. ... tsuih

f.

e ‘ .
Li.

5:...

,1...
0.3.- .

 

 

.51 t.
.7.
,.

«V

.9 fin.»

3.3%,. L

:11:

.sQ—u.
. ‘0

a
Invfl. .

t,
L‘-.‘C
13......
. $.41 :
.

Ii:
.«qfiltfit

.511!- s X. b
k I ‘8
.
v. vac

$1.!

gamma:

g o ‘8!
x .1...
.. .dv'rh. Ii."
:3 1
Z.
1
.139 . . ..
Lana. u a. .Jfiw
than 23.17.33!
(Ix l . x
.ml.
9.3?» “4%.
.vxr. Maggi: immu-
. .t Mill..- .0.
2.....25... nun
«dwzii.bux.. . .5
up»:
‘1!- :1
g.

x

3.1.

4. . H .1.
bu....,v..:i
_

1.

373
‘1 . ‘ix

 

 

_LIBRARY
Michigan State
University

 

This is to certify that the
dissertation entitled

BIOPOLYMER-BASED OCULAR DRUG DELIVERY

Ph.D.

SYSTEMS

presented by

WEIPENG LIU

has been accepted towards fulfillment
of the requirements for the

degree in CHEMICAL ENGINEERING

 

 

 

K Cum, coo; \J mummy»

T‘ Major Professor’sSignature d
No v 0902,; 0267 3/

Date

 

MSU is an Affinnative Action/Equal Opportunity Employer

PLACE 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KthroleccaPresIClRCIDateDue.indd

BIOPOLYMER-BASED OCULAR DRUG DELIVERY SYSTEMS

By

Weipeng Liu

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY

Chemical Engineering

2008

ABSTRACT
BIOPOLYMER-BASED OCULAR DRUG DELIVERY SYSTEMS
By
Weipeng Liu

A major problem in ocular therapeutics is the attainment of an optimal drug
concentration at the site of action. Because of the specific structure of the eye, the
treatment of ocular diseases is often a drug delivery problem. This dissertation represents
a systematic study of molecular structure-property—performance relationship of
chemically modified starch-based ocular drug delivery systems.

The specific polymers investigated were Dihydroxyl Starch (DHS) and DHS
Esters including acetates and propionates. The DHS was prepared through a two-step
method. Starch was first oxidized by NaIO4 to form Dialdehyde Starch (DAS) and DAS
was reduced to form the corresponding hydroxyls with NaBI-I4. Afier this modification,
polymers with increased solubility in water were obtained. By varying the ratio between
starch and oxidant in the first step, the content of Dialdehyde groups was controlled;
thereafter 20DHS, 60DHS, and lOODHS were prepared. Among them, lOODHS was
further esterified to obtain DHS acetates (DHSA) and DHS Propionates (DHSP). During
esterification, the degree of substitution (D8) of DHSA and DHSP was controlled and in
this work DHSA and DHSP with DS equal to 0.1 and 0.4 were prepared.

The specific aspects investigated in this research were surface property,
mucoadhesion, rheological behavior, controlled drug release, and their effects on ocular
drug delivery systems’ performance in animal tests. (1) Surface property. Previous work

was focused on lowering surface tension (ST) to help spreading. In this work, we took the

ST’s effect on both spreading and adhesion into account and suggested an ideal ST
should be moderately lower than that of the cornea. (2) Mucoadhesion. A strong adhesion
with mucus in the eye helps a drug delivery system to achieve a prolonged retention time
and form a strengthened network that sustains the release of drug. In this work, the
mucoadhesion of synthesized polymers was studied through a typical rheological
approach. (3) Rheological properties. Flow properties were thoroughly studied by other
researchers. However, it is rare for ophthalmic solutions to undergo only steady state
shear rate in situ. Thus, it was necessary to investigate their viscoelastic properties as well,
which were omitted by previous researchers. Therefore, in this work both steady shear
and dynamic behaviors of synthesized polymers were studied to investigate their effects
on ocular drug delivery systems. (4) Controlled drug release. The drug release profile is
important for achievement of sustained drug release and they were studied with a
standard USP dissolution method in this work. None of the four concepts we investigated
is new and some of them have been introduced to the study of ocular drug delivery
systems, however, no work has been done to study their effects systematically. As the
first work of systematic study of biopolymer-based ocular drug delivery systems, we
studied how these properties were affected by molecular structures, how they should be
designed by engineering chemical structures accordingly, and how they affected the
performance of ocular drug delivery systems, both theoretically and experimentally. The
work described in this dissertation helps fill the knowledge gaps, which exist for the

design of effective ocular drug delivery systems with modified starch.

ACKNOWLEDGMENTS

1 would like to first thank my advisor, Dr. Narayan, for the opportunity to work on
such an interesting project and for his guidance, help, and support for this work. Thank
Dr. Townsend for her help on the medical aspects of the project. Thank Dr. Steffe for
providing lab, equipments, and answering rheological questions. Thank Dr. Drzal for
help on surface tension analysis. I would also like to thank all of my friends in the
department for helping me in the many ways that you did. And finally I would like to
thank my parents and my wife, Tingting, for all of the support they have provided me as

well as my wife’s help with statistics.

TABLE OF CONTENTS

 

 

 

 

 

 

ACKNOWLEDGMENTS IV
LIST OF TABLES -- VIII
LIST OF FIGURES X
CHAPTER 1 INTRODUCTION AND BACKGROUND 1
1.] INTRODUCTION .......................................................................................................... 1
1.1.1 Objectives ........................................................................................................... 1
1.1.2 Organization of this Thesis ................................................................................ 2

1.2 BACKGROUND ............................................................................................................ 3
1.2.1 Anatomy of the eye ............................................................................................. 3
1.2.2 Common eye diseases ........................................................................................ 5
1.2.3 Drug diffusion in the eye .................................................................................... 7
1.2.4 Tear film ............................................................................................................. 9

1.3 RECENT RESEARCH TOPICS ON OCULAR DRUG DELIVERY ......................................... 10
CHAPTER 2 SYNTHESIS AND STRUCTURAL ANALYSIS - 13
2.1 EXPERIMENTS .......................................................................................................... 16
2.1.1 DHS preparation .............................................................................................. I 6
2.1.2 DHS Esterification ........................................................................................... I 7
2.1.3 Carbonyl content determination of DAS .......................................................... 18
2.1.4 NMR ................................................................................................................. 19
2.1.5 FT IR ................................................................................................................. 20
2.1.6 DS determination by Titration ......................................................................... 20

2.1. 7 Molecular weight measurement ....................................................................... 21

2.2 RESULTS AND DISCUSSION ....................................................................................... 23
2.2.1 Yield of DHS .................................................................................................... 23
2.2.2 Carbonyl content determination of DAS .......................................................... 25
2.2.3 ”CNMR ofDHS .............................................................................................. 25
2.2.4 F TIR of DHS .................................................................................................... 28
2.2.5 DS Determination of DHS Esters .................................................................... 31
2.2.6 F TIR of DHS esters .......................................................................................... 33

2. 2. 7 Molecular Weight ............................................................................................. 34
CHAPTER 3 PROPERTY CHARACTERIZATION 36
3.1 EXPERIMENTS .......................................................................................................... 36

3.1.1 Surface properties measurement ...................................................................... 36

 

