in solution were obtained at room temperature using a quartz cuvette with a lcm
path length. Polymer thin film spectra were recorded by front-face (22.5° ) detection.
Fluorescence
quantum
yields of polymers
in Tris buffer (100mM, pH 7.4) were
determined relative to solutions of coumarin 6 (F
=
0.78 in ethanol) as a reference. The
molecular weights of polymers were determined by using three PLgel 5m 105, 104, 103
(300 x 7.5 mm I.D) columns in series and a diode detector at 254nm at a flow rate of
1.0ml/min in THF or in DMF. The molecular weights were reported relative to
polystyrene or poly(ethylene oxide) standards purchased from Agilent Inc. Polymer thin
films on a cover glass (18 x 18 mm, pretreated with 1,1,1,3,3,3-hexamethyldisilazane)
were spin cast on an EC101DT photoresist spinner (Headway Research Inc.) using a spin
rate of 3000 rpm from a chloroform solution. Melting point (m.p.) determination was
performed using a Laboratory Devices MEL-TEMP instrument (open capillaries used)
and was uncorrected.
Materials.
All solvents were spectral grade unless otherwise noted. Morpholine and biotin were
purchased from Alfa Aesar and used as received. Fluorescein conjugated streptavidin,
rhodamine-conjugated
streptavidin,
Texas
Red-X
conjugated
streptavidin
and
46
sulforhodamine 101 were purchased from Molecular Probes Inc. and used as received.
All other chemicals were purchased from Aldrich Chemical In. and used as received. All
air and water sensitive synthetic manipulations were performed under a nitrogen
atmosphere using standard schlenk techniques.
47
(1): To a 250ml round bottom flask equipped with a reflux condenser containing 2,5diiodo-1,4-dihydroxybenzene (10.00g, 27.6mmol) was added 125 ml anhydrous N, N'dimethyl formamide (DMF) under nitrogen. The solution was cooled to 0°C, and nitrogen
was bubbled through the solution for 15 minutes. NaH as a 60% dispersion in mineral oil
(1.326g, 33.2mmol) was added and the resulting suspension was stirred for 20 min at 0°C.
Triethylene glycol monomethyl ether p-toluenesulfonate (9,94g, 31.2mmol) was then
transferred to the solution via syringe. The reaction was heated at 650 C for 14h under
nitrogen. A light clear brown solution was obtained. DMF was removed under reduced
pressure and the resulting brown oil was extracted with ethyl acetate (500 ml total)
against 200ml H2 0. The organic layer was washed with 50ml brine and the solvent was
removed under reduced pressure. The product was purified by column chromatography
with 6:4 hexane/ethyl acetate to afford a colorless oil which solidified to a white solid
upon standing (3.98g, 28%). m.p. 81-83 0 C. H NMR (300 MHz, CDC13): 7.38 (1H, s),
7.09 (1H, s). 5.27 (1H, s), 4.08 (2H, t, J=4.5Hz), 3.88 (2H, t, J=4.5Hz), 3.79 (2H, t,
J=4.5Hz), 3.69 (2H, t, J=4.5Hz), 3.67 (2H, t, J=4.5Hz), 3.38 (3H, s);
13 C
NMR (125 MHz,
CDC13): 152.6, 150.5, 125.0, 121.9, 87.8, 84.4, 72.1, 71.3, 71.0, 70.8, 70.5, 69.8, 59.3;
HR-MS (EI) calcd. For C1 3H 18 12 0 5 (M+): 507.9238, found: 507.9239.
(2): In a 250ml round bottom flask were combined 1 (2.00 g, 3.94mmol), K2 CO3 (1.632g,
11.81mmol), ethyl bromoacetate (0.567ml, 5.12mmol) and 100ml acetone. The flask was
fitted with a reflux condenser and the reaction mixture was refluxed for 12h. A pale
yellow suspension resulted. This was cooled, filtered and the solvent was removed under
reduced pressure. The residue was purified by column chromatography with 6:4
hexane/ethyl acetate and the product was isolated as a colorless oil which solidified upon
standing to a white solid (2.02g, 86%). m.p. 44-45°C. H NMR (300 MHz, CDC13 ): 7.26
(1H, s), 7.17 (1H, s), 4.61 (2H, s), 4.30 (2H, q, J=4.2), 4.13 (2H, t, J=3Hz), 3.88 (2H, t,
J=3Hz), 3.80 (2H, t, J=3Hz), 3.70 (2H, t, J=3Hz), 3.68 (2H, t, J=3Hz), 3.57 (2H, t,
J=3Hz), 3.39 (3H, s), 1.32 (3H, t, J=4.2Hz); 13C NMR (125 MHz, CDC13): 168.4, 153.9,
152.4, 123.9, 123.6, 86.7, 86.4, 72.2, 71.4, 71.0, 70.8, 70.4, 69.8, 67.7, 61.7, 59.3, 14.5.
HR-MS (EI) calcd. For C1 7 H2 41 20 7 (M+): 593.9606, found: 593.9625.
48
(3): In a 25() ml round bottom flask were combined 2 (2.00g, 3.36mmol) and KOH
(0.944g, 16.8mmol) in 70ml methanol. A reflux condenser was fitted and the reaction
was heated to reflux for 14h. The solvent was removed under reduced pressure. 45ml
10% HCl(aq)was added. The product precipitated and was isolated by centrifugation
followed by lyophilization. A white solid was obtained (1.79g, 94%). m.p. 74-760 C. H
NMR (300 MHz, DMSO): 7.38 (1H, s), 7.24 (1H, s), 4.72 (2H, s), 4.10 (2H, t, J=4.5Hz),
3.73(2H, t, J=--4.5Hz),3.62 (2H, t, J=4.5Hz), 3.53 (2H, t, J=4.5Hz), 3.52 (2H, t, J=4.5Hz),
3.42 (2H, t, J=4.5Hz), 3.23 (3H, s); 3C NMR (125 MHz, DMSO): 169.8, 152.7, 151.7,
123.0, 122.2, 86.7, 86.5, 71.3, 70.2, 69.9, 69.7, 69.6, 69.0, 66.1, 58.0; HR-MS (ESI) calcd.
For C15H2 0 12 0
7
(M+Na): 588.9191, found: 588.9182.
(4): In a 50ml round bottom flask equipped with a reflux condenser containing 3 (0.500g,
0.883mmol) was added 5ml SOC12. This was refluxed for 10h. The thionyl chloride was
then removed under reduced pressure to afford the acid chloride as a pale yellow oil
(0.521g, 0.883mmol). To this was then added 20ml CH2C12. Anhydrous NEt3 was then
added (0.185ml,
1.32mmol)
and the mixture was stirred for 5min. 1028 (0.329g,
1.32mmol) was added as a solution in 10ml CH 2C12. The reaction mixture was refluxed
for 12h. The solvent was removed under reduced pressure. The residue was dissolved in
100ml CHC:3 and washed with 30ml H2 0. The organic layer was washed with 15ml
brine, dried over MgSO4. The organic solvent was removed under reduced pressure to
afford a colorless oil which solidified upon standing to a white solid (0.560g, 80%). m.p.
81-83 0C. 'H NMR (300 MHz, CDC13): 7.28 (1H, br), 7.25 (1H, s), 7.17 (1H, s), 4.90 (1H,
br), 4.13 (2H., t, J=4.5Hz), 3.90 (2H, t, J=4.5Hz), 3.80 (2H, t, J=4.5Hz), 3.81-3.52 (16H,
m), 3.39 (3H, s), 3.32 (2H, t, J=5.1Hz), 1.45 (9H, s); 13C NMR (125 MHz, CDC13):
206.1, 167.4, 156.2, 154.1, 151.4, 123.3, 86.8, 86.3, 79.5, 72.2, 71.4, 71.0, 70.8, 70.6,
70.6,
70.5,
70.0,
69.8,
69.1,
59.3, 40.5,
39.1,
28.7; HR-MS
(ESI)
calcd.
For
C2 6H4 212 N2 010 (M+H): 797.1002, found: 797.1022.
49
(5): A 50ml round bottom flask containing 4 (0.487g, 0.61 lmmol) was loaded with 2ml
TFA. The clear yellow solution was stirred for 30min. The TFA was removed, 2ml H2 0
was added and was also removed under reduced pressure. The deprotected product was
dried under high vacuum. To this was added 5ml anhydrous DMF, NEt3 (0.450ml,
3.22mmol). This was stirred for 15min, then N-hydroxysuccinimido biotin29 (0.212g,
0.624mmol) was added. The pale yellow solution quickly became a thick white slurry
and was stirred at room temperature for 40h. The solvent was removed under reduced
pressure at 400 C and the reaction mixture was washed with 25ml H20. The product was
isolated by centrifugation and lyophilized to afford a white powder (0.525g, 94%). m.p.
175-1760 C. H-INMR (500 MHz, CDC13): 7.85 (2H, m), 7.39 (1H, s), 7.31 (1H, s), 6.43
(1H, s), 6.36 (1H, s), 4.52 (2H, s), 4.30 (1H, m), 4.11 (1H, m), 3.74 (2H, t, J=5.OHz), 3.62
(2H, t, J=5.0Hz), 3.54-3.30 (16H, m), 3.22 (3H, s), 3.18 (2H, m), 3.08 (2H, m), 2.80 (1H,
dd, J=12.5, 5.0Hz), 2.58 (J=12.5Hz), 2.06 (2H, t, J=7.5Hz), 1.62-1.57 (1H, m), 1.52-1.43
(3H, m), 1.32-1.26 (2H, m); 13C NMR (125 MHz, CDC13): 172.1, 167.2, 162.7, 153.0,
151.7, 123.3, 122.7, 86.9, 86.8, 71.3, 70.2, 69.9, 69.7,69.2, 69.0, 68.8, 61.3, 61.0, 59.2,
58.1, 55.5, 38.44, 30.37, 35.1, 28.2, 28.1, 25.3; HR-MS (ESI) calcd. For C31 H4 8 12 N4 0 10 S
(M+H): 923.1253, found: 923.1210.
Polymer 8: A 25ml schlenk flask was charged with 5 (0.0205g, 0.022mmol), 6 (0.0606g,
0.089mmol, synthesis to be reported in a forthcoming publication)
and 730 (0.050g,
0.11lmmol), Pd(PPh3)4 (6.41mg, 0.0056mmol) and CuI (1.06mg, 0.0056mmol) under N2.
To this was added 1.5ml freshly degassed morpholine under N2 . The reaction vessel was
sealed and heated at 60°C for 48h. 3ml H 20 was added and the reaction mixture was
dialyzed (cellulose membrane, MWCO 10000) against 1L deionized water for 2 days (6
water changes). The polymer was then lyophilized to afford an orange polymer (97mg,
95%). Mn= 130,000, PDI=1.48 for DMF soluble fraction. H NMR (500 MHz, DMF):
7.29 (20H, br), 6.39 (1H, s), 6.32 (1H, s), 4.78 (2H, s), 4.33 (38H, br), 3.94 (24H, br),
3.78-3.46 (160H, broad multiplet), 3.28 (33H, br), 1.60 (8H, br).
50
Polymer 9: A 25 ml schlenk flask was charged with 6 (0.0454g, 0.066mmol) and 7
(0.030g, 0.066mmol), Pd(PPh3) 4 (3.85mg, 0.00333mmol) and Cul (0.634g, 0.00333mmol)
under N2. To this was added 1.0ml freshly degassed morpholine under N2. the reaction
vessel was then sealed and heated at 600 C for 48h. 3ml H 2 0 was added and the mixture
was dialyzed against 1L deionized water for 2 days (6 water changes). It was then
lyophilized to afford an orange polymer (56mg, 96%). Mn=128,000, PDI=1.53 for DMF
soluble fraction.
H NMR (500 MHz, DMF): 7.30 (4H, s), 4.34 (8H, br), 3.95 (8H, br),
3.79-3.46 (32H, br), 3.29 (6H, s)
Polymer 13: A 25ml schlenk flask was charged with 5 (0.00796g, 0.00819mmol) and 11
(0.0214g, 0.0328mmol), 1211(0.020g, 0.418mmol), Pd(PPh3) 4 (2.367mg, 0.00205mmol)
and Cul (0.390mg, 0.00205mmol) under N2. 1.5ml of a freshly degassed, mixture of 4:1
toluene/diisopropylamine and 0.5ml freshly degassed DMF were added via syringe. The
reaction vessel was sealed and heated at 600 C for 5 days. The polymer was isolated by
precipitation into methanol followed by centrifugation. A yellow powder was obtained
(32mg, 83%). Mn=7700, PDI=2.04 for THF soluble fraction. H NMR (500MHz, CDC13):
7.66-7.47 (60H, broad multiplet), 7.05 (40H, br), 6.42 (1H, s), 6.39 (1H, s), 6.10 (20H,
br), 5.30 (2H, br), 4.68 (20H, br), 4.26 (22H, br), 3.83 (20H, br), 3.65 (20H, br), 3.55
(20H, br), 3.44 (20H, br), 3.31 (19H, br), 2.78 (2H), 1.40-1.25 (6H, broad multiplet).
Polymer 14: A 25ml schlenk flask was charged with 11 (0.020g, 0.0306mmol) and 12
(0.0149g, 0.0312mmol), Pd(PPh3) 4 (1.766mg, 0.00153mmol) and CuI (0.291mg,
0.00153mmol)
under
toluene/diisopropylamine
N2.
