z. t... s
z. r.
v .k:

1.32 3...?
1 .

Armin. , V . .A .
‘ , , n...”
. A . ‘ Wag!
9 5r ..

. L
in“.
u. .
.3,

.53

a... .u. ha

I)-
, ..
"3.... 3. ..
fit 1.: 31
$343.... .1
a: .1. 1:3,

unis 1:...
a .3me

an) e
fivflww;

LUL. .71.? .
7,1345”?!
134 to“ .

 

all .3. .2

 

 

I'Htgib‘

2003

-‘

This is to certify that the
dissertation entitled

DYNAMICS OF FLUORESCENT BIOLOGICAL PROBES IN
SOLUTION PHASE PROCESSES

presented by

Lee Kelepouris

has been accepted towards fulfillment
of the requirements for the

Ph. D. degree in Chemistrx

 

 

new

v Major Professor's Signature

October 4&002
Date

 

MSU is an Attinnative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

PIACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIHC/DatoDUOpGEI-p. 15

DYNAMICS OF FLUORESCENT BIOLOGICAL PROBES IN SOLUTION PHASE
PROCESSES

By

Lee Kelepoun's

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

2002

Abstract

DYNAMICS OF FLUORESCENT BIOLOGICAL PROBES IN
SOLUTION PHASE PROCESSES

By

Lee Kelepouris

In the past decade the use of fluorescence based methods has flourished in many
areas of science. Fluorescence is relevant to chemical, physical, and biological sciences
as a tool of investigation, analysis, control and diagnosis. The recent advances in light
sources, detectors, and compact fast electronics has prompted the advent of new
technology in these areas. Great advances in time-resolved spectroscopy have been made
in fluorescence lifetime imaging microscopy and fluorescence correlation spectroscopy.
Also, increased detection sensitivity allows single molecule detection in the restricted
field of a confocal microscope making possible studies of phenomena at the molecular
level. These numerous applicatons are possible due to the sensitivity of the emission
properties of organic chromophores to their molecular environment. Chromophore
emission may be affected the pH, polarity, dielectric constant, viscosity, quenchers and
many other factors.

The development and characterization of new fluorescent probes is needed. The
ability to tailor probes for specific environments is important. Many potential in viva
applications exist with the introduction of fluorescent residues amenable to protein
incorporation and biological systems.

This thesis is a study of fluorescent molecules in heterogeneous environments.

The study focuses on understanding the factors that determine a fluorescent probes ability

81'

de

I‘Cl

me

the:

[RI

env

to interact with and report on its molecular environment. The probes are derivatives of
the naturally occurring amino acid tryptophan and its non-natural analog 7-
azatryptophan. We began by studying 7-azaindole and 7-azatryptophan in saturated
solutions of aqueous adipic acid in an effort to detect the onset of crystallization. 7-
Azatryptophan has a carboxylic functionality in common the adipic acid and may act as a
site for aggregation enabling it to detect self-assembly of solutes preceding
crystallization. We monitor the steady state and time resolved emission and rotational
dynamics of the chromophores as a function adipic acid concentration to determine the
ability of the probe to sense solute events. The comparison of concentration-dependent
results reveal the probes unique interactions with solutes arising from their structural
differences.

We use the information gained on the properties of the 7-azaindole chromophore
in the adipic acid system and extend its use to the biological realm for the study of small
peptides. 7-Azatryptophan was attached covalently to a series of valine oligomers and
studied in water and in several aqueous micelle solutions to provide well defined
heterogeneous environments. This study establishes that a balance exists between ionic
and dispersion forces and mediates interactions with micelles. The study of probe
micelle interactions was continued using several tryptophan and 7-azatryptophan
derivatives. This work elucidated structural criteria for interaction with micelles and
revealed that the subtle differences in the indole and 7-azaindole chromophores lead to
measurably different interactions in these heterogeneous environments. The work in this
thesis provides a knowledge base to aid in the application and interpretation of
tryptophan and azatryptophan derivatives as probes of heterogeneous systems and peptide

environments.

To Eliag Kekertouptg

-iv-

M

DL
r0;
Ste
Sch.

alw

Acknowledgements

My Graduate school experience has been a pleasant one. This was due, in large
part, to the people I will acknowledge in the following paragraphs. Among these persons
are those I was fortunate enough to work with, those I developed friendships with, and
loved ones who encouraged and supported me throughout my academic career.

Of the people I must thank, my advisor, Gary Blanchard is at the forefront. Gary
has dutifully served as my research advisor. Of this, I am greatly appreciative. On many
occasions during my time in the group we have said Gary is “The Best”. Some of the
things earn him this description as an advisor and boss are his knowledge, patience,
generosity, humor, and the true interest he has in helping his students succeed as well as
others. Gary has not only taught me how to become a better scientist but I have learned
much about management, politics, and the scientific community in general through his
willingness to share his own experiences and endeavors.

I would also like to thank my committee members, John McCracken, Rob
Maleczka, and Simon Garrett. Simon deserves special mention for graciously stepping in
as my second reader on short notice.

I was lucky to have joined Gary’s group at a time when comradery was at a high.
During this time, I have made several very good friends. Joe Tulock showed me the
ropes of graduate school and the laser lab. Joining the group a mere half hour before me
Steve Bakiamoh was on a parallel course and was my partner though many graduate
school hurdles. Thanks Steve, it was nice to have someone to suffer with. You could

always count on Punit Kohli to be in the lab and ready for science or fun. These are the

gm

frie

futL

MIC

yom

 

group members with whom I was closest. We have had the opportunity to prove our
friendship many times over the years and I am sure we will continue to do so in the
future.

There are other group members both past and present I consider friends and could
recognize. I wish them the best, among them, the well deserving, Jaycoda Major. While
not a group member, I must thank Todd Bryden, for his friendship and for discussing
science in a manner that challenged and encouraged me.

I am indebted to my sister Nina and brother in-law Rob for providing tremendous
support and stability at just the right time during my undergraduate studies helping me to
perform at my potential. I likely would not have attended graduate school without them.

These people were all part of my successful graduate school career but the one
who helped most to make graduate school a pleasant journey is Carina, thank you for

your love, caring, and patience.

-vi-

7

TABLE OF CONTENTS

Page

List of Tables .............................................................................................................. ix
List of Figures .............................................................................................................. x
Chapter 1. Introduction ............................................................................................ 1
1.1 Literature Cited .......................................................................................... 8

Chapter 2. Lifetime and Reorientation Measurements of 7-Azaindole and 7-
Azatryptophan in Aqueous Adipic Acid Solutions. The Significance of
Pendant Functionalities in Solution Phase Association Processes ................... 10

2.11ntroduction 11

2.2 Experimental ............................................................................................ 14
2.3 Results and Discussion ............................................................................. 18
2.4 Conclusions .............................................................................................. 38
2.5 Literature Cited ....................................................................................... 39
Chapter 3. Dynamics of 7-Azatryptophan Derivatives in Micellar Media.
Elucidating the Interactions Between Peptide Oligomers and Micelles ............. 42
3.1 Introduction ............................................................................................ . 43
3.2 Experimental ............................................................................................ 46
3.3 Results and Discussion ............................................................................ 51
3.4 Conclusions .............................................................................................. 75

3.5 Literature Cited 76

Chapter 4. Dynamics of 7-Azatryptophan and Tryptophan Derivatives in
Micellar Media. The Role of Ionic Charge and Substituent Structure. . . . . 79

4.1 Introduction ............................................................................................ . 80
4.2 Experimental ............................................................................................ 84
4.3 Results and Discussion ............................................................................ 86
4.4 Conclusions .............................................................................................. 115

4.5 Literature Cited 116

vii

Chapter 5. Conclusions and Future Work ............................................... 119

5.1 Conclusrons 119
5.2 Future Work 122
5.3 Literature Cited ........................................................................................ 125

viii

 

 

 

Ta

Tu

Ta

Tat

Tat

 

Table 3.1

Table 3.2

Table 3.3

Table 4.1

Table 4.2

Table 4.3

Table 4.4

LIST OF TABLES

Solution and physical parameters of micelles used in this work ............. 48
Values of quantities extracted from the “Wobbling-in-a-Cone” model.... 72
Experimental diffusion constants for oligopeptides in water ................. 73
Solution and physical parameters of micelles used in this work ............. 87
Absorbance, emission, and fluorescence decay parameters for probes in

water and micelles. The (it-ratio (last column) is the ratio of the
prefactors for the two decay components ........................................ 89

Fitted decay time constants and zero-time anisotropies for the systems
indicated in the left column The experimental data were fitted to the
function: R(t) = R1(O)exp(-t/tl)+R2(O)exp(-t/‘tz). For probe/solution

systems with one decay constant indicated, R2(0) = 0 103
Values of quantities extracted from the hindered rotor model. 8’ is the
order parameter, determined from fits to Eq. 44 111

ix

Figl

Fl gu]

 

IIgur

 

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

Figure 2.9

LIST OF FIGURES

Structures of the chromophores, 7-azaindole (7A1) and 7-
azatryptophan (7AT), and solute, adipic acid ...................................

Schematic of the time correlated single photon counter used to
measure fluorescence lifetimes and rotational diffusion dynamics...

Linear absorption and emission spectra for the probes 7A1 (dotted
trace) and 7AT (solid line) in aqueous solution. Intensities of the
bands have been normalized for presentation .............................

Normalized emission spectra for (a) 7A1 and (b) 7AT as a function
of adipic acid concentration. Chromophore concentrations are 5
M for all solutions. The emission spectra shift progressively to the
red with increasing adipic acid concentrations. Shown are spectra
in water, 1%, 5%, and 100% of saturation adipic acid solutions.......

Calculated a-fractions for (a) 7A1 and (b) 7AT. The pH range of
adipic acid solutions (2.7 - 3.8) is boxed. The calculations were
made using ground state pKa values for each chromophore. For
7A1, pK = 4.5 and for 7AT, pKl = 2.70, pKz = 3.85 and pK3 =
9.35 ..............................................................................

Fluorescence lifetimes of (a) 7A1 and (b) 7AT as a function of
adipic acid concentration. Where error bars are not visible, the
uncertainty lies within the vertical dimension of the symbol............

Fluorescence lifetimes measured for aqueous solutions containing
(a) 5 11M 7A1 and (b) 5 u_l\_d 7AT in adipic acid (I), buffered (O),
and un-buffered (A), aqueous solutions as a function of pH. . . .

Stem-Volmer plots for 7A1 (0, bottom trace) and 7AT (I, top
trace) in citric acid/di tassium hydrogen phosphate buffer. For
7191 k,,=(2.5:|:0.1) x 10 M." s'1 and for 7AT kq= (2310.1) x 10° M."
s' ................................................................................

Reorientation data for 7AT (I, top) and 7A1 (0, bottom) in
aqueous adipic acid solutions ...............................................

12

17

19

21

22

25

27

3O

35

F1 gure
Figure

Figure

Figure

Figure

Table 1

Figure

FiSure

Figure .

Figlue :

 

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Table 3.6

Figure 3.7

Figure 3.8

Figure 3.9

Figure 3.10

Structures of 7-azatryptophan (7AT) and 7-azatryptophan-valine
methyl ester (7AT-Val“), n = 1-5

Structure of surfactants used in this work. Top: anionic SDS;
center: cationic CT AB; bottom neutral Thesit® ...........................

Fluorescence spectra of 7AT in water and surfactant solutions.

Spectra have been normalized and are offset for clarity of

presentation ....................................................................

Fluorescence spectra of 7AT and 7AT-valn oligopeptides in water.
Spectra have been normalized and offset for clarity of
presentation ...................................................................

Fluorescence lifetimes of 7AT and 7AT-Valn (n = 1-5) oligomers in
water(i:]), Thesit (0), CTAB (V), and SDS (A). For points where
error bars are not obvious, the uncertainty lies within the vertical
dimension ofthe symbol................... . ..

Dependence of 7AT-Van fluorescence lifetime on SDS
concentration. The vertical bar indicates the CMC for this system
(8.3 mM). Inset: Dependence of lifetime on low SDS
concentration (first three points in the figure) .....................................

Rotational diffusion time constants of 7AT and 7AT-Valn as a
function of n in water at 20°C. The slope of the best-fit line for
these data is 32.5 ps/valine unit ..............................................

Rotational diffusion time constants of 7AT and 7AT-Valn in SDS
(25 mM, 3 x CMC). The anisotropy functions of the oligomers
decay as a two component exponential and 7AT decays as a single
exponential .....................................................................

Rotational diffusion time constants of 7AT and 7AT-Valn in CT AB
(2.5 mM, 2.7 x cmc). The anisotropy functions of the oligomers n =
4 and 5 decay as a two component exponential while 7AT and 7AT-
Vall — 7AT-Val3 decay as a single exponential with a slope of 33.6
ps/valine unit ...............................................................

Rotational diffusion time constants of 7AT and 7AT-Val" in
Thesit® (1.0 mM, 10 x cmc). The anisotropy functions of the
oligomers n = 2-5 decay as a two component exponential while
7AT and 7AT-Val] decays as a single exponential.................

xi

45

45

52

53

55

57

61

65

69

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 5.1

Structures of the chromophores in this study. For 7AT X=N and
X=C for TRP ...................................................................

Normalized emission spectra of tryptophan derivatives in 25 mM
SDS solution. TRP = solid line, boo-TRP = dashed line, DD-TRP =
dotted line, NATA = alternating dot/dash line ............................

Normalized emission spectra of boc-7AT in aqueous micelle
solutions and water. boc-7AT in water = solid line, CT AB solution
= dashed line, SDS solution = dotted line, Thesit® solution =
alternating dot and dashed line ..............................................

Experimental anisotropy decay of DD-AT in water (data points) and
best-fit line to a single exponential decay function...................

Experimental anisotropy decay of DD-AT in Thesit® solution (data
points) and best fit line to a biexponential decay function ..............

Structures of (a) monomeric SUG and (b) polymeric SUG with
probe molecule incorporated ................................................

xii

82

90

91

100

101

124

of res
could
better
Phase
neat s
fluores
dmam
fluores

)‘Ieldin

SPOHILII
or)"Stall
and f0(
“fidersu
cOmmm
Problem

 

801mmn

Chapter 1

Introduction

Understanding the dynamics of solution phase interactions has been an active area
of research for several decades. A greater understanding of intermolecular interactions
could lead to the ability to predict and control chemical reactions and separations and to
better understand biological phenomena. A primary tool for the investigation of solution
phase events has been through transient and steady state spectroscopy of chromophores in
neat solvents. Many models have been used for the interpretation of such data and
fluorescence spectroscopy has proven to be a powerful method for understanding the
dynamics of chromophores, peptides, and proteins in solution. The time scale of
fluorescence decay and depolarization make it ideally suited to study these events,
yielding information on picosecond to nanosecond time scales.‘

A detailed understanding of the intermolecular interactions that lead to the
spontaneous self-assembly of solutes from solution could greatly benefit the process of
crystallization. Crystallization is an important technique in chemical, pharmaceutical,
and food industries for the separation and purification of many materials. A greater
understanding of crystallization would not only benefit these industries, but the scientific
community in general. A scientist investigating crystallization is faced with several
problems that must be solved in order to develop a general theory of crystallization so
that control and monitoring of the process may be implemented.

Many techniques have been used to study the structure of supersaturated

solutions. Attempts to correlate supersaturation with changes in physical properties as a

-1-

funcrion

discontin
solubllltj
glucose

ribratior
uamrni
is hinde
to the l

Suggest

prObe n
by stuc
Sensitix
mOlCCu
microe
in solv
t0 calc
quannj
mOlEC‘

Their 1

SOlute

function of concentration have been made. These bulk physical properties rarely show
discontinuities with concentration at the equilibrium saturation point, especially in low
solubility systems.2 Mathlouthi et al.3'5 used vibrational spectroscopies to study aqueous
glucose solutions. Measurements were made of concentration dependent changes in
vibrational frequencies of both solvent and solute. Their data showed that some of the
water molecules are coordinated in a tetrahedral manner and that H-O-H bending motion
is hindered in concentrated solutions. It was not clear what specific interactions gave rise
to the observed motional hindrance, but their investigations did provide evidence to
suggest some level of solute organization in supersaturated solutions.

There is precedent in the literature for studying crystallization with fluorescent
probe molecules. Berglund et al.6‘7 used fluorescent probes to understand crystallization
by studying the behavior of solvent molecules in crystallizing systems. A solvent
sensitive probe, pyranine, was used to study changes in the environment of water
molecules in sugar solutions. The authors reported that water experiences two
microenvironments in sugar solutions. In one environment water molecules are involved
in solvating sugar molecules and in the other they are not. The authors found it possible
to calculate the peak intensity ratio (PIR) from the steady state emission spectra of trace
quantities of pyranine in the sugar solutions. This ratio reflects the number of water
molecules in each environment and this PIR is sensitive to the degree of supersaturation.
Their results, while informative, were not sensitive to and did not provide a description of
solute organization.

1.89 performed investigations of aqueous systems of glucose. In this

Rasimas et a
work the fluorescent probes carminic acid and glycosylated resorufin were used to

examine solute environments. Both probes possess a pendant glucose moiety to allow for

-2-

incorpo
of the n
time-re:
rotation
concenl
rotutior
contain
several
carmini

glucose

mmmf
0f the
glycos§
study;
dlhami
FIOIOnz

IhCOrm

incorporation into aggregates of glucose molecules in solution. To gain an understanding
of the nature of the incorporation of carminic acid at the molecular level, steady state and
time-resolved spectroscopies were used. Measurements of fluorescence lifetimes and
rotational diffusion dynamics showed carminic acid to be sensitive to changes in glucose
concentration. This sensitivity was displayed in discontinuities of the lifetime and
rotational diffusion measurements near and at glucose saturation. These measurements
contained a wealth of information but were complicated due to the contributions of
several protonated forms of carminic acid. However, these findings indicated that
carminic acid could serve as a tool for understanding crystallization from solution in the
glucose system.

The glucose system was further studied by Rasimas et al. using glycosylated
resorufin as the probe. Glycosylated resorufin was chosen to study the effect the identity
of the probe molecule can have on the experimental results. Unlike carminic acid,
glycosylated resorufin has no labile protons. Three forms of the probe were used in this
study: glycosylated, protonated, and deprotonated resorufin. Rotational diffusion
dynamics were again utilized as a key means to understand crystallization. The
protonated species exhibited rotational diffusion dynamics that agreed well with
theoretical predictions due to bulk properties of the solutions studied, but the
deprotonated form exhibited slower rotational diffusion times than predicted from theory.
The slower rotational times of the deprotonated form highlighted the importance of
charge in describing the interactions of molecules in solution. Most importantly, the
reorientation times of glycosylated resorufin were shown to be independent of solution
viscosity at glucose concentrations near saturation. This was said to be evidence that the

glycosylated probe must reside in an environment that renders it insensitive to changes in

-3-

bulk sol

probes
l

environm

I}
trace amt
fluoresce
the solut:
response
Hmmx:
molecule
lnIOI'mm

distance,

by the

C1'I'Stalli
focuses
Study Clt
It is USE
It is em
"1031 C0
SOlUbilit
WOrk W

SOIUbill]

bulk solution viscosity near saturation. The authors concluded the non-glycosylated
probes were not involved in solute aggregation and that they sensed an aqueous
environment not associated with solution structuring.

This work by the Blanchard and Berglund demonstrated the effectiveness of using
trace amounts of fluorescent probes with a lock-and-key approach. This method uses a
fluorescent probe molecule that incorporates selectively into pre-crystalline aggregates of
the solute system via functionality in common with the crystallizing species. The optical
response of the probe is then monitored as a function of solution composition.
Fluorescence measurements are sensitive and rich in information about the immediate
molecular environment of the probe molecule. Fluorescent probe molecules may report
information on their immediate surroundings regarding concentration, pH, polarity,
distance, possible complexing, and electric field based on their optical response.10

The work reported in chapter 2 is a continuation of earlier experiments performed
by the Blanchard and Berglund groups to gain a mesoscopic understanding of
crystallization. While early experiments studied the crystallization of glucose, this work
focuses on the crystallization of dicarboxylic acids. In particular, adipic acid is under
study due to its economic importance and simplicity. Adipic acid has a variety of uses.
It is used in the manufacture of resins, plastics, urethane foams, and as a food additive'1
It is estimated that the annual global demand for adipic acid exceeds 1.9 x 109 kg with the
most common use being the production of nylon-6,6.'2 Adipic acid has a relatively low
solubility in water (1.9 g per 100 g water at 20° C).2 This serves in contrast to earlier

work with glucose where solubility is high (90 g per 100 g water at 20° C).2 In low

solubility systems, solution properties such as viscosity, density, and refractive index will

notbea

IFddlIlOl

behasio
(7AT) l
betseer
(7A1). '
concent
fluoresc
aggmga
chrome}
reorient.
thatane

l
with aql
biOph)S
Intermo‘
”IDIOPI
Includin
biologiC
as a bio.
Cell me:

Physioto

not be altered to a great extent, and therefore the ability to monitor solution saturation by
traditional means is difficult.

We employ this lock-and-key approach to study the pre-crystalline aggregation
behavior of adipic acid using the chromophores 7-azaindole (7A1) and 7-azatryptophan
(7AT) (fig. 2.1) as probes. The use of these two probes allows direct comparison
between a probe expected be included (7AT) in aggregation events and one that is not
(7A1). The steady state response of these probes was found to be sensitive to adipic acid
concentration due to a labile proton on the chromophore ring at position N7. Although
fluorescence lifetime measurements did not reveal direct information indicating solute
aggregation, opposite trends in lifetimes were evident revealing that the two
chromophores undergo very different interactions with adipic acid. The experimental
reorientation data showed evidence that adipic acid does form weak complexes with 7AT
that are not present with 7A1.

In chapter 3 we utilize 7AT incorporated into peptides to probe their interactions
with aqueous micelles. Protein fluorescence spectroscopy has proven to be a valuable
biophysical technique in the study of protein function, structure, dynamics and
intermolecular interactions in solution.”3 7AT has been used as an alternative to
tryptophan for peptide dynamics. Several factors render 7AT a useful biological probe

”"6 and has been shown to retain

including that it is amenable to protein synthesis
biological activity when incorporated into proteins. In addition to proteins, 7AT is useful
as a biological probe of peptides. The study of interactions between small peptides and

cell membranes is an advancing area of research. These interactions are important to

physiological functions like hormonal response and neurotransmission.l7 We covalently

attach 7

size affe

were ini
because

consists
charged
surfactzu
equilibn
Well-stut
PTOVide

interactit
bCIW‘een

IS mOde]

Charge 8
BCCEiUse
affOrdS I
Probes i
meaSUrei
llllCracm
Teonema

 

attach 7AT to a series of peptide oligomers though a peptide bond and study how peptide
size affects their ability to associate with micellar structures.

Micellar environments have been studied extensively for decades. These studies
were initially fueled by interest over the nature of the intramicellar microenvironment18
because they could serve as membrane mimetic systems. The simple model of a micelle
consists of a roughly spherical entity with hydrocarbon like core surrounded by a highly
charged layer (Stern layer) which contains the ionic headgroups of the individual

'9'20 This assembly is in dynamic

surfactant molecules, counterions, and water.
equilibrium with monomeric surfactant molecules in solution. Aqueous micelles are
well-studied organized molecular aggregates and we employ ionic and neutral micelles to
provide known and different heterogeneous environments in which to examine the
interactions of the 7AT tagged oligomers. The reorientation data point to a balance
between ionic and dispersion interactions of the peptides with micelles and this behavior
is modeled as a hindered rotor.