3.1.1.2 Method .......................................................................................................... 41
3.1.2 Mucoadhesion .................................................................................................. 42
3.1.2.1 Literature review ........................................................................................... 42
3.1.2.2 Method .......................................................................................................... 46
3.1.3 Rheology .......................................................................................................... 47
3.1.3.1 Steady shear rheological properties ............................................................. 48
3.1.3.2 Viscoelastic behaviors .................................................................................. 52
3.1.4 In vitro release test ........................................................................................... 55
3.1.4.1 Literature review ........................................................................................... 56
3.1.4.2 Method .......................................................................................................... 60
3.1.5 Stability study ................................................................................................... 60
3.2 RESULTS AND DISCUSSION ....................................................................................... 61
3.2.] Surface tension ................................................................................................. 61
3.2.2 Mucoadhesion .................................................................................................. 65
3.2.2.1 DHS solutions at the same concentration ..................................................... 65
3.2.2.2 DHS solutions with the same viscosity .......................................................... 68
3.2.2.3 DHSA Esters vs. lOODHS ............................................................................. 70
3.2.3 Rheological properties ..................................................................................... 73
3.2.3.1 Steady shear flow curve ................................................................................ 73
3.2.3.2 Viscoelastic behavior .................................................................................... 78
3.2.4 In vitro release profile of 0floxacin ................................................................. 87
3.2.4.] Release from isoviscous solutions ................................................................. 89
3.2.4.2 Release from solutions with different viscosity ............................................. 91
3.2.5 Stability Data ................................................................................................... 93
3.2.5.1 Storage stability ............................................................................................ 93
3.2.5.2 Effect of autoclaving ..................................................................................... 95
CHAPTER 4 PERFORMANCE CHARACTERIZATION 97
4.1 EXPERIMENTS .......................................................................................................... 97
4.1.1 EpiOcular in vitro test ..................................................................................... 97
4.1.2 In vivo irritation test ........................................................................................ 99
4.1.2.1 Assessment of ocular irritation after a single administration of the test
formulations .............................................................................................................. 99
4.1.2.2 Assessment of ocular irritation after multiple doses of the test formulations
................................................................................................................................. 1 00
4.1.3 Tear break-up time test .................................................................................. 100
4.1.4 In vivo Timolol release test ............................................................................ 101
4.1.4.1 In vivo drug release profiles ....................................................................... 101
4.1.4.2 Intra-Ocular Pressure test .......................................................................... 102
4.2 RESULTS AND DISCUSSION ..................................................................................... 103
4.2.1 EpiOcular ....................................................................................................... 103
4.2.2 In vivo irritation test ...................................................................................... 105
4.2.3 Tear Break-up time ........................................................................................ 105
4.2.3 In vivo Timolol release results ....................................................................... 109
4.2.3.1 In vivo Timolol release profiles .................................................................. 109

vi

4.2.3.2 IOP results .................................................................................................. 110

 

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 114
5.1 CONCLUSIONS ........................................................................................................ 1 14
5.2 FUNDAMENTAL CONTRIBUTION ................................. - ............................................ 1 15
5.3 RECOMMENDATIONS .............................................................................................. 115

REFERENCES 116

 

vii

LIST OF TABLES

TABLE 1: ESTERIFICATION OF lOODHS ............................................................................. 18
TABLE 2: DIALDEHYDE UNIT CONTENT OF DAS ................................................................ 25
TABLE 3: DS OF DHS ESTERS ............................................................................................ 32
TABLE 4: MOLECULAR WEIGHT BY GPC ............................................................................ 34
TABLE 5: INTRINSIC VISCOSITY .......................................................................................... 34
TABLE 6: TYPICAL BOND ENERGY ...................................................................................... 45
TABLE 7: SIMULATED TEAR SOLUTION .............................................................................. 60
TABLE 8: SURFACE TENSION RESULTS ................................................................................ 63
TABLE 9: REGRESSION OF STEADY SHEAR FLOW DATA ...................................................... 76
TABLE 10: VISCOELASTIC PROPERTY OF 20DHS ............................................................... 79
TABLE 1 1: VISCOELASTIC PROPERTY OF 60DHS ............................................................... 80
TABLE 12: VISCOELASTIC PROPERTY OF lOODHS ............................................................. 81
TABLE 13: VISCOELASTIC PROPERTY OF DHSA] .............................................................. 82
TABLE 14: VISCOELASTIC PROPERTY OF DHSAZ .............................................................. 83
TABLE 15: VISCOELASTIC PROPERTY OF DHSP] ............................................................... 84
TABLE 16: VISCOELASTIC PROPERTY OF DHSP2 ............................................................... 85
TABLE 17: REGRESSION DATA OF OFLOXACIN RELEASE FROM ISOVISOUS SOLUTIONS ..... 90

TABLE 18: REGRESSION DATA OF OFLOXACIN RELEASE FROM SOLUTIONS WITH DIFFERENT

VISCOSITY ................................................................................................................. 92
TABLE 19: T TEST RESULTS OF SOLUTIONS WITH DIFFERENT VISCOSITIES ....................... 92
TABLE 20: STORAGE STABILITY OF 3% 20DHS ................................................................. 93

viii

TABLE 21: STORAGE STABILITY OF 8.5% 60DHS .............................................................. 93

TABLE 22: STORAGE STABILITY OF 10% lOODHS ............................................................. 94
TABLE 23: STORAGE STABILITY OF 10% DHSAI .............................................................. 94
TABLE 24: STORAGE STABILITY OF 10% DHSA2 .............................................................. 94
TABLE 25: STORAGE STABILITY OF 10% DHSPl ............................................................... 94
TABLE 26: STORAGE STABILITY OF 10% DHSP2 ............................................................... 94
TABLE 27: VISCOSITY DROP AFTER AUTOCLAVING ............................................................ 95
TABLE 28: IOP CHANGE ................................................................................................... 1 12

ix

LIST OF FIGURES

FIGURE 1 ANATOMY OF THE EYE .......................................................................................... 3
FIGURE 2 SCHEMATIC REPRESENTATION OF TOPICAL DRUG DELIVERY MODEL .................... 8
FIGURE 3: SCHEMATIC REPRESENTATIONS OF THE STRUCTURE OF THE TEAR FILM. ............ 10
FIGURE 4: CHEMICAL STRUCTURE OF STARCH ................................................................... 13
FIGURE 5: PERIODATE OXIDAITON OF STARCH ................................................................... 14
FIGURE 6: BOROHYDRIDE REDUCTION OF DAS .................................................................. 14
FIGURE 7: CHEMICAL STRUCTURE OF GLUCOSE-C-DHS ..................................................... 15
FIGURE 8: ESTERIFICATION OF lOODHS ............................................................................ 16
FIGURE 9: CHEMICAL STRUCTURE OF DHS ESTERS ........................................................... 16
FIGURE 10: YIELD OF DHS ................................................................................................ 24
FIGURE 1 1: '3C NMR SPECTRUM OF CORN STARCH ........................................................... 26
FIGURE 12: ”C NMR SPECTRUM or 20DHS .................................................................... 26
FIGURE 13: 13C NMR SPECTRUM OF 6ODHS ..................................................................... 27
FIGURE 14: 13C NMR SPECTRUM OF lOODHS ................................................................... 27
FIGURE 15: FTIR SPECTRUM OF STARCH, 20DAS AND 20DHS ......................................... 29
FIGURE 16: FTIR SPECTRUM OF STARCH, 60DAS AND 6ODHS ......................................... 29
FIGURE 17: FTIR SPECTRUM OF STARCH, 100DAS AND lOODHS ..................................... 30
FIGURE I 8: FTIR SPECTRUM OF DAS WITH INCREASING DEGREE OF MODIFICATION ........ 31
FIGURE 19: 'H NMR SPECTRA or DHSA ......................................................................... 32
FIGURE 20: 1H NMR SPECTRA OF DHSP ........................................................................... 33