1.5ml
of
a
freshly
degassed
mixture
of
4:1
was added via syringe. The reaction vessel was sealed and
heated at 600 C for 5 days. The polymer was isolated by precipitation into ethyl acetate
followed by centrifugation. A yellow powder was obtained (21.3mg, 80%). Mn=14000,
PDI=2.02 for THF soluble fraction.
H NMR (500 MHz, CDC13): 7.53 (10H, broad
multiplet), 7.05 (8H, br), 6.20 (4H, br), 4.68 (4H, br), 4.26 (4H, br), 3.82 (4H, br), 3.65
(4H, br), 3.55 (4H, br), 3.44 (4H, br), 3.31 (6H, br)
51
General protocol for energy transfer assays in solution phase
7.51tl of a stock polymer solution (lmg/ml in Tris buffer, 40mM at pH7.4) was diluted
with the same Tris buffer to a total volume of 3ml in a fluorescence cuvette. To this was
added aliquots of dye-labeled streptavidin (1l of a lmg/ml solution) and fluorescence
emission was taken at each addition. Excitation wavelength at 440nm was chosen, and
emission spectrum was taken from 455-700nm.
General protocol for energy transfer assays in solid phase
Microscope coverslips were pretreated in 1,1,1,3,3,3-hexamethyldisilazane.31
Polymer solutions at lmg/ml in chloroform were spin-cast onto microscope coverslips at
a spin rate of' 3000rpm for 1 minute. The coverslips were put under vacuum for 2h, then
were incubated in a solution of dye labeled streptavidin or dye for lh. The coverslips
were then washed with deionized water, blotted dry and dried under vacuum for a
minimum of 5h. Excitation wavelength at 400nm was chosen, and emission spectrum was
taken from 415-700nm.
52
References
(1)
Kiessling, L. L.; Gestwicki, J. E.; Strong, L. E. Curr. Opin. Chem. Biol. 2000, 4,
696-703.
(2)
Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem. Int. Edit. 1998, 37,
2754--2794.
(3)
Wu, 1'. G.; Brand, L. Anal. Biochem. 1994, 218, 1-13.
(4)
Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Kluwer
Academic/Plenum Publishers: New York, 1999.
(5)
F6rster, T. Ann. Phys. 1948, 2, 55-75.
(6)
Song, X.; Shi, J.; Nolan, J.; Swanson, B. I. Anal. Biochem. 2001, 291, 133-141.
(7)
Song, X.; Shi, J.; Swanson, B. I. Anal. Biochem. 2000, 284, 35.
(8)
Song, X.; Nolan, J.; Swanson, B. I. J. Amer. Chem. Soc. 1998, 120, 11514-11515.
(9)
Song, X.; Swanson, B. I. Anal. Chem. 1999, 71, 2097-2107.
(10)
Zhou., Q.; Swager, T. M. J. Amer. Chem. Soc. 1995, 117, 7017-7018.
(11)
Yang., J.-S.; Swager, T. M. J. Amer. Chem. Soc. 1998, 120, 11864-11873.
(12)
Fan, C.; Plaxco, K. W.; Heeger, A. J. J. Amer. Chem. Soc. 2002, 124, 5642-5643.
(13)
Kushon, S. A.; Bradford, K.; Marin, V.; Suhrada, C.; Armitage, B. A.; McBranch,
D. W..; Whitten, D. G. Langmuir 2003, 2003, 6456-6464.
(14)
McQuade, D. T.; Hegedus, A. H.; Swager, T. M. J. Amer. Chem. Soc. 2000, 122,
12389-12390.
(15)
Kim, J.; McQuade, D. T.; Rose, A.; Zhu, Z.; Swager, T. M. J. Amer. Chem. Soc.
2001, 123, 11488-11489.
(16)
McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. Rev. 2000, 100, 2537-2574.
(17)
Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Amer. Chem. Soc. 2003, 125, 896900.
53
(18)
Liu, B.; Gaylord, B. S.; Wang, S.; Bazan, G. C. J. Amer. Chem. Soc. 2003, 125,
6707.6714.
(19)
Green, N. M. Methods Enzymol. 1990, 184, 51-67.
(20)
Green, N. M. Biochem. J. 1963, 89, 585-591.
(21)
Bayer, E. A.; Ben-Hur, H.; Gitlin, G.; Wilchek, M. J. Biochem. Biophys. Meth.
1986, 13, 103-112.
(22)
Leclerc, M.; Ho, H.-A. Synlett 2004, 2, 380-387.
(23)
Geiger, E.; Hug, P.; Keller, B. A. Macromol. Chem. Phys. 2002, 203.
(24)
Wilson, J. N.; Wang, Y.; Lavigne, J. J.; Bunz, U. H. F. Chem. Commun. 2003,
1626-1627.
(25)
Hamblett, K. J.; Kegley, B. B.; Hamlin, D. K.; Chyan, M.-K.; Hyre, D. E.; Press,
O. W.; Wilbur, D. S.; Stayton, P. S. Bioconjugate Chem. 2002, 13, 588-598.
(26)
Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 16, 4467-4470.
(27)
Yang, J.-S.; Swager, T. M. J. Amer. Chem. Soc. 1998, 120, 5321-5322.
(28)
Beer, P. D.; Cadman, J.; Lloris, J. M.; Martinez-Mdfiez, R.; Soto, J.; Pardo, T.;
Marcos, M. D. J. Chem. Soc., Dalton Trans. 2000, 11, 1805-1812.
(29)
Becker, J. M.; Wilchek, M. Biochim. Biophys. Acta 1970, 264, 165-170.
(30)
Kim, J.; Swager, T. M. Nature 2001, 411, 1030-1034.
(31)
Kim, J.; McHugh, S. K.; Swager, T. M. Macromolecules 1999, 32, 1500-1507.
54
Chapter 3
Mannose-Functionalized Polymers for the Detection
of Escherichia coli
Partially adapted from: Disney, M. D.; Zheng, J.; Swager, T.M.; Seeberger, P. H. J.
Amer. Chem. Soc. 2004, 126, 13343-13346.
55
Introduction
A variety of intercellular interactions are mediated by oligosaccharides. Mostly
located on the outer surface of the cell membrane, carbohydrates on glycoproteins are
important mediators for cell-cell recognition and have been implicated in processes such
as fertilization, cellular differentiation, aggregation of cells to form organs, and the
infection of cells by bacteria and viruses.'
For many bacteria, adherence to the host cell is a crucial first step to successful
infection. The attachment to cells can occur via bacterial lectins known as adhesins,
which specifically bind to certain host cell surface oligosaccharides. Adhesins can be
carried on hair-like organelles called pili or fimbriae that extend out from the bacterial
surfaces. They can also be directly incorporated into the microbial cell surface.2
Bacterial species belonging to the family Enterobacteriaceae are associated with
a range of human diseases such as cystitis, pyelonephritis, pneumoniae, meningitis,
bacteremia and diarrheal diseases.3 In gram-negative bacteria such as uropathogenic
Escherichia coli, different types of fimbriae such as P, S, Dr, and type 1 fimbriae may be
expressed and are classified with respect to their carbohydrate specificities.4 For example,
P fimbriae are equipped with the adhesin PapG, located at the tip of the pili, which
recognizes
aloc(1,4)Gal
moieties
on the glycolipids
of uroepithelial
cells. This
recognition mediates the adherence to the host and also induces mucosal inflammation. In
addition to P fimbriae, type 1 fimbriae are also important in binding to the host cell. They
manifest the mannose-binding FimH adhesin, which can bind to a number of cellular
targets such as human buccal cells, and epithelial cells in the bladder, lung and intestine.3
Type 1 fimbriae are heteropolymeric
mannose-binding
fibers and are expressed
by all members of the Enterobacteriaceae family. For E. coli, as many as 100 to 400 type
56
1 fimbriae can be found on their surfaces.5 The fibers are composed primarily of FimA
subunits (-18 kD), arranged in a helical manner with a diameter of 6-7 nm and an axial
hole of 20 to 25A.2 FimH is the mannose-specific adhesin (-32 kD). It is located at the
distal end of the fimbriae and is also distributed longitudinally along the organelle.3 As
these fimbriae are pervasive throughout the bacteria family, a method for bacteria
detection that makes use of mannose affinity may be extended to a wide variety of
bacterial species.
Figure3.1. . Long fimbriae protruding from the surface of E. coli (Reprinted with permission from
reference 6, copyright 2004 Nature Publishing Group)
In this chapter, a method for the detection of mannose-binding E. coli is presented.
This work was done in collaboration with Matthew Disney of the Seeberger group, in
ETH, Zfirich. While this work serves as a model for uropathogenic E. coli, we hope to
extend it to wider applications, such as enterohemorragic E. coli.
Current methods for E. coli detection
E. coli colonization of the human gastrointestinal tract typically occurs a few
hours after birth and this symbiotic relationship persists for decades to the benefit of both
partners. However, there are several strains of E. coli which can cause disease in healthy
individuals. The diseases caused by E. coli can usually be categorized as follows:
enteric/diarrheal disease, urinary tract infections and sepsis/meningitis. 6
Perhaps most familiar to public health concerns are enterohemorragic E. coli.
Foods contaminated by these bacteria are a major cause of infection outbreaks with
serious consequences. The bacteria are found principally in the bovine intestinal tract,
and the first outbreaks were attributed to undercooked hamburger meat. Since then, other
foods such as sausages, milk, lettuce, cantaloupe, apple juice and radish sprouts have also
been associated with the disease. 7 One of the largest outbreaks occurred in Japan in 1996,
where over 10,000 people were infected and 11 died.8 One potential reason for this
outbreak reaching these catastrophic proportions was the absence of testing food for
contaminants before public consumption.
Current methods that are used clinically for the detection of pathogenic bacteria,
such as E. coli, rely on selective growth of the bacteria from a contaminated sample,
which can take several days.'7 9"0 More recently, faster methods have been developed that
include pathogen recognition by fluorescently labeled antibodies, -13 DNA probes,'4' 15or
bacteriophages.7' 06
While fluorescent conjugated polymers have found use in a variety
of biological sensing applications,17 18 such as recognition of proteins by electrostatic
interactionsl and detection of pathogens by DNA hybridization,2 '0° 2 detection schemes
for whole cells have not been reported.
Carbohydrate-pathogen interactions such as E. coli with mannose, and influenza
virus with sialic acid, often occur via multivalent interactions,2 2 '23 resulting in higher
binding avidity compared to monovalent binding.24
By presenting the carbohydrate
moieties on a polymer scaffold, we hope to simulate and encourage the multivalent
interactions so prevalent in biology. In this chapter, a carbohydrate-functionalized PPE
that can be used for detection of E. coli by multivalent interactions is presented. In
contrast to previous examples of sugar-containing PPEs,25 28
the polymer is
functionalized after polymerization and provides a versatile scaffold for the rapid
attachment of a variety of different carbohydrates.
58
Results and Discussion
Coupling of the 2'-aminoethyl mannoside 1 29,30 and galactoside 2 31 to the PPE 4
was carried out in the presence of 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide
(EDAC) and N, N'-diisopropylethylamine
for 16-24 h. This was followed by quenching
of unreacted succinimide esters via addition of excess ethanolamine. Uncoupled reagents
were removed by dialysis of the reaction mixture against water for 2 days. A phenol
sulfuric acid test32 for carbohydrate loading showed that typically 25 % of the reactive
sites on the polymer were functionalized with glycosides.
HO OH
OH
°
O
Ho
H2
W
0OH
HO
1
o - _ NH2
3
2
Scheme 3.1. Structures of the carbohydrate derivatives used in the study.
4 R=OH;x:y=0:
5a R = OH or NH(CH 2)2 0H; x: y = 1: 1
sugar = mannose
5b R = OH or NH(CH2)2 0H; x: y = 1: 1
sugar = galactose
Scheme 3.2. Structures of the mannosylated polymers used in the study.
To insure that the carbohydrate moieties conjugated to the polymer retained their
ability to interact with carbohydrate binding lectins, a FRET experiment was carried out
between Alexa Fluor 594TMlabeled concanavalin A, a protein that recognizes mannose,33
and each of the sugar-functionalized
PPEs (Figure 3.3). Titration of labeled Con A into a
solution of mannose-functionalized polymer showed a concentration dependent decrease
59
in fluorescence signal (Figure 3.2). Experiments with galactose-functionalized polymer
showed no fluorescence change, as expected. Thus mannose binding lectins interact with
mannose displayed on the polymer without affecting binding selectivity. Furthermore, the
polymer does not exhibit any non-specific binding to Con A.
450
500
550
650
600
Wavelength (nm)
Figure3.2. Fluorescence emission spectra of polymer 5a with addition of Alexa Fluor 594-labeled Con A.
Energy transfer occurred from the polymer to the dye-labeled lectin.
2.5
fu
C
2.4
2.3
2.2
U:
8)
U,
2.1
2.0
LL.
1.9
1.8
· · · I·
10-7
·
1
1··111~
1
1
IL···ll
10-6
·
·
·
10-5
[Concanavalin A]
Figure3.3. Plot of the normalized fluorescence signal at 512 nm from addition of Alexa Fluor 594-labeled
Con A to a solution of mannose-functionalized 5a (0) or galactose-functionalized polymer 5b (o).