In chapter 4 we use several 7AT and tryptophan derivatives to study the role of
charge and substituent structure on the dynamics of these probes in micelle solutions.
Because 7AT is considered a biological probe, as an alternative to tryptophan, this study
affords us the opportunity to investigate the nature in which these structurally similar
probes interact with their surroundings. The steady state and fluorescence lifetime
measurements indicate partitioning of probes into micelles is highly dependent on ionic
interactions and governed by structural features in common with the micelles. The

reorientation data give evidence of subtle differences between interactions with micelles

for analogous 7AT and tryptophan derivatives.

pol)mer
fluoresc
examine
associat
molecul
environr

the resp<

 

Chapter 5 summarizes the findings in chapters 2-4 and proposes the use of
polymerized micelle assemblies as a medium in which to study the factors dictating
fluorescent probes interactions with their environment. These molecular micelles can be
examined with covalently attached intramolecular probes and non-bound freely
associating probes. These molecular micelles may serve a role in tailoring site-specific
molecular probes. These studies could provide a wealth of information regarding the
environment of polymerized micelles and could be useful systems in which to compare to

the response of probes in potentially larger and more highly ordered dendrimers.

1.1 Lit

Put

Rx)

Mu
3. Mat
4. Mat
5. Mai

81.

7. Pan

9- Ras
10. Tea
ll. The
13- Dra
13. Bee
14- Riel

P120
15- Ste

Cha

Am.

 

1.1 Literature Cited

1.

10.

ll.

12.

l3.

14.

15.

16.

Lakowicz, J. R. Principles of Fluorescence Spectroscopy; second ed.; Plenum
Publishers: New York, 1999.

Mullin, J. W. Crystallization; Butterworth-Heinimann: Oxford, 1993.

Mathlouthi, M.; Luu, D. V. Carbohydrate Res. 1980, 78, 225.

Mathlouthi, M.; Luu, D. V. Carbohydrate Res. 1980, 78, 203.

Mathlouthi, M.; Luu, C.; Meffrov-Biget, M.; Luu, D. V. Carbohydrate Res. 1980,
81, 213.

Chakraborty, R.; Berglund, K. A. J. Cryst. Growth 1992, 125, 81.

Pan, B.; Chakraborty, R.; Berglund, K. A. J. Cryst. Growth 1993, I30, 587.

Rasimas, J. P.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 11333.

Rasimas, J. P.; Berglund, K. A.; Blanchard, G. J. J. Phys. Chem. 1996, 100, 7220.
Teale, F. W.; Constable, D. Fluorescent Probes; Academic Press: New York, 1981.
The Merck Index; Merck & Co. Inc.: Rahway, 1989.

Draths, K. M.; Frost, J. W. J. Am. Chem. Soc. 1994, 116, 399.

Beechem, J. M.; Brand, L. M. Ann. Rev. Biochem. 1985, 54, 43.

Rich, R. L.; Negrerie, M.; Li, J.; Elliott, S.; Thronburg, R. W.; Petrich, J. W.
Photochem. Photobiol. 1993, 58, 28-30.

Steward, L. E.; Collins, C. 8.; Gilmore, M. A.; Carlson, J. E.; Ross, J. B. A.;
Chamberlin, A. R. J. Am. Chem. Soc. 1997, 119, 6-11.

Negrerie, M.; Bellefeuille, S. M.; Whitham, S.; Petrich, J. W.; Thomburg, R. W. J.

Am. Chem. Soc. 1990, 112, 7419-7421.

17. S;
18. Fe
19. T2

30. Ct

 

 

l7. Sargent, D. F.; Bean, J. W.; Schwyzer, R. Biophys. Chem. 1988, 31, 183.
18. Fendler, E. J .; Fendler, J. H. Adv. Phys. Org. Chem. 1970, 8, 271.
19. Tanford, C. J. Phys. Chem. 1972, 76, 3020.

20. Cationic Surfactants; Marcel Dekker, Inc: New York, 1991; Vol. 37.

Aza‘

(7AT)
hate 5
“diff
Thea:
3CKISR
menis
buffer
COHCen-
heist-0C
near an
7ATar.
hmeof

rcollent.

Chapter 2

Lifetime and Reorientation Measurements of 7-Azaindole and 7-
Azatryptophan in Aqueous Adipic Acid Solutions. The Significance of
Pendant Functionalities in Solution Phase Association Processes

Summary

In this chapter we report on the ability of the chromophores 7-azatryptophan
(7AT) and 7-azaindole (7A1) to sense aggregation of adipic acid in aqueous solution. We
have studied the fluorescence lifetime and reorientation dynamics of 7AT and 7A1 in
water and aqueous adipic acid solutions from subsaturation through supersaturation.
These probe molecules, differing by the presence (7AT) or absence (7A1) of an amino
acid side group, each possess a labile proton on the chromophore ring system that renders
them sensitive to changes in pH, and both chromophores are quenched collisionally in
buffer solution. The dependence of the 7A1 fluorescence lifetime on adipic acid
concentration differs fundamentally from that of 7AT, indicating complexation of the 7A1
heterocyclic nitrogen(s) with adipic acid. 7AT exhibits an increase in reorientation time
near and above adipic acid saturation concentration and this trend is absent for 7A1. The
7AT amino acid side group interacts with adipic acid with a characteristic persistence
time of < 50 ps. For 7A1, the interaction with adipic acid is too short-lived to be seen by

reorientation measurements.

-10-

2.1 Int

many ll
punfica
crystall'
the resu
molecul
crystalli

can and

aSSOciat
Crystalli
influenc
CTl'stalli
matter t
intermoi
intennol

SOlUIlOn

CTIICn' a.
moleCul

meane

 

2.1 Introduction

The process of crystallization from solution is of vital economic importance to
many industrial processes. Crystallization is the method of choice for separation and
purification in the chemical, pharmaceutical, and food industries. Controlling
crystallization from solution has a direct effect on the physical properties and purity of
the resulting product. To control the crystallization process, we must first understand the
molecular interactions between solution constituents that can ultimately lead to
crystallization. The goal of this work is to understand the role of probe molecules, which
can and do act as impurities, in mediating liquid phase self-assembly.

Crystallization from solution is a complex, dynamic process. The problems
associated with studies of crystallization include determining the structure of pre-
crystalline aggregates in supersaturated solutions, the nature of crystal nuclei, the
influence of impurities, and the relationship between the capture of impurities and the
crystallization conditions. The incorporation of impurities is perhaps the most significant
matter because of the consequences and the dependence of this process on the details of
intermolecular interactions in solution. For this reason we have focused on the
intermolecular interactions between (fluorescent) impurity molecules and species in
solution capable of crystallization.

The choice of the fluorescent probe molecules for this work is based on two
criteria. First, we are interested in working with biologically benign and relevant
molecules in aqueous environments because of the potential utility of such molecules to

the investigation of protein and peptide structure. Second, using probe molecules with

-11-

well-be
data. 1
amino :
of trypt
due to
[Mom
behatic

understr

(MT), 3

Slhthet}

7‘52an

 

 

well-behaved photophysical properties allows for easier interpretation of spectroscopic
data. For these reasons, tryptophan is one obvious choice. Tryptophan is an essential
amino acid with a well-characterized fluorescence response. The fluorescence properties
of tryptophan along with its chromophore moiety, indole, have been studied extensively
due to its use as a standard optical probe of protein structure and dynamics."3
Unfortunately, tryptophan is characterized by non-exponential fluorescence relaxation

behavior and, without the excited state properties of the chromophore being well

understood, interpreting transient fluorescence data in complex systems is problematic.M

 

7-azaindole 7-azatryptophan

O

HO
OH

adipic acid

Figure 2.1 Structures of the chromophores, 7-azaindole (7A1) and 7-azatryptophan
(7AT), and solute, adipic acid.

One consequence of this spectroscopic complication is the development of the
synthetic amino acid 7-azatryptophan (7AT) and its corresponding chromophore moiety

7-azaindole (7A1) as biological probes (Fig. 2.1).7'lo 7AT has been used as an alternative

-12-

to tryptophan in the study of protein structure and dynamicsg'”l2 because of its well-
behaved, single exponential fluorescence decay over a wide pH ranges“ The 7A1
chromophore is a prototypical model for excited state proton transfer, a fundamental
reaction in many chemical and biochemical processes. The dimer of the 7A1
chromophore is structurally similar to hydrogen-bonded DNA base-pairs and is regarded
as a model for the study of photoinduced mutagenesis.'3'l5

We use these two chromophores to study the pre-crystalline aggregation behavior
of aqueous adipic acid. This method is, in effect, a direct examination of impurity
incorporation because it uses trace amounts of a chromophore either with (7AT) or
without (7A1) a functionality in common with the crystallizing matrix. The time- and
frequency-domain optical responses of the probe are monitored as a function of solution
composition and these data reveal the details of the intermolecular interactions at work in
these systems. The adipic acid system has been studied previously using the optical
probes pyrene and pyrene carboxylic acid.16 Based on the results of that work, we expect
7AT to incorporate into adipic acid pro-crystalline aggregates through its carboxylic acid
functionality. In contrast, we expect 7A1 to provide the complementary, non-
incorporated response. Comparing the results we report here to those for the pyrene-
based probes in adipic acid solutions provides information not only on the role of

different probes but also allows comparison of amino acids to carboxylic acids for their

action as impurities in the crystallization of carboxylic acids.

-13-

2.2 E:

Sigma
obtain:
dipotas
Produc
Solutio

CODCCD'

maintai

Wfih a
Spectror
4500 or

using 2

the SP6
and 7A‘

With a t

 

3"? gent
100 p8

"m pui
Doublir

 

2.2 Experimental Section

Chemicals. 7-Azaindole (99+%) and D,L-7-azatryptophan were obtained from
Sigma Chemical Company and used without further purification. Adipic acid was
obtained from Aldrich (99%) and used as received. The buffer solutions were made from
dipotassium hydrogen phosphate and citric acid purchased from Spectrum Quality
Products, Inc. All solutions were prepared with HPLC grade water from Aldrich.
Solutions used for lifetime and reorientation measurements were 5 M in chromophore

concentration. The sample cuvette was temperature-controlled with a brass block holder
maintained at 20.0 :0.1 °C (Neslab RTE-221).

Steady-State Absorption and Emission Spectra. Absorption spectra were acquired
with a 2 nm band pass using an ATI-Unicam double beam UV-visible absorption
spectrometer. Emission spectra were recorded with one of two fluorimeters: a Hitachi F-
4500 operating with 5 nm excitation and emission bandpass or a Spex 1681 Fluorolog
using 2 nm bandpass for excitation and emission.

Time-Correlated Single Photon Counting (TCSPC) Spectrometer. A schematic of
the spectrometer used for fluorescence lifetime and reorientation measurements of 7A1
and 7AT is shown in figure 2.2 The light pulses used to excite the sample are produced
with a cavity-dumped, synchronously pumped dye laser (Coherent 702-2). Light pulses
are generated by a CW mode locked Nd:YAG laser (Quantronix 416) producing pulses of
100 ps duration FWHM at 1064 nm with ~9 watts average power at 80 MHz. The 1064
nm pulse train is frequency doubled with Quantronix SHG assembly model 324.
Doubling is achieved by angle tuning a type II KTP crystal yielding pulses at 532 nm

with an average power of ~800 mW. The 532 nm light pulses are directed for excitation

-14-

of 3 ca\
cavity c
of dye

pulses \
pulses i
obsen‘e

with 9t

   
  

creating

 

delay 8
focuse
introdu.
filtering
haITnOr
Laser (
88mm
S l m'
light “
(Orr)

pseuo.
(Amer
dean]

Hama
and 0

Chann

of a cavity dumped synchronously pumped dye laser (Coherent 702-2). The dye laser is
cavity dumped and its output can be tuned from ~550-1000 nm depending on the choice
of dye and optics used. For these experiments, the cavity was dumped at 4 MHz and
pulses were generated at 580 nm using Rhodamine 6G (Eastman Kodak) dye. The output
pulses in this configuration are ~90 mW average power with pulses of 5 ps FWHM as
observed by autocorrelation. The output of the dye laser is divided by a beam splitter
with 90% of the output being sent for excitation of the sample and 10% of the output
creating a reference pulse that is sent into a fiber optic delay line allowing for temporal
delay between signal and reference. The portion of the beam sent for excitation is
focused into an angle tuned 1 cm KDP crystal to produce 290 nm light. The KTP crystal
introduces and astigmatism that is compensated by a plano-convex lens followed by
filtering through a colored glass filter to remove the fundamental from the second
harmonic light. The polarization is then rotated by a fused silica half-wave rhomb (CV1
Laser Corporation) to 0° (vertical). The polarized UV is focused into the sample with an
88mm focal length bi-convex fused silica lens resulting in average power at the sample of
S 1 mW. For all experiments emission is collected at 90° with respect to the incident
light with a reflecting microscope objective (Ealing x36/0.5) followed by a Gian-Taylor
(G-T) prism for selection of polarization. The selected polarization is then sent thought a
pseudo polarization scrambler and focused into a subtractive double monochromator
(American Holographic DBlO-S). The detection electronics have been described in
detail elsewhere,” and we review the salient features briefly here. For detection, a
Hamamatzu (R3809U-50) cooled dual stage microchannelplate photomultiplier is used
and operated at a bias of 3.1 kV. The signal produced from the detector is sent to one

channel of a four channel constant fraction discriminator (CFD, Tennelec TC455) for

-15-

process.
the TM
start. T
delay I:
terminull
whose

referen.
channei
helps trl

nmge

(Tektro

 

(RICA,
maxi m ;
Fluoreg
respeer
instmr
to~1l
Teone.
Anal)
timeS

1'10”“

processing prior to input to the time-to-amplitude converter (TAC, TC864). We operate
the TAC in “reverse mode”, as such the input signal from the MCP-PMT serves as the
start. The stop signal originates from the portion of laser pulse split into the fiber optic
delay line. This custom made optical delay consists of varying lengths of fiber
terminating in detection by an avalanche photodiode detector (PD, Hamamatsu 82381)
whose signal is routed into a second channel of the CFD. The output of the CFD
reference signal is electronically delayed (T ennelec TC412A) and sent to the TAC stop
channel. This configuration of start and stop channels is termed “reverse mode” and
helps to eliminates pulse pile-up and ensure operation of the TAC in a linear conversion
range. The valid conversions of the TAC are output and displayed on an oscilloscope
(T ektronix 2230) and rate meter (T ennelec TC525) and sent to a multichannel analyzer
(MCA, Tennelec PCA-II) for collection. Emission was monitored at the fluorescence
maxima of the chromophores using a 20 nm bandpass on the collection monochromator.
Fluorescence intensity decays were collected at polarizations of 0°, 54.7° and 90° with
respect to the (vertical) polarization of the incident excitation pulse. The typical
instrument response function is 35 ps and the chromophore lifetimes range from ~400 ps
to ~1 ns. No deconvolution of the response function from the data was required. For the
reorientation data, where TOR ~ 30-50 ps, deconvolution may be thought to be important.
Analysis of the experimental data, however, show deconvolution to produce reorientation
times that are the same to within the experimental uncertainty as those extracted from

non-deconvoluted data.

-l6-

       
   
 
     
    
 
  

‘ ' d ‘h . ‘
gpur l
pip in: 1.1;:
. - . ."x-rh -.
I y ‘- s
‘. I r'. '. 1.-
,

2o)

pulse
diagnostics

    

cavity dumped dye laser

fiber optic
switched

560 nm — 900 nm 0 delay

 

polarization
control I

temperature
controlled cuvette

  

reflective G-T monochromator

objective prism

Figure 2.2 Schematic of the time correlated single photon counter used to measure
fluorescence lifetimes and rotational diffusion dynamics.

-17-

2.3 Results and Discussion

The objective of the work we report here is to understand how each probe
molecule interacts with adipic acid in aqueous solution. We are most concerned with the
influence of these chromophores in the concentration region centered around adipic acid
saturation. The sensitivity of these probes to their environment depends on their
chemical structures and the identity of the solvent and saturable constituent. A
prerequisite to these measurements is the development of an understanding of the probe
molecule spectroscopic properties under comparatively well-controlled conditions. The
steady state linear response, fluorescence lifetime and reorientation time dependence on
adipic acid concentration for both 7A1 and 7AT provide the information we are interested
in. We consider the steady-state optical properties of these probes first.

Steady-State Spectroscopy of 7A1 and 7AT. The steady state spectroscopy of 7A1
and 7AT has been investigated previously and there is a substantial understanding of the
electronic structure of these molecules. 7"2 We use the information gained from this
literature to aid our understanding of the data we report here. The 7A1 chromophore is
characterized by the presence of several electronic excited states in close energetic
proximity, with the energetic spacing and transition strengths depending on the
protonation condition of the heterocyclic nitrogens. At chromophore concentrations well
above those we use in this work, there is some spectroscopic evidence for the formation
of H-bonded oligomers.18 For the experiments we report here, we do not see evidence for

H-bonded dimers that can lead to H+-exchange.

-13-

2“
O

.o
co

.9
ox

 

normalized linear response (a.u.)

 

 

 

 

wavelength (nm)

Figure 2.3 Linear absorption and emission spectra for the probes 7A1 (dotted trace) and
7AT (solid line) in aqueous solution. Intensities of the bands have been normalized for

presentation.

Linear response data provide information on the nature of the local environment
sensed by the probes. Steady state absorption and emission measurements were obtained
for 7A1 and 7AT in aqueous adipic acid solutions. The absorption and emission maxima
of aqueous 7A1 are 288 nm and 389 nm, respectively (Fig. 2.3). In adipic acid solutions,
absorption band shifts to the red by ~2 nm, and for adipic acid concentrations higher than
10% of saturation, the fluorescence maximum of 7A1 shifts to 441 nm (Fig. 2.4a). This
significant emission red shift is nearly complete by 5% of saturation and is thought to
arise from protonation of the azaindole chromophore at position N7, The ground state

pKa of 7A1 is 4.5'9 and the pH for the solutions of adipic acid we use are in the range of

-19-

that o

 

 

 

variet
amine
Mth t

that t.

 

reSped
a func
soluti
acid 5
$68 fr
adipi
max
deer
POsi
of ‘
(lift

Val

pH 3.80 at 1% of saturation to 2.68 at 110% of saturation. A species distribution plot for
7A1 (Fig. 2.5a) indicates that the dominant species interrogated spectroscopically is the
protonated form.

The sensitivity of 7AT to its local environment is similar, but more complex than
that of 7A1. The presence of the amino acid side group on 7AT gives rise to a wider
variety of (likely stronger) interactions with adipic acid than those seen for 7A1. The
amino acid side chain of 7AT can exist in cationic, zwitterionic, and anionic forms along
with the labile proton of the chromophore at N7. Potentiometric titrations of 7AT show
that the ground state pKa values for COzH, N7, and NH3+ are 2.70, 3.85 and 9.35,
respectively.20 With these pKa values, we calculated a species distribution plot of 7AT as
a function of pH (Fig 2.5b). Three species coexist in the pH range of the adipic acid
solutions used here. The experimental absorption maxima of 7AT aqueous and adipic
acid solutions were 290 nm and 292 nm, respectively, showing the same behavior as we
see for 7A1. The fluorescence maximum of aqueous 7AT is at 399 nm (Fig. 2.3) and in
adipic acid solutions greater than 20% of the saturation concentration, the fluorescence
maximum shifts to 450 nm (Fig. 2.4b). As with 7A1, the fluorescence spectral shift with
decreasing pH is the result of protonation of the azaindole chromophore at the N7
position. The emission spectral shift is more gradual for 7AT than it is for 7A1 because
of the lower N7 pKa of 7AT (3.85) compared to that for 7A1 (4.5). Because of this
difference, solutions of higher adipic acid concentration are required to reach the pH
values where the emission spectrum is dominated by the N7-H protonated species. For
7AT, there is also a change in the protonation of the carboxylic acid side chain in this pH

range. Because the amino acid side group is not conjugated with the azaindole

-20-

 

Fig‘
3de

Cont
acid

.o
4;

p
N

 

normalized fluorescence intensity

 

P
o

 

’—
O
I

(b)
0.8 -

1% 5% 100%
0.6 -

0.4 -

 

0.2 -

normalized fluorescence intensity

 

 

350 400 450 500 550

wavelength (nm)

Figure 2.4 Normalized emission spectra for (a) 7A1 and (b) 7AT as a function of
adipic acid concentration. Chromophore concentrations are 5 M for all solutions.
The emission spectra shift progressively to the red with increasing adipic acid
concentrations. Shown are spectra in water, 1%, 5%, and 100% of saturation adipic
acid solutions.

-21-

 

Fig,
:1ch
Valu
and

 

 

1.0 _—'\

 

0,3. HAI’

0.6 -

0.4 -

 

 

 

 

 

 

 

 

 

 

Figure 2.5 Calculated a-fractions for (a) 7A1 and (b) 7AT. The pH range of adipic
acid solutions (2.7 - 3.8) is boxed. The calculations were made using ground state pK.
values for each chromophore. For 7A1, pK = 4.5 and for 7AT, pK. = 2.70, pKz = 3.85
and pK3 = 9.35.

-22-

chrom

functtt

inforn
Due t
excite
local r'
organ
Intcra
terms
For 7.
the IN
SUITOI
this ill
HUOrfi
be as
Possil

the C
a‘fii l2
“0an

deL

 

chromophore, we do not expect the protonation status of the amino acid carboxylic acid
functionality to have a pronounced effect on the spectroscopic response of the probe.
Fluorescence Lifetimes of 7A] and 7AT. Lifetime measurements yield
information on solution composition and may be sensitive to solute local organization.
Due to the number of intermolecular and intramolecular factors that can influence the
excited state lifetime of a given chromophore, the dependence of fluorescence lifetime on
local environment remains in the realm of phenomenology for most systems. For many
organic molecules, the excited state lifetime depends on site-specific intermolecular
interactions, such as H-bonding or complexation, and this effect can be understood in
terms of the influence of the interaction on the electronic structure of the chromophore.
For 7A1, the most likely locations for site-specific intermolecular interactions to occur are
the NH and N7 functionalities.18 The 7A1 N l-H moiety will interact with its
surroundings through hydrogen bonding and there is some evidence that the strength of

this interaction is related to the chromophore fluorescence lifetime.'8 The relatively short
fluorescence lifetime (In) of ~ 900 ps for 7A1 in water at room temperature is thought to

be associated with internal conversion mediated by the Nl-H stretching mode or,
possibly, dissociation of the N -H bond.'9'2'

Because of the labile and environmentally sensitive nature of the N-H bond(s) in
the chromophore, 7A1 and 7AT are relatively sensitive to the proton content and
availability of their immediate surroundings. The impact of the Nl-H functionality on

nonradiative decay channels is illustrated by methylation of N7. When N. is methylated,

modulation of the nonradiative pathways accessed by specific interactions at the N 7 site is

-23-

 

tuock

7AT l

conce
aquer
1163i
flLlOIL
lfighe
eithef
acki
fluort
chané
thatt

Saturl

and,
Idetn
PFOU3
am]
OIher

that .

l

Unde

adipi

 

 

blocked, resulting in a longer radiative lifetime. Specifically, the fluorescence lifetime of
7AT is 800 ps and that of 1-methyl-7AT is 21 ns.'9

Lifetimes for 7A1 were measured in aqueous adipic acid solutions over the
concentration range of 0 to 110% of saturation of adipic acid. The lifetime of 7A1 in
aqueous solution is 912 i 3 ps and it increases in adipic acid solutions to ~ 1150 ps (Fig.
2.6a). The rise in lifetime follows the adipic acid concentration-dependence of the
fluorescence spectral shift; it is nearly complete at 1 % saturation and remains constant at
higher adipic acid concentrations. In principle, this lifetime change can be the result of
either protonation of the chromophore at the N7 position or complexation with adipic
acid. We will return to a discussion of this point below. Under careful examination, the
fluorescence lifetime of 7AI decreases slightly at higher adipic acid concentrations. This
change is seen below saturation and continues though saturation. There is no indication
that the fluorescence lifetime of 7A1 is sensitive to any changes in local environment near

saturation.
The single exponential fluorescence lifetime of 7AT, In, is 796 :i: 3 ps in water

and, in contrast to the behavior seen for 7A1, the addition of adipic acid reduces the
lifetime of 7AT (Fig. 2.6b). The change in lifetime of 7AT could also be the result of
protonation of the azaindole chromophore because it follows the concentration-
dependence of the steady state fluorescence spectral shift, or it could arise from some
other effect. The lifetime remains constant through saturation of adipic acid. The fact
that the lifetime decreases for 7AT in adipic acid solution while that of 7A1 increases
under the same conditions is a clear indication that the nature of their interactions with

adipic acid are different.