FIGURE 21 : THE FTIR SPECTRUMS OF DHS ESTERS .......................................................... 34

FIGURE 22: SCHEMATIC DIAGRAM SHOWING THE INTERFACIAL TENSIONS INVOLVED IN

SPREADING AN OPHTHALMIC SOLUTION OVER THE CORNEA ....................................... 37
FIGURE 23: SURFACE TENSION OF SOLUTIONS .................................................................... 64
FIGURE 24: MUCOADHESION OF 3% 20DHS ...................................................................... 66
FIGURE 25: MUCOADHESION OF 3% 60DHS ...................................................................... 66
FIGURE 26: MUCOADHESION OF 3% lOODHS .................................................................... 67
FIGURE 27: MUCOADHESION COMPARISON OF 3% DHS SOLUTIONS .................................. 67
FIGURE 28: MUCOADHESION OF 8.5% 6ODHS ................................................................... 68
FIGURE 29: MUCOADHESION OF 10% lOODHS .................................................................. 69
FIGURE 30: MUCOADHESION COMPARISON OF ISOVISCOUS DHS SOLUTIONS .................... 69
FIGURE 31: MUCOADHESION OF 10% DHSA] ................................................................... 70
FIGURE 32: MUCOADHESION OF 10% DHSA2 ................................................................... 70
FIGURE 33: MUCOADHESION OF 10% DHSPI .................................................................... 71
FIGURE 34: MUCOADHESION OF 10% DHSP2 .................................................................... 71
FIGURE 35: MUCOADHESION COMPARISON OF 10% DHS ESTER SOLUTIONS ..................... 72
FIGURE 36: RHEOGRAM OF 20DHS .................................................................................... 73
FIGURE 37: RHEOGRAM OF 60DHS .................................................................................... 73
FIGURE 38: RHEOGRAM OF lOODHS .................................................................................. 74
FIGURE 39: RHEOGRAM OF DHSA] ................................................................................... 74
FIGURE 40: RHEOGRAM OF DHSA2 ................................................................................... 74
FIGURE 41: RHEOGRAM OF DHSPl ................................................................................... 75
FIGURE 42: RHEOGRAM OF DHSP2 ................................................................................... 7 5

FIGURE 43: VISCOSITY COMPARISON OF DHS SOLUTION AT THE SAME CONCENTRATION” 78

Xi

FIGURE 44: VISCOSITY COMPARISON OF DHS ESTER SOLUTIONS AT THE SAME

CONCENTRATION ........................................................................................................ 78
FIGURE 45: LOSS TANGENT OF 20DHS .............................................................................. 79
FIGURE 46: LOSS TANGENT OF 60DHS .............................................................................. 80
FIGURE 47: LOSS TANGENT OF lOODHS ............................................................................ 81
FIGURE 48: LOSS TANGENT OF DHSA] ........................................................ 82
FIGURE 49: LOSS TANGENT OF DHSA2 ............................................................................. 83
FIGURE 50: LOSS TANGENT OF DHSPl .............................................................................. 84
FIGURE 51: LOSS TANGENT OF DHSP2 .............................................................................. 85
FIGURE 52: STRUCTURE OF OFLOXACIN ............................................................................. 87
FIGURE 53: UVN IS SPECTRUM OF OFLOXACIN ................................................................. 88
FIGURE 54: CALIBRATION CURVE FOR OFLOXACIN ............................................................ 89
FIGURE 55: RELEASE PROFILES OF OFLOXACIN FROM ISOVISCOUS SOLUTIONS ................... 89

FIGURE 56: RELEASE PERCENTAGE FROM ISOVISCOUS SOLUTIONS VS. SQUARE ROOT OF
RELEASE TIME ........................................................................................................... 90

FIGURE 57: RELEASE PROFILES OF OFLOXACIN FROM lOODHS AT DIFFERENT
CONCENTRATIONS ..................................................................................................... 91

FIGURE 58: NON ISOVISCOUS RELEASE PERCENTAGE vs. SQUARE ROOT OF RELEASE TIME

................................................................................................................................... 92
FIGURE 59: EPIOCULAR RESULTS OF 20DHS AND lOODHS ............................................ 104
FIGURE 60: EPIOCULAR RESULTS OF DHS ESTERS .......................................................... 104

FIGURE 61: TBUT COMPARISON OF DHS SOLUTIONS WITH DIFFERENT CONCENTRATIONS

................................................................................................................................. 106
FIGURE 62: TBUT OF ISOVISCOUS SOLUTIONS ................................................................. 106
FIGURE 63: TBUT OF COMMERCIAL PRODUCTS AND POLYMERS ...................................... 108
FIGURE 64: FIRST SET OF TIMOLOL RELEASE TEST ........................................................... 1 10

xii

FIGURE 65: IOP CHANGE vs. TIME ................................................................................... l 12

xiii

Chapter 1 Introduction and Background

1.1 Introduction

Drug delivery is usually a problem in ocular therapeutics because Of the specific
structure of the eye. Eye diseases can cause patients anxiety and discomfort, or even
worse, loss of vision or facial disfigurement. One of the most important concerns in
ocular therapeutics is the attainment of an optimal drug concentration at the site Of action.
Poor bioavailability of active agents from ocular dosage forms is mainly due to two major
constraints1 the precomeal loss factions and the relative impermeability of the corneal
epithelial membrane. Due to these physiological and anatomical barriers, only a very
small fraction of the drug, usually 1—5% or even less of the instilled dosez, is effectively
absorbed. While there are many topical ophthalmic solutions, all of them have very poor
bioavailability. Therefore, there is a need for ocular drug delivery systems with long
retention times, relatively high bioavailability, ease of use, ease of manufacture, and
overall acceptance by the patient.
1.1.1 Objectives

This project was focused on fundamental aspects of the ocular drug delivery
process as well as structure-property performance relationship of ocular drug delivery
systems using modified starch-based biopolymers. As such, this is the first work tO
systemically study modified starch-based ocular drug delivery systems. Generally, the
ocular drug delivery process can be divided into four subprocesses: (l) spreading or
wetting; this subprocess is important for minimizing irritancy and further forming of
mucoadhesion. It is affected and can be optimized by controlling the interfacial properties

Of the ocular drug delivery system. (2) Mucoadhesion; this subprocess is important for

prolonging the retention time in the eye, which is the bottleneck of most current ocular
drug delivery systems. (3) Retention: this subprocess is significantly affected by
rheological properties Of the ocular drug delivery system, which are important for both
reducing irritancy and increasing the retention time. (4) Optimized drug release; this
subprocess is important to achieve an effective treatment and to avoid any side effects.
Based on these design principles, 3 series of polymers with tailored chemical structures
have been synthesized to achieve Optimized properties for the starch-based ocular drug
delivery. Successful completion of this work should lead to improved design of modified
starch-based ocular drug delivery systems and will also help understanding fundamental
aspects of ocular drug delivery process and their relationships with molecular structures.
1.1.2 Organization of this Thesis

The thesis is divided into five main parts. In Chapter 1, the need for new ocular
drug delivery systems is addressed. This chapter also includes comprehensive reviews
and background information of current problems as well as current research activities in
this field.

In Chapter 2, detailed experiment procedures are given including the synthesis of
these novel polysaccharide-based polymers, which are water soluble and suitable for
Ophthalmic solution preparation. This chapter also includes detailed structural analyses of
the products using FT-IR, NMR, and other chemical methods.

In Chapter 3, property characterizations of ophthalmic solutions made of
synthesized polymers are described. Properties investigated include surface tension,
mucoadhesion, rheological properties, drug release, and stability.

In Chapter 4, the focus is on animal tests, including irritancy, tear break-up time,

and drug release tests to characterize the performance of selected candidates.

Chapter 5 draws conclusions based on the finished work and suggests future
research plans.
1.2 Background

The fact that drug delivery to the eye is an important and active research topic has
attracted many researchers into this field. In order to further understand this topic, in this
section, the anatomy of the eye, common eye diseases, drug diffusion model and the tear
film structure in the eye are discussed.
1.2.1 Anatomy of the eye

TO understand the unique drug delivery issues associated with the eye, the
structure of the eye is the first consideration. Figure 1 shows a schematic cross section of

the eye.