Two bacterial strains that differ only in their mannose binding properties were
obtained from Prof. Orndorff's group at North Caroline State University and were used to
assess whether the mannose-functionalized PPE 5a can fluorescently stain E. coli. In
addition to a strain that binds to mannose, a second strain that is mutated in its FimH
protein to abolish mannose binding was used. 34 The non-functionalized polymer 4, the
mannosylated polymer 5a, and 2'-fluorescein aminoethyl mannoside 3 were individually
incubated with these bacterial strains. After incubating a 1 ml bacterial suspension at an
OD6 00 of 1.0 (108 cells) with the appropriate polymer or dye-labeled mannose for 30
minutes, the suspensions were centrifuged to pellet the cells. The supernatant solution
containing unbound polymer or dye-labeled mannose was discarded and the cells were
washed twice with phosphate buffered saline (PBS, pH 7.2). The bacteria were then
resuspended in PBS. Neither polymer 4 nor the 2'-fluorescein aminoethyl mannoside 3
appeared to bind either E. coli strain. The mannose-functionalized polymer, 5a, however,
imparted a strong fluorescent label to wild type E. coli (Figure 3.4) that was not removed
even upon separation and rinsing. The resuspended and rinsed non-mannose binding E.
coli gave no polymer fluorescence after incubation with 5a.
Figure 3.4. UV-lamp illumination of mutant, left, and mannose-binding, right, E. coli strains after
incubation with mannose-functionalized 5a. Approximately 10 gtg of polymer was incubated with 1 ml
bacterial solution at 1.0 ODroo.
Binding of mannose- and galactose-functionalized polymers to bacteria were
microscopically imaged. After incubation with 5a, the mutant bacteria remained as
individual cells that did not bind to polymer (Figure 3.5A), while the wild type bacteria
formed clusters with fluorescent centers where the polymer was bound to many cells
(Figure 3.5B).
These brightly fluorescent clusters were formed by thirty to several
thousand bacteria (Figures 3.5B and 3.6). The larger clusters had the strongest
fluorescence signal while single cells in the culture exhibited little fluorescence. Though
aggregation of Jurkat cells has been previously observed with Con A attached to mannose
functionalized ROMP polymers,35 such direct polymer-cell clustering has not been
reported with the much smaller E. coli nor has it been used for detection purposes.
Additionally, the fluorescence emission spectrum of the polymer in the bacterial clusters
exhibited a more red-shifted and aggregated behavior (increased emission at 550nm) than
spectra in PBS solution (Figure 3.7). This was consistent with increased n-stacking
interactions between the polymer strands as they were brought into closer proximity by
the bacteria. After incubation with 5b, neither mutant nor the wild type bacteria were
fluorescently stained, which is expected since E. coli does not bind to galactose.
Figure3.5. Laser scanning confocal microscopy image of(A) mutant E. coli that did not bind to mannose.
Individual cells observed with no aggregation. (B) A fluorescent bacterial aggregate due to multivalent
interactions between the mannose binding bacterial fimbriae and 5a. (Superimposed fluorescence and
transmitted light images).
Figure3.6. Fluorescence microscopy image of a large fluorescent bacterial cluster.
0
r:
a
.-
0
0,
o•
o
_=
450
500
550
600
650
700
Wavelength (nm)
Figure3.7. Fluorescence emission spectra of 5a in PBS and normalized fluorescence spectra of a bacterial
cluster obtained using confocal microscopy.
Serially diluted solutions of wild type E. coli were incubated with 5a, washed to
remove unbound polymer, and imaged using fluorescence microscopy to determine the
detection limit. Results show that fluorescently stained clusters of cells can be observed
with as little as 104 bacteria (Figure 3.8). This is similar to the detection limit that is
observed using fluorescently labeled antibodies. 9 Furthermore, the number of cells that is
present in the clusters decreases as the number cells decreases.
108
107
106
105
104
10 3
Figure 3.8. Detection limit for bacterial staining using 5a. Bacteria number is indicated above each image.
These results suggest that multivalent interactions were critical for detection,
since the mannosylated PPE allowed for fluorescent detection of E. coli while 2'fluorescein aminoethyl mannoside 3 did not. The multivalent binding nature of 5a was
demonstrated by testing this polymer for inhibition of Con A-induced hemagglutination
of sheep erythrocytes. The concentration of mannose displayed by the PPE to inhibit
hemagglutination was over 500-fold less than for the monomeric mannose derivatives,
indicating that polymers bind Con A in a multivalent manner.
The observed
enhancement is similar to that reported with polymers prepared by ring-opening
metathesis polymerization (ROMP).3 6
Table 3.1. Inhibition of sheep erythrocyte hemagglutination. Concentrations correspond to those of
mannose units.
Compound
Inhibiting Dose, M
Mannose
0.02
2'-Aninoethyl mannoside
0.01
2'-Fluorescein aminoethyl mannoside 3
0.01
Mannose conjugated polymer 5a
16 x 10-6
Non-functionalized polymer 4
N.D.
Competition experiments were also completed to determine the concentration of
D-mannose that inhibited the binding of 5a to the wild type E. coli. Experiments were
completed with 10 jtg of 5a or 2.9 x 10-9 moles of mannose conjugated to the PPE and
increasing concentrations of D-mannose. Results showed that a 10 mM concentration of
D-mannose was needed to completely inhibit binding of 5a to E. coli. At concentrations
of mannose that were less than 10 ,tM, the size of the bacterial aggregates was not
affected.
Thus, the enhancement in binding due to the multivalency of 5a was 3.5 x 106-
fold. Binding of the polymer to E. coli was significantly enhanced, when compared to Dmannose, by :multivalency.
As a proof of principle, energy transfer for detection of bacteria was carried out
with the analogous mannosylated polymer 7 (with a longer linkered mannoside 6) and a
carboxy-X-rhodamine (ROX)-labeled mannoside 8. Several variations of the experiments
were carried out. When the mannose binding wild type bacteria was incubated with both
7 and ROX-labeled mannoside 8, large clusters of bacteria was observed under confocal
microscopy (Figure 3.9A). Fluorescence emission of the cluster when excited at 364 nm
exhibited energy transfer from the polymer to the ROX dye (Figure 3.9B), due to colocalization of the polymer donor and the dye-labeled mannose acceptor. No clusters and
65
no energy transfer were observed when the wild type bacteria were incubated with only
the dye-labeled mannose. When the mutant bacteria were used, neither cluster formation
nor energy transfer occurred upon incubation with both 7 and the dye-labeled mannose,
or only with the dye-labeled mannose. Thus, co-localization of the energy transfer
partners is necessary for energy transfer. This was facilitated by the multivalent nature of
the bacteria-carbohydrate interactions.
OH
HO
HO6
0
N
8
7 R = OH or NH(CH 2 )2 0H
sugar = mannose
NH2
100I
N
H
O
Man
Scheme 3.3. Polymer donor 7 and dye-labeled acceptor 8 used in energy transfer studies.
C
.2
w
E
C
0
U.
400
450
500
550
600
650
700
750
vavelengthm nm)
Figure3.9. Laser scanning confocal microscopy image of (A) wild type E. coli when incubated with both
ROX-labeled mannose 8 and 7, (B) fluorescence spectrum obtained from the bacteria cluster showing
energy transfer from the polymer to the dye-labeled mannose.
A catalogue of carbohydrate-pathogen interactions are known in the literature.24
Some of these interactions, however, are not specific for one type of pathogen, an
example of this is the cross-reactivity of mannose towards Samonella enterica and E. coli.
The limitation of having ligands with imperfect selectivity can be overcome through the
use of cross-reactive sensor analysis.3 7 In these experiments the presence of a ligand is
determined through the binding of many different analytes, such a detection scheme is
used by the nose. In conjunction with the possibility of using energy transfer as a
detection scheme, many different carbohydrates can be coupled to polymers and analyzed
in parallel, perhaps in a 96-well plate format, should allow detection of the presence of a
single or multiple pathogens within complex mixtures.
Conclusions
A new method for fluorescent detection of bacteria based on water-soluble
fluorescent conjugated polymers has been developed. Glycosides displayed on the
surface of the polymers retain their ability to interact with known carbohydrate-binding
lectins. Incubation of the polymers with E. coli shows that the polymers bind to the
bacteria and yield brightly fluorescent cell clusters. This aggregation is due to multivalent
interactions between the mannosylated polymer and mannose receptors located on the
bacterial pili, which was corroborated by microscopy, hemagglutination, and competitive
binding experiments. This multivalency and resulting cell aggregation is essential for
detection. In contrast to methods for pathogen detection that use selective growth in
liquid media or on plates, which can take several days, carbohydrate-functionalized PPEs
can detect the presence of a pathogen in as little as 10 to 15 minutes. The preference of
different bacteria to bind to specific carbohydrates allows the potential sensing of a range
of pathogens such as cholera in water and other sources.
67
Experimental.
General. H and 13CNMR spectra for monomers and polymers were recorded on a
(Varian 300 MHz) or on a Varian VXR-500 (500 MHz) instrument. The chemical shift
data for each signal are given in units of 6 (ppm) relative to tetramethylsilane
(TMS)
where 6 (TMS) = 0, and referenced to the solvent residual. High-resolution mass spectra
were obtained on a Finnigan MAT 8200 system using sector double focus and an electron
impact source with an ionizing voltage of 70V, and with a Bruker DALTONICS APEX II,
3 Tesla, FT-ICR-MS with ESI source or EI/CI source. UV-visible absorption spectra
were measured with a Cary 50 UV/visible spectrometer. Fluorescence spectra were
measured with a SPEX Fluorolog-2 fluorometer (model FL112, 450W xenon lamp). The
spectra in solution were obtained at room temperature using a quartz cuvette with a lcm
path length. The molecular weights of polymers were determined by using three PLgel
5pm 105, 104', 103 (300 x 7.5 mm I.D) columns in series and a diode detector at 254nm at
a flow rate of 1.0ml/min in THF or in DMF. The molecular weights were reported
relative to polystyrene or poly(ethylene oxide) standards purchased from Agilent Inc.
Melting point (m.p.) determination was performed using a Laboratory Devices MELTEMP instrument (open capillaries used) and was uncorrected.
Materials.
All solvents were spectral grade unless otherwise noted. Morpholine and biotin were
purchased from Alfa Aesar and used as received. Alexa Fluor 594-labeled concanavalin
A was purchased from Molecular Probes, Inc. All other chemicals were purchased from
Aldrich Chemical In. and used as received. All air and water sensitive synthetic
manipulations were performed under a nitrogen atmosphere using standard schlenk
techniques.
68
Carbohydrate Synthesis.
The glycosides 1 and 2 and 6 were synthesized according to published procedures. 29-31
The 2'-fluorescein ethylamino mannoside (3) was synthesized by adding 9.8 mg of 2'ethylamino-mannoside to a solution of 21 mg of 5-(and-6)-carboxyfluorescein,
succinimidyl ester (Molecular Probes, Eugene OR) in 2 mL of 25% aqueous DMF with
10 pL of N,N'- diisopropylethylamine. The reaction was stirred for 2 h and the reaction
was purified by silica gel choromatography using a gradient of methanol in chloroform.
The product was analyzed by ESI mass spectrometry, M=582.1 (M + H+).
The ROX-labeled mannoside (8) was synthesized by adding mannoside 6 (3.0 mg, 0.011
mmol) to a solution of 5-(and-6)-carboxy-X-rhodamine, succinimidyl ester (7.09 mg,
0.011 mmol) in
diisopropylethylamine.
1.0 mL of 25%
aqueous DMF
with 2.5
tL of N,N'-
The reaction was stirred for 20 h and the reaction was purified by
reverse phase HPLC, fractions containg the desired product were combined and
lyophilized to give a deep purple powder. The product was analyzed by MALDI-TOF
mass spectrometry, M = 784.6 (M+H+).
Polymer Synthesis
Polymer 4 was synthesized according to published procedures.3 8
Polymer 5a and 5b and 7: A solution containing 1.8 mg of acid functionalized polymer
was dissolved in 2 mL of DMF and 8.8 mg of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDAC) and 5 mg of N-hydroxysuccinimide are added. The reaction was
stirred for 4h., then 18 mg of 2' aminoethyl glycoside was added in water. The solution
became slightly cloudy and an additional portion of DMF added until the solution
69
became clear. After 2 h, 500 tL of N,N-diisopropylethamine was added and the reaction
stirred overnight. To quench unreacted succinimide esters, 500 pL of ethanolamine was
added and the reaction stirred for at least 3 h. This solution was dialyzed against water
for 48 h in an 8,000-11,000 MWCO dialysis bag. After dialysis, the polymer solution
was lyophilized to dryness, resuspended in water, and stored at -20 °C until use.
Phenol-Sulfuric Acid Test for Carbohydrate Loading onto the Polymer.3 2 Approximately
60 jg/mL of polymer in 1 mL of water was mixed with 333 ptLof a 5% aqueous solution
of phenol and then 1.33 mL of H2 SO4 added. The reaction was incubated at room
temperature for 30 min and then the absorbance measured at 490 nm. The increase in
absorbance relative to a blank solution that contained the same amount of polymer, as
determined by UV-Vis spectroscopy, was used to quantify the increase in absorbance due
to carbohydrate loading. The amount of carbohydrate was determined by comparison to
the increase in absorbance observed from a standard curve generated by testing mannose.
Results show that the carbohydrate typically reacted with - 25 % of the reactive sites.
Fluorescence Resonance Energy Transfer Experiments (FRET). Spectra were recorded at
25
C with a SLM Aminco® Bowman Series 2 Luminescence
Spectrometer.
excitation wavelength was 440 nm and the emission scan was from 455-650 nm.