-24-

1200 r- (a)
1150 - . . r . I

L i li'll
1100 -

1050 r-

1000

950 -

Fluorescence Lifetime (ps)

 

llllllLL+kllllllLJlLlll|

0 10 20 3040506070 8090100110

 

800 - .
. (b)
750 ~

§

650~ ,

i

550 - I

Fluorescence Lifetime (ps)
LII
8

450 L ' - - ' ""'

4 l 1 I 1 I s l A l 1 l #1 n l A l l l 1 l J I

010 20 30 40 50 60 70 80 90100110
Adipic Acid Concentration (% saturation)

 

Figure 2.6 Fluorescence lifetimes of (a) 7A1 and (b) 7AT as a function of adipic acid
concentration. Where error bars are not visible, the uncertainty lies within the vertical

dimension of the symbol.

-25-

hare

scpar

huen

 

bufi‘c'
NICK
phos
7AJ
pilil
exhil
sohn

solut

 

foH

base

Pr0t
SOlL
COrt
fire
shC
lik

C0

In an effort to elucidate the specific role of adipic acid in these measurements, we
have also measured the lifetimes of 7A1 and 7AT in buffer solutions. It is important to
separate the adipic acid and W contributions to the lifetime dependence because we are
interested in determining the role of adipic acid interactions with the chromophores. The
buffer solutions spanned the same pH range as the adipic acid solutions. We used
McIlvaine’s buffer,22 a system comprised of citric acid and dipotassium hydrogen
phosphate such that the sum of the constituent analytical concentrations is 0.1 M. The
7A1 lifetime data are plotted for both the adipic acid the buffer solutions as a function of
pH in Fig. 2.7a. A striking feature of these data is the difference in the trends they
exhibit and the large differences between lifetimes for 7A1 in buffer and adipic acid
solutions. For 7AT, the data, shown in Fig. 2.7b, exhibit similar behavior for both
solutions, but with an offset in lifetime at lower pH. We understand these data as
follows.

The differences we observe for 7A1 in buffer and adipic acid solutions cannot be
based solely on pH. The direct interaction or complexation of adipic acid with 7A1 must
be responsible for the data shown in Fig. 2.6a. The dependence of the 7A1 lifetime on
protonation at the N7 position is manifested in the buffered solutions. The buffered
solutions exhibit the same absorption and emission spectra as the adipic acid solutions at
corresponding pH values, so the existence of a strong complex between adipic acid and
the chromophore heterocyclic nitrogens is not evident. Thus, the formation of a weak, or
short-lived complex between adipic acid and 7A1 based on H-bonding interactions is
likely responsible for the form of the data shown in Fig. 2.6a. This conclusion is

consistent with the reorientation data for 7A1, which we consider below.

-26-

1150 _ (a)

. I adipic acid
1100 - '

1
I

n
n
c
n

I

1050

.5

950 ' water

0

8
.j.

0

fluorescence lifetime (ps)

‘c
‘. buffer ”MO- ------------

850

r
.0"
0

 

 

D
8“) l A l A l a J l l A l

i water 0))

00
8
' I
r.

s
Q.‘
s
‘Q
‘-
Q
‘Q

“a
so

'I‘I
O

buffer

O\

LII

O
r

adipic acid

l
I
I
I
l
I
l
I
I
0’
I
I
I
I

9. P P
r ‘\ [e
450 ‘ 6
ago“

\
b .n‘
‘

LII

LII

O
I

 

fluorescence lifetime (ps)
LII
8 §

 

N
L»)
A
LII
O\
\I

Figure 2.7 Fluorescence lifetimes measured for aqueous solutions containing (a) 5

11M 7A1 and (b) 5 11M 7AT in adipic acid (I), buffered (o), and un-buffered (A),
aqueous solutions as a function of pH.

-27-

mirrt
deter
acid
chro
llll’OU

lifeti

The
DUTit

the r

SO-c.
com
of q-
flu0p
not l

and

 

 

For 7AT, the adipic acid-lifetime dependence and buffer solution pH-dependence
mirror one another, and these data show that the lifetime of this chromophore is
determined largely by protonation at the N7 position. If there is an interaction with adipic
acid for this chromophore, it is with the amino acid side chain and not at the
chromophore heterocyclic nitrogens. Again, this is an issue that can be addressed
through reorientation measurements (vide infra). The buffer solution pH-dependent
lifetime data for 7A1 and 7AT also mirror one another but are offset in pH (Figs. 2.7).
We understand this offset as resulting from the different N7 pKas for these molecules.
The data in Fig. 2.7b show a systematic offset in lifetime between the adipic acid and
buffer solutions for 7AT. This offset is the result of collisional quenching interactions in
the buffer solution.

There are two possible mechanisms for quenching in these systems. The first is
so-called static quenching because it is mediated by the formation of a ground state
complex, where only the noncomplexed species exhibits a radiative response. This type
of quenching is manifested by a decrease in fluorescence intensity but with no change in
fluorescence lifetime of the radiatively active chromophores. Our experimental data are
not consistent with this quenching mechanism. It has been shown previously that 7A1
and 7AT are quenched by K1 and acrylamide,23 and our buffer system is seen to alter the
fluorescence lifetime of the chromophores. The key question here is to what extent the
pH-dependent changes in lifetimes are due to [H] and to what extent the data can be
understood in terms of collisional quenching at a constant pH? We can address this issue
by measuring the chromophore lifetimes as a function of buffer concentration at a

constant pH.

-23-

que

“in
hfet
binr
solu
was
fit it
intn
and
phat
1.16
and

~10r

 

adip

Chro

With

alle

 

 

pH .
pH.

 

The fluorescence lifetime of a chromophore in the presence of a collisional

quencher is expected to behave according to the Stem-Volmer relationship,24 given by

Eq. 2.1.
1% = 1+ 1csv [Q] = 1+ quo[Q] (2.1)

where to is the lifetime of the chromophore in the absence of quencher, 1: is the observed
lifetime in the presence quencher, Ksv is the Stem-Volmer constant, and kq is the
bimolecular quenching rate constant. Stem-Volmer plots for 7A1 and 7AT in buffer
solutions are shown in Figs. 2.7. For these measurements, the total buffer concentration
was varied and the pH held constant at 3.2. The Stem-Volmer plot for 7A1 gives a linear
fit with the slope and intercept yielding the bimolecular collision rate constant and the
intrinsic fluorescence lifetime, respectively. These values, kq = (2.5i0.1) x 109 M'l s’l
and to = 1053 21:20 ps result in a calculated value of K... = 2.63 M". The Stem-Volmer
plot for 7AT yields a linear fit with kq = (2.33201) x 109 M'I s", to = 512 :6 ps and Ks. =
1.164 M". The extrapolated values for to are in good agreement with experiment for 7A1
and are approached closely for 7AT in adipic acid at pH 3.2, where tn ~ 1150 ps. The
~10% difference between the Stem-Volmer to value and the experimental In value in
adipic acid solution is likely due to the difference in dielectric response sensed by the
chromophore in the adipic acid and buffer solutions.

The Stem-Volmer plots indicate that the two probes are quenched collisionally
with similar efficiency in the citric acid/dipotassium hydrogen phosphate buffer system,
at least at pH 3.2. While the physics of the quenching process may be the same over the
pH range of interest here, there is more than one species present for each probe at this

pH, and it is possible that one form of each chromophore may be more susceptible to

-29-

1' (ps')

FlgL
citril
and

QUeJ

 

and
mu]
Whit
iSSui

qUEr

25071103 F

 

7AT
2.25x10’3c
"'3. 2007:1035:
Te 7A1
1.25x10‘3-
1.00xlO-31.1.1.1.1.1.1.1.1

 

0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22

buffer concentration (M)

Figure 2.8 Stem-Volmer plots for 7A1 (0, bottom trace) and 7AT (I, top trace) in
citric acid/dipotassium hydrogen phosphate buffer. For 7A1 k,,=(2.5:0.1) x 109 M'1 s’1
and for 7AT k,= (2.3101) x 10° M" s' .

quenching than another. 7A1 at pH 3.2 exists primarily (~95%) in its protonated form
and 7AT is ~80% protonated at the chromophore N7 site. In addition to the issue of
multiple probe molecule species being present, it is not clear from the data we report here
which of the buffer constituents is responsible for the quenching. Regardless of these
issues, recovery of the same value of kq for each chromophore suggests the same

quenching mechanism for each.

-30-

Reorientation of 7A1 and 7AT. The steady state and fluorescence lifetime
measurements provide important information regarding the local environment of the
probe molecules. These data are useful but cannot, by themselves, provide a complete
picture of adipic acid interactions with 7A1 and 7AT in solution. To better understand the
nature of the local organization present in aqueous adipic acid solutions, we have
examined the rotational diffusion dynamics of 7A1 and 7AT in those systems. Rotational
diffusion measurements can provide information about intermolecular interactions over
the length scales comparable to and greater than the dimensions of the probe
molecule?!"26 Comparing the reorientation behavior of 7A1 with that of 7AT provides
direct insight into the existence and time scale of intermolecular interactions between the
probe molecules and adipic acid in solution.

The theoretical framework for the interpretation of rotational diffusion data is

d,”27 and both neat and binary solvent systems have been modeled

well establishe
effectively. Reorientation measurements sense solvent-solute interactions over a time
scale in which a large number of molecular collisions occur, and the information
extracted from the data is relevant to the average environment experienced by the
reorienting moiety. For many binary systems, such as those reported in this work, the
molecular-scale interactions responsible for the motional behavior of the probes are not
modeled accurately using only the bulk solution properties and, in many cases, the results
are consistent with site-specific intermolecular interactions. For polar systems, dielectric
friction,28 dipolar interactions,29 and the formation of solvent-solute complexes30 can all
contribute to the reorientation behavior of a given probe molecule. We consider the

reorientation behavior of 7A1 and 7AT and the role that such interactions play in

determining the motional properties of these probe molecules.

-31-

pots
confl

oner

chm)
the;

to fi

obse

CXp(

“he
rota

Spex

re p (‘7

[he

 

CqU

 

dec.

 

To obtain reorientation data, we use our TCSPC system to excite the samples with
a vertically polarized light pulse and collect the time-resolved emission response at
polarizations parallel (1"(t)) and perpendicular (Ii(t)) to that of the excitation pulse. The
combination of these time-domain data according to Eq. 2.2 yields the induced

orientational anisotropy function, R(t).

I"(t)—11(t)

R(t) =
In“) + 2110)

(2.2)

 

The form of the experimental R(t) function is modeled to infer information on the
chromophore local environment. The form of R(t) depends on the optical properties of
the probe and its interactions with its local environment. In principle, R(t) can contain up
to five exponential decays, depending on the angle between the excited and emitting
transition dipole moments and the effective rotor shape of the probe.3| It is unusual to
observe more than two decays in R(t) and the most common case is to recover a single
exponential decay

R(t): R(0)exp(t/IOR) (2.3)

where Ton is the orientational relaxation time, related to the Cartesian components of the
rotational diffusion constant, and R(O) is the zero time anisotropy, related to the intrinsic
spectroscopic properties of the chromophore. For the reorientation measurements we
report here, R(t) decays as a single exponential, as indicated in Eq. 2.3.

Extracting chemical information from reorientation data relies on relating ton to
the molecular properties of the system. The modified Debye-Stokes—Einstein (DSE)
equation (Eq. 2.4) is the starting point for the extraction of chemical information from the

decay of R(t).”34

-32-

The
prot

accc

vise

 

IhEs

Can

phe

 

pTEL
11.. I
that

roltf

 

 

= "Vf (2.4)

The term 11 is the bulk viscosity of the medium, V is the hydrodynamic volume of the
probe molecule, k3 is the Boltzmann constant, T is temperature, S is a shape factor to
account for the non-spherical shape of the probe, and f is a friction coefficient. The bulk
viscosity can be measured directly, and the quantities S 33 and V 35 are determined by
models and are based on the molecular structure of the probe. The term f is, effectively,
the solvent-solute frictional coefficient and is estimated based on the shape of the probe.34
In the stick limit, f = l, and this quantity can vary from 0 to 1 in the slipping boundary
condition. The values of Ton predicted by the DSE equation can be compared to those
obtained experimentally as a means to gauge the strength of interaction between the
probe molecule and its immediate surroundings. It is important to note, however, that
there is substantial uncertainty in the determination of probe or system properties with the
DSE equation. Specifically, estimation of the hydrodynamic volume, V, solute shape, S,
and solvent-solute frictional term, f, all rest on model parameterizations, and the
relationship between bulk and local solvent viscosity remains an open issue. Despite
these limitations, comparison of experimental data to the predictions of the DSE model
can provide some insight into the nature of intermolecular interactions.

Despite the simplicity of this model and the intrinsic complexity of the
phenomena under consideration, the modified DSE equation provides a useful, qualitative
prediction of the reorientation behavior of small molecules in polar solvents. Based on
the experimental observation of a single exponential anisotropy decay and the assumption
that the dominant electronic transition is long-axis polarized,36 we model 7A1 as a prolate

rotor. For 7A1 we calculate V = 109 A3 using the method of van der Waals increments.”

-33-

 

 

 

Tozi
axis
ring

calc

1.1th

solu

 

also
their:
moi
thEr
(Fig.
Peon

”ILLk

 

To approximate the non-spherical shape of 7A1 in the context of a prolate rotor, the long-
axis and short-axis molecular dimensions of the optimized geometry in the plane of the
ring were used. The axial ratio was held constant and dimensions were scaled to give the
calculated hydrodynamic volume. Penin’s equation33 gives a shape factor S = 0.82 for a
prolate ellipsoid with a major axis of 8 A and a minor axis of 5 A. Using these
parameters and assuming a sticking limit boundary condition (f = 1), appropriate for polar
solvents and solutes, we calculate a reorientation time of 32 ps for 7A1 in water. This
value agrees well with the experimental data for 7A1 water and in aqueous adipic acid
solutions (Fig. 2.9). The reorientation time for 7A1 is constant to within experimental
uncertainty for all adipic acid solutions. From our previous work we know that the bulk
solution viscosity change over this range of adipic acid concentration is negligible.16 We
also know from that work that, near saturation concentration, there are oligomeric adipic
acid species present in solution. The reorientation data for 7A1 suggest that this probe
molecule does not associate strongly with the adipic acid. We do, however, know that
there is a detectable interaction between 7A1 and adipic acid as seen in the lifetime data
(Fig. 2.6a). The fact that we do not see any adipic acid concentration-dependence to the
reorientation data (Fig. 2.9) shows that the lifetime of the 7AI-adipic acid interaction is
much less than the reorientation time of 30 ps.

We present our reorientation data for 7AT in Fig. 2.9. The results for this probe
molecule are qualitatively different than those we recover for 7A1. Modeling the shape
of 7AT is not as straightforward as it is for 7A1. As for 7A1, the anisotropy decay is
observed experimentally to be single exponential and because of the similarity of the
chromophore portion of 7A1 and 7AT, we model 7AT as a prolate rotor with a van der

Waals volume of 179 A3.35 The effective rotor shape of 7AT will not necessarily

-34-

ll

1 (ps)

 

Figu
adipi

The
leng
di m,
Pro]
With

time

5, m HHI

Ton (p8)

3“} i i i sit

 

 

7A1
20 -
10 1 . 1 . 1 . 1 . 1 . 1
0 20 40 60 80 100

adipic acid concentration (% saturation)

Figure 2.9 Reorientation data for 7AT (I, t0p) and 7A1 (0, bottom) in aqueous
adipic acid solutions.

be approximated well by its optimized geometry due to the labile nature of its side chain.
The one dimension that will not change is the long axis of the chromophore. We use the
length of 8 A for this axis. The hydrodynamic volume of 7AT dictates a second axis
dimension of 6.5 A. With these values, the shape factor S is 0.94 and modeling 7AT as a
prolate rotor in the stick limit yields TOR = 47 ps. This calculated time constant falls
within the range of experimental reorientation times for 7AT in aqueous adipic acid
solutions (Fig. 2.9). The adipic acid-concentration dependence of the 7AT reorientation

times possesses a form that is discemibly different than that seen for 7A1. The

-35-

COIlC
cn\i

WC I'l

led t
are l
to it
obse
ascri
rang.
pK.
data
101' ]
inter

betu

that
COm

TeOn'

 

C0”-C
With

that

 

reorientation time increases from Ton = 43 :I: 3 ps at 80% of adipic acid saturation to tog
= 52 t 2 ps at 105% of saturation. This increase in reorientation time with adipic acid
concentration near saturation demonstrates that 7AT senses changes in its molecular
environment that are not reflected by the bulk solution. Comparing these results to those
we reported previously for a different complexing system is instructive.

Previous reorientation studies of l-pyrenecarboxylic acid in adipic acid solution
led to the conclusion that adipic acid oligomers formed in solution.‘6 Carboxylic acids
are known to form ring-bound dimers by hydrogen bonding.37 For this type of complex
to form, both carboxylic acid moieties must be protonated. The fact that we do not
observe the same behavior for 7AT as was seen for l-pyrenecarboxylic acid can be
ascribed to the fact that the 7AT amino acid side group is largely deprotonated in the pH
range we work in here. The pKa of l-pyrenecarboxylic acid is 4.038 in comparison to a
pKa of 2.70 for the carboxylic acid functionality of 7AT. The experimental reorientation
data do not exhibit a step-wise increase in reorientation time near saturation, as was seen
for pyrenecarboxylic acid. We conclude from this finding that the details of the
intermolecular interactions between 7AT and adipic acid are different than those seen
between 1-pyrenecarboxylic acid and adipic acid.

The continuous trend in the reorientation time of 7AT near saturation suggests
that the 7AT-adipic acid system does indeed form complexes but the lifetime of these
complexes is short relative to the reorientation time of the molecule. If we use the
reorientation time of 50 ps as an estimate, at room temperature this upper bound
corresponds to a complexation energy of ~ 3-5 kcal/mol, an energy entirely consistent
with hydrogen bonding. Because this effect is seen for 7AT and not for 7A1, we infer

that the interaction responsible for the behavior we observe is with the 7AT side group.

-36-

'Thei

zuit

pron
finer
shou
facto
an:r
quhe
Chan
cone
enus

rang

 

 

There can be a number of H-bonding interactions between adipic acid and the
zwitterionic amino acid group, and our experimental data do not allow resolution of this
matter.

It is also possible that, near saturation, the fraction of 7AT molecules that are fully
protonated increases and such a condition could lead to stronger solvent-solute
interactions, accounting for the reorientation data. The ground state or-fraction plot,
shown in Fig. 2.5b would seem to be consistent with this possibility, but there are two
factors that serve to make it unlikely. The first is that the excited state pK, values, which
are relevant to the measurements we report here, are not known but are almost certainly
quite different than those for the ground state of 7AT. The second factor that makes a
change in species distribution unable to account for these data is that, in the adipic acid
concentration range around saturation, there is very little change in the steady state
emission spectrum, suggesting no substantial variation in species distribution over the

range where the reorientation times of 7AT vary.

-37-

2.4 Conclusions

We have examined the steady state and time-resolved responses of 7A1 and 7AT
in adipic acid solutions over the subsaturation to supersaturation concentration range.
Our lifetime data indicate the existence of a weak complex between 7A1 and adipic acid.
For 7AT, the change in lifetime with adipic acid concentration can be accounted for in
terms of protonation at the N7 position. The reorientation data for 7AT reveal
interactions with adipic acid that have a characteristic persistence time somewhat less
than 50 ps. The interactions seen in the 7A1 lifetime data are too short-lived to be
detected with reorientation measurements. The 7AT amino acid side group interacts with
adipic acid in solution more strongly than do the chromophore heterocyclic nitrogens.
The choice of the probe molecule for studies involving association with short-lived
species can be made with reasonable certainty based on the chemical structures and
properties. This work also underscores the importance of probe molecule identity and
functionality when examining multicomponent, heterogeneous systems. When this work
is viewed in the context of impurity incorporation in crystallizing systems, it is clear that
the chemical identity of the impurity is of central importance, consistent with the well-
known structural selectivity of the crystallization process. With knowledge of the
relationship between chemical structure and role in mediating crystallization, judicious
choice of impurity could, in principle, be used to control crystal properties such as habit

and growth rate.

-33-

h)

10.

ll.

13.

 

2.5 Literature Cited

1

10.

ll.

12.

13.

Alexander, J.; Ross, B.; Rousslang, K. W.; Brand, L. Photochem. Photobiol. 1981,
20, 4361.

Chen, L. X. Q.; Petrich, J. W.; Fleming, G. R. Chem. Phys. Lett. 1987, 20, 55.
Cockle, S. A.; Szabo, A. G. Photochem. Photobiol. 1981, 34, 23-27.

Petrich, J. W.; Chang, M. C.; McDonald, D. B.; Fleming, G. R. J. Am. Chem. Soc.
1983, 105, 3824-3832.

Szabo, A. G.; Rayner, D. M. J. Am. Chem. Soc. 1980, 102, 554-563.

Robbins, R. J .; Fleming, G. R.; Bedded, G. 8.; Robinson, G. W.; Thistlethwaite, P.
J .; Woolfe, G. J. J. Am. Chem. Soc. 1980, 102, 6271-6279.

Negrerie, M.; Bellefeuille, S. M.; Whitham, 8.; Petrich, J. W.; Thomburg, R. W. J.
Am. Chem. Soc. 1990, 112, 7419-7421.

Rich, R. L.; Negrerie, M.; Li, J.; Elliott, S.; Thronburg, R. W.; Petrich, J. W.
Photochem. Photobiol. 1993, 58, 28-30.

Rich, R. L.; Smimov, A. V.; Schwabacher, A. W.; Petrich, J. W. J. Am. Chem. Soc.
1995, 117, 11850-11853.

Rich, R. L.; Chen, Y.; Neven, D.; Negrerie, M.; Gai, F.; Petrich, J. W. J. Phys.
Chem. 1993, 97, 1781-1788.

Hogue, C. W.; Szabo, A. G. Biophys. Chem. 1993, 48, 159-169.

Brennan, J. D.; Clark, I. D.; Hogue, C. W. V.; Ito, A. S.; Juliano, L.; Paiva, A. C. M.;
Rajendran, B.; Szablo, A. G. Appl. Spectrosc. 1995, 49, 51-59.

Guallar, G.; Batista, V.; Miller, W. J. Phys. Chem. 1999, 110, 9922.

-39-

 

14.

15.

16.

17.

18.

 

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

Taylor, C. A.; El-Bayoumi, M. A.; Kasha, M. Proc. Natl. Acad. Sci. USA. 1969, 63,
253.

Fiebig, T.; Chachisvilis, M.; Manger, M.; Zewail, A. H.; Douhal, A.; Garcia-Ochoa,
1.; de La Hoz Ayuso, A. J. Phys. Chem. A 1999, 103, 7419-7431.