 

Iris

  

Cornea

a“... /

i . _,“ (”A .7 77
humor // \
Lens Retina

Conjunctlva

 

 

 

Figure 1 Anatomy of the eye
The three areas of interest in the eye with respect to topical application of drugs
are the cornea, the conjunctiva, and the nasolachrymal drainage system3. The cornea is

the main pathway for the permeation of drugs into the eye4. In terms Of drug delivery, the

cornea consists of three main barriers in series: the epithelium, the stroma, and the
endothelium. The outer epithelium is lipophilic in nature and consists of 5-6 layers of
cells. It is the most significant barrier to hydrophilic drug delivery. The stroma accounts
for approximately 90% of the corneal thickness and is a relatively Open hydrophilic
region. The final important region, the inner endothelium, consists of a single layer of
flattened cells and is in direct contact with the anterior chamber. Because of both the
hydrophilic and hydrophobic regions, the cornea provides an effective barrier to drug
transport.

The second major region Of the eye with importance to drug delivery is the
conjunctiva, which is a thin mucous membrane that lines the posterior surface of the
eyelids and the outer regions of the cornea. It is involved in the formation and the
maintenance of the precomeal tear film and it is involved in the protection of the eye. The
epithelium of the conjunctiva is somewhat different from that of the cornea. The human
conjunctiva is 15-25 times more permeable to hydrophilic drugs than the cornea and it
provides an alternative absorption path for drugs applied topicallyl.

The nasolachrymal drainage system accounts for most of the drug loss in the
precomeal region. It consists of three parts: the secretory system, distributive system, and
the excretory system. It is thought that tears are largely absorbed by the mucous
membranes of the ducts and that only a small amount reaches the nasal passagess. This
absorbtion by the mucous membranes provides another route of systemic absorption of
the applied drugs. The cul-de-sac of the eye normally holds 7-9 pl of tears and can
contain up to 20-30 ul if care is taken not to blink. The normal tear flow rate is 1 pl per

minute and the pH is maintained at 6.5-7.6. The high turnover of tear fluid and the

limited capacity of the sacs can lead to a high clearance rate of drugs once applied3.
1.2.2 Common eye diseases

In this section, we will review some common eye diseases which need topical
therapies by targeting the outer surface or anterior chamber of the eye. They are
introduced as follows.

Dry eye. According to the definition by Brewitt and Sistanié, dry eye is a disease
of the ocular surface attributable to different disturbances of the natural function and
protective mechanism of the external eye, leading to an unstable tear film during the open
eye state. This disease can be caused by tear deficiency or excessive evaporation and is
associated with symptoms of discomfort. The definition of dry eye has been further
expanded by Albietz to include any functional or component anomaly of the lids or
glands associated with tear production in which the quality and/or quantity of the tear
film is adversely affected and there is an inability to maintain a healthy ocular surface].
Clinically dry eye disease can be assigned to two major classes: aqueous deficient dry
eye, due to a reduced aqueous tear secretion; and evaporative dry eye. No matter what the
initial cause is, chronic dryness of the ocular surface results in an inflammatory reaction,
which is the key mechanism of chronic ocular surface injury. Patients often complain
about grittiness, foreign body sensation, burning, soreness, stinging, scratchiness,
dryness, blurry Vision, a “film over the eyes,” paradoxical reflex tearing, and
photophobia. Absolute tear deficiency can lead to blindness due to severe ocular surface
disease and attempts of surgical reconstruction frequently fail in this situations. The

global features in common for both forms of dry eye are (1) a set of characteristic

symptoms, (2) ocular surface damage, (3) reduced tear film stability, and (4) tear
hyperosmolarityg.

The goals of dry eye treatment are to reduce symptoms, to improve tear film
quantity and quality, and to reverse the ocular surface damage. Therapeutical approaches
include (1) tear substitution, (2) pharmacological stimulation of tear secretion, and (3)
tear preservation through reduction of tear evaporation or drainage"). The latter two
approaches are beyond the scope Of this study. We will focus our discussion on the tear
substitution approach.

The goal of using tear substitutes is to increase humidity at the ocular surface and
to improve lubrication. This approach is currently the most widely used therapy for dry
eye. It includes a variety of components to formulate a considerable number of
commercially available preparations. Cellulose ethers are most commonly used in dry eye
solutions and have good retention time on the ocular surface”, sodium hyaluronate has
been found particularly beneficial in corneal wound healing '2 , carbomers provide
excellent adhesive behavior and higher retention time,13 and recently lipid containing
drops aim to rebuild the lipid layer”.

Glaucoma is a group of disorders characterized by progressive damage to the eye
at least partly due to intraocular pressure damaging the optic nerve. Because glaucoma
comes in many forms, there is not a universal treatment for it. Treatments can be
categorized by either increasing outflow or by decreasing aqueous production.

Conjunctivitis is inflammation of the conjunctiva or the mucous membrane

surrounding the eye, also known as pinkeye.

Keratitis is inflammation Of the cornea characterized by loss of luster and
transparency, as well as cellular infiltration.

Iritis (anterior uveitis) is inflammation of the iris; can be caused by systemic
diseases (such as rheumatoid arthritis), systemic infections (such as measles, syphilis, and
tuberculosis), trauma, or idiopathic (unknown) sources.

1.2.3 Drug diffusion in the eye

For ailment of the eye, topical administration is usually preferred. The topically
administered ocular drugs have to reach inner parts of the eye to obtain therapeutic
effects. The transcomeal penetration is believed to be the major route for ocular drug
absorption. Diffusion is thought to be the process by which most drugs penetrate the
cornea.

As shown in Figure 2, topically applied drugs are cleared from the precomeal area
in three different ways: tear drainage, diffusion across the conjunctiva, and diffusion
across the cornea. Both tear drainage and diffusion across the conjunctiva account for
drug loss in the precomeal area.

Eyedrops mix with the tear fluid when they are administered to the precomeal
area. After administration the extra solution volume rapidly flows from the precomeal
area into the nasolachrymal drainage system. The drainage of instilled solution is rapid.

Iand

The drainage rate constant in rabbits increases with instilled volume to 0.31min'
0.82 min'1 for eyedrops of 5 ul and 50 ul, respectively. Futherrnore, the rate of solution
drainage decreases with elevated solution Viscosity and mucoadhesiveness.”

Another important route of drug loss from the precomeal area is drug diffusion

across the conjunctiva. Since the conjunctival permeability is fairly high compared to that

of the cornea most drugs diffuse across the conjunctiva easily.

 

 

 

Prec omeal Area

 

 

 

.I

 

 

 

 

 

 

 

 

 

 

 

 

  
 

 

 

i
Conjunctiva
Cornea Tear
Drainage
Anterior
V v

 

 

Systemic circulation

 

 

Figure 2 Schematic representation of topical drug delivery model

According to research done by A. Urtti's, a mass balance equation is induced to

describe the drug release in the precomeal area.

dQ/dt = Cpc (Cltf-t Clcj + Cleo)

where

dQ/dt: drug release rate from polymer matrix

Cpc: precomeal drug concentration

Cltf‘. drug clearance via tear turnover

Clcj: drug clearance from fluid to conjunctiva (Clcj = Permeabilitycj * Areacj)

Clco: drug clearance from fluid to cornea (Clco = Permeabilityco * Areaco)

Based on the Equation (1.1), the quantity of drug diffusing across the cornea can be
figured out as,
Cpc Clco = dQ/dt — (Cpc Cltf + Cpc Clcj) (1.2)

Therefore, the quantity of the drug absorbed through the cornea can be optimized
by controlling drug release rate from polymer matrix and/or by decreasing drug loss
through tear drainage and conjunctival absorption. In summary, the strategies to increase
the bioavailability of topically applied drug to the eye, involve (1) increasing contact time
between drug and cornea, and (2) increasing corneal permeability without a
correspondence increase in the conjunctival permeability.