The
A
solution of 6 Clgof polymer was dissolved in PBS, pH 7.2 buffer. Aliquots of Alexa
Fluor
5 94TM labeled
concanavalin A (Con A) were added to the solution, the
concentration of Con A is given in amount of monomer (MW = 25,000). After each
addition, the sample was allowed to equilibrate for 2 min prior to recording a spectrum.
Additions of Con A were continued until no change in the fluorescence signal was
observed.
Fi,corr =
The signal was corrected for the dilution factor, according to the equation:
Fiobs X Vi/V0 , where Fi,cor is the corrected intensity for point I, Vi is the volume
after the
ith
addition, and V0 is the initial volume (typically 300 pL). The fluorescence
maximum, or 512 nm, was then plotted.
70
Cell Growth and Incubation of Polymer with Cells. The strains used in this study were
kindly donated by Prof. Orndorff and are denoted ORN178 for the mannose binding
strain and 0RN208 for the mutant strain that does not bind mannose.34 Cells were grown
in LB media overnight at 37
C until they reached an approximate
OD 6 0 0
of 1.0. The
culture was then centrifuged and cells washed twice with PBS buffer. A 10 ptgaliquot of
the polymer or an equivalent amount of 2'-fluorescein ethylamino mannoside in terms of
the amount of carbohydrate was added to a 1 mL aliquot of these cells in PBS buffer with
1 mM CaC12 and 1 mM MnC12. The suspension was then incubated for 30 min at room
temperature with gentle shaking and centrifuged to pellet the cells.
Pellets were
resuspended the same buffer by disruption of the pellet by pipetting with a pipet tip or
vortexing, and centrifuged. This procedure was repeated twice for each culture. Cells
were then visualized under a transilluminator and little fluorescence was observed with
ORN208 (mutant) strain while high fluorescence signal was observed with strain
ORN178 (mannose binding).
Fluorescence Microscopy. Cells were incubated with polymer as described above and
were resuspended in a 1 mL of PBS with
mM of CaC12 ,
mM of MnC12, and 10%
glycerol. A 10 lL of the culture was removed and spotted onto a glass slide. Cells were
imaged using fluorescence microscopy either with a Zeiss Axiovert 100M or with a
Nikon Eclipse TS100 inverted fluorescence microscopes. Cells were imaged using a
100X oil immersion or a 40X objectives and fluorescence images obtained with the FITC
filters. Typical exposure times used were 0.2, 0.6, and 1 seconds.
Incubation of E. coli with Galactose-Functionalized Polymer. 10,tg of the galactosefunctionalized polymer, 5b, was incubated with 1.0 OD600 of either wild type of mutant E.
coli as described above.
Cells were then imaged using a fluorescence microscope.
Results show that 5b does not stain either bacterial strain
71
Laser Scanning Confocal Microscopy. Cells were prepared as described for fluorescence
microscopy. Images were taken with a Leica TCS SP2 with AOBS laser scanning
confocal microscope. An oil immersion 63X objective and 8X zoom, with excitation at
351 and 364 nm using an UV laser from Coherent Enterprise was used.
Fluorescence spectra. Solution fluorescence spectrum of 5a in PBS (used to compare to
the confocal spectrum obtained from the cell clusters, Figure 3D) were measured using a
Jobin-Yvon SPEX Fluorolog-r3 fluorometer (model FL312, 450W xenon lamp),
excitation at 364 nm.
Hemagglutination Experiments.
sheep red blood cells (Sigma).
Hemagglutination experiments were completed with
Cells were resuspended in PBS buffer and were tanned
with a 1/10,000 tannic acid solution in PBS by incubation of the cells at 37 °C for 10 min.
Blood cells were then centrifuged and washed with 40 mL of PBS buffer. A solution of
cells (final concentration 3%) was placed into individual wells in a 96-well plate.
Serially diluted concentrations of Con A in PBS were added to each well and the cells
were incubated at room temperature for at least 1 h. The concentration of Con A that
caused complete agglutination was then increased 4-fold for experiments used to
determine the hemagglutination activity of mannose derivatives and polymers. These
experiments were completed as described above except that 2-fold serial dilutions of
inhibitors were added (mannose derivative and various polymer solutions).
Each
experiment was completed in at least duplicate and the results are the average of these
independent measurements.
Energy transfer experiments with E. coli
Cells were grown in LB media overnight at 37
OD600
C until they reached an approximate
of 1.0. The culture was then centrifuged and cells washed twice with PBS buffer.
A 5 g aliquot of the polymer 7 and 2 pl of a lmg/ml solution of the ROX-mannoside
were added to 0.75 mL aliquot of these cells in PBS buffer with 1 mM CaC12 and 1 mM
72
MnC12. The suspension was then incubated for 2 hours at room temperature with gentle
shaking and centrifuged to pellet the cells. Pellets were resuspended the same buffer by
disruption of the pellet by pipetting with a pipet tip or vortexing, and centrifuged. This
procedure was repeated twice for each culture. The cells were imaged by confocal
microscopy.
73
References
(1)
Voet, D.; Voet, J. G. Biochemistry; Second edition ed.; John Wiley & Sons, Inc.:
New York, NY, 1995.
(2)
Soto, G. E.; Hultgren, S. J. J. Bacteriol. 1999, 181, 1059-1071.
(3)
Jones, C. H.; Pinkner, J. S.; Roth, R.; Heuser, J.; Nicholes, A. v.; Abraham, S. N.;
Hultgren, S. J. P. Natl. Acad. Sci. USA 1995, 92, 2081-2085.
(4)
Connell, H.; Agace, W.; Klemm, P.; Schembri, M.; Marild, S.; Svanborg, C. P.
Natl. Acad. Sci. USA 1996, 93, 9827-9832.
(5)
Lindhorst, T. K.; Kieberg, C.; Krallmann-Wenzel, U. Glycoconjugate J. 1998, 15,
605-613.
(6)
Kaper, J. B.; Nataro, J. P.; Mobley, H. L. T. Nat. Rev. Microbiol. 2004, 2, 123140.
(7)
Deisingh, A. K.; Thompson, M. J. Appl. Microbiol. 2004, 96, 419-429.
(8)
ww.citv.sakai.osaka.jp/city/info/o 157rprt/index.html.
(9)
Willis, R. C. Mod. Drug Discovery 2004, 7, 36-42.
(10)
L6pez, M. M.; Bertolini, E.; Olmos, A.; Caruso, P.; Gorris, M. T.; Llop, P.;
Ram6n, P.; Cambra, M. Int. Microbiol. 2003, 6, 233-243.
(11)
Yu, L.. S. L.; Reed, S. A.; Golden, M. H. J. Microbiol. Meth. 2002, 49, 63-68.
(12)
Nakamura, N.; Burgess, J. G.; Yagiuda, K.; Kudo, S.; Sakuguchi, T.; Matsunaga,
T. Anal. Chem. 1993, 65, 2036-2039.
(13)
Yamaguchi, N.; Sasada, M.; Yamanaka, M.; Nasu, M. Cytometry 2003, 54A, 2735.
(14)
Jung, W.-S.; Kim, S.; Hong, S.-I.; Min, N.-K.; Lee, C.-W.; Paek, S.-H. Mat. Sci.
Eng. C-Biomim. 2004, 24, 47-51.
74
(15)
Stender, H.; Oliveira, K.; Rigby, S.; Bargoot, F.; Cooll, J. J. Microbiol. Meth.
2001, 45, 31-39.
(16)
Goodridge, L.; Griffiths, M. Food Res. Int. 2002, 35, 863-870.
(17)
McQuade, D. T.; Pullen, A. E.; Swager, T. M. Chem. Rev. 2000, 100, 2537-2574.
(18)
Jelinek, R.; Kolusheva, S. Chem. Rev. 2004, 104, 5987-6015.
(19)
Fan, C.; Plaxco, K. W.; Heeger, A. J. J. Amer. Chem. Soc. 2002, 124, 5642-5643.
(20)
Liu, B.; Bazan, G. C. Chem. Mater. 2004, 16, 4467-4476.
(21)
Dor6, K.; Dubus, S.; Ho, H.-A.; Lvesque,
I.; Brunnette, M.; Corbeil, G.;
Boissinot, M.; Boivin, G.; Bergeron, M. G.; Boudreau, D.; Leclerc, M. J. Amer.
Chem. Soc. 2004, 126, 4240-4244.
(22)
Karlsson, K.-A. Curr. Opin. Struc. Biol. 1995, 5, 622-635.
(23)
Karlsson, K.-A. Adv. Exp. Med. Biol. 2001, 491, 431-443.
(24)
Mammen, M.; Choi, S.-K.; Whitesides, G. M. Angew. Chem. Int. Edit. 1998, 37,
2754-2794.
(25)
Kim, I.-B.; Erdogan, B.; Wilson, J. N.; Bunz, U. H. F. Chem. Eur. J. 2004, 10,
6247-6254.
(26)
Kim, .-B.; Wilson, J. N.; Bunz, U. H. F. Chem. Commun. 2005, 1273-1275.
(27)
Erdogan, B.; Wilson, J. N.; Bunz, U. H. F. Macromolecules 2002, 35.
(28)
Lavigne, J. J.; Broughton, D. L.; Wilson, J. N.; Erdogan, B.; Bunz, U. H. F.
Macromolecules 2003, 36, 7409-7412.
(29)
Chernyak, A. Y.; Sharma, G. V. M.; Kononov, L. O.; Krishna, P. R.; Levinsky, A.
B.; Kochetkov, N. K.; Rao, A. V. R. Carbohyd. Res. 1992, 223, 303-309.
(30)
Corbell, J. B.; Lundquist, J. J.; Toone, E. J. Tetrahedron-Asymmetr
2000, 11, 95-
111.
(31)
Ni, J.; Singh, S.; Wang, L.-X. Bioconjugate Chem. 2003, 14.
(32)
Saha, A. K.; Brewer, C. F. Carbohyd. Res. 1994, 254, 157-167.
75
(33)
So, L. L.; Goldstein, I. J. J. Biol. Chem. 1968, 243, 2003-2007.
(34)
Harris, S. L.; Spears, P. A.; Havell, E. A.; Hamrick, T. S.; Horton, J. R.; Omdorff,
P. E. J. Bacteriol. 2001, 183, 4099-4102.
(35)
Gestwicki, J. E.; Strong, L. E.; Cairo, C. W.; Boehm, F. J.; Kiessling, L. L. Chem.
Biol. 2002, 9, 163-169.
(36)
Mortell, K. H.; Weatherman, R. V.; Kiessling, L. L. J. Amer. Chem. Soc. 1996,
118, 2297-2298.
(37)
Albert, K. J.; Lewis, N. S.; Schauer, C. L.; Sotzing, G. A.; Stitzel, S. E.; Vaid, T.
P.; Walt, D. R. Chem. Rev. 2000, 100, 2595-2626.
(38)
Zheng, J.; Swager, T. M. Chem. Commun. 2004, 2798-2799.
76
Chapter 4
Amphiphilic Polymers at the Air-Water Interface
77
Introduction
The behavior of molecules at the air-water interface has long been a topic of
interest. In antiquity, the Babylonians used to practice divinity by studying the spreading
behavior of oil on water.' Throughout the ages, monolayers have been used by the
Japanese in printing2 and were described by Pliny the Elder as having calming effects on
water.3'4 However, the first scientific investigation into monolayers was not until 1774,
by Benjamin Franklin,5 where he made the following report: "At length at Chapman
where there is, on the common, a large pond, which I observed to be one day very rough
with the wind, I fetched out a cruet of oil, and dropped a little of it on the water. I saw it
spread itself with surprising swiftness upon the surface. The oil, though not more than a
teaspoonful, produced an instant calm over a space several yards square, which spread
amazingly and extended itself gradually until it reached the leeside, making all that
quarter of the pond, perhaps half an acre, as smooth as a looking glass." Had Franklin
done a more quantitative
analysis, he would have come to the realization that one
teaspoon of oil covering an area of half an acre would lead to a coating of about 1 nm
thick.6
In 1865, Lord Rayleigh was the first to propose that the oil film on water would
be equivalent to a layer of one molecule thick.7 The true confirmation of this suspicion
was not until the experiments of Agnes Pockels.8 Conducted in a modified kitchen sink
which eventually served as the model for Langmuir troughs, she described the water
surface contamination as a function of surface area for different oils. Based on these
experiments, Raleigh later determined the thickness caster oil to be 1 nm, which
corresponded to a monolayer.9 Systematic studies were performed by Irving Langmuir in
the 1910s, with a series of fatty acid salts, where he demonstrated that the size, shape and
orientation of the molecules could be determined using a Langmuir trough.10 Katherine
Blodgett then continued the work and reported the sequential transfer of monolayers onto
substrates.
78
Surface tension and surface pressure
At the air-water interface, the surface water molecules experience an imbalance of
forces - they have a stronger attraction toward the liquid phase than towards the gas
phase.3 '
2
This result in a net attractive force towards the bulk and the interface will be
drawn to the interior of the water, contracting and minimizing its area in the process. The
excess energy originating from the difference in environment between the surface
molecules and those in the bulk is called surface free energy and is quantified as a
measurement
of energy/area.
13
One can also describe the liquid as having a surface
tension which is quantified as force/length measurement. The units for surface tension are
mN/m (or dynes/cm).
When an amphiphile dissolved in a volatile organic solvent is placed on a water
surface, it quickly spreads to cover the available area. A monolayer is formed when the
solvent evaporates. At a large surface area, the distance between neighboring amphiphiles
is large and their interactions are weak. One can consider this monolayer to be a two-
dimensional gas and the amphiphiles have minimal effect on the surface tension.