Tulock, J. J.; Blanchard, G. J. J. Phys. Chem. B 1998, 102, 7148-7155.

DeWitt, L.; Blanchard, G. J .; LeGoff, E.; Benz, M. E.; Liao, J. H.; Kanatzidis, M. G.
J. Am. Chem. Soc. 1993, 115, 12158-12164.

Bulska, H.; Grabowska, A.; Pakula, B.; Sepiol, J.; Waluk, J.; Wild, U. P. J. Lumin.
1984, 29, 65-81.

Chen, Y.; Rich, R. L.; Gai, F.; Petrich, J. W. J. Phys. Chem. 1993, 97, 1770-1780.
Chen, Y.; Gai, F.; Petrich, J. W. J. Phys. Chem. 1994, 98, 2203-2209.

Chon, P. T.; Martines, M. L.; Cooper, W. C.; Collins, S. T.; McMorrow, D. P.;
Kasha, M. J. Phys. Chem. 1992, 96, 5203.

Day, R. A., Jr.; Underwood, A. L. Quantitative Analysis, 6th ed.; Prentice Hall:
Englewood Cliffs, NJ, 1991.

Petrich, J. W.; Rich, R. L.; English, D. S. Photochem. Photobiol. 1998, 67, 76.

Ingle, J. D.; Crouch, S. R. Spectrochemical Analysis, Englewood Cliffs: NJ, 1988.
Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1994, 98, 9411-9416.

Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 7904-7912.

Gudgin-Templeton, E. F. K.-W., G. A. J. Phys. Chem. 1986, 2896.

Kivelson, D.; Spears, K. G. J. Phys. Chem. 1985, 89, 1999-2001.

Blanchard, G. J. J. of Phys. Chem. 1989, 93, 4315-4319.

Blanchard, G. J. Anal. Chem. 1989, 61, 2394-2398.

Chuang, T. J.; Eisenthal, K. B. J. Chem. Phys. 1972, 57, 5094-5097.

-40-

32.

33.

34.

35.

36.

37.

38.

Debye, P. Polar Molecules New York, 1929.

Perrin, F. J. Phys. Radium 1934, 4, 497.

Hu, C.-M.; Zwanzig, R. J. Chem. Phys. 1974, 60, 4354-4357.

Edward, J. T. J. Chem. Ed. 1970, 47, 261.

Ilich, P. J. of Molec. Struc. 1995, 354, 37-47.

Pauling, L.; Brockway, L. 0. Proc. Natl. Acad. Sci. USA. 1934, 20, 336.

Donckt, E.; Dramaix, J .; Nasielski, J.; Vogels, C. Trans. Faraday Soc. 1969, 65,

3258.

-41-

 

 

llOl

pol

 

anir
use
Witl
dyi
hea.
in [l
mC-C
dvn

d

the l

Chapter 3

Dynamics of 7 -Azatryptophan Derivatives in Micellar Media.
Elucidating the Interactions Between Peptide Oligomers and Micellss

Summary

We have investigated the motional and population relaxation dynamics of the
non-natural amino acid 7-azatryptophan (7AT) in free (zwitterionic) form and bonded to
polyvaline oligomers in aqueous micelles. We have studied the oligomers in solutions of
anionic, cationic, and neutral surfactants above their critical micelle concentrations. The
use of these peptide oligomers enables the study of charge and structure on interactions
with aqueous micelles. The 7AT fluorescence population decay and reorientation
dynamics were monitored in aqueous and micellar solutions as a function of micellar
head group charge and oligopeptide chain length. The lifetime data on 7AT are discussed
in terms of its local environment, and the data point to partitioning of probes into micelles
mediated by ionic and dispersion interactions with the micelles. The reorientation
dynamics are indicative of persistent 7AT-micelle interactions and can be understood in

the context of the hindered rotor model.

-42-

 

SD

21113

 

Our
P6P
obt:
dep

nan

que

Pro

CQI
Ext

SUC

3.1 Introduction

Understanding the dynamics of intermolecular interactions in solution has proven
to be a substantial challenge to the chemistry community over the past several decades.
Solution phase interactions are important because of their central role in mediating a
number of phenomena, ranging from organic synthesis to chemical separations and a
wide range of biologically important processes. We are interested in understanding the
interactions of selected chromophores with entities that exhibit some level of self-

organization in solution. "5

Such systems include species undergoing crystallization
(phase separation), liquid crystalline materials, and micelles, which are the focus of this
work. To interrogate these micelles we have synthesized polyvaline oligomers and have
attached the non-natural amino acid 7-azatryptophan (7AT) for use as a chromophore.
Our intent is to understand a simplistic mimic of a biomembrane and the interactions a
peptide oligomer will have with such a system. We use fluorescence spectroscopy to
obtain information on 7AT-tagged peptide oligomers in solution. Fluorescence
depolarization gives information on restricted rotational motions that proceed on
nanosecond and picosecond time scales in these systems.6

Tryptophan is a fluorescent amino acid that possesses a significant fluorescence
quantum yield and a strong absorbance near 280 nm.7‘8 Tryptophan is not an ideal optical
probe, however. It is characterized by a non—exponential fluorescence population decay
and, for many systems, the presence of multiple tryptophan residues can serve to
complicate the site-specificity of the information garnered from such experiments.

Extrinsic fluorescent probes such as dansyl have been used in the study of proteins but

such probes often attach in a non-specific manner and disrupt native structure. Site-

.43-

directed mutagenesis can help overcome these limitations by selective removal of
tryptophan residues and or replacement by a single fluorescent chromophore whose
spectral properties are distinguishable from all other intrinsic species.9 In recent years, an
effort has been made to develop and characterize the non-natural amino acid 7AT for use
as a biological probe.“H3 This chromophore has been used to study protein structure and
dynamics"“6 because its spectral response is shifted relative to that of tryptophan, its 81
population decay is characterized by a single exponential over a wide pH range, and the
time constant of the decay is sensitive to the immediate environment of the
chromophore.'°"6

A large amount of research has gone into understanding the structure and
dynamics of micelles because these labile structures serve as model systems for
biological membranes, as a medium in which chemical reactions and excitation transfer
studies can be performed and, due to their ability to solubilize proteins and other organic
compounds, micelles have found use in chemical separations. Many investigations of
micellar systems utilize the fluorescence of (presumably incorporated) dye molecules to
provide information on micellization. We report here on our study of the dynamics of
small peptide fragments composed of the fluorescent non-natural amino acid 7AT linked
via peptide bond to valine oligomers (Figure 3.1) in aqueous micelles formed by anionic,
cationic and neutral surfactants (Figure 3.2). Our experimental data suggest that the
polyvaline oligomer interacts significantly with the hydrophobic portion of the micelle,
that the 7AT chromophore protrudes from the micellar structure and its motional
dynamics are mediated by the identity and charge of the micelle constituent polar head

group. We interpret these results in the context of the hindered rotor model.

   

’1
N OCH3
O n
H3C CH3
7-azatryptophan (7AT) 7-azatryptophan-valine methyl ester AT-(Val).,

Figure 3.1 Structures of 7-azatryptophan (7AT) and 7-azatryptophan-
valine methyl ester (7AT-Vain), n = 1-5.

\/\/\/\/\/\/0\ //O +
//S\ -Na
0 0

sodium dodecyl sulfate (SDS)
Me Br.

4.
{-Me
Me

 

cetyltrimethylammonium bromide (CT AB)

.MA...

polyethylene glycol 400 dodecyl ether (Thesit)

 

Figure 3.2 Structure of surfactants used in this work. Top: anionic SDS;
center; cationic CT AB; bottom neutral Thesit®.

-45—

3.2 Experimental

All reagents were obtained from Sigma-Aldrich, Inc. and used as received. N-t-
boc-7-azatryptophan was prepared by the procedure described by Petrichl3 using D,L—7-
azatryptophan hydrate, BOC-ON [2-(tert-butoxycarbonyloxyimino)-2-
phenylacetonitrile], and triethylamine in water/dioxane solvent.

Synthesis of N—7—azatryptophan-(I -5) valine methyl ester peptides ( 7AT- Val 7 -
7AT-Vols). A series of five peptides containing one 7AT and multiple valine (Val)
residues was synthesized. In the final form, these peptides all had a free amine N-
terrninus and a methyl ester C-terrninus. All amino acid couplings were performed using
the water soluble carbodiimide, l-[3-(Dimethylamino)propyl]-3-ethylcarbodiirnide,
hydrochloride (EDCI)17"8 along with the coupling catalyst hydroxybenzotriazole (HOBt).
Throughout the peptide couplings t-boc or-amino protection/deprotection chemistry was
used.'9 Peptides were prepared by sequential condensations of N-t-boc-Valine with L-
Valine methyl ester, followed by addition of N-t-boc-7-azatryptophan to deprotected and
isolated valine peptides. A description of the coupling and deprotection procedure is as
follows: EDCI (1.92 g, 10 mmol) and triethylamine (TEA) (1.4 mL, 10 mmol) was
added to a stirred solution of N-t-boc-Valine (2.17 g, 10 mmol) in 20 mL N,N-
dimethylformamide (DMF) at 0° C. The solution was stirred for five minutes and HOBt
(1.35 g, 10 mmol) was added and stirred for an additional 30 minutes. In a separate flask
L-Valine methyl ester hydrochloride (1.67 g, 10 mmol) and TEA (1.4 mL, 8 mmol) was
added to 10 mL of DMF and stirred for five minutes, then filtered to remove
triethylamine hydrochloride salt. The filtrate was added to the N-t-boc-Valine vessel and

stirred for 1 hour at 0° C, then at room temperature until completion of the reaction as

determined by thin layer chromatography (TLC). The reaction was acidified with
aqueous 5% citric acid and a white precipitate formed. Enough ethyl acetate was added
to dissolve and extract the reaction product. The ethyl acetate fraction was washed with
citric acid solution, saturated NaCl, saturated NaHC03, and again with saturated NaCl
solutions, then dried over anhydrous N32804. Solvent was removed under reduced
pressure to yield product N-t-boc-Valine-Valine methyl ester (3.09 g, 66%).

Deprotection of t-boc-modified peptides. 2.5 g N-t-boc-valine-valine methyl ester
was added to a stirred mixture of 10 mL trifluoroacetic acid (TFA) and 10 mL of dry
dichloromethane (DCM). Thioanisole was added as a scavenger when deprotecting
peptides containing 7AT to protect the aromatic ring from attack by liberated
carbocations. The progress of the reaction was followed by TLC in CHCl3zMeOH (9:1
v/v). DCM was removed in vacuo and dry ether added to precipitate the trifluoroacetate
salt. The salt was collected by filtration and washed several times with dry ether. The
product was dried over P205/KOH.

Micelle Preparation. Sodium dodecyl sulfate (SDS), cetyltrimethylammonium
bromide (CTAB) (99%+), and Thesit®, (polyethylene glycol 400 dodecyl ether) were
obtained from Sigma-Aldrich, Inc., checked for fluorescent impurities, and used as
received.

Methods. All aqueous solutions were prepared with water from a Milli-Q water
filtration system. All solutions used for dynamical measurements were ~5 uM in 7AT.
Upon preparation, surfactant solutions were sonicated for five minutes, stored overnight
and measured the next day. The sample cuvette was temperature-controlled with a brass

block holder maintained at 20.0 21:01 °C (Neslab).

To ensure micelle formation, the concentrations of surfactants in this study were
several times their critical micelle concentration (CMC). Certain physical parameters and
solution properties of the surfactants and micelle solutions must be known for proper
interpretation of the data and to ensure that changes in the optical response of the probes
are not due to changes in bulk solution properties. The physical parameters of the
surfactant solutions used are given in Table 3.1. One of the relevant physical parameters
needed for interpretation of the data is the hydrodynamic radius (rh) of the micelles. For
ionic surfactants this value is the sum of the micelle core radius, the head group radius,
and two layers of water.2°’2' For ionic surfactants this method of determining n. has been
found to match values found from light scattering experiments. The structure of the non-
ionic surfactant Thesit® micelle is not as well ordered as CT AB and SDS. The
hydrophilic portion of Thesit® is made up of a polyethyleneglycol chain and not a bulky
head group. This structural condition produces a hydrated mantle resulting in a less well

defined surface aqueous region and n, is estimated at 35 A.

Table 3.1 Solution and physical parameters of micelles used in this work.

 

 

Surfactant“ Charge Viscosity pH Concentration r). 2M
(cP) (mM)/number (A) (us)

of CMC
CTAB + 1.00 :i: .01 6.2 2.5 / 0.92 25.7 17.6
SDS - 1.03 i .01 6.6 25.0/ 8.3 20.7 9.2
Thesit® neutral 0.99 4- .01 7.1 1.0 / 0.1 35 44.4

" CTAB = Cetyltrimethylammonium bromide; SDS = sodium dodecyl sulfate;
Thesit® = poly(ethylene glycol) 400 dodecyl ether.

-48-

The azaindole chromophore is sensitive to solution pH because it can be
protonated at the N. position. This sensitivity is manifested as a significant emission red
shift and decreased excited state lifetime below pH ~5.22 Thus the position of the 7AT
emission spectrum and the fluorescence lifetime are both sensitive to local environment.
Rotational diffusion time constant(s) are related to “viscosity” according to the Debye-

23'” The viscosities of surfactant

Stokes-Einstein equation and its modifications.
solutions were determined by bulk measurements using a viscometer and the pH was
determined using a pH meter, with all measurements being controlled at 20°C. The small
changes in bulk solution properties associated with the formation of the micelles will not
affect the optical response of the chromophores substantially. The viscosity changes
associated with the addition of amphiphiles to form micelles are negligible and the pH
changes do not occur in a range that affects the protonation of the chromophore or the
polyvaline peptides.

Steady-State Absorption and Emission Spectra. Absorption spectra were acquired
using Cary Model 300 double beam UV-visible absorption spectrometer. Absorption
data were collected with 2 nm resolution. Emission spectra were recorded with a Spex
Fluorolog 3 spectrometer, with a 5 nm bandpass for excitation and a 2 nm bandpass for
emission collection.

Time-Correlated Single Photon Counting (TCSPC) Spectrometer. The
spectrometer we used for the lifetime and dynamical measurements in this work was
described in detail in chapter 2. The light pulses used to excite the sample are produced
with a synchronously pumped, cavity-dumped dye laser (Coherent 702-2). The dye laser

is excited by the second harmonic of the output of a continuous wave (CW) mode-locked

-49-

Nd:YAG laser (Quantronix 416). Samples were excited with 290 nm light (Rhodamine
6G dye, Kodak, KDP Type I SHG). Emission was monitored at the florescence
maximum of 7AT in the solution of interest using a 20 nm collection bandpass.
Fluorescence was collected at 0°, 54.7°, and 90° with respect to the polarization of the
incident pulse. The instrument response function for this system is typically 35 ps
FWHM and lifetimes measured range from ~600 ps to ~12 ns. We did not need to
deconvolute the instrument response function from the experimental data. The shortest
reorientation times we measure are on the order of 50 ps, and because we can recover the
entire anisotropy decay, some loss of the initial portion does not affect the accuracy of
our determinations. Because of the size of the chromophores used (vide infra), it is not

realistic to expect any anisotropy components with time constants less than ~ 50 ps.

-50-

3.3 Results and Discussion

The objectives of these experiments are to understand the optical response of a
chromophore incorporated into a peptide fragment as a function of chain length and
determine how these peptides interact with aqueous micelles. The peptide chromophore
senses and reports on changes in its local environment. In this study changes of
environment are dictated by intramolecular interactions of 7AT with associated valine
residues and by specific interactions of the peptide oligomers with surfactant molecules
and micellar structures. The optical response of these probes will be used as a means to
detect and understand factors that lead to partitioning and surfactant association of probes
in micellar environments. We consider the steady state and time-resolved spectroscopic
data separately.

Steady-State Spectroscopy of 7-azatryptophan and 7AT-(Val)7.5 methyl ester. The
absorbance spectrum of 7AT in aqueous solution is centered at 290 nm and the emission
maximum is 399 nm in both pure water and in surfactant solutions (Figure 3.3). When
7AT is linked covalently though its carboxyl functionality by formation of a peptide bond
with valine, the absorbance spectrum remains unchanged. The emission maxima of the
7AT-(Val)1-5 methyl ester series exhibit a small blue shift of 2 — 3 nm in pure water as a
function of number of valine residues (Figure 3.4). 7AT is known to be sensitive to its
immediate environment which, in this case, is the primary and secondary structure of the
peptide to which it is bound. The incorporation of 7AT into peptide fragments is known
to cause emission red-shifts; the 7AT emission band is centered at 414 nm in Nae-Pro-

7AT-Asn-NH2, 11 402 nm for Lys-D-7AT-Lys and 399 nm for Lys-L—7AT-Lys” in the

-51-

emission intensity (a.u.)

Thesit
CTAB

SIB

11.41111. W
300 350 400 450 500 550

wavelength (nm)

 

 

Figure 3.3 Fluorescence spectra of 7AT in water and surfactant solutions. Spectra
have been normalized and are offset for clarity of presentation.

-52-

7AT-Va15

7AT-Var4

emission intensity (a.u.)

7AT-Var3
7AT-Var2

7AT-Val]

 

 

71XT‘

l L l g L A l

300 350 400 450 500 550

wavelength (nm)

Figure 3.4 Fluorescence spectra of 7AT and 7AT-valn oligopeptides in water. Spectra
have been normalized and offset for clarity of presentation.

-53-

pH range of ~5 — 8, where the azaindole chromophore is not protonated. From these
studies and the information we report here, it is clear that the primary structure of a
peptide can have a substantial effect on the 7AT emission response and that different
enantiomeric forms of a single peptide can cause small shifts in the 7AT emission
spectrum. It is not surprising that the 7AT-(Val)n peptides differing only in the number
of valine residues exhibit small spectral shifts despite the fact that the chromophore is
adjacent to the same residue in all cases. The emission spectra for the peptide series in
the cationic surfactant CT AB solution and the neutral surfactant Thesit® solution are
identical to those in water. We see modest spectral shifts for 7AT in (anionic) SDS
solution, but in SDS the 7AT-(Val).-5 oligomers show a larger blue shift of ~13 nm.
Emission shifts in organic dye molecules in surfactant solutions are common and are
attributed typically to the dye partitioning into micelles. For example, spectral blue shifts
of various indoles have been attributed to incorporation into the micellar core of Brij-35
at concentrations far above the CMC. 27 The steady state absorption spectra are thus
consistent with our expectations and provide some level of insight into 7AT-micelle
interactions.

Fluorescence Lifetimes of 7 -azatryptophan and 7A T-( Val) 7 . 5 methyl ester.
Excited state lifetime measurements can provide information on solution composition and
may also contain information on the location of probe relative to the micelle. The
fluorescence lifetime of 7AT is sensitive to local environment. 7AT is zwitterionic at pH
7 and its emission decays as a single exponential with a time constant of ~800 ps. As

noted above, 7AT has been incorporated into peptide fragments and oligomers and

-54-

generally decays as a single exponential in a nominally uniform environment, with the
exception of the octapeptide as reported by Petrich. 28

In pure water and in micellar solutions, the emission intensity decays of 7AT and

all of the AT-Valn oligomers are best fit to single exponentials and these decays time

 

 

12CK)-
o
D
1100 - g
D
U
.flx 1CKX)- ' i' .
é a 9
. g '
i
' SKID - .9
£3 a
0 ' T c
g d
g 800
8
5
a:
71K) -
.s
.5
L
(1]) - ‘i “
l J J I l
() l 2 3 4. 5
number of valine residues

Figure 3.5 Fluorescence lifetimes of 7AT and 7AT-Valn (n = 1-5) oligomers in water(D),
Thesit (0), CTAB (V), and SDS (A). For points where error bars are not obvious, the
uncertainty lies within the vertical dimension of the symbol.

-55-

constants are shown in Figure 3.5. Goodness of fit was determined by x2 criterion and
visual inspection of residuals of the fit. Single exponential lifetimes indicate that one
fluorescent species is present in a single environment. Upon inspection of the lifetime
data, it is obvious that the lifetime of zwitterionic 7AT (0 valine residues) depends little
on the presence of or proximity to micelles. This finding would seem to indicate that
7AT experiences a similar environment in water and micellar solutions and that it does
not interact appreciably with surfactant molecules or micelles. When comparing
lifetimes of the peptide oligomers in surfactant solutions to those in water, large changes
are evident. The largest differences are noted in solutions of SDS. The 7AT peptide
oligomer lifetimes decrease significantly in SDS solution.

To illustrate this sensitivity, lifetimes were measured for 7AT-Val) in solutions of
varying SDS concentration (Figure 3.6). The observed decreases in lifetime with
increasing SDS concentration are not the result of simple quenching, nor does the
decrease set in only above the CMC. These data show a ~30% decrease in lifetime to
~760 ps at a surfactant concentration of 0.5 mM SDS, followed by a constant lifetime
until the CMC is exceeded. Above the CMC, the fluorescence lifetime of 7AT-Val.
decreases again in a stepwise manner to ~600 ps. For other micellar solutions, slight
oscillations in observed lifetimes were recovered that are analogous to steady-state shifts
in emission.

To better understand the nature of interactions between the probe and micelles,
we have investigated the rotational diffusion dynamics of our probes in aqueous solutions
of cationic, anionic and neutral micelle-forming surfactants. Rotational diffusion

measurements can provide information about intermolecular interactions over the length

-56-

1200 -

 

 

 

 

1100 - I »
E A 1100. .
I- : g l
E 1000- E E 1000;
8 900 - : 800-
§ ‘ ’ '
8 3 70° 0.00001‘o.02‘o.03‘0.04‘0.05‘
.- 800 - ' -
g 3 SDS concentration (mM)
IE _ I! I E
a" 3
>. 700- 5
Ed 3
< ' g
l‘ a
600 r 3 I I I I
t 3
11....1.IJ:.1..A.1....1....J
0 5 10 15 20 25

SDS concentration (mM)

Figure 3.6. Dependence of 7AT-Val) fluorescence lifetime on SDS concentration. The
vertical bar indicates the cmc for this system (8.3 mM). Inset: Dependence of lifetime
on low SDS concentration (first three points in the figure).

scale of the probe molecule. ”'30 Reorientation data can distinguish in an unambiguous
manner whether or not probes are experiencing interactions with micelles. Comparing
the dynamical response of 7AT and the peptide oligomers in pure water and surfactant
solutions will provide direct information on probe-surfactant association.

There is a large body of literature on the rotational diffusion of organic

30.31

chromophores in single solvent and binary systems. Micellar systems represent a

-57-

special subset of binary systems and the issue of reorientation in a micellar environment
has been examined before. 32'3“ Rotational diffusion has found wide use as a means of
probing intermolecular interactions and transient solution-phase organization because the
theory is relatively well established and the relevant molecular interactions proceed on a
time scale that can be accessed experimentally. The interactions responsible for
determining the reorientation dynamics of a chromophore in solution occur rapidly and,
over the course of the chromophore’s reorientation, many such intermolecular
interactions contribute to the overall experimental observation. A common conclusion of
these measurements made in binary solutions is that the molecular-scale interactions
responsible for the dynamical behavior of the probes are not dictated by the bulk
properties of the solvent. For polar systems the relevant intermolecular interactions may
include dielectric friction, 35 dipolar interactions, and under certain circumstances,
formation of solvent-solute complexes. We consider our data on the reorientation of
7AT-Val oligomers in micellar environments. Our measurements point to the strong
interaction between the oligopeptide(s) and the micelles and we interpret our
reorientation data in the context of a hindered rotor.