Based on the above discussion, we concluded that both precomeal loss and
corneal permeation barrier account for poor performance with current ocular drug
delivery systems. In order to significantly improve bioavailability the drug delivery
system should have a prolong drug/comea contact time and/or high corneal permeability.
Some synthesized polymer candidates that can be utilized as drug delivery matrices to
potentially improve the bioavailability are discussed later.

1.2.4 Tear film

The production and turnover of tears is important for maintaining the health of the
ocular surface. Tears clean, lubricate, and nourish the surface of the eye and provide
physical and immune protection against infection and mechanical trauma. A small change
in tear film stability and/or volume will result in a significant alteration of the quality of
the retinal image; thus, maintenance of a stable tear film is essential to healthy vision.
The tear film is composed of three main components mucin, water, and lipids. More than

98% Of the total tear volume is water”.

The inner mucin layer is produced by conjunctival goblet cells and epithelial cells
of the conjunctiva and cornea. It allows for the wetting of the ocular surface and
stabilizes the tear film against the stresses exerted by blinking. The aqueous layer is
produced by the main and accessory lacrimal glands. It is responsible for carrying
essential growth factors to the epithelium and washing away the epithelial debris, toxic
elements, and foreign bodies. The outer lipid layer contains a variety of lipids that protect
the tear film against evaporation. In the classical View, the film is thought to have a three-
layered architecture. More recently, an aqueous—mucus gel with a mucin gradient has

been proposed.16

 

 

Air Air
Lipid layer Lipid layer
Aqueous layer
Mucln
conc. Aqueseu-Inucln gel

-Iliucus layer. .

     
 

Ocular surface epithelium Ocular surface epithelium

Figure 3: Schematic representations of the structure Of the tear film.
1.3 Recent research topics on ocular drug delivery
Scientists involved in ophthalmic pharmaceutical are facing the challenge to
improve ocular drug bioavailability from less than 1-5% to at least 15-20%.
Investigations aimed at improving topical bioavailability are being pursued along the
following principles. First is to minimize precomeal drug loss and maximize corneal

absorption by controlling drug release profile, prolonging drug contact time with ocular

10

surface, and/or transiently changing corneal structure. Second is the use of solid matrices
and drug delivery devices, which provide the controlled and continuous delivery of
ophthalmic active agents to the pre- and intra-ocular tissues. In this section, some typical
ocular drug delivery systems are discussed.

Aqueous gels/Hydrogels: Hydrogels, colloidal gels with water as the dispersion
medium, are of particular interest because they can offer controlled drug release
according to Specific needs. However, difficulty in sterilization and/or easy bacterial
contamination has limited their large-scale production and clinic use].

Bioadhesive polymer: Bioadhension refers to the attachment of a drug carrier to
a specific biological tissue for drug delivery purposes. Coating the external surface of the
globe of the eye is a thin film of glycoprotein referred to as mucus. Therefore,
bioadhension is also referred to as mucoadhension. Increasing the contact time in the
precomeal area appears to be governed by both the mucoadhensive agent as well as the
Viscosity effects of the polymer. Therefore, in designing the ocular drug delivery systems
using mucoadhensives, a vehicle needs to be found that imparts good mucoadhensive
strength as well as appropriate Viscosity at a low concentration.

Microparticles and nanoparticles: Particulate systems have the potential to
become promising for ophthalmic drug delivery by offering an approach to combine
extended drug release and improved patient compliance. However, formulation stability,
control of particle size, control of the rate of drug release, and large-scale manufacturing

of sterile preparations are still major issues in the development of Ophthalmic particulate

systems.

11

Penetration enhancers: In order to design a more specific penetration enhancer,
it is necessary to have a better understanding of membrane transport, physiology of tight
junction, etc. In addition, other approaches such as increasing residence time and
inhibition of metabolizing enzymes should be taken into account in conjunction with a
penetration enhancer.

Noncorneal route: Relatively high conjunctival and scleral permeability make it
possible to put the conjunctival/scleral pathway for the intra-ocular entry of drugs. Much
progress has been made in understanding the fiindamental basis of drug penetration Via
those noncomeal pathways. The greatest potential for the concept appears to be the intra-
Ocular delivery of drugs to treat posterior segment eye diseases, which currently has not
been effectively treated by topically administered drugs.

Drug delivery devices: Besides the “pulse entry” type Of drug release systems,
some systems providing controlled and continuous ocular drug delivery have been
developed. These systems can achieve therapeutic action with a smaller dose and fewer
Side-effects. Typical systems include inserts, implantable systems, and contact lenses.
However, patient resistance to surgery and placing an object in the precomeal region
should be given particular attention in order to improve overall patient acceptance.
Although a number of polymers have been utilized for ocular drug delivery, none of them
provides an optimal approach for this application. In this project, we synthesized a series
Of starch-based polymers based on the fundamental understanding of the ocular drug
delivery process and characterized their properties and performance through both in vitro

and in ViVO methods.

12

Chapter 2 Synthesis and Structural Analysis
The objective of this project was to synthesize a safe, water soluble, and
biocompatible material that would have the ability to prolong the drug-eye contact time
and give controlled drug release profile. Some natural polysaccharides exhibit some of
these characteristics, such as biocompatibility and water dispersibility, but they are not
optimized for the use of ophthalmic solutions. Waxy starch was chosen as the starting
material because of its abundance and current acceptance in pharmaceutical applications.

Its chemical structure is shown in Figure 4.

 

 

 

F- e '—
CHZOH CH20H CHZOH
O
OH OH
OH OH
OH OH
a: amylose
side chain I ' ' '
CHZOH Cl-lel-l
O
OH OH
main
I I I I I I I
chain
OH

 

b: amylopectin

Figure 4: Chemical structure of starch

13

In this study, our objective was to obtain a water soluble product with Opened
glucose rings in the starch backbone. This was achieved in two steps: (1) Oxidation of
starch with sodium periodate to produce Di-Aldehyde-Starch (DAS) and (2) Reduction of
DAS by sodium borohydride to the corresponding hydroxyls. In the periodate oxidation
step, the starch ring was opened between the C-2 and C-3, which formed a dialdehyde
structure. Next, the dialdehyde was reduced and hydroxyl groups were formed at C2 and

C3. The product was denoted as DHS (Di-Hydroxyl-Starch).

 

z
2:.
10
h
\
‘o
o
I
I
/ \
o
I
\

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H H
/° H NaBI—I4 /° H
o O .
/ o ’ >/ ’
/ H \ / Ho H
o OH
_ H _ n H n

 

 

 

Figure 6: Borohydride reduction of DAS

l4

By effectively controlling the periodate oxidation step, copolymers were formed
that contain both the structure of the glucose ring and the flexibility of the open ring
structure with C—OH groups on them. In this work a number along with DAS or DHS is
used to represent products with different degrees of modifications. For example, 20DAS
means periodate oxidized starch with 20% of its ring open and the borohydride reduction
product of 20DAS is called 20DHS accordingly. In this project three final products,

20DHS, 60DHS, and lOODHS, were synthesized.

 

 

 

 

 

Figure 7: Chemical structure of glucose-c-DHS
Among the three final products, lOODHS was further modified to achieve altered
surface activity. The product, lOODHS, was esterified with acetic and propionic
anhydride to obtain grafted ester groups with different chain length. In addition, by
controlling the amount of anhydride added, esters with various degree of substitution

were achieved.

15

 

\

 

 

Acetic or
Propionic
Anhydride

 

 

R = OH, OCOCH3 or OCOCH2CH3

\

 

 

\

HO

 

i—

H

Figure 8: Esterification of lOODHS

m

 

 

R = OH, OCOCH3 or OCOCHZCH3

2.1 Experiments

2.1.1 DHS preparation

water with sodium periodate at ambient temperature in the dark. Sodium periodate with
1.2 times as much as the theoretical amount (12.8g) was used for preparation of
completely oxidized starch (lOODAS). Partially oxidized starch was also prepared by the
addition of sodium metaperiodate of 0.6 and 0.2 times of theoretical amount (2.1 and 6.4g
respectively). The oxidant concentration of each solution prepared was 0.025 M, 0.075

M, and 0.15 M, respectively. The oxidation was performed under magnetic stirring for

DAS was prepared by oxidizing 8.1 g starch powder suspended in 400 mL of

16

HR

 

Figure 9: Chemical structure of DHS Esters

 

 

six hours. Oxidized products were recovered by filtering and were washed at least 3 times
with 400 ml distilled water to remove inorganic salts.