However, when the surface area is decreased, there is repulsion between the molecules.
This surface pressure, it, is described by:
n = y -y
(Equation 4.1)
Where Yo is the surface tension in absence of a monolayer and y is the surface tension
with the monolayer.
14
Surface pressure - area isotherms
The Langmuir balance in use nowadays is largely similar to that used by
Katharine Blodgett. A Teflon trough holds the subphase, whose temperature may be
controlled by circulating water in channels under the trough. The monolayer is deposited
between two movable Teflon barriers, which can change the area between them. The
surface pressure is measured by the Wilhelmy plate method, which is based on
79
quantifying the force on suspended thin platinum or paper plate that is partially
submerged in the subphase.
The forces acting on the plate are gravitational force and surface tension, which
drag the plate down, and the buoyancy effect which lifts the plate up. For a rectangular
plate of length L, width W, thickness T and immersed in water to a depth D, the net force
downwards F is described by:1 4
F = (ppLWT)g + 2(W+T)ycosO - (pLDWT)g
Where pp and PL are the densities of the plate and liquid, respectively,
(Equation 4.2)
is the surface
tension, 0 is the contact angle of the liquid to the plate and g is the acceleration due to
gravity. For a pressure reading that is zeroed and a plate that is always kept at a constant
level, the equation is simplified to:
F = 2(W+T)ycosO
(Equation 4.3)
For a paper plate, where the contact angle of the liquid to the plate is 0, the equation is
further reduced to:15
F = 2(W+T)y or
= F/2(W+T)
(Equation 4.4)
Knowing the force exerted on the Wilhelmy plate, the surface tension, , is easily
measured and consequently the surface pressure can be determined.
The behavior of films at the air-water interface is described by the surface
pressure-area isotherm. An amphiphile, such as a fatty acid, is dissolved in an organic
solvent and spread onto the water in a Langmuir trough, in between the barriers. Upon
compression of the monolayer by moving the barriers, the molecules orient themselves at
the surface. A representative isotherm
16
is shown below (Figure 4.1):
80
401
35
30
25
zZ 20
15
10
C,
-
I
I
20
30
·
I
I
I
40
50
§0
Area/Molecule(angstromst
Figure 4.1. Typical pressure-area isotherm of a fatty acid undergoing three phases (gas, liquid and gas),
each marked by a distinctive change in slope value. (Reproduced from reference 16, with permission from
the author)
Regions with different slopes correspond to various phases of the monolayer. At
low compression, the amphiphile molecules have minimal interactions with one another
and correspond to the gas phase. At medium compression the molecules are interacting
with one another and this corresponds to a liquid phase. At high compression, there is
packing of the amphiphile molecules and they are vertically oriented with respect to the
interface, this is manifested as a solid phase in the isotherm.
Conjugated polymers at the air water interface
Small molecule amphiphiles are usually studied at the air-water interface. It is
possible, however, to use polymers in Langmnuir-Blodgett (LB) experiments. For the
deposition of LB films, there are two major routes to their preparation. In the first case,
preformed
polymers
can be used. In the second case, monomeric amphiphiles
are
deposited at the air-water interface. The monomers can be transferred to a substrate and
polymerized, or they can be polymerized at the water surface and then transferred. Here,
we will primarily consider the first scenario, where preformed polymers are used.
There are a variety of polymers that can be used for LB experiments. A few
flexible polymers are poly(dimethyl siloxane), 17 1 8 poly(vinyl alcohol), 19 poly(octadecyl
81
methacrylate).2 0 Rigid rod polymers include polyglutamates,21 '
polyisocyanides.
24
poly(p-phenylene),
Conjugated
2 829'
polymers
30
polyaniline,
° 31
22
polysaccharides23 and
such as polypyrrole, 2 5 polythiophene, 26' 2 7
poly(p-phenylenevinylene),
32
and
azo
polymers 3 3 34 have all been used to prepare LB films.
In our group, the Langmuir-Blodgett technique has been used to probe the
fundamental properties and behavior of conjugated polymers. An early study used the
Langmuir-Blodgett technique to deposit oriented poly(p-phenylene ethynylene)s films
onto substrates,3 5 as rigid rod monolayers tend to align perpendicular to the direction of
compression and parallel to lines of flow between the water and the substrate in the
dipping process. Polymers with polar macrocycle and non-polar side chains
spontaneously reorganized into aligned fibrillar structures following transfer onto a
hydrophobic surface, minimizing energy by displaying the non-polar side chains on the
outside of the fibrils and shielding the polar groups in the interior. A second film
deposition at right angle to the first film resulted in a final grid-like structure after
rearrangement.
Interchain distance effects on the photophysical properties of conjugated
polymers were investigated by synthesizing polymers with side chains of varying bulk
and coating them onto glass substrates by drop casting, spin-coating or using the
Langmuir-Schaefer method.36 Side chains that induced the largest interchain distance
gave rise to higher quantum yield, as the polymer was least aggregated. Side chains that
gave the smallest interchain distance showed marked aggregation in both the emission
and the absorption spectra, and yielded the smallest quantum yield. In this report, the
Langmuir-Schaefer method provided a way of depositing alignd polymer onto substrates
to provide well-organized re-stackedchains that made these studies possible.
The unparallel control over polymers using the Langmuir-Blodgett technique was
perhaps best illustrated by a report published in 2001.37Here, four PPEs were synthesized
82
with amphiphilic monomers which have different orientations at the air-water interface.
The polymers displayed face-on, alternating face-on and edge-on (zipper), or edge-on
conformations at the interface with the possibility of switching between these
conformations upon application of surface pressure (Figure 4.2). The polymer phases
were monitored using pressure-area isotherms, fluorescence, and UV-Vis absorption
spectra. For the face-on polymer 1, the planar and highly conjugated structure at the airwater interface resulted in a red-shifted absorption spectrum when compared to the
solution spectrum. At high surface pressure when the polymer folded into a multilayer, an
aggregation peak was observed in the UV-vis spectrum, which indicated interpolymer cofacial
c-n7
interactions. Polymer 2 exhibited a reversible phase change from face-on to
zipper conformation upon mechanical compression, as seen in the pressure-area isotherm.
This was corroborated by spectroscopic data, where a blue shift was observed both in the
absorption and fluorescence. Again, an aggregation peak was observed when the polymer
folded into multilayers at high surface pressure. Polymer 3 switched reversibly from a
native zipper conformation to an edge-on conformation upon compression. The edge-on
phase was confirmed by the UV-vis and fluorescence spectra, which showed a red-shifted
0.014
P
Polymer 1
Polymer
1
kn
I.
I
416
'
Polymer 2
?
I
!.
0.006
Water
Al
rce-on
0,0020
147
o.•
iIte
0.014
420
0.014
Polymer
Is.
I
0.01
'4
Polymer 3
/
I
l.)
458
44,'ý
0.0%•2
435
0.0(20 .46
MAps
_rCIM
;Polymer 3
421
I
Z
I
ii
00145
oo
- -• - - • 00°1°°-
421
r
455•
450
Polymer4
0.06
Polymer4
C
0.04
4W3
0.03
0.01
I,
O~dge-on
Q
1
02....,0
A300
350
400
50
SW0
Wavelength (nm)
Figure 4.2. PPEs at the air-water interface. Phase changes in the polymer occurring upon mechanical
compression (left) were mirrored in the behavior of the UV-vis absorption and fluorescence emission
spectra (right). (Reprinted with permission from reference 37, copyright 2001 Nature Publishing Group)
aggregation absorption band concurrent with a quenched fluorescence. Polymer 4 had an
edge-on structure and extensive -7zcofacial interactions. It was increasingly quenched
and aggregated with mechanical pressure. The data from each of the polymers
complemented one another, and showed that the conformation of the polymers had
important consequences on the photophysical properties. This study was further
elaborated upon in a later article.3 8
Protein-ligand interactions at the air-water interface
Protein-ligand interactions at the air-water interface mimic the cell surface and
serve as a good model for understanding biological phenomena. In addition to precise
control over the components, the Langmuir technique can accommodate a versatile array
of methods to monitor changes. Binding interactions may be interrogated using the
surface pressure, which changes when a protein in the subphase binds to the monolayer.
The magnitude of the surface pressure change may used to compare the strength of
protein-ligand interactions.39 The monolayers may also be imaged using microscopic
techniques such as Brewster angle microscopy or epifluorescence microscopy, the latter
when fluorescently labeled components are used. Spectroscopic methods such as IR,
fluorescence and absorption may also be used. Previous studies have probed binding of
streptavidin to a mixed monolayer of biotinylated lipophilic fullerenes,4 0 binding of
streptavidin/avidin to biotin-labeled lipids,4 1 42 and streptavidin binding to both endlabeled biotinylated polystyrene at the monolayer and biotinylated ferritin in the
subphase.43 While protein-small molecule amphiphilic monolayers have been studied
with great interest, conjugated polymers have not found extensive use in the study of
protein-ligand interactions.
An early report investigated the molecular recognition between concanavalin A
(Con A) and mannoside diacetylene lipid monolayer.4 4 Con A was a protein isolated
from the jack bean and could bind to saccharide-containing receptors, such as D-glucose
or D-mannose containing molecules. In this study, when the protein was added to the
subphase of a mannoside diacetylene lipid monolayer at 10mN/m, the molecular area
84
increased over time. If f-mannosidase, an enzyme which cleaved mannose from the lipid,
was injected into the subphase prior to the addition of Con A, only a slight increase in
molecular area was observed. This indicated that Con A was binding to the mannose
residues of the lipid monolayer. Further proof was provided by adding fluoresceinlabeled Con A to the subphase. Fluorescence decreased with increasing incubation time,
as more Con A bound to the monolayer. This was due to increased concentration of the
protein at the interface, which in turn led to fluorophore aggregation and consequent
quenching.
Other studies have focused on organized diacetylene lipid monomers, which
could polymerize upon UV irradiation at the Langmuir-Blodgett trough and subsequently
be transferred onto glass substrates. Numerous protein binding and biological detection
schemes using these supported polymer films have been carried out by other groups.4
5-49
Results and discussion
In this chapter, amphiphilic biotinylated polymers were used to provide insight
into ligand-protein interactions at the air-water interface. Biotinylated monomers 7 from
an earlier study and 4 were synthesized with different linkers to study the effect of linker
length on binding (Scheme 4.1). These were then polymerized with hydrophobic
monomer 5 to generate amphiphilic polymers 6 and 8 (Schemes 4.2 and 4.3).
85
0
0
H
0
H
I
N\
0
H2N NON0/
~
i. TFA
io2 o
~
2.
f,,
1
H
H
/NS
HN
0
1
H
1
H
0
3
H
H~~
H
HN NH
O
4
Scheme 4.1. Synthesis of biotinylated monomer 4.
o
0°
H
H
-N
XH
•
HN
,
C18 H37 0
0
OC1 8H37
/
"H
C18H37 0
/
C1 8H37
Pd(PPh
3 )4, Cul
0
a
Morpholine,60 C
NH
4
oo
5
0
HN'
HNXNH
NHjrzRNN
6 0
''I Cul
Scheme 4.2. Synthesis of biotinylated amphiphilic polymer 6 with a short linker to biotin.
H
0°>
0
O-(>
I `~H
N
H
C18H370
N HH
NH
OC
1 8H37
C18H370
Pd(PPh
3)4,
-
Morpholine,
60C
0
7
OC
1 8H37
,
0~(
5
1
or/S
O
HN
HN
1.s 0
0
8
HNY NH
°
Scheme 4.3. Synthesis of biotinylated amphiphilic polymer 8 with a long linker to biotin.
Polymers 6 and 8 were soluble in chloroform and formed a monolayer when
deposited onto Tris buffer at pH 7.5 in a Langmuir-Blodgett trough. Based on previous
work by Jinsang Kim, it was expected that the polymers would form a zipper-phase due
to the alternating face-on and edge-on co-monomers. Upon compression, the polymers
changed to a edge-on phase that was highly compressible (Figure 4.4). The pressurearea isotherms and spectroscopic data for the two polymers showed a distinct change in
slope at the phase change (Figures 4.5 and 4.8). In addition, the area occupied per repeat
unit was larger for polymer 8 with the longer linker to the biotin, possibly due to more
disorder of the biotinylated groups in the water subphase.
86
The polymers showed aggregation behavior when compressed to the edge-on
phase as evidenced by the UV-vis absorption and fluorescence data. Aggregation peaks at
441 nm and 447 nm arose for polymers 6 and 8, respectively, at surface pressures
between 10 and 15mN/m. Concurrently, fluorescence was quenched at the transition
between zipper and edge-on phases and continued to quench when the polymer was
further compressed. This was consistent with increased interchain n-stacking interactions.
/
.·I
NH
0
HN
O"NH
0
/
HN
S
NH
HN--,
Figure 4.3. Conformational change from zipper to edge-on phase at the air-water interface upon barrier
compression, for biotinylated polymer 6 (and analogously, 8).
35
30
25
E 20
Eb15
10
5
0
40
60
80
100
120
140
160
18(
2
Arealrepeat unit (A )
Figure 4.4. Pressure-area isotherms for polymer 6 with no avidin added to the subphase, after three
annealing cycles. Change from zipper to edge-on at around 15 mN/m.
0.025
0.020
8
0.015
_e
S0.010
0.005
0.000
380
360
400
420
440
460
Wavelength (nm)
480
500
520
540
Figure4.5. UV-vis absorption spectra of polymer 6, at various surface pressures. Appearance of a redshifted aggregation peak starting from 15 mN/m.