We excite the ensemble of 7AT chromophores in solution with a (vertically)
polarized light pulse and monitor the time evolution of emission intensity polarized
parallel and perpendicular to the excitation polarization. We combine these data to

generate the induced orientational anisotropy function, R(t),

I"(t)-Il(t)
1,,(r)+ 21,0)

 

R(t) = (3.1)

where I"(t) and 11(t) are the time-dependent emission intensities. The function R(t)

represents the re-randomization of the ensemble of excited molecules and the central

-53-

issue in such measurements is the application of models and theories to relate the
functionality of the R(t) decay to the immediate environment of the chromophore. In
principle, R(t) can contain up to five exponential decays, depending on the angle between
the transition moments and the shape of the volume swept out by the rotating
chromophore. 36 In homogeneous environments, it is unusual to observe more than two
decays and the most common case is a single exponential decay. R(t) decays as a single
exponential for 7AT and the 7AT-Val.-5 methyl ester peptide series in water. The
experimental anisotropy functions for the same chromophore containing peptides in

micellar solutions decay with two exponential components,

R(t) = R,(0)exp('%l)+R2(0)exp(‘%2) (3.2)

It is significant that we observe different functionalities for the anisotropy decay in neat
solvent (water) and in micellar solutions. This finding points to the incorporation of the
probe(s) in micellar environments and we thus need to consider two different models to
account for our experimental findings.

For the case of reorientation in water, a one component decay is consistent with
our earlier findings22 and we can interpret these data in the context of the modified

Debye-Stokes-Einstein (DSE) model.

R(t) = R(0)exp(—t/1:0R)
(3.3)

I = M
0" kBTS

where tog is the reorientation time constant, 17 is the solution bulk viscosity, V is the

hydrodynamic volume37 of the reorienting moiety, f is a friction coefficient to account for

the solvent-solute boundary condition and S accounts for the non-spherical shape of the

-59-

volume swept out by the reorienting molecule. The experimental data show a linear
dependence of the reorientation time on the number of valine residues (Figure 3.7),
consistent with incremental increases in the hydrodynamic volume of the reorienting
moiety. A linear fit of the reorientation times of the positively charged 7AT-(Val)|-5
methyl ester series yields a slope of 32.5 :1: 1.2 ps/valine. This result is consistent with
the predictions of Eq. 3.3 in the stick limit, where the volume of a valine methyl ester
moiety is taken to be 123 A3 (S = l), yielding a predicted slope of 30 ps/valine.37
Although these results follow the DSE model, this is not necessarily an expected result.
Proteins are known to display segmental motion caused by fast local reorientation of a
fluorescent residue and a longer-time component associated with the motion of the
peptide as a whole. 3840 For the short oligomers we use in this work, segmental motion
is not necessarily expected, but it is interesting to note that the data fit well to a single
exponential decay.

We now consider the reorientation of the peptide oligomers in micellar solutions.
The first issue to address is whether or not the chromophore is incorporating in some
fashion into the micelle. We assert that the change in functionality of R(t) upon micelle
formation indicates incorporation. In cases where a single exponential anisotropy decay
is seen in micellar solution, the location of the chromophore can be open to question.
With the assumption that the oligomers incorporate into micelles, there is an issue of how
to model the reorientation of these species. The fluorescence depolarization of a probe
molecule in a micelle can be described by up to three independent motions. These
motions may include movement of the probe in a restricted region within the micelle,

translational motion of the probe as it is adsorbed to the surface of the micelle, and

225 F

200 - l,’

 

175 '- I, db

p—i

LII

O
I

125 -

reorientation time (ps)
8
I
1:...

75 - f,

 

 

50-7111:
_ i
25 -
l l l l J l
0 1 2 3 4 5

number of valine residues

Figure 3.7 Rotational diffusion time constants of 7AT and 7AT-Valn as a function of
n in water at 20° C. The slope of the best-fit line for these data is 32.5 ps/valine unit.

rotation of the micelle as a whole. There are several models that may be applicable to
the interpretation of R(t) decays in an environment where rotation of a species is
hindered. 4"“ Two models predict R(t) to decay as a single exponential. In one of these

models, the chromophore is attached to the micelle as a rigid entity, with the anisotropy

-6l-

decay time constant representing the motion of the (spherical) micelle as a whole, and we
designate this time constant m. The second model that predicts a single exponential
decay of R(t) models the chromophore as being located in the core of a micelle and free
to rotate within that confined volume. In this model, the anisotropy is expected to decay
as a single exponential with a rate constant of (tM'l + If)" where 13 is the decay time
constant corresponding to chromophore motion in the (homogeneous) core of the micelle.
The experimental data suggest that neither of these models are appropriate for the
systems we report on here.

There are more complex models that predict anisotropy decays with two or three
exponential components, depending on the details of the model. For the case where the
chromophore is bound near the surface of the micelle, a biexponential decay is expected,
with time constants of t“, arising from restricted motion analogous to that of a hindered
rotor and m for overall reorientation of the cone equal to micellar motion. ‘2 If
translational motion of the chromophore about the surface of the micelle is considered, a
third decay component, rd, is expected. 32 The anisotropy function for this model is
shown in Eq. 3.4,

R(t) = 1?.(0)[s'2 +(1—S'2)exp(—t/tw)]exp(—t(l/td 44/1,,» (3.4)
where R(O) is the zero time anisotropy, S’ is an order parameter which is a measure of the
equilibrium orientational distribution of the chromophore transition moment. It is clear
from the above discussion that, depending on the complexity of the model used in the
interpretation of the experimental data, there is the potential for over-interpretation. We

apply Ockham’s razor to our treatment of the data.

-62-

The experimental data for reorientation of (zwitterionic) 7AT in anionic SDS
micellar solution (25 mM, 3x CMC) shows that R(t) decays to be single exponential to
within our ability to resolve. The time constant of ton of 54 :l: 5 ps is measurably longer
than that of 7AT in water (Ion = 41 :1; 4 ps) in water, despite the fact that the bulk
viscosity of the SDS solution is essentially the same as that of water. This finding
suggests that there is indeed some interaction of the 7AT with the micelle, but the nature
of the interaction and its persistence time cannot be addressed directly by these data
alone. The reorientation dynamics of the 7AT-(Val)ll oligopeptides in SDS are
qualitatively different. For all of the (cationic) oligopeptides in SDS, the experimental
R(t) function decays with two exponential components (Figure 3.8). The average time
constants of the two decays are 73 ps and 491 ps and there is no clear dependence of
these time constants on oligomer length. It is tempting to assume that the interactions of
the cationic peptides are mediated by ionic interactions with the anionic SDS micelle
head groups based on a comparison of these data with that for zwitterionic 7AT. Based
on these findings, it is tempting to presume that the two time constants correspond to fast,
possibly restricted motion within the micelle and slower global motion of the micelle as a
whole. The difficulty with that interpretation, however, is that neither of the time
constants correspond to the expected values for these motions. The most glaring
mismatch between experimental data and model prediction is for the longer time
constant. The overall motion of SDS micelles is predicted by Eq. 3.3 to occur on the
several nanosecond timescale. The ~500 ps decay time constant we recover appears to be
too fast to account for global micelle motion. The peptides do not display single

exponential dynamics or time constants that would be consistent with the probe being

-63-

 

 

 

 

700 7-
r "l’
600 - I
. T V
500 -
a i l ‘
E 400 - V T
:3
g 300 -
200 r
100 - 7AT Q
. . 9 i ° 9
O 1 l l L l L
0 1 2 3 4 5
number of valine residues

Figure 3.8 Rotational diffusion time constants of 7AT and 7AT-Valu in SDS (25 mM,
3 x cmc). The anisotropy functions of the oligomers decay as a two component
exponential and 7AT decays as a single exponential.

bound rigidly in the micelle or free to rotate within the core. Double exponential decay
dynamics could result if the oligopeptides reside in two distinct environments, partitioned
between the micelles and aqueous phase of the solution. If this were the case, the short

time components should correspond to the reorientation values in water and the long

-64-

 

 

71X) -
600 - I
* l
SCK) -
g .
g 400 -
8 P
a:
2
,0 300 r
E
22
l
21X) -
100 - 7AT
Q
I- D Q
0 l 1 l l 1 l
() 1 2. 3 4~ 5
number of valine residues

Figure 3.9 Rotational diffusion time constants of 7AT and 7AT-Van. in CT AB (2.5
mM, 2.7 x cmc). The anisotropy functions of the oligomers n = 4 and 5 decay as a two
component exponential while 7AT and 7AT-Val. — 7AT-Va13 decay as a single
exponential with a slope of 33.6 ps/valine unit.

component should be much longer if it is attached rigidly to or contained within a
micelle. The fact that single exponential fluorescence decays are recovered in some cases

supports the notion that the probes are located primarily in a single environment. The

-65-

anisotropy could yield double exponential dynamics if the probes are intercalated in the
micelle but the long component recovered for R(t) is much less than the expected m for
SDS of ~9 ns. 34 The slow time constant is not consistent with the motion of the free
oligopeptide either (Fig. 3.7), leaving quasi-translational motion of the chromophore
along the outer extent of the micelle as the only physical motion consistent with this time
constant. We will return to a discussion of this point after considering the reorientation
dynamics of the oligopeptides in cationic and neutral micelles.

The reorientation dynamics of the 7AT-Valine methyl ester series in (cationic)
CI‘ AB surfactant solution (1.0mM, 10x CMC) are shown in Figure 3.9. The most
striking feature contained in these data is the change in reorientation dynamics with
increasing oligopeptide length. We observe that the functionality of R(t) is a single
exponential decay in 7AT-Val] - 7AT-Va13 and R(t) decays as a double exponential for
7AT-Val4 and 7AT-Va15. This behavior is qualitatively different than the dynamics seen
in SDS, where all oligopeptides produced double exponential anisotropy decays. For the
oligopeptides 7AT-Val. - 7AT-Va13, the slope of the volume-dependence for
reorientation is 33.6 i 1.6 ps/valine residue, the same to within the experimental
uncertainty as that recovered from the reorientation dynamics for the 7AT-Va11 - 7AT-
Vals oligomers in water. In addition, the reorientation time constants for 7AT-Val. —
7AT-Val3 peptides in CTAB match those measured in water. These data indicate that the
7AT-Val] - 7AT-Va13 peptides do not associate with CT AB micelles to a significant
extent; they are essentially free in solution. The longer peptides 7AT-Val. and 7AT-Va15
exhibit a two component anisotropy decay CT AB micellar solution, with the average

time constants for the two components being 76 ps and 533 ps. It is important to note

-66-

that the two components are characterized by the same time constants for both 7AT-Val4
and 7AT-Va15.

It is clear that there is a substantial interaction between the longer oligopeptides
and the CT AB micelles and the dynamics recovered suggest that the interaction is quite
similar to that seen for these probes in SDS micelles. We attribute this peptide length-
dependent behavior to the balance between electrostatic and van der Waals interactions in
these systems. The cationic oligopeptides will experience ionic (electrostatic) repulsion
by the CT AB cationic head groups and the strength of this interaction is such that it
requires the van der Waals interactions between four or more valine residues and the
micelle lipophilic region to overcome electrostatic repulsion. While it is not possible to
quantitate the strength of this interaction, we can speculate that the van der Waals
interactions will be similar in magnitude to those seen in alkanethiol monolayers on gold,
where the interchain interaction energy is taken to be ~ 300 cal/mol-CHZ. Clearly, the
valine structure is more complicated and the nature of the micellar structure is more
fluxional than that of SAMs. For the valine units, there are two CH groups and two CH3
groups per monomer. If we assume the interaction strength to be the same for these
groups, on average, as for CH2 groups, then four valine residues would produce a
strength of interaction roughly equivalent to a C16 aliphatic chain, or ~ 4.8 kcal/mol. The
conesponding disorder-to-order transition in alkanethiols occurs between 9 and 10
carbons, or 2.7 — 3 kcal/mol, but these systems are capable of a much greater degree of

organization than valine oligomers in micelles. This crude calculation suggests, however,
that the electrostatic repulsion term for CTAB — 7AT-Valn interactions is on the order of

a few kcal/mol at most.

-67-

For the neutral surfactant Thesit®, the valine oligomer length dependence of the
reorientation dynamics is presented in Figure 3.10. Not surprisingly, these data present a
case that appears to be intermediate between the anionic SDS and cationic CI‘ AB micelle
systems. These data show a change in reorientation behavior from single to double
exponential anisotropy decays for the oligomers, with the onset of a two-component
decay beginning at 7AT-Va12. For Thesit®, the dipeptide 7AT-Va11 displays a single
exponential anisotropy decay with a time constant the same as that measured in water.
By analogy to the data for CT AB, the absence of electrostatic contributions to ThesitQ
interactions with the probes reduces the barrier to valine-micelle interactions. Because of
the different structures of CT AB and Thesit®, the strength of valine-micelle interactions
will clearly not be the same for the two systems. For the longer oligopeptides, where a
two-component anisotropy decay is seen, the time constants are independent of peptide
length and the slow time constant is slightly smaller than the conesponding data for
CT AB and SDS micelles.

We assert that the systems that exhibit two-component anisotropy decays with
decay time constants in the ~75 ps and ~500 - 600 ps time windows can be explained
using a single formalism. The fundamental physics responsible for the data we observe is
essentially the same for all of the micellar systems, with differences in the details being
associated with the specific identities of the micellar constituents. We summarize our
findings as follows: (1) 7AT exhibits only small changes in tog for each surfactant
solution compared to its behavior water, indicating that the probe molecule itself does not
associate with micelles to a significant extent. This finding is consistent with the lifetime

data. (2) Peptides either interact strongly with micelles and manifest a two-component

-68-

 

4:.
8
I
l—<1——l

 

 

 

v
g _
.3 300 -
-§ 1
S
5 200
.5. _
§
100 e
7AT '3 Q 9 g Q
r :1
O I I I I I I
0 l 2 3 4 5

number of valine residues

Figure 3.10 Rotational diffusion time constants of 7AT and 7AT-Valn in Thesit® (1.0
mM, 10 x cmc). The anisotropy functions of the oligomers n = 2-5 decay as a two
component exponential while 7AT and 7AT-Val. decays as a single exponential.

anisotropy decay, or they do not interact with the micelles to a measurable extent. (3) For
systems manifesting a two component decay, the decay time constants are independent of

oligopeptide length in a given micellar system and similar for all three micelles studied

-69-

here. For this reason, we use the average of fitted decay time constants for a given
micelle in the modeling we discuss below.

Wobbling-in-a-cone with translational diffusion and micelle rotation. As
discussed above, the expected reorientation time constant for a micelle is several
nanoseconds and we do not observe any such long time anisotropy decay in our data.
The possible reasons for this absence are either that we are not sensitive to such a long-
time component, which is not the case, or that the persistence time of the micelle-
oligopeptide association is substantially shorter than the reorientation time of the micelle.
This explanation is consistent with relatively weak interactions between the probe and
micelle, in keeping with our estimate of the interaction energies of several kcal/mol for
these systems. We consider that the interaction between the oligopeptide and the micelle
structure is characterized by interactions between the lipophilic portion of the micelle and
the valine peptide tails, and that the 7AT head group is in closest proximity to the nricelle
head groups, where it is characterized by some freedom to translate or move about on the
surface of the micelle. The translational diffusion of a chromophore on a spherical
surface will manifest itself as a reorientation of emission dipoles. Eq. 3.4 describes the
general case of the anisotropy decay for three independent motions and we apply the
appropriate simplifications to bring this model into consistency with the experimental
data,

l/Id =1/rslow —l/rM (3.5)
l/tw = l/rfast —1/td -1/'cM = New - l/tslow (3.6)

We determine values for m from the DSE equation at 20 °C or we assume it to be

infinitely long, but in either case we are insensitive to this quantity experimentally. The

-70-

translational diffusion coefficient may be related to 1,. using the value for r). (Table 3.1)
the radius of the surface of the micelle over which the probe diffuses by Eq. 3.7. 3‘

D, = 1;} I67, (3.7)
This diffusion constant is related to the slower of the two motions we sense. The
wobbling-in-a-cone model“;44 is characterized by a molecule constrained within a cone
of semi-angle 00 with the motion of the molecule in the confined environment being
described by a wobbling diffusion coefficient (Dw). We can extract the value of Dw from

the experimental data using the quantities I... and S ’,

Dw ={rw(1— S'2)}"[— cos2 00(1 + cos00)2{ln[(l+ cos00)/2]
+(1—cost90)/2}{2(1—cosflo)}"+(1—cos00) (3.8)
x(6 + 8cos€o - cos2 00 ~12cos3 00 — 7cos4 00)/24]

In this model, the quantity 00 is related to the order parameter through Eq. 3.9.
S' = 0.5cos 00(1 +cos 00) (3.9)

The values for S’, 15.0., and em were extracted from the experimental data using Eq. 3.4
and from this information, the quantities 61), D... and D. were calculated for each
surfactant system (Table 3.2). The values Dw for all the peptides in water are given for
comparison in Table 3.3. In water, Dw is calculated using Dw = mean)" based on the
assumption that in water the order parameter is zero and the equilibrium orientational
distribution is random because the probe rotates freely.

The values of the order parameter, S ’, vary from 0.50-0.73 for reorientation of the

7AT-Valn oligopeptides in the micelles. The order parameter is a measure of the time-

averaged orientational distribution of the peptides and can take on values from zero for a

fully random orientational distribution, to one for completely ordered systems. The

-71-

Table 3.2 Values of quantities extracted from the “Wobbling-in-a-Cone” model.

 

 

Surfactant 5’ 0° (degrees)“ Dw x 109 (s") Dt x 10-9 (m2/s)
SDS 0.73 36 i 3 1.2 0.86
CTAB 0.58 46 t 2 1.7 0.91
Thesit® 0.50 52 i 2 2.3 1.1

“00 calculated according to Eq. 9. 90 = COS" (0.5(41 + 35' -1)). Uncertainty in the
determination of S’ is taken to be 1: 5%.

values recovered for the order parameter indicate that the peptide oligomers do indeed
have a preferential orientation with the micelle surface. In SDS (0) = 36°) they are
considerably more confined than in CI‘ AB (0) = 46°) and Thesit° (a) = 50°). In SDS
there is clear evidence for order, but the proximity of the cone angle observed in ThesitGD
to the magic angle suggests substantial disorder.

The diffusion coefficients for the peptides in water decrease with peptide size,

according to the increase in correlation times, (Table 3.3), and this is an expected result.
The recovered parameter Dw for the 7AT-Valn oligomers demonstrates that the 7AT

chromophore is free to reorient about its tethering bond to the valine chain. The
wobbling diffusion constant, D... reveals substantial confinement of the 7AT
chromophore in micellar environments. The value of Dw in micelles is two to three times
slower than that for 7AT in bulk water. It is important to keep in mind that the free
volume accessible to the 7AT chromophore in the micelle is substantially less than that of

7AT in water. We calculate from the cone angles 00 that the volume accessible to the

-72-

Table 3.3 Experimental diffusion constants for oligopeptides in water.

 

 

Dw x 10° (s")
7AT 4.1
7AT-Val. 2.3
7AT-Val; 1.6
7AT-Val3 1.2
7AT-Val. 0.96
7AT-Va15 0.83

 

7AT chromophore in SDS is 7% of a full sphere, for CT AB, 9% and for Thesit” 9.4%.
When the DW data are corrected for accessible volume, we recover an effective viscosity
of 47 cP for the SDS micelle, 26 cP for the CT AB rrricelle and 18 cP for the Thesit®
micelle. While these viscosities may seem high, they are not substantially different than
that of bulk l-dodecanol or ethylene glycol (~20 cP). In the glycols, hydrogen bonding is
thought to dominate the viscous flow properties while in the n-alcohols, van der Waals

interactions are thought to be responsible.

It is also useful to consider the magnitude of D.. The oligopeptides exhibit
moderately fast translational diffusion on the surface of the micelles. The self-diffusion
of the ionic surfactants in this study has been investigated previously and are reported to
be on the order of (0.2-1.5) x 10'10 m2 s". 45’“ The translational diffusion constants we
recover for the oligopeptides used in this study are ~ 10’9 m2 s", a factor of ~10 faster
than surfactant self-diffusion. This is not necessarily a surprising result because of the
inability of the oligopeptides to incorporate into the micellar structure with the same

efficiency as a constituent surfactant molecule. The largest translational diffusion

-73-

constant we measure is for the neutral Thesit® micelles. We note that Thesit® is the least
ordered micelle used in this study and the intrinsic disorder of the micelle may be related
to efficient translational diffusion. The loosely organized hydrophilic ethylene glycol

chains create the surface and outer layer of Thesit® micelles.

-74-

3.4 Conclusions

We have studied the steady-state and time-resolved optical response of 7AT and
7AT-Valn peptide oligomers in water and aqueous micelle solutions. The charge and

chain length of the peptides dictate interactions with micelles. The reorientation
dynamics of the peptides were consistent with the model of wobbling-in-cone dynamics
with translational diffusion of the dye along the surface of the micelle with rotation of the
micelle as a whole. Our experimental data reveal a balance between electrostatic or ionic
interactions and van der Waals interactions for polyvaline oligopeptides and selected
micellar structures. For the case of anionic micelle headgroups and cationic
oligopeptides, we see the strongest interactions because ionic and van der Waals forces
are both attractive. For the cationic micelle CT AB, we observe the balance between ionic
and van der Waals interactions because of the opposing nature of these forces. This
balance allows us to estimate the strength of ionic interactions between the probe and the
micelle to be on the order of several kcal/mol. Using a neutral surfactant Thesit‘ID
provides an intermediate case, as expected. Our experimental reorientation data indicate
transient interactions between the micelle and the probe oligopeptides, with limited
structural freedom for the probe molecules in the apparently viscous rrricellar

environment.

-75-

35 Literature Cited

1. Tulock, J. J .; Blanchard, G. J. J. Phys. Chem. A 2000, 104, 8340-8345.

2. Rasimas, J. P.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 11333-11338.

3. Rasimas, J. P.; Berglund, K. A.; Blanchard, G. J. J. Phys. Chem. 1996, 100, 17034-
1704(L

4. Rasimas, J. P.; Berglund, K. A.; Blanchard, G. J. J. Phys. Chem. 1996, 100, 7220-
7229.

5. Tulock, J. J .; Blanchard, G. J. J. of Phys. Chem. B 1998, 102, 7148-7155.

6. Lakowicz, J. R. Principles of Fluorescence Spectroscopy, Second Edition ed.; Plenum
Publishers: New York, 1999.

7. Cockle, S. A.; Szabo, A. G. Photochem. Photobiol. 1981, 34, 23-27.

8. Ross, J. B.; Rousslang, K. W.; Brand, L. Biochemistry 1981, 20, 4361-4369.

9. Steward, L. E.; Collins, C. 8.; Gilmore, M. A.; Carlson, J. E.; Ross, J. B. A.;
Chamberlin, A. R. J. Am. Chem. Soc. 1997, 119, 6-11.

10. Negrerie, M.; Bellefeuille, S. M.; Whitham, S.; Petrich, J. W.; Thomburg, R. W. J.
Am. Chem. Soc. 1990, 112, 7419-7421.

11. Rich, R. L.; Negrerie, M.; Li, J.; Elliott, S.; Thronburg, R. W.; Petrich, J. W.
Photochem. Photobiol. 1993, 58, 28-30.

12. Rich. R. L.; Chen, Y.; Neven, D.; Negrerie, M.; Gai, F.; Petrich, J. W. J. Phys. Chem.
1993,97,]781-1788.

13. Rich, R. L.; Smimov, A. V.; Schwabacher, A. W.; Petrich, J. W. J. Am. Chem. Soc.

1995. 117, 11850-11853.