Oxidized product (DAS) was added to a solution containing sodium borohydride,
which was 1.5 times as much as the theoretical amount. The reduction was performed at
room temperature with magnetic stirring for two hours and the excess borohydride was
destroyed with acetic acid. Both lOODHS and 60DHS were completely soluble in water,
whereby 20DHS was only partially soluble and thus was separated through filtration into
the supernatant solution and an undissolved portion. All the solutions were dialyzed
against distilled water to remove inorganic salts and were then concentrated and dried to
recover the products. The undissolved portion was suspended in distilled water, filtered,
and thoroughly washed with distilled water, and then dried to constant weight. The final
weights of the dried samples (20DHS, 60DHS, and lOODDHS) were recorded for yield

calculation and further analysis.

2.1.2 DHS Esterification

In this project, DHS esters were synthesized by reacting lOODHS with acetic or
propionic anhydride in alkalic aqueous solution. The procedure was as follow:

1. 8 g of lOODHS was dissolved in a beaker with 80 ml of distilled water.

2. Then was stirred on a magnetic plane for 30 minutes until complete

dissolution.

3. The beaker was transferred with lOODHS solution into a water bath with
temperature set at 15°C.

4. Acetic anhydride (AA) or propionic anhydride (PA) was added into the

solution in droplets.

l7

5. Meanwhile, 10% Sodium Hydroxyl solution was added in droplets to keep the
pH between 8 and 9.
6. After all AA or PA was added, the solution was transferred into a dialysis tube
and was left the tube in distilled water to remove inorganic salts.
7. The dialyzed solution was dried to recover final products: DHS Acetate (DHSA)
and DHS Propionate (DHSP).
To obtain products with various DS, AA, or PA was added as shown in Table 1.

Table l: Esterification of lOODHS

 

 

 

 

 

Sample N(Anhydride)/N(DHS) AA or PA, gram
DHSA] 0.1 0.255
DHSA2 0.4 1.021
DHSP] 0.1 0.325
DHSP2 0.4 1.301

 

 

 

 

 

2.1.3 Carbonyl content determination of DAS

Measurement of the aldehyde groups in DAS was achieved in a procedure by
quantitative reduction with sodium borohydride. The relatively rapid method used by
Lindberg and Misiomyl7 for estimation of reducing monosaccharides was found to be
convenient and applicable to the analysis of the DAS sample. The procedure was based
on the hydrogen consumed in the conversion of carbonyl (C=O) to alcohol (C-OH)
groups. The difference between the hydrogen evolved on hydrolysis of a sodium
borohydride blank containing no DAS and of a mixture after reaction of excess reagent
with the DAS sample was the amount of hydrogen required for reduction of the carbonyl

groups in the dialdehyde units of DAS.

18

A DAS sample of 0.10 g was placed in a flask with 0.0280 g sodium borohydride,
which was more than the theoretical amount. A 100 ml buret filled with water was
connected with the flask to measure the volume of evolved hydrogen. The reduction was
performed at room temperature with magnetic stirring for two hours and the stirring was
stopped and hydrogen volume was recorded before it was released. Excess acetic acid
was added to the flask to destroy unreacted sodium borohydride and the volume of
evolved hydrogen was recorded. The sum Of both recorded hydrogen volumes was the
volume of hydrogen generated by the portion of sodium borohydride, which did not react
with DAS. In a separate flask, a blank experiment without DAS was run under the same
conditions and the total volume of evolved hydrogen was recorded as well. The
difference in total hydrogen volume between the two runs gave the amount of sodium

borohydride reacted with DAS, therefore carbonyl content.

2.1.4 NMR

The 13C NMR Spectrums Of 20DHS, 60DHS, and 100DUS were recorded with a
Varian 500 MHz superconducting NMR-Spectrometer operating at 499.738 MHz
interfaced with a Sun Microsystems UltraS UNIX console. Measurement ofNMR spectra
was performed in D20 (99.9% purity) at 25°C. The 3-(Trimethylsilyl)- Propionic acid-
D4, sodium salt (TSP) was used as a reference, with a peak at 0 ppm.

Also, Proton NMR was used to determine the D8 of DHS esters. This was done
on the Varian 500 also. The solvent used was Dimethyl sulfoxide (DMSO).

Triflouroacetic acid was added into the test samples to remove the hydroxyls.

19

2.1.5 FTIR

A Perkins Elmer System 2000 FT IR was used to characterize all samples. The
samples were pressed in KBr pellets and run for various amount of scans to achieve high

quality spectra. The wavelength range was between 4000 cm'I to 450 cm".

2.1.6 DS determination by Titration

The degree of substitution (DS) indicates the average number of substitutions per
anhydroglucose unit in starch. The highest possible DS is three since there are three OH
groups available per anhydroglucose unit. The DS of esterified starch can be determined
by hydrolyzing substituted groups with 0.1 N NaOH and then titrating back with 0.1 N
HCl to the original pH prior to the NaOH addition.H3

Ten grams of the sample were added to a 250 ml conical flask covered with 25 ml
of distilled water. The mixture was conditioned in a Tecator 1024 shaking water bath for
1 hour at 30°C and then the pH of the mixture was measured. The pH of the samples
ranged from 6 to 7. Next, 150 ml of 0.1 N NaOH were added to each flask. The sample
was then conditioned for 48 h at 50°C to hydrolyze the fatty acids substitutes. The excess
NaOH of the samples was titrated with HCI back to the original pH. The DS was

calculated as follows:
DS=(MFA*MW my I W 'MFA(MW FA-MW H20)] (2- 1)
where DS=degree of substitution; W=weight of the sample (g); MFA=mols of titrated fatty

acid (HCI); MWFA=molecular weight of the fatty acid (HCI); MWH20=molecular weight

of water (1 8); and MW AN=molecular weight of a repeat unit (164).

20

2.1.7 Molecular weight measurement

In this project the molecular weight distribution of all synthesized polymers was
determined by Gel permeation chromatography (GPC) and their intrinsic Viscosity was
measured to estimate their Viscosity average molecular weight.

GPC involves passing a dilute polymer solution through a tubular column packed
with polymeric gel (crosslinked) beads. Under high pressure flow some of the polymer
chains are forced into the pores of the gel, while others pass by the gel beads. The
residence time Of a given polymer chain in the packed column depends on the path it
takes through the gel. The output of a GPC reflects the number of chains at a given
retention volume, which is directly related to the molecular weight of the sample. From a
GPC output, the number average molecular weight, weight average molecular weight and
polydispersity index were calculated in comparison to standards. The polydispersity
index (PDI) is a measure of the distribution of molecular mass in a given polymer
sample. The PDI calculated is the weight average molecular weight divided by the
number average molecular weight. The PDI indicates the distribution of individual
molecular masses in a batch of polymers. The PDI value is always greater than 1, but as
the polymer chains approach is uniform chain length, the PDI approaches unity ( 1). The

PDI from polymerization is denoted as:

PDI = Mw/Mn (2.2)

M", the number average molecular weight, is the total weight Of all the polymer

molecules in a sample, divided by the total number of polymer molecules in a sample.

MW, the weight average molecular weight, is based on the fact that a bigger molecule

21

contains more of the total mass of the polymer sample than the smaller molecules do.