I
I
I
i
-
7
m
m
m
m
m
m
I
I
450
500
,
·
I
I
550
600
Wavelength (nm)
I
I
650
I
I
700
Figure 4.6. Fluorescence emission spectra of polymer 6, at various surface pressures. The fluorescence
switched to a quenched, broad spectrum corresponding to aggregated polymers starting from 15 mN/m.
No avidin in subphase
35
30
25
20
E
15
10
5
0
''"''''''''
60
80
100
120
140
160
Area/repeat unit (A2)
180
Figure4.7. Pressure-area isotherm of polymer 8 with no avidin added to the subphase, after three annealing
cycles. Change from zipper to edge-on phase at around 15 mN/m.
0.022
0.020
0.018
0.016
0.014
0.012
0.010
0.008
0.006
0.004
0.002
0.000
360
380
400
420
440
460
480
500
520
540
Figure4.8. UV-vis absorption spectra of polymer 8 at various surface pressures. Appearance of a redshifted aggregation peak starting from 15 mN/m.
400
450
500
550
600
650
Wavelength (nm)
Figure4.9. Fluorescence emission spectra of polymer 8, at various surface pressures. The fluorescence
switched to a quenched, broad spectrum corresponding to aggregated polymers starting from 15 mN/m.
To test the ability of the polymers to interact with avidin and streptavidin, a stock
solution of the protein avidin was made at 0.5mg/ml. A series of avidin concentrations
were tested at the Langmuir-Blodgett trough to determine the amount of protein that
could be added to the subphase without significant increase in surface pressure. An
amount of 0.25mg for a 60ml subphase was deemed adequate for experiments. With the
barriers open, this amount of avidin in 0.5 ml Tris buffer was added to the subphase of
the annealed polymer monolayer and incubated for lhr.
The pressure-area isotherms for polymer 6 following avidin incubation and
annealing were shown below (Figure 4.10). There was an increase in surface area
occupied by the monolayer throughout the isotherms and the phase transition was now
less defined.. Compared to the original polymer without added protein, there was more
hysteresis on expansion at surface pressures from 10 to 35mN, corresponding to the edge-
on conformation.
Ar~
35
30
25
z
20
E
, 15
10
5
0
60
80
100
120
140
160
180
Area/repeatunit (A2)
Figure 4.10. Pressure-area isotherm of polymer 6 when avidin was added to subphase, reproducible
isotherms after initial annealing cycles.
Spectroscopic measurements were also performed (Figure 4.11). Less aggregation
resulted from avidin binding as evidenced by the fluorescence and UV-vis spectra.
Compared to the native polymer, the fluorescence was less quenched at 15mN/m and
90
higher surface pressures. The aggregation peaks were also less pronounced for polymer 6
at the edge-on phase.
'' "' "
0.014
0.012
0.010
cc 0.008
0
*
0.006
0.004
0.002
0.000
-0.002
360
380
400
420
440
460
Wavelength (nm)
480
500
520
540
450
500
550
600
650
Wavelength (nm)
Figure4.11. Left: UV-vis absorption. Right: fluorescence emission spectra of polymer 6 after incubation
with avidin, at various surface pressures. The protein-complexed polymer was not as aggregated at high
surface pressures when compared to the native polymer.
Polymer 8 exhibited analogous behavior when incubated with avidin in the
subphase (Figures 4.12 and 4.13). The fluorescence spectra in particular showed a less
aggregated phase when compared to the native polymer. One possible explanation may
be that the globular protein avidin, measuring 56 x 50 x 40A,50 prevented efficient
7t-71
cofacial interactions between adjacent polymer chains due to steric interactions of the
proteins in the subphase. The hysteresis in the pressure-area isotherms on expansion of
the monolayer may also correspond to a drag effect that the bound protein has on the
polymer, as it increased the bulk of the side chain and its hydrophilicity, thereby
preventing complete recovery to the zipper phase. This also could explain the lack of
definition in the slope change as the polymer changes form zipper to edge-on phase.
In
4U
zeroed and equilibrated
35
30
25
E 20
E
15
10
5
80
60
100
120
140
Area/repeat unit (A2)
160
180
Figure4.12. Pressure-area isotherms of polymer 8, after initial annealing cycles. Hysteresis on expansion
from edge-on to zipper phase.
U.U012
I(
0.010
0.008
_ 0.006
0
0.002
0.000
-0.002
-0.004
360
380
400
420
440
460
480
500
520
Wavelength (nm)
540
450
500
550
600
650
Wavelength (nm)
Figure4.13. Left: UV-vis absorption. Right: fluorescence emission spectra of polymer 8 after incubation
with avidin, at various surface pressures. The protein-complexed polymer was not as aggregated at high
surface pressures compared to the native polymer.
The relative quantum yields for the polymers at 35 mN/m (monitored at 500 nm)
were calculated for both 6 and 8. Following incubation with avidin, the quantum yields
were 1.58 and 1.69 times the original quantum yield of the native polymers (without
avidin) for 6 and 8, respectively. Control experiments were carried out using biotin presaturated avidin to preclude biological recognition. These latter relative quantum yields
remained constant, at 1.01 and 1.05 times the original values for the native polymers 6
and 8, respectively.
Avidin was therefore binding to the polymers and hindering the
efficient edge-on aggregation.
To verify that the biotin groups bound to the polymer monolayer could interact
with proteins, energy transfer assays were carried out with 0.040 mg Texas Red-Xlabeled streptavidin injected into the subphase. Following film annealing, dye-labeled
streptavidin was added to the subphase and incubated for approximately lh. After
addition and incubation of the labeled streptavidin to the subphase with the barriers open,
a minimum amount of energy transfer was observed with polymer 6, while polymer 8
displayed significantly higher energy transfer under similar conditions (Figures 4.14 and
4.16, respectively). This may be related to the linker length of the biotin to the polymer.
The longer linker provided better access to the deep biotin binding pocket of streptavidin
located at the center of each
-barrel. For the shorter linker, biotin cannot bind as
efficiently to the streptavidin, resulting in a weak energy transfer signal. In addition,
energy transfer for 6 was more efficient when the polymer was in the zipper phase (O to
15 nM/m). As there were less interchain interactions in the zipper phase, the polymer was
less likely to undergo quenching through non-radiative processes and chances of energy
transfer were increased. Conversely,
at the edge-on conformation
(15 to 35 mN/m),
quenching through polymer aggregation competed with energy transfer and it was thus
less efficient: (Figure 4.14).
93
.o
-71
E
w
C
sO
o_)
0
IL
450
500
550
600
650
700
Wavelength (nm)
Figure4.14. Fluorescence emission spectra of polymer 6, when incubated with Texas Red X-labeled
streptavidin, at various surface pressures. Less energy transfer observed at higher surface pressures, when
the polymer is in the aggregated edge-on phase.
Similar results were observed with polymer 8, where energy transfer diminished
as the polymer was compressed due to competition from fluorescence quenching. In
addition, Texas Red-X-streptavidin was injected into the subphase in two manners: with
the barriers closed (Figure 4.15) and barriers open (Figure 4.16). More energy transfer
was observed when the streptavidin was added to the trough with the barriers open,
indicating increased binding of the protein with the polymer. This was not surprising
considering that at minimum surface pressure conditions and open barriers, the polymer
was in the zipper phase and the biotinylated side chains were located further apart in the
subphase, allowing binding of the streptavidin. However, at the edge-on phase with
closed barriers, the biotinylated side-chains were closely packed in the subphase and did
not allow efficient binding of the protein. The system was not in equilibrium and this was
reflected in the protein binding.
0
0
E
In
4)
0
UL
400
450
500
550
600
650
700
Wavelength (nm)
Figure 4.15. Fluorescence emission spectra of polymer 8, when incubated with Texas Red X-labeled
streptavidin with closed barriers, at various surface pressures. Less energy transfer observed at higher
surface pressures, when the polymer is in the aggregated edge-on phase.
400
450
500
550
600
650
700
Wavelength (nm)
Figure 4.16. Fluorescence emission spectra of polymer 8, when incubated with Texas Red X-labeled
streptavidin with open barriers, at various surface pressures. Less energy transfer observed at higher surface
pressures, when the polymer is in the aggregated edge-on phase. More energy transfer signal observed
when compared to Figure 4.14.
Excitation spectra were measured in order to ascertain that the Texas Red dye was
indeed fluorescent due to energy transfer (Figure 4.17). The spectra corresponded largely
to the polymer's absorption profile. Furthermore, direct excitation at 400 nm (where the
polymer would be irradiated) of only Texas Red X-labeled streptavidin in the Langmuir
trough failed to produce any fluorescence. The observed fluorescence was therefore due
to energy transfer.
D
,0
oi
0
C
U
LI
350
400
450
500
550
Wavelength (nm)
Figure4.17. Excitation spectra of Texas Red X-labeled streptavidin, when incubated with polymer 8.
The affinity of the polymers for the free Texas Red dye (not conjugated to any
protein) was also monitored (Figure 4.18). While some energy transfer occurred due to
non-specific binding for both polymers, they do not explain the tremendous amount of
energy transfer observed for the protein experiments.
450
500
550
Wavelength (nm)
600
650
450
500
550
600
650
Wavelength (nm)
Figure4.18. Incubation with sulforhodamine 101 (Texas Red parent dye). Left: polymer 6. Right: polymer
8. Difference between the polymers was not as great as when Texas Red X-labeled streptavidin was used.
Conclusions
Amphiphilic polymers with pendant biotin moieties were synthesized. Their
behavior at the air-water interface was studied using the Langmuir trough. The linker
length suggested that subtle changes in the structure of the polymer can have large
consequences for analyte recognition. Energy transfer was more marked for the
biotinylated polymer with the longer linker, suggesting better binding with streptavidin.
97
Experimental.
General.
H and
3C NMR spectra for monomers and polymers were recorded on a
(Varian 300 MHz) or on a Varian VXR-500 (500 MHz) instrument. The chemical shift
data for each signal are given in units of 6 (ppm) relative to tetramethylsilane
(TMS)
where 6 (TMS) = 0, and referenced to the solvent residual. High-resolution mass spectra
were obtained on a Finnigan MAT 8200 system using sector double focus and an electron
impact source with an ionizing voltage of 70V, and with a Bruker DALTONICS APEX II,
3 Tesla, FT-ICR-MS with ESI source or EI/CI source. The molecular weights of
polymers were determined by using three PLgel 5m 105, 104 , 103 (300 x 7.5 mm I.D)
columns in series and a diode detector at 254nm at a flow rate of 1.0ml/min in THF or in
DMF. The molecular weights were reported relative to polystyrene or poly(ethylene
oxide) standards purchased from Agilent Inc. Melting point (m.p.) determination was
performed using a Laboratory Devices MEL-TEMP instrument (open capillaries used)
and was uncorrected.
Monolayer studies were performed on a NIMA 201 M Langmuir-Blodgett trough with a
quartz window. The in situ UV-visible absorption spectra were measured with a Cary 50
UV/visible spectrometer with fiber optics. Fluorescence spectra were measured with a
SPEX Fluorolog-2 fluorometer (model FL12,
450W xenon lamp) equipped with a
bifurcated fiber optic cable oriented at about 600 relative to the flat surface of the
subphase.
Materials.
All solvents were spectral grade unless otherwise noted. Morpholine and biotin were
purchased from Alfa Aesar and used as received. Texas Red-X conjugated streptavidin
and avidin were purchased from Molecular Probes Inc. and used as received. All other
chemicals were purchased from Aldrich Chemical In. and used as received. All air and
water sensitive synthetic manipulations were performed under a nitrogen atmosphere
using standard schlenk techniques.
98
(4): To 1 (0.367 g, 0.618 mmol) was added 5 ml CH 2Cl 2 . Anhydrous NEt 3 was then
added (0.129 ml, 0.927 mmol) and the mixture was stirred for 5min. 251 (0.129 g, 0.804
mmol) was added as a solution in 5 ml CH2C12. The reaction mixture was refluxed for
12h. The solvent was removed under reduced pressure. The residue was dissolved in 50
ml CHC13 and washed with 15ml H2 0. The organic layer was washed with 15ml brine,
dried over MgSO4. The resulting yellow oil was eluted with 7:3 ethyl acetate/hexane
through a plug of silica. Fractions containing the product were combined. The organic
solvent was removed under reduced pressure to afford a pale yellow solid (0.228g, 52%)..
M.p.: 53-54C.'H
NMR (500 MHz, CDC13): 7.24 (1H, s), 7.19 (1H, br), 7.16 (1H, s),
4.95 (1H, br), 4.45(2H, s), 4.13 (2H, t, J=4.5Hz), 3.90 (2H, t, J=4.5Hz), 3.80 (2H, t,
J=4.5Hz), 3.70-3.66 (4H, m), 3.56 (2H, t, J=4.5Hz), 3.54 (4H, m), 3.38 (3H, s), 3.33 (2H,
m), 1.43 (91-, s);
13C
NMR (125 MHz, CDC13 ): 167.9, 156.3, 154.1, 151.2, 123.4, 123.2,
86.8, 86.2, 79.8, 72.1, 71.3, 70.9, 70.8, 70.4, 69.7, 69.1, 59.3, 40.6, 39.3, 28.6; HR-MS
(ESI) calcd. For C2 2H34 12 N2 08 (M+H): 708.32, found: 709.0.