-75-

14. Brennan, J. D.; Clark, 1. D.; Hogue, C. W. V.; Ito, A. S.; Juliano, L.; Paiva, A. C. M.;
Rajendran, B.; Szabo, A. G. Appl. Spectrosc. 1995, 49, 51-59.

15. Broos, J .; Ter Veld, F.; Robillard, G. T. Biochem. 1999, 38, 9798-9803.

16. Hogue, C. W.; Szabo, A. G. Biophys. Chem. 1993, 48, 159-169.

17. Sheehan, J. C.; Cruickshank, P. A.; Boshart, G. L. J. Org. Chem. 1961, 26, 2525.

18. Nozaki, S. Chem. Lett., 1997; 1, 1-2.

19. Kocienski, P. J. Protecting Groups; Thieme: New York, 1994.

20. Robson, R. J.; Dennis, E. A. J. Phys. Chem. 1977, 81 , 1075.

21. Tanford, C. J. Phys. Chem. 1972, 76, 3020.

22. Kelepouris, L.; Blanchard, G. J. J. Phys. Chem. A 2000, 104, 7261-7267.

23. Hu, C.-M.; Zwanzig, R. J. Chem. Phys. 1974, 60, 4354-4357.

24. Debye, P. Polar Molecules; New York, 1929.

25. Penin, F. J. Phys. Radium 1934, 4, 497.

26. Dewitt, L.; Blanchard, G. J .; Legoff, E.; Benz, M. E.; Liao, J. H.; Kanatzidis, M. G.
J. Am. Chem. Soc. 1993, 115, 12158-12164.

27. Ashby, K. D.; Das, K.; Petrich, J. W. Anal. Chem. 1997, 69, 1925-1930.

28. English, D. 8.; Rich, R. L.; Petrich, J. W. Photochem. Photobiol. 1998, 67, 76-83.

29. Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1994, 98, 9411-9416.

30. Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1995, 99, 7904-7912.

31. Gudgin-Templeton, E. F.; Kenny-Wallace, G. A. J. Phys. Chem. 1986, 90, 2896.

32. Chou, S. H.; Wirth, M. J. J. Phys. Chem. 1989, 93, 7694-7698.

33. Quitevis, E. L.; Marcus, A. H.; Fayer, M. D. J. Phys. Chem, 1993; 97; 5762-5769.

-77-

34

35.

36.

37.

38.

39.

40.

41.

42.

43.

45.

46.

Maiti, N. C.; Krishna, M. M. G.; Britto, P. J.; Periasamy, N. J. Phys. Chem. B 1997,
101,11051-11060.

Kivelson, D.; Spears, K. G. J. Phys. Chem. 1985, 89, 1999-2001.

Chuang, T. J .; Eisenthal, K. B. J. Chem. Phys. 1972, 57, 5094-5097.

Edward, J. T. J. Chem. Ed. 1970, 47, 261.

Hanson, D. C.; Yguerabide, J .; Schumaker, V. N. Biochem., 1981; 20; 6842-52.
Kroes, S. J.; Canters, G. W.; Gilardi, G.; van Hoek, A.; Visser, A. J. Biophys J.,
1998; 75; 2441-2450.

Lakshmikanth, G. S.; Krishnamoorthy, G. Biophys. J., 1999; 77; 1100-1106.
Lakowicz, J. R.; Cherek, H.; Maliwal, B. P.; Gratton, E. Biochem. 1985, 24, 376.
Lipari, G.; Szabo, A. Biophys. J. 1980, 30, 489-506.

Szabo, A. J. of Chem. Phys. 1984, 81, 150.

. Kinosita, K., Jr.; Kawato, 8.; Ikegami, A. Biophys. J. 1977, 20, 289-305.

Ahlnas, T.; Soderman, O.; Hjelm, C.; Lindman, B. J. Phys. Chem. 1983, 87, 822.
Nery, H.; Soderrnan, 0.; Canet, D.; Walderhaug, H.; Lindman, B. J. Phys. Chem.

1986, 90, 5802.

-73-

Chapter 4

Dynamics of 7-Azatryptophan and Tryptophan Derivatives in Micellar
Media. The Role of Ionic Charge and Substituent Structure

Summary

We have investigated the motional and population relaxation dynamics of several
derivatives of 7-azatryptophan (7AT) and tryptophan (TRP), and N-acetyl-
tryptophanamide O‘IATA) in anionic, cationic, and neutral surfactant solutions above the
surfactant critical micelle concentrations. The specific derivatives of 7AT and TRP used,
and NATA were chosen to elucidate the roles of chromophore charge and side group
structure in mediating interactions with micelles. Fluorescence lifetime data are sensitive
to local environment, revealing significant differences in the ability of different
chromophores to interact with the micelles. Reorientation dynamics of these
chromophores indicate persistent probe-micelle interactions that are determined by a

balance between ionic charge(s) and dispersion interactions.

-79-

4.1 Introduction

The dynamics of peptides and proteins in solution is of fundamental importance to
many essential biological processes and fluorescence spectroscopy has proven to be a
versatile tool for examining such phenomena. The characteristic time scale of
fluorescence population decays is matched well to the speed of many important
dynamical phenomena in protein and peptide systems.1 One of the key issues in the study
of biological systems is the choice of a fluorescent probe that will not itself serve to
perturb the structure of the molecular system under investigation. Tryptophan (TRP) is a
widely used chromophore for investigations of peptide and protein behavior because it is
one of a limited number of fluorescent amino acids; tyrosine and phenylalanine have
been studied less extensively owing to their lower absorbance cross sections,
fluorescence quantum yields and absorption and emission band positions. The indole
chromophore present in tryptophan is sensitive to its local environment, and is
characterized by an emission maximum between 308 nm and 355 nm, depending on its
immediate environment.

While TRP is found naturally in many biologically active peptides and proteins, it
is not an ideal probe because its fluorescence population decay is non-exponential, and
this anomalous behavior is thought to arise from efficient charge transfer behavior
between multiple stable conformations of the indole chromophore to the side chain. 2
Within the framework of this interpretation, the lifetime of the chromophore is
conformation-dependent, making the interpretation of fluorescence lifetime data difficult.

In addition, the presence of multiple tryptophan residues in a protein can preclude the

extraction of site-specific information from TRP probes.

-30-

In recent years, several non-natural amino acids have been investigated as
alternatives to TRP as biological probes, including 7-azatryptophan (7AT). 3'6 The
spectral response of 7AT is red-shifted relative to that of TRP, allowing its use in the
study of protein structure and dynamics. 7'9 7AT is characterized by a single exponential
population decay of its 87 state over a wide pH range, and the time constant of the decay
is sensitive to the immediate environment of the chromophore. 6‘9 The single exponential
population relaxation dynamics of 7AT is an expected result for organic chromophores
and its behavior stands in contrast to that of the structurally similar TRP. This difference
in relaxation dynamics has been attributed to a relatively low barrier to the formation of a
charge transfer excited-state in 7AT. '0

We use steady-state and time-resolved fluorescence spectroscopy to obtain
information on molecules containing the 7-azaindole and indole chromophores in
micellar solutions. We are interested determining the effects of chromophore and micelle
head group charge, and probe molecule structure on their interactions. We use three
substituted forms of each chromophore in this comparative study (Figure 4.1). At neutral
pH, they are the zwitterionic amino acids 7-azatryptophan (7AT) and tryptophan (TRP),
anionic species formed by the addition of a tert-butoxy group at the N-terrninus of each
amino acid, named boo-7AT and boc-TRP, and cationic species produced though a
peptide bond with dodecylamine at the C-terrninus of the amino acid residues, designated
DD-7AT and DD-TRP. The latter chromophores possess an aliphatic moiety that will
allow significant dispersion interactions with micelles. Tryptophan octyl ester has
attracted considerable interest as a suitable model for TRP in the case of membrane

”-15

proteins, and the fluorophores DD-7AT and DD-TRP may also serve this purpose.

-31-

 

“’ Q7901

7AT, TRP Boc-AT, Boo-TRP

 

DD-AT, DD-TRP

NH2

-1 M

N-acetyltryptophanamide
(N ATA)

Figure 4.1 Structures of the chromophores in this study. For 7AT X=N and X=C for
TTKP.

-32-

We have also examined n-acetyl-tryptophanamide (NATA) in the same micelles. NATA
is a neutral species and is of interest because it is an indole-containing molecule that has
been used before as a mimic of TRP contained in a peptide backbone. It is characterized
by a single exponential 8. population decay. '6'” We are interested in micelles because
they are a reasonably well understood class of heterogeneous systems that can function as
simplistic models for biomembrane structures. Surfactants and micelles have been
investigated extensively due to their many uses, ranging from acting as media in which
chemical reactions and excitation transport studies can be performed, to systems for
enhancing protein solubility and effecting chemical separations.

We report here on our study of the dynamics natural and non-natural amino acid
derivatives in aqueous micelles formed by anionic, cationic and neutral surfactants. Our
experimental data suggest that probe interactions with the surface or interior of the
micelles can dominate, depending on the charge of the chromophore used and the identity

of its pendant moiety.

-83-

 

4.2 Experimental

Materials. The surfactants sodium dodecyl sulfate (SDS),
cetyltrimethylammonium bromide (CT AB) and polyethylene glycol 400 dodecyl ether
(Thesit®) were obtained from Sigma-Aldrich, Inc., checked for fluorescent impurities,
and used as received. All chromophores and reagents needed for derivatization were also
obtained from Sigma-Aldrich, Inc. The molecules D,L-7-azatryptophan hydrate, D,L-
tryptophan hydrate, N-acetyl-tryptophanamide, and N-t—boc-tryptophan were used as
received. N-t-boc-7-azatryptophan was prepared by the procedure described by Petrich‘S
using D,L-7-azatryptophan hydrate, 2-(tert-butoxycarbonyloxyimino)-2-
phenylacetonitrile (BOC-ON), and triethylamine in water/dioxane solvent.

DD—7AT and DD-TRP were prepared by condensations of the N-t-boc-amino acid
with dodecylamine followed by removal of the boc group. The coupling of amino acids
to dodecylamine was performed using the water soluble carbodiimide, 1-[3-
(dimethylamino)propyl]-3-ethylcarbodiimide, hydrochloride (EDCI) 18"9 along with the
coupling catalyst hydroxybenzotriazole (HOBt). A description of the coupling and
deprotection procedure has been described before. The deprotection procedure was
modified slightly. The DD-boc-X compounds (X = 7AT, TRP) were stirred into 9:1
trifluoroacetic acid:dimethyl sulfide and the progress of the reaction was followed by thin
layer chromatography and worked up as indicated previously. 2°

All aqueous solutions were prepared with water from a Milli-Q water filtration
system. All solutions used were 35 11M in chromophore concentration and spectroscopic

measurements were performed within 24 hours of solution preparation. The sample

-34-

temperature was controlled using a brass block cuvette holder maintained at 20.0 :1: 0.1 °C
(Neslab).

Steady-State Absorption and Emission Spectra. Absorption spectra were acquired
using a Cary Model 300 double beam UV-visible absorption spectrometer set at 2 nm
resolution. Emission spectra were recorded with a Spex Fluorolog 3 spectrometer using a
5 nm bandpass for excitation and a 2 nm bandpass for emission collection.

Time-Correlated Single Photon Counting (TCSPC) Spectrometer. The
spectrometer we used for the lifetime and dynamical measurements has been described in
detail before. 2' The light pulses used to excite the sample are produced by a
synchronously pumped, cavity-dumped dye laser (Coherent 702-2, 5 ps pulses, 4 MHz
repetition rate). The dye laser is excited by the second harmonic of the output of a
continuous wave (cw) mode-locked Nd:YAG laser (Quantronix 416, 100 ps pulses, 80
MHz repetition rate). Samples were excited at their absorption maxima (Pyromethene
567 dye, Exciton, KDP Type I SHG). Emission was monitored at the chromophore
florescence maximum in the solution of interest using a 20 nm bandpass for collection.
Fluorescence was collected at polarizations of 0°, 54.7°, and 90° with respect to the
vertically polarized excitation pulse. The instrument response function for this system is

typically 35 ps FWHM and lifetimes measured range from ~600 ps to ~85 ns.

-35-

4.3 Results and Discussion

The objectives of this work are to understand the similarities and differences in
the Optical response of related chromophores that arise from structural modifications, and
to elucidate how these modifications influence probe interactions with their immediate
environment. In this study, micelles serve as a heterogeneous solution environment in
which to examine these issues. The interactions of these probes with solvent, surfactant,
and micelles will be mediated by their charge, substituent structure and functionality, and
specific chromophore characteristics. We consider the steady-state and time-resolved
spectroscopic data separately, following a brief discussion of solution and chromophore
properties important to both groups of measurements.

The surfactant concentrations used were several times their critical micelle
concentration (CMC) to ensure micelle formation. In order to interpret the data and be
certain that changes in the optical responses of the probes are not due to changes in bulk
solution properties, several physical parameters and solution properties of the surfactants
and micelle solutions must be known. The relevant physical parameters of the surfactant
solutions used are listed in Table 4.1. The micelle hydrodynamic radius, rh, is among the
quantities needed to interpret the data. For ionic surfactants, this value is estimated by
summing the micelle core radius, the head group radius, and two layers of water. ”'23
This method of determining n. for ionic surfactants has been found to be in agreement
with values obtained from light scattering experiments. The micelles formed by the
neutral surfactant Thesit® are not as well ordered as those made using ionic CT AB and

SDS amphiphiles. The hydrophilic portion of these two classes of micelles differ

significantly. The hydrophilic portion of SDS and CTAB are bulky sulfate and

-86-

Table 4.1 Solution and physical parameters of micelles of micelles used in this work.

 

 

Surfactant“ Charge Viscosity pH Concentration r), 1M
(cP) (mM)/number (A) (us)

of CMC
CT AB + 1.00 :t .01 6.2 2.5 / 0.92 25.7 17.6
SDS - 1.03 i .01 6.6 25.0/ 8.3 20.7 9.2
Thesit neutral 0.99 :l: .01 7.1 1.0 l 0.1 35 44.4

a CT AB = Cetyltrimethylammonium bromide; SDS = sodium dodecyl sulfate;
Thesit = poly(ethylene glycol) 400 dodecyl ether.

ammonium groups, respectively, while Thesit® is comprised of a poly(ethylene glycol)
chain. This latter structural condition results in a hydrated mantle with a less well
defined surface aqueous region, and rh is estimated at 35 A. 2°

The dominant forms of the free amino acids TRP and 7AT depend on solution pH
due to the presence of carboxylic acid and amine functionalities. Amino acids can take
on a positive or negative charge or be zwitterionic, depending on the solution pH. In
addition, 7AT can be protonated at the N7 position, resulting in a significant emission red
shift and decreased excited state lifetime below pH ~5. 24 TRP does not possess a labile
proton on its indole moiety, but has been reported to exhibit an additional fluorescence
decay component above pH 9. The existence of multiple different protonated states and
low energy excited electronic states in TRP and 7AT accounts for the sensitivity of their
emission spectra to local environment.

Viscosity affects rotational diffusion time constant(s) according to the modified

25-27

Debye-Stokes-Einstein equation. The bulk surfactant solution properties of viscosity

-37-

 

and pH were measured at 20 °C with a Canon-Fenske viscometer and a pH meter,
respectively. The addition of surfactants at concentrations above their CMC in this study
has a negligible effect on bulk solution viscosity. We measure changes in pH in the
solutions but these pH changes do not occur in a region that affects the state of
protonation of the chromophore or the overall charge of probe molecules. The changes in
bulk solution properties associated with the formation of the micelles are small and will
not influence the optical response of the chromophores substantially.

Steady-state spectroscopy. The experimental absorbance and emission maxima of
the chromophores studied here are given in Table 4.2. Representative emission spectra
are shown in Figure 4.2 for the TRP derivatives in SDS and Figure 4.3 for boc-AT in
water and aqueous micelle solutions. Inspection of the absorbance maxima indicate the
ground states of the chromophores are not perturbed to a great extent by covalent
modification of the chromophore side groups or by the presence of micelles in solution.
The structure- and environment-dependent changes in emission maxima for these
chromophores are more pronounced. Both 7AT and TRP are known to exhibit emission
spectral shifts when incorporated into peptides and proteins, with the spectral shifts seen
for TRP being larger than those for 7AT. The first comparison we make is between the
free residues, the N-boc modified chromophores, and C terminus modified dodecylamine
forms, all in water. Compared to the free residues, emission of the boo-modified forms
are red shifted by 5 nm (7AT) and 4 nm (TRP), and the dodecylamine-modified
chromophores are blue-shifted by 2 nm (7AT) and 8 nm (T RP). The spectral shifts we

observe for these chromophores are not due to changes in bulk solution properties, but

-33-

 

Table 4.2 Absorbance, emission, and fluorescence decay parameters for probes in water
and micelles. The (rt-ratio (last column) is the ratio of the prefactors for the two decay
components.

 

 

 

Averbs max Avern max 1111 (PS) 7112 (PS) (xi/“2
AT Aqueous 290 399 80019
CTAB 290 399 8151
SDS 290 399 8291
Thesit® 290 399 7701
boc-7AT Aqueous 290 404 637:1:9
CTAB 290 397 1505113
SDS 289 404 642111
Thesit® 289 404 660112
DD-7AT Aqueous 290 397 901:1:10
CTAB 290 397 1467111
SDS 290 384 57819
Thesit® 290 387 924111 2407170 9.4
TRP Aqueous 279 359 8721100 3470150 0.11
CTAB 279 359 634147 3384180 0.13
SDS 279 359 8851150 3100170 0.15
Thesit 279 359 689168 3473180 0.11
boo-TRP Aqueous 279 363 1033:39 8450126 0.14
CTAB 282 352 1858133 5923133 0.45
SDS 279 363 792:1:36 8361118 0.25
Thesit® 279 363 1000127 8383120 0.27
DD-TRP Aqueous 279 351 18491109 30491102 0.49
CTAB 281 351 1253132 3634156 0.59
SDS 279 339 948135 2827125 0.52
Thesit® 281 345 28491113 67831192 1.67
NATA Aqueous 279 360 3153::18
CTAB 279 358 3067122
SDS 279 358 1035190 2746140 0.2
Thesit® 279 360 3168160

 

-39-

 

 

 

 

 

NJ ‘
I'. \
.' , 1 1 ——TRP
0.8 - .- ,’ “, - - - -boc-TRP
: ' ’I ‘\ ------ DD-TRP
g 1 -. X ----- NATA
° 0.6 - ’ '- .
, \
E . 1 . ~
E 0 4 -' ' " ‘
a '- s I ‘. \
o ,. p . \
r: r \
I \
I \
0.2 '- I \
i ’ \ \
t. . I \
o ’ ° 6 . . . ‘ s
0.0 -- ' - ’ .............
I L I l I 1 I u I
300 350 400 450 500

wavelength (nm)

Figure 4.2 Normalized emission spectra of tryptophan derivatives in 25 mM SDS
solution. TRP = solid line, boc-TRP = dashed line, DD-TRP = dotted line, NATA =
alternating dot/dash line.

arise from substituent effects on the excited states of the chromophores. It is known that
the charged groups of zwitterionic TRP influence its excited state(s) and the emission
shifts we observe are related to the stabilization of different charge distributions on the
substituted molecules.

In micellar solutions, the behavior of 7AT and TRP are similar to that seen in

aqueous solutions. The steady state responses of the amino acids show no change in

1.0 - ,-

 

water
0.8 - 1 1 - - - - CTAB“

0.6 - 1,: ‘

 

0.4 -

normalized emission

 

0.2 - .14 . '

0.0 ""°

1 l I 1 l n I 1 I 1 I

300 350 400 450 500 550

wavelength (nrn)

 

Figure 4.3 Normalized emission spectra of boc-7AT in aqueous micelle solutions and
water. boo-7AT in water = solid line, CT AB solution = dashed line, SDS solution =
dotted line, Thesit® solution = alternating dot and dashed line.

emission maxima in micellar solution, indicating that the presence of the micelles is not
altering their local environment to a significant extent. The hoe-modified derivatives are
anionic above ~pH 3 by virtue of the presence of a free carboxylate functionality, and
exhibit emission blue shifts relative to water in solutions containing the cationic micelle
CTAB. The chromophores modified at their C terminus with dodecylamine are cationic

below pH 9 and these chromophores also exhibit a significant blue shift in solutions

-91-

containing anionic SDS micelles. Emission spectral shifts of organic dye molecules in
surfactant solutions are common and are typically attributed to partitioning of the
chromophore into the micelles. The spectral blue shifts seen for various indoles in
micellar environments has been attributed to incorporation of the chromophores into the
micellar core of Brij-35 but at concentrations far above the CMC, 28 similar blue shifts for
indole are reported to occur only at high concentrations of SDS, while the negatively
charged tryptophan derivative N-acetyltryptophan (NATA) is reported to blue shift with
decrease in quantum yield around the CMC of CTAB. 2° Large blue shifts have been
reported for TRP residues that reside in hydrophobic protein pockets. The steady state
emission data of the chromophores we report on here are thus useful for gauging
interactions between the chromophores and micellar structures in solution.

Fluorescence lifetimes of 7A T, TRP and NATA. Excited state lifetime
measurements can provide information on solution composition and may also contain
information on the interactions between the probe and the micelle. For the same reason
that the emission spectra of the chromophores are sensitive to local environment, their
fluorescence lifetimes reflect intermolecular interactions. 7AT is zwitterionic at pH 7
and its emission intensity decays as a single exponential with a time constant of ~800 ps.
As noted above, 7AT has been incorporated into peptide fragments and oligomers, and its
excited state population decays as a single exponential in uniform environments, with the
exception of the octapeptide, as reported by Petrich. 3° The excited state population
decay of TRP is more complex. TRP typically displays a biexponential fluorescence
intensity decay in its zwitterionic form and when incorporated into peptides and proteins.

NATA has been used as a simple model for tryptophan incorporated into a peptide

-92-

 

173—"

backbone. This is not necessarily a good model since NATA is anomalous among
indole-containing molecules; it is known to exhibit a single exponential fluorescence

'6'” We consider the population relaxation dynamics of these three

intensity decay.
chromophores separately.

The fitted parameters for the emission intensity decays of the probes used in this
study are listed in Table 4.2. All fluorescence lifetimes fit well to either a single or a
double exponential decay. Goodness of fit was evaluated by x2 criterion and visual
inspection of the residuals of the fitted function to the data. A single exponential
intensity decay indicates that the fluorescent species exists in a single environment or that
there is rapid exchange between multiple environments over the time scale of the
measurement. Deviations from single exponential decay behavior indicate the
fluorescent species exists in multiple stable environments where exchange is slow, or
simultaneous relaxation from multiple excited electronic states. Probe molecules
containing the 7-azaindole chromophore exhibit a time-domain response fitted best to a
single exponential decay, except for DD-7AT in Thesit®. Inspection of the lifetime data
reveals that the lifetime of zwitterionic 7AT depends little on the presence of or
proximity to micelles, suggesting that 7AT experiences a similar environment in water
and micellar solutions and that it does not interact appreciably with surfactant molecules
or micelles. Modifying the amino and carboxylic acid functionalities of 7AT affects the
steady state emission properties of the chromophore and has a pronounced effect on the

fluorescence lifetime. In water, the fluorescence lifetimes range from ~640 ps for boc-

7AT to ~800 ps for 7AT and ~900 ps for DD-7AT.