These average molecular weights are calculated based on the following two equations:

 

 

Molecular weight can also be calculated from the viscosity of a polymer solution
whereby bigger polymers’ molecules make the solution more viscous than small
polymers. The molecular weight obtained by measuring the Viscosity is different from
either the number average or the weight average molecular weight. It is between Mn and
Mw and it is usually closer to the weight average than the number average molecular
weight. Viscosity average molecular weight is determined by intrinsic Viscosity and the

Mark Houwink equation:

[n1=KM.°‘ (2.5)

where [n] is intrinsic Viscosity, K and or are constants and M" is the experimental

Viscosity average molecular weight.

Here, or and K are constants for a specific polymer/solvent/temperature system. At
the theta condition or should approach 1/2, for non-theta conditions c>1/2 and for good-
solvent scaling or is expected to be 3/5. a varies from the values of 1/2 or 3/5 due to

short-range interactions and their implied effect on the definition of M. Branched

22

polymers can have a value of a less than 1/2. Values greater than 0.6 are usually
associated with the chain rigidity and asymmetry of the coil due to features such as

helical coiling.

The viscometer used to measure dilute solutions in this project was an Ubbelohde
capillary viscometer. In this viscometer the Poiseuille equation for laminar pressure flow
in a capillary tube is used. The volumetric flow rate, Q, under gravity for constant volume

is given by Poiseuille's law:
4
Q = new /(8n) (2.6)

Where p is density, g is gravitational constant, r is the capillary radius, and n is the
absolute Viscosity. Since the flow time is proportional to the Viscosity (t = kn) the

specific Viscosity can be calculated as following:
nsp= [t " to] /to (2.7)

Where t is the efflux time Of the solution and to is the efflux time of the solvent. The

intrinsic Viscosity can be calculated from the equation derived by Solomon and Ciuta'g.

[n] = [201... — mm... + mum/c (2.8)

2.2 Results and discussion

2.2.1 Yield of DHS

In the absence of any side reactions and provided all the products were collected

with no loss, 8.08 g 20DHS, 8.04 g 60DHS or 8.00 g lOODHS should have been

23

recovered given 8.10 g starch was used as the starting material. In practice the actual
yield was derived from the weight of collected products divided by the theoretical weight.

The yields of DHS are shown in Figure 10.

 

 

 

 

 

 

Yield of Products with Different Degree of Modification
100.00
80.00
:5
R
a
60.00 —
40.00 -
10013-18

 

 

 

 

Figure 10: Yield ofDHS

It is apparent that the actual yields were between 70-78%. These low yields can
be explained by the following reasons: (1) Moisture in the starch. According to the
information provided by the manufacturer, there is approximately 11% moisture in the
starch. Although it was dried before reaction, there could still have been moisture left,
which contributed weight to the starting material. (2) Side reactions. Degradation caused
by side reactions were observed during both periodate oxidation and borohydride
reduction steps”. The second reason also explains why the yield of a high modification

product is lower than that of a low modification product.

24

2.2.2 Carbonyl content determination of DAS

The results of carbonyl contents determination are given in Table 2. The carbonyl
contents of 20DAS and 60DAS are slightly higher than the theoretical values because of
the moisture in starch. However, the carbonyl content of 100DAS is slightly lower than
100%, most likely due to the fact that the oxidation of starch was not completely finished
in the six hours the reaction was run. However, considering the production cost and time,
98.66% yield is acceptable for this study.

Table 2: Dialdehyde unit content of DAS

le # aIO4 AG

 

2.2.3 13C NMR of DHS

The Spectrums of starch, 20DHS, 60DHS and lOODHS are shown in Figure 11,
12, 13 and 14. Herein, we focus our discussion on starch and lOODHS, since 20DHS and
60DHS can be seen as copolymers with anhydride-glucose unit and Di-Hydroxyl-
Glucose unit, which are the repeat units of starch and lOODHS respectively, and their
spectra are the combinations of starch and lOODHS. The '3 C NMR spectrum of lOODHS
agrees with the work previously done by Narayanz'. It shows four carbon Signals, two of
which integrate to two carbons each. The most downfield signal at 104 ppm is obviously
the C-1 carbon, which is at 100 ppm in the spectrum of starch. The next signal appears at
78 ppm and is assignable to the C-4 and C-5 carbons. The next two signals appear at 63

ppm representing C2 carbon, and 60.5 ppm, which is assigned to carbons C-3 and C-6.

25

c3 02

 

 

06
05
c1 ‘-
04
I I I I I I I I I I I I Ifi I I I I I I I I I l
100 90 80 70 60 ppm

Figure 11: 13C NMR spectrum of corn starch

 

 

 

 

 

 

fiIIIIITIIIIIrIIIIIrIIIII IIIIIIIIIIIIIMIIIIIIIIIIIIIIIIIIIIII

100 70 00 ”'3

 

Figure 12: 13c NMRspectrum of20DHS

26

 

 

 

 

 

 

 

I I I I I I 1 I ' I I I I I I I I I I I 1 I I I I I I
100 90 80 70 .0 PP”

Figure 13: 13C NMR spectrum of 60DHS

 

 

 

 

 

06
04,
c5 02
’01
100 90 80 70 60 ppm

Figure 14: '3C NMR spectrum of lOODHS

27

2.2.4 FTIR of DHS

In Figure 15, 16, and 17 the Spectrum of starch as the starting material is shown at
the top. The wide band observed at 3348 cm'1 can be attributed to the O-H stretching of
the amylopectin and its width is ascribed to the formation of inter- and intra-molecular
hydrogen bonds. The bands at 2935 and 2887 cm'l are attributed to the asymmetric
stretching of C-H, the band at 1656 cm'1 is ascribed to absorbed water and the bands at
1421 and at 1357 cm"I ascribed to the angular deformation of C-H. The C-O ether bond
Shows stretching at l 156 cm", while the C-0 alcohol bond shows stretching at 1015 cm".

The Spectrums of the products after periodate oxidation are shown in the middle
of Figure 15, 16, and 17. A band around 1730 cm'l represents stretching of the C=O
group was observed. According to Narayan’s work25 most of the aldehyde groups were
not in the free state, but probably involved in the formation of hemialdol structure. That
is why the peak is not as strong as a regular aldehyde group. Also the C-0 alcohol peak at
l015 cm'l is smaller than the corresponding peak Of starch. This confirms the structure
change from C-O alcohol to C=O carbonyl group.

The Spectrums of products after reduction by borohydride are at the bottom of
Figure 15, 16, and 17. It is apparent that the carbonyl peak at 1730 cm'l disappeared and
the C-0 alcohol peak changed back to be as strong as that of starch. Both of these peaks

confirmed the borohydride reduction from carbonyl group to hydroxyl group.

28

 

 

IR Spectra of Starch, ZODAS and ZODHS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6
5 ..
4 -l
Starch
.3 3 l —-—2ooAs
<
—-200Hs
2 .1
1 .
0 - . . . . . .
3950 3450 2950 2450 1950 1450 950 450
Wavelength, 1Icm
Figure 15: FT IR spectrum of starch, 20DAS and 20DHS
IR Spectra of Starch, 600AS and BODHS
e
5 .
4 1
» -—Slarch
2 3 . —60DAs
Aldehyde ——eooHs
2 . at
1 .
0

3950 3450 2950 2450 1950 1450 950
Wavelength, ‘llcm

 

450

 

Figure 16: FTIR spectrum of starch, 60DAS and 60DHS

29

 

 

 

 

 

IR Spectra of Starch, 1OODA8 and 1000HS

 

 

 

 

 

 

 

 

5 l
4 .
a -—Starch
2 31 ——-100DAs
2 Aldehyde ‘ --—1OODHS
1 l
O I f I I I I I
3950 3450 2950 2450 1950 1450 950 450

Wavelength, 1Icm

 

 

 

Figure 17: FTIR spectrum of starch, 100DAS and lOODHS
In Figure 18 the spectra of DAS with different degrees of modification are shown. The
height differences of carbonyl peak indicate that various degrees of modification were
achieved as we expected. The 20DAS shows a very weak carbonyl peak, because only
20% repeat units Of starch was oxidized. The 60DAS shows a stronger carbonyl peak and

the 100DAS has the strongest carbonyl peak.