(5): A 50ml round bottom flask containing 4 (0.228 g, 0.322 mmol) was loaded with 5ml
TFA. The clear yellow solution was stirred for 30min. The TFA was removed, 2ml H2 0
was added and was also removed under reduced pressure. The deprotected product was
dried under high vacuum. To this was added 3ml anhydrous DMF, NEt 3 (0.049 ml, 0.483
mmol). This was stirred for 15min, then N-hydroxysuccinimido
biotin5 2 (0.lllg,
0.325
mmol) was added. The pale yellow solution quickly became a thick white slurry and was
stirred at room temperature for 12h. The solvent was removed under reduced pressure at
400 C and the reaction mixture was washed with 10 ml H2 0. The product was isolated by
centrifugation and lyophilized to afford a white powder (0.244 g, 91%). M.p.: 203-205 0 C.
'H NMR (500 MHz, DMSO): 7.88 (2H, m), 7.39 (1H, s), 7.31 (1H, s), 6.43 (1H, s), 6.36
(1H, s), 4.49 (2H, s), 4.28 (1H, m), 4.11 (2H, m), 3.74 (2H, t, J=5.OHz), 3.62 (2H, t,
J=5.OHz), 3.53 (4H, m), 3.42 (2H, t, J=5.OHz), 3.23 (3H, s), 3.19 (2H, t, J=5.OHz)), 3.14
(2H, t, J=5.OHz), 3.08 (2H, m), 2.80 (1H, dd, J=12.5, 5.0Hz), 2.58 (1H, J=12.5Hz), 2.06
(2H, t, J=7.5Hz), 1.62-1.57 (1H, m), 1.52-1.43 (3H, m), 1.32-1.26 (2H, m); 3C NMR
(125 MHz, ]DMSO): 172.3, 167.3, 162.6, 153.0, 151.9, 123.6, 122.7, 87.1, 86.7, 71.3,
99
70.1, 69.8, 69.7, 69.6, 69.1, 68.9, 61.0, 59.2, 58.0, 55.4, 38.3, 38.1, 35.2, 28.2, 28.0, 25.2;
HR-MS (ES[) calcd. For C2 7H401 2N 40 8 S (M+H): 834.50, found: 835.0.
Polymer 6: A 25ml schlenk flask was charged with 4 (20.00 mg, 0.0240 mmol), 5
(15.892 mg., 0.0240 mmol), Pd(PPh 3) 4 (1.39mg, 0.00120 mmol) and CuI (0.206 mg,
0.00120 mmol) under N2. To this was added 1.0 ml freshly degassed morpholine under
N 2 . The reaction vessel was sealed and heated at 600 C for 48h. The polymer solution was
then precipitated into -125ml methanol and collected by centrifugation, followed by
drying on high vaccum. Yield: 25mg, 86%. Mn = 15000, PDI = 1.78 for THF soluble
fraction. H NMR (500 MHz, CDC13, ppm): broad peaks at 7.1, 4.6, 4.4, 4.2, 3.9, 3.8, 3.6,
3.5, 3.4, 3.1, 2.9, 2.8, 2.0, 1.8, 1.6, 0.9
Polymer 8: A 25ml schlenk flask was charged with 7 (20.00 mg, 0.0217 mmol), 5
(14.375mg,
0.0217 mmol), Pd(PPh 3) 4 (1.25mg, 0.00108 mmol) and CuI (0.206 mg,
0.00108 mmol) under N2. To this was added 1.0 ml freshly degassed morpholine under
N 2 . The reaction vessel was sealed and heated at 600 C for 48h. The polymer solution was
then precipitated into -125ml methanol and collected by centrifugation, followed by
drying on high vaccum. Yield: 25mg, 89%. Mn = 9600, PDI - 1.41 for THF soluble
fraction. H NMR (500 MHz, CDC13): broad peaks at 7.1, 4.6, 4.3, 3.9, 3.8, 3.6, 3.5, 3.4,
3.1, 2.9, 2.7, 2.2, 1.8, 1.3, 0.9.
100
References
(1)
Tabor, T. J. Colloid. Interface Sci. 1980, 75, 240-245.
(2)
Terada, T.; Yamamoto, R.; Watanabe, T. Sci. Papers. Inst. Phys. Chem. 1984, 23,
173-184.
(3)
Adamson, A. W. Physical Chemistry of Surfaces; Fifth ed.; John Wiley & Sons,
Inc.: New York, NY, 1990.
(4)
Tanford, C. Ben Franklin Stilled the Waves: An Informal History of Pouring Oil
on Water: With Reflections on the Ups and Downs of Scientific Life in General;
Duke University Press: Durham, NC, 1989.
(5)
Franklin, B. Phil. Trans. R. Soc. 1774, 64, 445-460.
(6)
Roberts, G. G. Adv. Phys. 1985, 34, 475-512.
(7)
Rayleigh, L. Proc. Soc. 1890, 47, 364-367.
(8)
Pockels, A. Nature 1891, 43, 437-439.
(9)
Rayleigh, L. Phil. Mag. 1899, 48, 321.
(10)
Langmuir, I. J. Amer. Chem. Soc. 1917, 39, 1848-1906.
(11)
Blodgett, K. B. J. Amer. Chem. Soc. 1935, 57, 1007-1022.
(12)
Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry;
Third ed.; Marcel Dekker, Inc.: New York, NY, 1997.
(13)
www.ksvltd.com/sitellite/pdf/lb.pdf.
(14)
Birdi, K. S. Lipid and Biopolymer Monolayers at Liquid Interfaces; Plenum Press:
New York, NY, 1989.
(15)
www.nima.co.uk/equipment/ps/wilhelmy.htm.
(16)
www.shef.ac.uk/physics/research/molec/mol-elec/index.php?page=lbmono.
(17)
Mirley, C. L.; Koberstein, J. T. Langmuir 1995, 11, 1049-1052.
(18)
Shapovalov, V. L. Thin Solid Films 1998, 327-329, 816-820.
101
(19)
Seki, T.; Ichimura, K.; Fukuda, R.-I.; Tanigaki, T.; Tamaki, T. Macromolecules
1996, 29, 892-898.
(20)
Kim, Y. K.; Sohn, M. H.; Sohn, B. C.; Kim, E.; Jung, S. D. Thin Solid Films 1996,
284-285, 53-55.
(21)
Mabuchi,
M.; Kobata,
S.; Ito, S.; Yamamoto,
M.; Schmidt, A.; Knoll, W.
Langmuir 1998, 14, 7260-7266.
(22)
Mabuchi, M.; Ito, S.; Yamamoto, M.; Miyamoto, T.; Schmidt, A.; Knoll, W.
Macromolecules 1998, 31, 8802-8808.
(23)
Alexmndre, S.; Drue,
V.; Gomes-Ferreira,
S.; Huguet, J.; Valleton, J.-M.
Langmuir 1999, 15, 7708-7713.
(24)
Teerenstra, M. N.; Vorenkamp, E. J.; Schouten, A. J.; Nolte, R. J. M. Thin Solid
Films 1991, 196, 153-162.
(25)
Antunes, P. A.; Santana, C. M.; Aroca, R. F.; Oliveira, O. N.; Constantino, C. J.;
Riul, A. Synth. Met. 2005, 148, 21-24.
(26)
de Boer, B.; van Hutten, P. F.; Ouali, L.; Grayer, V.; Hadziioannou,
G.
Macromolecules 2002, 35, 6883-6892.
(27)
Arslanov, V. V. Russ. Chem. Rev 2000, 69, 883-898.
(28)
Bo, Z.; Rabe, J. P.; Schluter, A. D. Angew. Chem. Int. Ed. 1999, 38, 2370-2372.
(29)
Sawada, H.; Kita, H.; Yoshimizu, M.; Kyokane, J.; Kawase, T.; Hayakawa, Y.;
Yoshino, K. J. Fluorine. Chem. 1997, 82, 51-54.
(30)
Dhanabalan, A.; Riul, A.; Constantino, C. J. L.; Oliveira, O. N. Synth. Met. 1999,
101, (90.
(31)
Ramanathan, K.; Ram, M. K.; Malhotra, B. D.; Murthy, A.; Surya, N. Mat. Sci.
Eng. C-Biomim. 1995, C3, 159-163.
(32)
Ram, M. K.; Bertoncello, P.; Nicolini, C. J. Mater. Sci 2003, 38, 4951-4956.
(33)
Kimkes, P.; de Jong, A.; Oostergetel, G. T.; Schouten, A. J.; Challa, G. Thin Solid
Films 1994, 244, 705-709.
102
(34)
Feng, C. L.; Zhang, Y. J.; Jian, L.; Yan, L. S.; Lian, Y. X.; Gui, R. Q.; Lei, J.; Dao,
B. Z. Langmuir 2001, 17, 4593-4597.
(35)
Kim, J.; McHugh, S. K.; Swager, T. M. Macromolecules 1999, 32, 1500-1507.
(36)
Langmuir, I.; Schaefer, V. J. J. Amer. Chem. Soc. 1938, 60, 1351-1360.
(37)
Kim, J.; Swager, T. M. Nature 2001, 411, 1030-1034.
(38)
Kim, J.; Levitsky, I. A.; McQuade, D. T.; Swager, T. M. J. Amer. Chem. Soc.
2002, 124, 7710-7718.
(39)
Brockman, H. Curr. Opin. Struc. Biol. 1999, 9, 438-443.
(40)
Maierhofer, A.; Braun, M.; Vostrowsky, O.; P.Hirsch, A.; Langridge, S.; Bayerl,
T. M. J. Phys. Chem. B. 2001, 105.
(41)
Blankenburg, R.; Meller, P.; Ringsdorf, H.; Salesse, C. Biochemistry 1989, 28,
8214--8221.
(42)
Samuelson, L. A.; Miller, P.; Gallotti, D. M.; Marx, K. A.; Kumar, J.; Tripathy, S.
K.; Kaplan, D. L. Langmuir 1992, 8, 604-608.
(43)
Hannink, J. M.; Comelissen, J. J. L. M.; Farrera, J. A.; Foubert, P.; De Schryver,
F. C.; Sommerdijk, N. A. J. M.; Nolte, R. J. M. Angew. Chem. Int. Ed. 2001, 40,
4732-.4734.
(44)
Wang, S.; Leblanc, R. M. Biochim. Biophys. Acta 1999, 1419, 307-312.
(45)
Charqych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D. Science 1993, 261,
585-588.
(46)
Charych, D. H.; Cheng, Q.; Reichert, A.; Kuziemko, G.; Stroh, M.; Nagy, J. O.;
Spevak, W.; Stevens, R. C. Chemistry and Biology 1996, 3, 113-120.
(47)
Ma, B.; Fan, Y.; Zhang, L.; Kong, X.; Li, Y.; Li, J. Colloid Surface B 2002, 27.
(48)
Song, J.; Cheng, Q.; Zhu, Z.; Stevens, R. C. Biomed. Microdevices 2002, 4, 213221.
(49)
Cheng, Q.; Stevens, R. C. Chem. Phys. Lipids 1997, 87, 41-53.
103
(50)
Pugliese, L.; Coda, A.; Malcovati, M.; Bolognesi, M. J. Mol. Biol. 1993, 231,
698-7 10.
(51)
Beer, P. D.; Cadman, J.; Lloris, J. M.; Martinez-Mtfiez,
R.; Soto, J.; Pardo, T.;
Marcos, M. D. J. Chem. Soc., Dalton Trans. 2000, 11, 1805-1812.
(52)
Becker, J. M.; Wilchek, M. Biochim. Biophys. Acta 1970, 264, 165-170.
104
Juan Zheng
Education
Ph. D. 2005, organic chemistry
Massachusetts Institute of Technology, Cambridge, MA
Thesis title: Poly(phenylene ethynylene)s in biosensor applications
Advisor: Prof. Timothy M. Swager
B.Sc. Hon. 2000, bioorganic chemistry
University of Toronto, Toronto, ON, Canada
Thesis title: Fluorinated BINOLs for asymmetric catalysis
Advisor: Prof. Andrei K. Yudin
Experience
Graduate Research Assistant
2000 - present, Massachusetts Institute of Technology
* Developed model biosensor consisting of biotin-functionalized electronically
conjugated water-soluble polymers
* Developed amphiphilic fluorescent conjugated polymers for air-water interface and
liposomal studies
* Prepared carbohydrate-functionalized fluorescent conjugated polymers for bacterial
sensing (collaboration with Peter Seeberger's group, Zirich)
· Co-authored book chapter on conjugated polymers in chemosensing and biosensing
· Teaching assistant activities:
* teaching assistant for advanced chemistry laboratory
* recitation leader in introductory and advanced organic chemistry, assisted in exam
preparation and grading
Undergraduate Research Assistant
Advisor: Prof. Andrei K. Yudin
1999 - 2000, University of Toronto
Synthesized and analyzed fluorinated BINOLs for asymmetric catalysis
Advisor: Prof. Ian Manners
Summers 1998 and 1996, University of Toronto
Synthesized and characterized ferrocene containing inorganic polymers
Advisor: Prof. Andrew Woolley
1996-1997, University of Toronto
Investigated caged gramicidin for model ion channels
Awards
NSERC post-graduate fellowship (2000, declined)
NSERC undergraduate research award (1999)
Sarah Cusick Gollop and William George Gollop memorial scholarship in
chemistry (1999)
Dickinson-Cartwright 3T0 Scholarship (1999)
AstraZeneca undergraduate research poster award (1999)
105
Publications
1. Zheng, J.; Swager, T. M. Poly(arylene ethynylene)s in chemosensing and
biosensing. Advances in Polymer Science. Springer-Verlag Heidelberg, 2004,
151-179
2. Zheng, J.; Swager, T. M. Biotinylated poly(p-phenylene ethynylene): using
energy transfer for the detection of biological analytes. Chem. Comm. 2004,
2798-2799
3. Disney, M. D.; Zheng, J.; Swager T. M. Seeberger P. H.; Visual detection of
bacteria with carbohydrate-containing fluorescent polymers. J. Am. Chem.