-93-

 

The fluorescence decays for boo-7AT in water, SDS, and Thesit‘D are similar but,
in CT AB, the lifetime more than doubles to 1.5 ns. It is unusual to recover lifetimes this
long for the 7AT chromophore in aqueous media. Lifetimes for 7-azaindole of 1.4 ns are
seen in heptane, 3' and ~1.7 ns in cyclohexane. 32 As mentioned previously, boc-7AT
exhibits an emission spectral blue shift in CT AB relative to water, consistent with the
chromophore residing in a nonpolar environment. The interior of a CT AB micelle is less
polar than bulk water and, in heptane, the boc-7AT emission blue shift is greater than in
micellar media. For the 7-azaindole chromophore, two emission bands are seen in
nonpolar, aprotic solvents. In heptane a dominant band (~326 nm) is associated with the
“normal” species and secondary Stokes shifted tautomer emission (~480 nm) arising
from excited state double proton transfer of the dimer. 3' The presence of small amounts
of water disrupts the dual emission behavior. For boc-7AT in CT AB there is no evidence
of dual emission and it would not be expected unless the chromophore were sequestered
in the center of the micelle, since water is known to penetrate significantly into the
mantle of ionic micelles.

The lifetime of DD-7AT depends sensitively on its local environment. In SDS
micelles, DD-7AT has a lifetime of ~580 ps. Its lifetime increases to ~1.5 ns in CT AB,
nearly the same as for boc-7AT in CT AB, with which DD-7AT shares an essentially
identical steady state emission response. This finding suggests that the substituted
chromophores may reside in similar environments and states of solvation despite their
opposite charge and different pendant structures. DD-7AT displays a second slow
fluorescence decay component in the neutral micelle Thesit® that contributes

approximately 10% to the total decay as given by the prefactors of the fitted exponential

-94-

 

function, with the time constants being 924 ps (90%) and 2.4 ns (10%). There are several
possible explanations for these data. The first is that we are measuring partitioning
between free and micelle-bound probes. This explanation is consistent with the fast
decay component being similar to that of DD-7AT in water but the steady state emission
is shifted 10 nm with respect to water. The second possibility is that the probe is
predominantly bound to the micelles, with the micelle presenting two distinct
environments to the chromophore. This explanation would require slow exchange
between the two environments during the lifetime of the excited chromophore, and we
believe this situation to be physically unrealistic. A third possibility is that the solvation
state induced by the Thesit® micelles causes dual exponential decay due to a small
fraction of probes in a proper state of solvation to undergo interactions with the terminal
alcohol groups of Thesit®.

The fluorescence intensity decay of TRP is known to be biexponential, with time
constants of ~0.8 ns and ~3.5 ns in water. When TRP is placed in micellar solutions,
only small changes in the time constants are observed. This finding suggests that the
presence of micelles has a limited effect on the chromophore excited state, consistent
with our findings for 7AT. The fluorescence intensity decay of hoe-TRP in water is also
biexponential, with time constants of ~1 ns and ~8.5 ns (Table 4.2). The slow decay time
constant is significantly longer for boc-TRP than for TRP and one explanation for this
finding is that the TRP long lifetime is shortened due to self-quenching by the charged
amino group at neutral pH. The lifetime of boc-TRP is similar in water, SDS, and
Thesit"D but with time constants of ~2 ns and ~ 6 ns in CT AB. For boo-TRP, the short

lifetime component shows a significant increase in lifetime in CT AB, like the 7AT

-95-

 

derivatives, while the long component decreases with respect to its value in water. While
the interactions of boc-TRP are likely mediated by electrostatic forces, interpretation of
its lifetimes is complicated by the complex and opposing manner in which the two
components are manifested. It is interesting to note at this point that bromide ions of
CT AB are a reported to quench the fluorescent probes HANS and DANS. 33 In contrast,
we have noted significant increases in lifetimes for the AT probes in CT AB with respect
to water and a less well defined effect on the lifetimes of the TRP probes in CT AB. This
means that either the probes reside within the micelles where bromide counterions do not
have access or that the chromophores are not highly susceptible to bromide quenching.
We believe that for the AT probes, the latter explanation is more likely; their emission
maxima indicate a highly polar environment because the probes reside near the
headgroup region of the micelles, an area rich in bromide counterions.

DD-TRP displays the most environmentally sensitive fluorescence response, with
unique fluorescence decay time constants for each system measured. The decay times of
boo-TRP and DD-TRP do not appear to be correlated to one another, in contrast to the
data for boo—7AT and DD-7AT. In comparison to the other indole containing species, the
fluorescence response of DD-TRP in Thesit® is the most unusual as was DD-7AT for the
azaindole probes. The short and long time components of DD-TRP in Thesit® are about
twice their value(s) in water (Table 4.2). In CTAB, the short time constant decay
comprises the majority of the fluorescence decay, in contrast to the behavior seen for 7-
azaindole-based probes. We note that this anomalous behavior is not seen for the
reorientation data (vide infra) and further investigation will be required to identify the

basis for these findings.

-96-

 

The fluorescence of the neutral chromophore NATA exhibits single exponential
decay kinetics with a decay time constant of ~3.1 ns in water, CT AB and Thesit® (Table
4.2). In SDS-containing solutions, NATA exhibits a two-component population decay.
The lifetimes recovered for NATA and DD-TRP in SDS are similar. We believe this
situation to be coincidental; the indole chromophores are not experiencing the same
environment, as seen by the large difference in the steady state emission (19 nm) and the
relative contributions of each lifetime component for these two probes.

To better understand the interactions between the probes and micelles, we have
investigated the rotational diffusion dynamics of the chromophores in aqueous solutions
of micelle-forming surfactants. Time resolved fluorescence depolarization experiments
provide information on the rotational diffusion of the probe molecules. These
measurements are sensitive to and can provide information about intermolecular
interactions over the length scale of the probe molecule. 34'” Reorientation data provides
direct information on probe-surfactant association. We can distinguish whether or not
probes are interacting significantly with micelles by comparing the dynamical response
of the probes in water and micellar solutions.

There are many studies in the literature of the rotational diffusion of organic
chromophores in neat and selected binary solvent systems. ”'36 Micellar environments

37““ and are a special subset of binary systems. The theory of

have been examined before
rotational diffusion is well established and, because the relevant molecular processes
proceed on a time scale that can be accessed experimentally, it has found wide use as a

means of probing intermolecular interactions and transient solution-phase organization.

The time scale of the events determining reorientation dynamics are short and, as a result,

-97-

many interactions occur over the course of the chromophore’s reorientation resulting in
the overall experimental observation being a representation of the average molecular
environment experienced by the probe. For reorientation measurements in binary
solutions it is often found that the molecular-scale interactions responsible for the
dynamical behavior of the probes are not dictated by the bulk properties of the solvent.
In polar systems, the relevant intermolecular interactions may include dipolar
interactions, dielectric friction, ‘2 and under certain circumstances, formation of solvent-
solute complexes. We consider our data on the reorientation of NATA, and 7AT and
TRP derivatives in micellar environments. Our measurements point to the existence of
strong interactions between certain of the probes and the micelles and we interpret our
reorientation data in the context of a hindered rotor model shown previously to be
appropriate for such systems.

We use a pulse of vertically polarized light to excite an ensemble of
chromophores in solution and monitor the time-dependent decays of the polarized
components of emission. We combine the emission intensity data polarized parallel and
perpendicular to the excitation polarization and generate the induced orientational

anisotropy function, R(t),

I"(r) - I 1 (r)
I"(r) + 211(1)

 

R(t) = (4.1)

where Iu(t) and 11(t) are the time-dependent emission intensities. The function R(t)
represents the re-randomization of the ensemble of excited molecules. In these
measurements the data is compared to various models and theories in an effort to relate
the functionality of the R(t) decay to the immediate environment of the chromophore.

Theoretically, R(t) can contain up to five exponential decays and its form depends on the

-93-

shape, size, flexibility of the labeled molecule, and the angle between the transition

43 In simple homogeneous environments a single

moments of the chromophore.
exponential decay is commonly recovered and it is unusual to observe more than two
decays. The anisotropy of all the chromophores we examine in this study decay by a
single exponential in water and the anisotropy decay of DD-AT in water is shown in
Figure 4.4. These same probes displayed marked changes in their anisotropy decay in at
least one type of micelle solution with exception of the free amino acids that remain

single exponential. The experimental anisotropy functions are seen to change from single
to decays with two exponential components (see Figure 4.5) in micellar solutions,

R(t) = R1 (0) exp (-t/2'l ) + R2(0) exp (—t/2'2) (4.2)
and a central issue is the assignment of origin of the biexponential behavior. It is
significant that we observe different functionalities for the anisotropy decay in water and
in micellar solutions. This finding points to the incorporation of the probe(s) into
micellar environments and we thus need to consider two different models to account for
our experimental findings.

In water, a one-component reorientation decay is consistent with our earlier
findings for AT derivativeszo‘24 as well as one-component decays reported for tryptophan
and NATA. 44 These simple decay data may be interpreted in the context of the modified
Debye-Stokes-Einstein (DSE) model, 25°27

R(t) = R(O) exp(-t/r0R)
(4.3)

I = M
0" kBTS

-99-

 

where ton is the reorientation time constant, 77 is the solution bulk viscosity, V is the
solute hydrodynamic volume, 45 f is a friction coefficient to account for the solvent-solute
boundary condition and 8 accounts for the non—spherical shape of the volume swept out

by the reorienting molecule.

0.100 r

0.075 -

0.050 -

R(t)

0.025 -

 

0.000

 

 

 

 

Figure 4.4 Experimental anisotropy decay of DD-AT in water (data points) and best-fit
line to a single exponential decay function.

We now consider the reorientation of the probes in rrricellar solutions. Comparing
the reorientation data gathered in water to the data for the micellar solutions allows us to

determine whether or not the chromophores are associating or incorporating in some

-100-

0.14 r-

0.12 '- l

0.10 l-

0.08 -

0.06 -

R(t)

0.04 -

0.02 -

0.00

 

 

'0'02 T 1 1 1 1 J

l L l

0 2000 4000 6000 8000 10000

 

 

 

timc(ps)

Figure 4.5 Experimental anisotropy decay of DD-AT in ThesitO solution (data points)
and best fit line to a biexponential decay function.

manner into micelles. In cases where a single exponential anisotropy decay is seen in
micellar solution, the location of the chromophore can be open to question. We assert a
change in the functionality of anisotropy in micellar solutions indicates incorporation into
micelles. We observed that certain probes display biexponential anisotropy decays
signaling incorporation into micelles, and we need to choose an appropriate model to aid

interpretation of the observed dynamics.

-101-

'ZOI‘

ul S98ueqo ,(q Aldurts .10} p91unoooe 9q rouueo saouaiajjlp 98an pue ‘suotrnlos rumoejsns
ul soqord 98an .10; sruersuoo curl) uotreruauoar out ut soouarajjlp alqernse9ur are
9.191”, 'suorrnlos [[9 ul sKeoop .(dorrosrue [elru9uodxo 918uts llqtho (ml pue .LVL 891:)st
oluououlmz GILL “5°17 alqel ul u9A18 are @118911 .1. pm: 8V 1;) ‘sqs SOHOOILU our pue 1918M
ur saqord 91p jo uotreruauoar .10; saldorroslue 9urn-oraz pun slumsuoo 9mg 9141,
'919q uo nodal 9M SLUQJSAS our .10} areudordde sl
[9poul r9qu9u 1qu 18988ns sopnrlufieur .1;qu pue 19110991 9111 sruersuoo cum [mu9uru9dX9
9111 '9[[9otur 91p jo 9109 9111 111 uoliour aroqdoulorqo 1mm poreloosse ruersuoo curl:
Aesop 91p 8; 81 9.1911116 ‘patruauodxa 918uls e ‘ l.(lfilr + l.Wr) jo tuetsuoo own a qum sKeo9p
,(dorrostue 9q1 ‘lopow slur u] '9urnlon pouguoo 1qu utqum 910101 or 99.1} St pue ouaorur
a jo 9100 our ut p9reool st aroqdourorqo out ‘9seo puooos 9111 111 W; ‘ruersuoo own a ,(q
pozuaroereqo st uotroul slur pun Kiltuo plfiu e se 9llaorur our or paqoerre st oroqdourorqo
91p 919111111 9883 out it; alloorul (1891.19qu) 91p jo uolrour 91p sruosardar rumsuoo
curl) Recap ,(donoslue 9141 '898129 om Sutmouq 91p ur [etru9uodx9 alfiuls e se Aesop or
parotpard 9q ueo (1)3 or» 'p919pulq sl uonoru [euolteror 9.1qu IU9tuuorlAu9 ua ut sfieoop
(1)3 jo uolrerordsarut out or alqeoudde 9q Item 1qu slapour [2.191198 are 9.191”,
'9.tmon.ns mjnoojour 9n9ortu purl 9qo.1d uo
Klaxlrlsuas pu9d9p sotureufip OSOILL '9u9oltu 91p jo 9912an our 880.198 .10 ul 9qo.1d paroqrar
.IO p9qrospe 9m jo uolrour [euotrelsuen pue 9[[9:>ltu 91p jo uolfiar p919u1891 e u; 9qo.1d 9qr
jo ructuonour 9pn19u1£eur 9119mm 91p ut solumufip 9qo.1d GILL '9[n99[our 9qo.1d pateroosse
91p jo sctureufip 91p jo 1u9pu9d9put st uottnlos ut 9[[99ttu 91p jo uolrmor 9m '9[[9oltu
our ulqum 9qord 9111 go sotureurfp JB[1'109[OLLI our pue anooltu 91p jo uolreror our qroq

worj srlnsar oncotur 0 ul alnoojoul 9qo.1d a 50 uotrezlmlodap 99uoosaronu 9111‘

Table 4.3 Fitted decay time constants and zero-time anisotropies for the systems
indicated in the left column The experimental data were fitted to the function:
R(t) = R.(0)exp(-t/1.'.)+R2(0)exp(-t/1:2). For probe/solution systems with one decay
constant indicated, R2(0) = 0.

 

 

 

R1(0) TOR (ps) R2(0)(PS) TOR; (ps)
AT Aqueous .063 1.01 4115
CTAB .065 1.005 6318
SDS 0.07 1.005 5415
Thesit® 0.064 5014
boc-7AT Aqueous 0.0611005 6318
CTAB 0.0411003 256118 0.051.002 21221120
8138 0.0641003 6616
Thesit® 0.061.005 7317
DD-7AT Aqueous 0.0661002 9815
CTAB 0.0331002 196117 0.0311002 1337197
SDS 0.0321004 85119 0.0461004 498147
Thesit® 0.0471002 196110 0.0881002 2975181
TRP Aqueous 0091.01 5115
CTAB 0.11.01 7017
SDS 0.081.01 6717
Thesit® 0.08101 4716
boc-TRP Aqueous 0.1 11.02 8018
CTAB 0.05101 146121 0.07101 13411166
SDS 0.14102 124110
Thesit® 0.082102 101115 0.068102 5291.04
DD-TRP Aqueous 0091.01 1 18115
CTAB 0.0511005 205117 0.0521005 1213156
SDS 0.061.008 223125 0.051.005 1557166
Thesit® 0.0351004 207143 0.091.003 1204157
NATA Aqueous 0. 11.01 7217
CTAB 0.111.01 7117
SDS 0.09101 76112 0.041.006 630186
Thesit” 0.091.01 73110

 

-103-

 

bulk solution properties; the bulk viscosities of the micelle solutions are essentially the
same as that of water. There must be some interaction of 7AT and TRP with the
micelles, but the nature of the interaction and its persistence time cannot be addressed
directly by these data alone. The single exponential anisotropy decays seen for these
chromophores suggest that the persistence time of any probe-micelle interaction is short
relative to the reorientation time of the micelle.

The reorientation dynamics of substituted 7AT and TRP, and with NATA in
several of the micelles are fundamentally different than those seen for TRP and 7AT. For
the oppositely charged probe/micelle systems (boc-X'ICTAB+ or X-DDVSDS’), the
experimental R(t) function is a biexponential decay. It is tempting to interpret this
functionality as representing the interactions between probe and micelle being facilitated
by ionic interactions. While such interactions are likely important, we see evidence in the
data that (neutral) NATA also interacts significantly with ionic micelles. Thus ionic
interactions alone cannot account for the probe/micelle interactions. Additionally, when
the chromophores have structural features in common with the surfactants, such as C12
aliphatic chain, ionic interactions appear not to dominate the probe/micelle interactions.
This finding is consistent with the existence of a double layer structure at the interface
between amphiphile head groups and the bulk medium. Screening from such a double
layer structure can serve to mediate the importance of ionic interactions.

The anisotropy functions of the cationic probes DD-7AT and DD-TRP exhibit
two-component exponential decays in all micellar solutions. The existence of two
reorientation time constants could result either from the probe partitioning into two

distinct environments or from the probe residing in a single environment and the two-

-104-

component reorientation dynamics resulting from the effective rotor shape of the probe.
An examination of the fluorescence lifetime data in conjunction with the reorientation
data allows these possibilities to be distinguished. The azaindole-containing probes
exhibit single exponential emission intensity decays in water. The fluorescence intensity
decay of DD-7AT in Thesit® displays a major component (90%) with a time constant of
924 1 11 ps and a minor component (10%) with a time constant of 2407 1 70 ps. The
short component for this system is the same as that recovered in pure water, 901 1 10 ps.
The lifetime data suggest a small fraction of probe being incorporated into micelles, with
the majority of the probe in water. Thus the reorientation data are indicative of the probe
partitioning into two environments; one on or within the micelle and the other dominated
by bulk water. The fact that the fast reorientation time of DD-7AT in the micellar system
(~200 ps) is twice that measured in water (~100 ps) indicates that the probe resides in an
aqueous environment that is influenced substantially by the presence of the micelles. It is
possible that the fast reorientation time constant for this system is the result of rapid
association/dissociation kinetics for probe/micelle interactions. DD-7AT exhibits a two-
component anisotropy decay in CT AB and SDS micellar solutions, as does boo-7AT in
CT AB. The variations in anisotropy decay time constants for these systems point toward
there being significant and relatively short-lived probe interactions with the micelles.

The analogous indole probes display double exponential anisotropy decays in
micellar solutions. The recovered time constants for the indole and azaindole probes
show similar trends but considerably different time constants in their reorientation
dynamics. Intuition suggests that these chromophores would exhibit similar reorientation

behavior. The structural similarity of TRP and 7AT derivatives is central to the use of

-105-

 

7AT as an alternative to TRP as a probe in biological systems. While 7AT has been used
effectively in situations where biological activity is retained, the rates of activity changed
significantly upon substitution. Another indication of the significant differences 7AT—
substitution introduces into biological environments is the experimental finding that
certain 7AT-substituted proteins can be crystallized, where the corresponding non-mutant
tryptophan-containing proteins have not been crystallized. 50 It appears that the presence
of the ring-bound N at the 7-position in 7AT plays an important structural as well as
spectroscopic role in determining the properties of this molecule.

When two decays are seen in a micellar system, it is reasonable to assume that the
decays correspond to fast, restricted motion within the micelle or at its surface, and
slower global motion of the micelle. The recovered short time constants span a range
from 85 to 260 ps and the long components vary from 0.5 to 3 ns. The probes are not
bound rigidly in the micelle or free to rotate within the core because they do not display
single exponential dynamics or possess time constants consistent with these scenarios.
The short component depends on the specific probe/micelle system and the time
constants recovered are not inconsistent with those expected for restricted motion in or on
a micelle. For all of the systems we report here, the long decay time constant never
approaches that expected for global micelle motion if the probes are attached rigidly to or
contained within a micelle. The correlation times expected for the micelles, m, are listed
in Table 4.1. These correlation times were calculated using Eq. 4.3 in the stick-limit (f=l,
S=l) using n. for the radius of the micelles to determine volume. These predicted
correlation times range from several ns to tens of ns, depending on the size of the micelle.

The slow time constants we measure experimentally are not consistent with the motion of

-106-

 

the free probes. These findings leave quasi-translational motion of the chromophore
along the outer extent of the micelle as the only physical motion consistent with this time
constant. 2°

Owing to the structures of the probes and micelles used here, there are two
significant types of interactions that can occur. These are ionic and dispersion
interactions, and they can act either in concert with one another or in an opposing
manner, depending on the specific system. It is evident from the experimental data that
there is substantial interaction between the oppositely charged probes and micelles and
when the probes share common structure with micellar constituents. For the cationic
probes DD-7AT and DD-TRP, we observe qualitatively similar behavior in cationic,
anionic and neutral micelles, with the differences in the time constants between systems
likely being related to the characteristic persistence times for the dominant interactions.
We take these findings to indicate that for C12 aliphatic chains interacting with micelles
of similar length, the dispersion forces compete effectively with ionic forces and
dominate the observed dynamics. This finding can be reconciled if we consider that the
nominally charged probe and micelle moieties are likely screened efficiently by spectator
ions, leaving dispersion interactions as the dominant forces in these systems. For the
anionic boc-7AT and boa-TRP derivatives, we recover two—component anisotropy decays
in cationic CT AB micelles, single component decays in anionic SDS micelles and in
water, and probe-dependent behavior in the neutral Thesit® micelle. We understand the
behavior of these probes in CTAB and SDS as being mediated primarily by ionic
interactions, and the different behavior of the two probes in Thesit® results from the

subtle structural difference between the chromophores. We have discussed this issue

-107-

 

I

above. The neutral probe NATA appears to remain predominantly in the aqueous phase
of all micellar systems, with measurable contributions from micelle-incorporated
chromophore being seen only for SDS.

We apply a single formalism to the systems that exhibit two—component
anisotropy decays to explain their physical behavior. The fundamental physics
responsible for the data we observe is essentially the same for all of the micellar systems,
with differences in the details being associated with the specific identities of the micellar
constituents. We summarize our findings as follows: (1) 7AT and TRP exhibit only
small changes in tea for each surfactant solution compared to their behavior in water,
indicating that the probe molecule itself does not associate substantially with the micellar
structures. This finding is corroborated by the lifetime and steady state emission data.
(2) The number of anisotropy decays seen for any of the probes studied here is an
indicator of the extent of probe/micelle interactions for that system. Two-component
anisotropy decays indicate strong probe/micelle interactions and one-component
anisotropy decays suggest that the probe resides primarily in the aqueous environment.
(3) The details of the probe/micelle interactions are determined by the balance between
ionic and dispersion interactions, where the double layer serves to attenuate ionic
interactions.

We consider next the time constants we obtain experimentally and how to
reconcile these data with the structural features of the systems we have investigated. For
the case where the chromophore is bound near the surface of the micelle, a biexponential
decay is expected, with time constants 1w arising from restricted motion analogous to that

of a hindered rotor, and W for overall reorientation of the micelle. ‘8 If translational

-108-

motion of the chromophore about the surface of the micelle is considered, a third decay

component, rd, is expected. 38 The anisotropy function derived from this model is given

by Eq. 4.4,

R(t) = 11(0)[s'2 +(l-S'z)exp(-t/1:w)]exp(—t(l/td +l/1M)) (4.4)
where R(O) is the zero time anisotropy, S’ is an order parameter which is a measure of the
equilibrium orientational distribution of the chromophore transition moment. Because
the micelles are relatively large the additional orientational relaxation imposed on the
probe only affects the slow decay component due to translational diffusion. These
motions take place independently and their decay rates can be added thus Eq. 4.4 does
not include these motions explicitly. Depending on the complexity of the model used in
the interpretation of the experimental data, it is possible to over-interpret the information
present. We are cognizant of this possibility and we choose to use the simplest model
that accounts for the features of the data.