4.
Soc. 2004, 13342-13346
Thieme K, Bourke SC, Zheng J., MacLachlan M. J.; Zamanian F.; Lough A.
J.; Manners I. Synthesis, characterization, and structures of zircona- and
boracyclosiloxanes with ferrocenyl substituents. Can. J. Chem. (2002), 80 (11):
5.
1469-1480
Chen Y.; Yekta, S; Martyn, L. J. P.; Zheng, J.; Yudin, A. K. Regioselective
substitution of fluorine in F8BINOL as a versatile route to new ligands with
axial chirality. Org. Lett. (2000), 2(22), 3433-3436
6.
MacLachlan, M. J.; Zheng, J.; Thieme, K.; Lough, A. J.; Manners, I.; Mordas
C.; LeSuer, R.; Geiger, W. E.; Liable-Sands, L. M; Rheingold, A. L. Synthesis,
characterization, and ring-opening polymerization of a novel
[1]silaferrocenophane with two ferrocenyl substituents at silicon. Polyhedron
7.
(2000), 19 (3), 275-289
Yudin, A. K.; Martyn L. J. P.; Pandiaraju, S.; Zheng, J. Lough, A. J.
F8BINOL, an electronically perturbed version of BINOL with remarkable
8.
configurational stability. Org. Lett. (2000), 2(1), 41-44
MacLachlan, M. J.; Zheng, J.; Lough, A. J.; Manners, I.; Mordas, C.; LeSuer,
R.; Geiger, W. E.; Liable-Sands, L. M.; Rheingold, A. L. Ferrocenylsiloxane
chemistry: synthesis and characterization of hexaferrocenylcyclotrisiloxane and
9.
tetraferrocenyldisiloxanediol. Organometallics (1999), 18(7), 1337-1345
MacLachlan, M. J.; Ginzburg, M.; Zheng, J.; Knoll, O.; Lough, A. J.;
Manners, I. Ring-opening addition of hydrogen chloride to monocyclic and
spirocyclic[l]ferrocenophanes: a convenient and controlled route to
ferrocenylchlorosilanes
1415
and germanes. New J. Chem. (1998), 22(12), 1409-
106
Acknowlegments
I joined the Swager group by a stroke of luck in 2000. That year, Prof. Peter Seeberger
was overwhelmed by the number of students who wished to join his research group. To
accommodate everyone, he asked me if the collaborative project on carbohydratefunctionalized polymer sensors, which was to be a major part of my thesis, could be based in
Tim's group. I promptly agreed and have since enjoyed every moment of my time as a
member of the Swager group.
Tim had a tremendous amount of influence over my graduate experience that most
likely will extend beyond these five brief years at MIT. His astuteness, boundless love of
research, and many brilliant ideas inspired me and reminded me time and again of the reasons
why I enjoy chemistry. I don't think I could have made it without his gentle encouragement,
support, and trust. In his integrity, tolerance, generosity and kindness, I saw what it meant to
be a decent person even when under extreme pressure, and I observed how this brought out
the best of those around him. He's been a great advisor and teacher; and I've learned many
valuable lessons from him that I will carry with me long after I leave here. I have been blessed
and truly lucky to have known him.
I am grateful to the professors and graduate students I worked with during my
undergraduate years at the University of Toronto. Professors Manners, Woolley, Ozin, and
Yudin introduced me to research in chemistry. Mark MacLachlan took me under his wing and
infused me with enthusiasm for working with the flasks of orange and red, and sometimes
pyrophoric substances. Without them I could not have made it to MIT.
As for my colleagues, I couldn't have asked for better people to work with and I will
sorely miss interacting with them. I am most thankful to my collaborators, Matt Disney and
Prof. Seeberger, who have been tremendously helpful and patient. I am indebted to the both of
them. My first benchmate, Kenichi, helped me with many aspects of working in the lab and
with our group's chemistry when I first started. My biosensor benchmates, Jordan, Gigi, and
Jess, patiently put up with Pink Floyd, the Nightmare before Christmas, Sarah McLachlan (it's
so depressing!) on perpetual replay and were always generous in their advice and help. Paul
Kouwer, whom I've converted to the Jazz Oasis and in turn, introduced me to Manu Chao,
gave me much technical advice. I had a great time working and sharing space with all them.
Other people have helped me in many ways and I am thankful to them. Jean (the synthesis
god) gave me countless compounds and was unfailingly generous with suggestions and his
time. Justin Hodgkiss of the Nocera group helped me with fluorescence lifetime
measurements. Sam provided answers to my questions about photophysics. Becky kept the
group in smooth running order (what would we do without you? And 80's nights!), along with
Richard Lay, Karen Warren and Simone Nakhoul. The DCIF staff: Dave, Mark, and Li Li
assisted me with NMR spectroscopy and mass spectrometry. Others in the group: Andrew (the
chats and tequila), Tae-Hyun (you flirt!), Evgueni (the vodka, of course!), Elena (the great
pastry chef), Jess (hot-pot, yum), Anne (the references), Aimee and Steffen (the great advice),
107
Karen and Phoebe (the boisterous conversations), Paul Byrne, Youngmi and Craig (my fellow
classmates), and many more (everybody!) all made the lab home for me.
I also could not have such a happy experience at graduate school without my good
friends at Ashdown (Chi, Hsiang-Wei, Rob, An, Shounak). I will miss our talks, outings and
shared meals. And of course, my friend Chiaki, who has been a willing ear and offered me
much support in trying times. It would be a lonely life without them.
Lastly, none of this would be possible without my family. My Great-Aunt, GreatUncle and their family sponsored my parents and me to immigrate to Canada nearly 20 years
ago and got us on our feet - without them I may not even be at MIT today. My cousin
Michael and his extended family have offered me much counsel and helped me recover from
illness. I am thankful to all of them.
My parents have given up much in immigrating to Canada. They left their siblings and
parents behind, and struggled through many hardships to build a life for us and to provide me
with better opportunities. They have been my safety net and my support. Difficult as it was,
they have let me fly far from the nest to find my own life. I am grateful for their unconditional
love and for their sacrifices. In our absence from China, three of my grandparents passed on. I
especially miss my grandma, who loved the little terror that I was and demanded nothing in
return, and I wish that my grandparents could all somehow see me graduate.
And of course, I must thank Josh. In addition to his frequent roles as technical
consultant, psychologist, [copy editor], cook, laundry-boy, chauffeur, to name a few, he has
patiently stood by me through thick and thin since the very first day of graduate school. It's
been such a fun journey. I can't wait to start the next phase of our lives together, hand in hand.
108
Appendix
109
------
.
I-1
I
I
,7 ,
7
8
8
6
3
4
5
2
1
0
ppm
'H NMR (CDC13 ) of 1 (Chapter 2).
I I .
I
.
"i
.1,
,,.
i
,i
ild
.I . 1
I
.
.... ,
.
ilil,
7MT77
o1
---
16
1. 11. 1Mir
11,11 Im
.- . .I I , , '7711
'. -- 77 71
11
-FMIMT
.- ..' ' 7717r-11771mlj I !MT; ,
. .
140
lZ0
100
I
I
I TI
80
60
40
20
. i7
ppm
3C NMR (CDC13) of 1 (Chapter 2).
110
i
_
--
8
7
6
5
4
_
_
.
I
J_
_
--
--
0
2
3
1
0
ppm
'H NMR (CDC13 ) of 2 (Chapter 2).
~"_;
F
..' l .Jl
iT.,.,'"L::'
....
T
iv: q- r ''l'q
I
'-Iq, ll
181]
I,
luh.
q'' ' Irl*l
-T
.TI
160
.....................
MWANNAWAM
, i ,, ] - ii1
Irwlllr--r~o I
.......
I I
T
I
T
140
T' I
12 0
T
1~
. , .. llr
T
'r r i,.
*rJ.i'"llWmllmi, d
I'r'r~
w**w~L~rr
w*A*"viwiI~·UY~u
1Wm....
T
100
80
60
40
20
ppm
' 3C NMR (CDCI 3) of 2 (Chapter 2).
111
I
j
i
T
-
1H NMR
-·
, -
-----
---------
I--
-
,
-
0Tp
1
2
43
5
6
7
8
iI
(DMSO-d 6 ) of 3 (Chapter 2).
-
·- '"
*---
,
-180
.ll_i ,
i.,....
..-
-r-X-- *18T 16
160
-,
-
-
,
· ·-·
...
140
{1
1
-l-ll
^---rrrrrnrrrrrr··r-----n-*·-----rr120
100
I
I ... .
L.- ·.--
·- - r--l-·- · r···-i
80
....
i
-rm
60
I
r---m--77Tr
40
·· ·-- --Tmr-mlm
20
· -.
ppm
'3C NMR (DMSO-d6) of 3 (Chapter 2).
112
r
r TVTT
4
§
-
w
-
2
3
X
.
r
1
r
0
ppm
'H NMR (CDC13) of 4 (Chapter 2).
.. ]
IjdYLII
I
-
i.i
,
A k
~rm~~r
,,.,,....,.
i1 l-l.r
, ,-, . . -..
180
.Y~r~L~
ll7ln~r
Im~lr'l~~
160
I
LIL--rlll,~~~·~,
,..,.,,.,.-.,
140
1.
- ,
· 1lY~~~~
-- m-71-47-T
,,,,,.
120
,,
I
I~~.,IILI~
7-lr-PIMM-9,prrmv'rlmp, nm~
.......,
100
ril
80Gn
L
-L
' II I ·
r IVMFP
FIMPIn
'
I rll''
·I-lr·*,·T,·n
4n
wli,b-AL~;hL.u
lira
r·rrll-·ll-pns?nl
TI
?
nnm
' 3 C NMR (CDC13 ) of 4 (Chapter 2).
113
-
T --
- --
-
'--
T
--
I
,
;
5
6
7
8
4
3
2
1
0
ppm
'H NMR (DMSO-d 6 ) of 5 (Chapter 2).
~~-r·-'-
I
I
I
· LI··L-U-·II
--.-
LIL·_III.- -I·.-I ---
·II-·I·--yUI_-_Y.II_._I1_
kv·
IY1·J0Ij· l l-
I I
Y~
I I. ,
I 1,..,,,. -1
.~.,,
'
r
....
rI
18(1
160
I,,'
.,
r
140
120
100
80
6 An
aA
n
13CNMR (DMSO-d6) of 5 (Chapter 2).
114
,1
.
, ------- - -8
'H NMR (DMF-d
7
7
7
r
6
I
5
,
4
4
I
I
i
t3
3
cL
_..._
ppl
2
1
0
Ppm
and D 2 0) of 8 (Chapter 2).
115
I
or
F
ri~-
7
8
lH NMR (DMF-d
7
6
-
I
5
TV-
4
---
As
I
-
3
1 2
2
1
1
T
r
l0
0
ppmr
ppm
and D 2 0) of 9 (Chapter 2).
116
AJ
r-
A8
-1/
v
1
-r
7
X
I-
-
6
rl
~
.
-
.
-
5
FAj
Y-A
\
Ju-r
T
E
4
I
3
a
-
J
I
I
c
,
2
1
0
ppm
'H NMR (CDC13 ) of 13 (Chapter 2).
117
ol__
-T
8
r-
7
ix J
li\ AII
__._ ,
I
-
6
_
'
,
5
,
4
4
-
--
3
3
I
2
-5-
1
1
,
1
0
ppm
IH NMR (CDC13) of 14 (Chapter 2).
118
11
1 ,
8
~--,
~
-4
T,
-
7
-%i
6
5
I
~
I
4
i
I
I
3
i
[-
I
2
1
0
ppm
'H NMR (CDC13 ) of 3 (Chapter 4).
.
I_-1,11--
.
[
i
I
i1,~-~]11,~[
18
)
.11
1
'
I
....
I I
i
140
13CNMR (CDC13 ) of 3 (Chapter 4).
m~
J-~~~-aci
h
~
160
II
fq
-
1
rTT~-T-Tr
120
WAI..
i. i 1
F-
rT
r
,T
100
TT-TT
- TTI
I
I-
J-
1
-
I
Icrmrr
TT
80
60
40
20
nnm
119
1,
AI
8
7
6
I
.
5
4
4J
T KT
-
p
3
r
IT
2
1
0
ppm
'H NMR (DMSO-d6 ) of 4 (Chapter 4).
""
--
'
-------
180
'
-----
'''
-.---
""
160
-- ___
----- '----
140
-- "
1
.------
120
~~
_
100
.
80
"'
1
1I lI
60
, 1
.
1.
40
1 I1
I1
20
nnm
' 3C NMR (DMSO-d 6 ) of 4 (Chapter 4).
120
I
i!
l
I
8
7
6
5
71
i
4
2
'
-
'
·
·
1
0
PP
ppm
'H NMR (CDCI3) of 6 (Chapter 4).
121
I
I
I
I
ii~1-8
i
III
1
7
-
ij L0
-
1111I1
6
1
_, 177
S
4
3
2
1
0
ppm
'H NMR (CDC13) of 8 (Chapter 4).
122