The expected reorientation time constant for a micelle is several nanoseconds and
we do not observe any such long time anisotropy decay in our data. The possible reasons
for this finding are either that our measurements are not sensitive to such a long-time
component, which is not the case, or that the persistence time of the micelle-probe
association is substantially shorter than the reorientation time of the micelle. The latter
explanation is consistent with interactions between the probe and the micelle that are
characterized by a relatively small driving force. We consider that the interaction
between the probes and the micelles is dominated by ionic interactions for the smaller
charged probes and by dispersion-mediated penetration of aliphatic moieties with the

lipophilic portion of the micelle for DD-7AT and DD-TRP. In all cases, the

~109-

 

chromophore is in closest proximity to the micelle head groups, and this portion of the
probe is capable of motion at or near the surface of the micelle. We treat this motion as
translational diffusion on a spherical surface. Eq. 4.4 describes the general case of the
anisotropy decay for three independent motions and we can apply the following
simplifications to bring this model into consistency with the experimental data,

“1,, =l/Islow -l/1:M (4.5)
l/‘rw = l/tfast — 1/1‘I —l/1:M = 1/1:fast - l/tslow (4.6)

We determine values of m from the DSE equation at 20 °C, or we assume 1M = no, but in
either case we are insensitive to this quantity experimentally. The translational diffusion
coefficient, D., can be related to rd through the radius of the micelle, rh, (Table 4.1) over
which the probe diffuses, by eq 4.7. 4

D, = 132/614 (4.7)
This diffusion constant is related to the slower of the two motions we sense. The fast
time constant is related to the rotational motion of the chromophore about its tethering

146'48 assumes the molecule is

bond to the micelle. The wobbling-in-a-cone mode
constrained within a cone of semi-angle 00 with the rotational motion of the molecule
being described by a wobbling diffusion coefficient, Dw. The value of DW is related to

our experimental data through the quantities 1w and 5’,

DW = {1w (1— S'z)}"[—cos2 60(1 + c0860)2{ln[(1+ cos60)/2]
+(l-cost90)/2}{2(1—00860)}"+(1—c0860) (4.8)
x(6+ 8cosl90 —cosz 190 —l2cos3t9o - 7cos‘ 60)/24]

where the cone semi-angle 00 is related to the order parameter through

S ' = 0.5 cos 60(1 + cos 60) (4.9)

-110-

 

Table 4.4 Values of quantities extracted from the hindered rotor model. 8’ is the order
parameter, determined from fits to Eq. 4.4.

 

 

 

. , 00 1).. x 109 D. x 10'9
probe solutron 8 (degrees) (S4) (mzls)
AT Aqueous 90 4.07

CTAB 90 2.65
SDS 90 3.09
Thesit® 90 3.33
boo-7AT Aqueous 90 2.65
CTAB 0.741 35 0.335 0.456
SDS 90
Thesit® 90 2.65
DD-7AT Aqueous 90 1 .70
CTAB 0.696 39 0.496 0.761
SDS 0.768 33 0.858 1.36
Thesit® 0.807 30 0.350 0.640
TRP Aqueous 90 3.27
CTAB 90 1.67
SDS 90 1.34
Thesit® 90 3.55
boc-TRP Aqueous 90 2. 10
CTAB 0.764 34 0.546 0.760
SDS 90 1.34
Thesit® 0.673 40 0.976 3.81
DD-TRP Aqueous 90 1.41
CTAB 0.71 1 38 0.440 0.845
SDS 0.674 40 0.467 0.381
Thesit® 0.849 26 0.232 1.65
N ATA Aqueous 90 2.3 1
CTAB 90 2.35
SDS 0.555 48 1.97 1.15
Thesit® 90 2.28

 

 

00 calculated according to Eq. 4.9.

190 =cos' 0.5 1+8S —l . Uncertainty
1 J—"

in the determination of S’ is taken to be 1 5%.

-lll-

The values for S’, 15.0“, and If“! were extracted from the experimental data using Eq. 4.4
and from this information, the quantities 61). DW, and D‘ were calculated for each
surfactant system (Table 4.4). The values of Dw for probes yielding single exponential
decay were calculated using Dw = (61011)". The order parameters in these solutions are
zero and the equilibrium orientational distribution is random because the probes are
assumed to reorient freely.

The values of the order parameter, S’, vary from 0.56 to 0.85 for reorientation of
the probes in the micelles. The order parameter is a measure of the time averaged
orientational distribution of the chromophores and can take on values from zero for a
fully random orientational distribution, to one for rigid, highly ordered systems such as
crystals. The values recovered for the order parameter indicate the probes do have
preferential orientation with respect to the micelle surface. On the basis of the S’ values
we extract from our data, the most motionally restrictive environment is formed by
Thesit® for the probes DD-TRP and DD-7AT. These probe/micelle systems displayed
among the largest changes in reorientation times compared to their behavior in water.
These probes likely interact significantly with the polyethylene glycol mantle of the
Thesit® micelles. The lowest value of S’ is seen with the NATA/SDS system.
NATA/SDS also yields the largest wobbling diffusion coefficient, approaching that of
NATA in water. These complementary pieces of information point to NATA being only
weakly associated with the micelle, with the most likely interaction occurring at the
micelle/water interface.

The recovered diffusion coefficients for the probes in water vary inversely with

probe size as expected. It is also useful to consider the magnitude of D., the translational

-112-

 

diffusion coefficient along the micelle/water interface. The probes exhibit moderately
fast translational diffusion at the micelle/water interfaces. The self-diffusion behavior of

the ionic surfactants used in this study has been investigated previously, with values for
D1 the lateral diffusion of surfactant molecule in micelles reported on the order of (0.2-
1.5) x 10"0 m2 s". 5'52 The translational diffusion constants we recover for the probes
used in this study range from ~ 10'9 -10'10 m2 s", with the smallest value for the
oppositely charged probe/micelle systems. The smallest value of D. we recover is for
DD-TRP in SDS. This probe can interact significantly with the nonpolar interior regions
of micelles, with the strength of interaction being similar to that of a constituent
surfactant molecule. The combination of strong dispersion interactions with favorable
ionic interactions leads to a probe/micelle system that are relatively tightly bound. We
note also the relatively slow translational diffusion constants seen for the boc-derivatized
probe/CTAB systems, which we attribute to the existence of attractive ionic interactions.
The largest translational diffusion constants we measure are for probes in (neutral)
Thesit® micelles. The order parameter data in Thesit® vary considerably depending on
probe. For boo-AT a low order parameter and large D. value both point to the absence of
extensive order in the micelle a result consistent with the physical picture of Thesit®
micelles. DD-Trp has the highest order parameter indicating a restrictive environment
immediately surrounding the probe but also the second fastest translational diffusion in
the study. We explain this apparent contradiction by suggesting that the chromophore is
indeed in a highly ordered environment, drawn relatively deeply into the Thesit® micelle
by its aliphatic moiety and with little opposing force due to the probe’s overall

hydrophobic nature and absence of ionic interactions competing near the surface mantle

-ll3-

region of this neutral micelle. As a result, the experimental reorientation time is higher
than expected since, in this calculation, it is assumed that the probe is at or near the
surface of the micelle. The DD-AT probe yields differing results from that of its TRP
analog. It too has a high order parameter but with much lower translational diffusion
constant than DD-TRP. This is a manifestation of the differing physical interactions of
these probes induced by the N7 in the ring of 7AT. While structurally very similar, this
subtle difference in the chrom0phore results in its association with the hydrophilic
alcohol terminated, polyethylene glycol portion of the ThesitO micelle. The Azaindole
chromophore is well known to hydrogen bond in alcohols and differences in D. of a
factor of two have been attributed to its propensity for hydrogen bonding between the
chromophore DPP with TX-lOO micelles as compared to the structurally similar non
hydrogen bonding DMDPP.53 The experimental data, in conjunction with the model we
use, allow the interpretation of our findings based on a balance between ionic and

dispersion interactions.

-114-

4.4 Conclusions

We have studied the steady-state and time—resolved optical response of several
indole and azaindole derived probes in water and aqueous micelle solutions. The charge
and structure of the probes dictate interactions with micelles. The reorientation dynamics
of the probes were consistent with the model of wobbling-in-cone dynamics with
translational diffusion of the probe along the micelle/water interface for SDS and CT AB,
and within the mantle of Thesit® micelles, along with slower rotation of the micelle in all
cases. Our experimental data reveal the ability of ionic interactions to mediate
probe/micelle interactions and that van der Waals interactions can overcome these forces.
For the case of oppositely charged probe/micelle systems we see the strongest
interactions because ionic and van der Waals forces are acting in concert. Probes with
structure in common (aliphatic chain) interact with micelles, regardless of probe charge.
Our experimental reorientation data indicate transient interactions between the micelle
and the probe oligopeptides, with limited structural freedom for the probe molecules in
the apparently viscous micellar environment. Taken as a whole, this work demonstrates
that there is a clear balance between ionic and dispersion interactions in amphiphilic
systems, and that these interactions can be controlled in a straightforward and intuitive

manner through the structure of the probe and surfactant.

-115-

4.5 Literature Cited

1.

8.

9.

Lakowicz, J. R. Principles of Fluorescence Spectroscopy; second ed.; Plenum
Publishers: New York, 1999.

Petrich, J. W.; Chang, M. C.; McDonald, D. B.; Fleming, G. R. J. Am. Chem. Soc.
1983, 105, 3824-3832.

Rich, R. L.; Chen, Y.; Neven, D.; Negrerie, M.; Gai, F.; Petrich, J. W. J. Phys. Chem.
1993, 97, 1781-1788.

Rich, R. L.; Negrerie, M.; Li, J.; Elliott, S.; Thronburg, R. W.; Petrich, J. W.
Photochem. Photobiol. 1993, 58, 28-30.

Rich, R. L.; Smimov, A. V.; Schwabacher, A. W.; Petrich, J. W. J. Am. Chem. Soc.
1995, 117, 11850-11853.

Negrerie, M.; Bellefeuille, S. M.; Whitham, S.; Petrich, J. W.; Thomburg, R. W. J.
Am. Chem. Soc. 1990, 112, 7419-7421.

Brennan, J. D.; Clark, I. D.; Hogue, C. W. V.; Ito, A. S.; Juliano, L.; Paiva, A. C. M.;
Rajendran, B.; Szablo, A. G. Appl. Spectrosc. 1995, 49, 51-59.

Broos, J .; Ter Veld, F.; Robillard, G. T. Biochem. 1999, 38, 9798-9803.

Hogue, c. W.; Szabo, A. G. Biophys. Chem. 1993,48, 159-169.

10. Chen, Y.; Gai, F.; Petrich, J. W. J. Phys. Chem. 1994, 98, 2203-2209.

11. de Foresta, B.; Gallay, J .; Sopkova, J .; Champeil, P.; Vincent, M. Biophys. J. 1999,

77, 3071-3084.

12. Chattopadhyay, A.; Mukherjee, S.; Rukmini, R.; Rawat, S. S.; Sudha, S. Biophys. J.

1997, 73, 839-849.

13.Ladokhin, A. S.; Holloway, P. W. Biophys. J. 1995, 69, 506-517.

-ll6-

 

l4. Sengupta, B.; Sengupta, P. K. Biochem. Biophys. Res. Commun. 2000, 277, 13-19.

15. Tortech, L.; Jaxel, C.; Vincent, M.; Gallay, J .; de Foresta, B. Biochim. Biophys. Acta-
Biomembr. 2001, 1514, 76-86.

16. Borkman, R. F.; Douhal, A.; Yoshihara, K. Biophys. Chem. 1993, 203-211.

17. Beechem, J. M.; Brand, L. M. Ann. Rev. Biochem. 1985, 43.

18. Nozaki, S. Chem. Lett. 1997, 1-2.

19. Sheehan, J. C.; Cruickshank, P. A.; Boshart, G. L. J. Org. Chem. 1961, 26, 2525.

20. Kelepouris, L.; Blanchard, G. J. J. Phys. Chem. B 2002, 106, 6600-6608.

21. DeWitt, L.; Blanchard, G. J .; LeGoff, B.; Benz, M. B.; Liao, J. H.; Kanatzidis, M. G.
J. Am. Chem. Soc. 1993, 115, 12158-12164.

22. Tanford, C. J. Phys. Chem. 1972, 76, 3020.

23. Robson, R. J .; Dennis, E. A. J. Phys. Chem. 1977, 81 , 1075.

24. Kelepouris, L.; Blanchard, G. J. J. Phys. Chem. A 2000, 104, 7261-7267.

25. Hu, C.-M.; Zwanzig, R. J. Chem. Phys. 1974, 60, 4354-4357.

26. Perrin, F. J. J. Phys. Radium 1934, 497.

27. Debye, P. Polar Molecules New York, 1929.

28. Ashby, K. D.; Das, K.; Petrich, J. W. Anal. Chem. 1997, 69, 1925-1930.

29. Eftink, M. R.; A., G. C. .J. Phys. Chem. 1975, 80, 486-493.

30. English, D. S.; Rich, R. L.; Petrich, J. W. Photochem. Photobiol. 1998, 67, 76-83.

31. Guha Ray, J .; Sengupta, P. K. Chem. Phys. Lett. 1994, 230, 75-81.

32. Chapman, C., F.; Maroncelli, M. J. Phys. Chem. A 1992, 96, 8430.

33. Matzinger, S.; Hussey, D. M.; Fayer, M. D. J. Phys. Chem. B 1998, 102, 7216-7224.

34. Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1994, 98, 9411-9416.

-ll7-

35

36.

37.

38.

39.

41.

42.

43.

45.

47.

48.

49.

50.

51.

52.

53.

Jiang, Y.; Blanchard, G. J. 1995, 99, 7904-7912.

Gudgin-Templeton, E. F.; Kenny-Wallace, G. A. J. Phys. Chem 1986, 2896.

Chou, S. H.; Wirth, M. J. J. Phys. Chem. 1989, 93, 7694-7698.

Quitevis, E. L.; Marcus, A. H.; Fayer, M. D. J. Phys. Chem. 1993, 97, 5762-5769.
Maiti, N. C.; Krishna, M. M. G.; Britto, P. J .; Periasamy, N. J. Phys. Chem. B 1997,

101,11051-11060.

. Klein, U. K. A.; Haar, H. P. Chem. Phys. Lett. 1978, 531.

Blatt, B.; Ghiggino, K. P.; Sawyer, W. H. Chem. Phys. Lett. 1985, 47.
Kivelson, D.; Spears, K. G. J. Phys. Chem. 1985, 89, 1999-2001.

Chuang, T. J .; Eisenthal, K. B. J. Chem. Phys. 1972, 5 7, $094-$097.

. Davis, D. M.; McLoskey, D.; Birch, D. J. S.; Gellert, P. R.; Kittlety, R. S.; Swart, R.

M. Biophys. Chem. 1996, 63-77.

Edward, J. T. J. Chem. Ed. 1970, 47, 261.

. Kinosita, K., Jr.; Kawato, S.; Ikegami, A. Biophys. J. 1977, 20, 289-305.

Lakowicz, J. R.; Cherek, H.; Maliwal, B. P.; Gratton, E. Biochem. 1985, 24, 376.
Lipari, G.; Szabo, A. Biophys. J. 1980, 30, 489-506.

Szabo, A. J. Chem. Phys. 1984, 81 , 150.

Evrard, C.; Fastrez, J .; Declercq, J .-P. Acta Crystal., Section D: Biological
Crystallography 1999, D55, 430-435.

Ahlnas, T.; Sodennan, 0.; Hjelm, C.; Lindman, B. J. Phys. Chem. 1983, 87, 822.
Nery, 11.; Soderrnan, 0.; Canet, D.; Walderhaug, H.; Lindman, B. J. Phys. Chem.
1986, 5802.

Dutt,G. B., J. Phys. Chem. B. 2002, 106, 7398-7404.

-118-

Chapter 5

Conclusions and Future Work

5.1 Conclusions

This body of work has demonstrated the application of fluorescence probes to
interrogate solution phase interactions. We chose to use the biological probes 7-
azatryptophan and 7-azaindole because of their well-behaved photophysics, and
biological compatibility amenability. These molecules were used to investigate solutions
of aqueous adipic acid and their steady state and fluorescence emission decays were both
found to be sensitive to solute concentration. The time domain emission decays of the
two probes were fundamentally different and through comparison of lifetime
measurements in adipic acid solution and buffered solutions, it was evident that 7A1
forms a short-lived complex with adipic acid. The buffered solution measurements also
lead to the finding that both molecules are sensitive to collisional quenching in citric
acid/dipotassium hydrogen phosphate buffer.

Reorientation measurements showed that these molecules reorient in the stick
limit, in accordance with the DSE equation, and revealed the ability of 7AT to interact
with adipic acid, presumably through the carboxylic acid moieties. An increase in the
reorientation time of 7AT was observed near and above adipic acid saturation
concentration. 7AT was determined to form complexes with adipic acid though its
common functionality but the persistence time of these complexes is short relative to the

reorientation of the molecule.

-ll9-

‘ I

The information gathered for 7AT and 7A1 in the adipic acid solutions
demonstrates the utility of these probes for sensing and reporting on their environment.
The importance of probe structure in determining the ability to sense spontaneous solute
self-assembly was evident in the orientational relaxation response of these probes. These
data, along with previous knowledge from this group on the adipic acid system,
demonstrates the ability of the lock-and-key approach and criteria necessary to choose
appropriate probes for successful interrogation of carboxylic acid systems.

The steady-state optical response of 7AT incorporated into peptide oligomers was
investigated in water and aqueous micelles. The sensitivity of 7AT to the peptide
environment was evident in the steady state and fluorescence decays. This sensitivity in
water was indicated by slight changes in emission maxima and fluorescence lifetime with
increasing length of the 7AT-Val oligomer. Fluorescence lifetimes were sensitive to
surfactant species and a clear discontinuity in lifetime was observed in SDS above the
CMC, indicating peptide association with micelles.

The rotational diffusion time constants of 7AT-(Val)u peptide series were found to
depend linearly on the number of valine residues in the peptide oligomers. This result
agreed well with predicted values from the modified DSE model in the stick limit for the
theoretical volume of these peptides, when the oligomers are modeled as a sphere. This
result and the apparent lack of segmental motion agrees with the notion that the
hydrophobic valine residues fold back over the chromophore, restricting its independent
motion and affecting its molecular environment in water. This is supported by the shifts
noted in the steady state response and the time-resolved emission decays of 7AT
contained within the oligomers chain. The dynamics revealed a balance between ionic

and van der Waals forces that serve to mediate interactions between peptides and micelle

-120-

 

structures. The strong interactions were between the anionic SDS headgroups and the
cationic 7AT-Val.-5 probes. Conversely, only the longest cationic 7AT-Val“, oligomers
interact with the cationic CT AB, owing to their greater lipophilic valine structure
overcoming repulsive forces.

Investigating the balance between ionic and dispersion forces in probe-micelle
interactions led to further investigations where chromophore charge and structure were
varied systematically. Derivatives of tryptophan and 7AT were produced to yield probes
with net positive or negative charge and varying physical structure. The emission
characteristics of these probes varied greatly in micellar solutions compared to water.
Substantial blue shifts were seen for all oppositely charged micelle-probe combinations.
The rotational dynamics revealed that the indole and azaindole probes, although very
similar in structure, yield significantly different responses in certain micelles and are
likely due to the ability of 7AT to hydrogen bond with surfactant molecules.

The probes examined in this thesis, when interacting with micelles, were found to
exhibit motion consistent with hindered rotor dynamics and translational diffusion along
the surface of the micelle.M The application of this theory to the experimental data helps
to elucidate the location and environment of the probes within or at the surface of the
micelles. These results demonstrate the utility this class of biological probes possess for
investigation of a variety of systems if fictionalized prudently or incorporated into

peptides and proteins.

-121-

 

5.2 Future Work

Chapters 3 and 4 of this thesis have described the optical response of various
indole and azaindole containing peptides and analogs in micellar media. Through the
lifetime and reorientation data, along with application of theory related to guest probes in
micelles, we have gained insight into the factors effecting location and environment of
these probes in various micelles. Certain probe and micelle combinations displayed
significant interactions, although the reorientation times of the probes never approach that
of the expected reorientation times for the micelles as a whole. We attribute this mainly
to lateral diffusion of the probes and significant local rotational freedom of the
chromophore when contained in or at the micelle surface. This is a physically reasonable
picture of probe micelle interactions and although micelles serve as a convenient medium
in which to study peptide and probe interactions, they also introduce some ambiguity.
The micelles used in this study have been well characterized and are dynamic entities
resulting in micelle fragments, free surfactant, and an overall average size distribution of
whole micelles.

In order to understand the contribution of micelle dynamics and better understand
translational diffusion, molecular micelles (polymerized micelles) could be used.
Molecular micelles have been of utility as a psuedostatioary phase for both achiral and
chiral separations in capillary electrophoresis (CE).5'8 Polymerized micelles are widely
applicable to many separations as a great array of molecular micelles may be synthesized
to promote specific separations by tailoring of the head groups. They are produced in
solution by polymerization of micelle forming surfactants like N-undecylenyl-L-glycinate

(SUG) that have vinylic groups at their hydrophobic end as illustrated in Figure 5.1.5

~122-

 

Polymerized micelles have enhanced stability and rigidity and are controllable in size due
to covalent bonding between surfactants eliminating the normal dynamic equilibrium
between surfactant monomers and micelles. These properties are of great utility in that
unambiguous placement of probe molecules by covalent incorporation into the micelles
could be accomplished. Polymerizing probes into micelles characterized by varying
lengths of aliphatic chains could result in depth profiling of the micelles. Comparison of
probes covalently attached to those unattached would give a great deal of information on
the location of free probes and a more direct measure of translational diffusion in these
systems.

Continued systematic investigation of nonnatural amino acids in peptides is also
of significant interest. Specifically the effects of intramolecular environment on the
optical response of chromophores. Useful experiments would include probe species
situated centrally within an oligopeptide as well as at the N and C termini with varying

charge on the peptide.

-123-

 

M O'Na*
‘3’ 11%

(b)

Na*0’\(O

O
Na*0”k

{1‘2

Na+0‘ 0

NH 0

05/]

O'Na+

O
’Na*
0
NH - 1
. Na
0
°\ NH
0 O’Na+
NH
O
O
HN\ //0
Wows
+H3N
/ / \
O
O'Na+

Figure 5.1 Structures of (a) monomeric SUG and (b) polymeric SUG with probe

molecule incorporated.

-124-

 

5.3 Literature Cited

1. Periasamy, N. J. ofFluores. 2001, 11, 161-166.

2. Maiti, N. C.; Krishna, M. M. G.: Britto, P. J.; Periasamy, N. J. Phys. Chem. B 1997,
101, 11051-11060.

3. Kelepouris, L.; Blanchard, G. J. J. Phys. Chem. B 2002, 106, 6600-6608.

4. Krishna, M. M. G.; Das, R.; Periasamy, N.; Nityananda, R. J. Chem. Phys. 2000,
112, 8502.

5. Wang, J .; Warner, I. M. Anal. Chem 1994, 66, 3773-3776.

6. Palmer, C., P.; Terabe, S. J. Microcolumn Sep. 1996, 8, 115-121.

7. Shamsi, S. A. Anal. Chem. 2001, 73, 5103-5108.

8. Tabor, D., G.; Underwood, A. L. J. Chromatogr. 1989, 463, 73.

-125-

. . . . . . . . L . . . . . ,
.11 : .. ..L _. y. . . r. .1 . . 1. .. .._
whrtfitfiyfrrnrlrmwfl3’11.wa . R. 1.. . .L . It.“ 09.4.”..1 p15.” ”.3 a”

~

a 1“: 14.10,: .n 1.1 . Jam-is. .1... .1qu ..
o . .. . €12...— vr
.w V4.1... 1313...!

.111 :- .4l. .4. .. Q ZI-‘ulll. Int , . . 1
“~01... ... :o T .3 . .. a. a. 14 .r,....4.£.\«..w~.1ol..

.4 R

 

1.1.1.1. 1141...... ..
.. .. ion-t; a. .

 

. Eur...-

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

1111.111111111111111111