30

 

IR Spectra of DAS

 

 

-— 1OODAS
— BODAS
— ZODAS

Abs

iii

 

 

 

 

 

 

 

 

r

1551 1751 1051 1551 1451
Wavelength,1lcm

 

 

 

Figure 18: FTIR spectrum of DAS with increasing degree of modification

2.2.5 DS Determination of DHS Esters

The DS was determined by two methods: titration and NMR. It was observed
(Table 3) that the measured DS were very close to the theoretically calculated values,
especially for products with low DS. Although the esterification was done by following a
similar method for esterification of starch in aqueous environment, we obtained much
higher DS than that of starch esters. According to the review of Tessler and Billmersn,
typical restriction of an aqueous esterification process is a low substitution level with a
DS less than 0.2. This is believed to be because of primary hydroxyl group content
difference. In starch there is only one primary hydroxyl group in each anhydrous glucose

unit. However, all three hydroxyl groups in the repeat unit of lOODHS are primary.

3]

Although the efficiency of lOODHS esterification was much higher than that of starch, we

still need to point out that the efficiency decreases as the DS increases, which is similar to

the esterification of starch.

Table 3: DS of DHS esters

 

 

 

 

 

 

 

 

Sample DS(Calcu|ation) DS(TItration) STDEV(TItration) DS(NMR) STDEV(NMR)
DHSAI 0.10 0.11 0.027 0.10 0.010
DHSA2 0.40 0.38 0.026 0.37 0.014
DHSP1 0.10 0.08 0.034 0.09 0.016
DHSP2 0.40 0.37 0.035 0.36 0.016

 

 

 

 

 

In addition, compared to the titration method the NMR method offered a few advantages.

As can be seen from the above table, NMR gave much more reliable results with lower

standard deviations. Also, this method was much easier and less time consuming.

0H3

 

 

 

 

Figure 19: ‘H NMR Spectra of DHSA

32

'CH3

'CH2

 

___._/\\. .-_ -J L

8 6 4 2 0 2 4 ppm

 

Figure 20: ‘H NMR Spectra of DHSP

2.2.6 FTIR of DHS esters

The spectra of DHS esters (DHSA, DHSP and lOODHS) are shown Figure 21. A
band around 1730 cm'| represents stretching Of the C=O that was Observed in all of the
esterified products. Also we can see the peak height difference between products with
different DS (DHSAl vs. DHSA2. and DHSP1 vs. DHP2). This further confirms the

presence of C =0 from ester groups.

33

 

IR Spectrum of DHS eaters

 

 

M —100DHs
/ — DHSA1
———~—~ —DHSA2
fm ~— DHSP1
“—4 —DHsP2

 

 

 

f.
i

 

 

 

 

 

1951 1751 1351 1551 1451
Wavelength, 110m

 

 

 

Figure 21: The FTIR Spectrums of DHS Esters

2.2.7 Molecular Weight

The calculated average molecular weights and Viscosity of starch, DHS, DHSA
and DHSP are shown in Tables 4 and 5:

Table 4: Molecular weight by GPC

 

 

 

 

 

 

 

 

 

 

 

Sample an, Da le, Da PDI

ZODHS 878,000 1,047,400 1.19
SODHS 151,800 351,600 2.32
100DHS 139,000 371 ,600 2.67
DHSA1 138,900 847,000 2.50
DHSA2 133,100 344,400 2.59
DHSP1 151.400 379.300 2.51
DHSP2 146,200 367,100 2.51

 

 

 

Table 5: Intrinsic Viscosity

 

 

 

 

 

 

 

 

 

 

Intrinsic

Sample Viscosity (dl/g)
starch 88.24

20DHS 82.60

60DHS 19.10

lOODHS 18.14

DHSA1 9.00

DHSA2 10.68

DHSP1 1 1.10

DHSP2 l 1.94

 

 

 

Since K and a are not known for our products, it is not possible to calculate a
Viscosity average molecular weight. However, the data indicate good correlation between
the results from the GPC and the intrinsic Viscosity, particularly for DHS. The molecular
weight of DHS further confirmed that some degradation occurred during the modification
of starch as was also Observed from the yielded data. Furthermore, the degree Of
modification Of the starch was directly proportional to the extent of degradation and both
GPC and intrinsic Viscosity results showed the same trend. During esterification of
lOODHS, the molecular weight of DHS esters was shown to be the similar or greater than
that Of lOODHS. This means there was little or no degradation during the reaction, which
agreed with our stability test results that DHS was stable in basic solutions. Although the
intrinsic Viscosity of DHS esters is much lower than that of lOODHS, it is believed that
the K and a constants, not molecular weight, contributed to the difference.

In summary, the modified products, DHS and DHS esters, were synthesized and
characterized by FTIR, NMR, and chemical methods to analyze the degree of
modification. The degree of modification is controllable by varying the reaction
conditions and the products are water dispersible or soluble, depending on the degree of

modification.

35

Chapter 3 Property Characterization

In this chapter some key properties of our starch-based polymers are reported as
they relate to the intended application., The surface tension was measured to predict the
spreading ability on the ocular surface and the adhesion strength. Mucoadhesion testing
was used to predict the retention time in the eye and Rheological analysis was done to
study the flow characteristics and the Viscoelastic properties, which are important
indicators for comfort and retention time. In vitro release study was studied to estimate
drug release profiles. In addition, the effect of storage and autoclaving were also
measured.

3.1 Experiments

3.1.1 Surface properties measurement

Surface tension (ST) is defined as the force acting on the surface Of a liquid that tends
to minimize the area of the surface or the force that appears to act across a line of unit length
on the surface. It is also known as interfacial force, interfacial tension, or surface tensity. The
cohesive forces between liquid molecules are responsible for this phenomenon. The
spreading or coating ability of the substance can be characterized to some extent by
measuring the ST. The ability of ophthalmic solutions to spread on the cornea is important
for of two reasons. ( 1) It helps to lower irritation and (2) It helps to develop an intimate
contact with the mucus layer and improves the adhesion to the mucus in the eye. Given
this, it is Obvious that ST is an important physical property of Ophthalmic solutions as it
influences the formation and stability of the preocular tear film.

Previous research was focused on obtaining low ST for Ophthalmic solutions,

however, an ophthalmic solution with low ST also means low thermodynamic adheSion

36

work and therefore, low adhesion strength, which can result in short retention time in the
eye. In this work we take both Spreading and adhesion into account and study the ST’S
effect on the performance.

3.1.1.1 LITERATURE REVIEW
Wetting

The wetting theory, which was developed predominantly in regard to liquid

adhesives, was in this study to predict spreading by using interfacial tensions.

915%!) Air

Solution

 

9193) 9105) Cornea

 

 

 

Figure 22: Schematic diagram showing the interfacial tensions involved in spreading an
ophthalmic solution over the cornea
Figure 22 is an example Showing schematically an Ophthalmic solution spreading
over the cornea. The contact angle, 0, which should be zero or near zero for proper

spreading, is related to interfacial tensions through Young’s equation:

gm = gm +gsa c0319 (3.1)

where the subscripts c, s, and a represent the cornea, solution, and air. For spontaneous

wetting to occur 0 must be equal to 0, and therefore

37

gca >/= gcs + gsa (3-2)

The spreading coefficient, S, of an ophthalmic solution over the cornea can be used to

predict spreading and can be determined as
S = gca ' gcs — gsa (3-3)

For an ophthalmic solution to spread over the cornea, S must be positive. Therefore, it is
advantageous to maximize gca while minimizing gcs and gsa.

The corneal epithelium was previously23 considered as a hydrophobic surface on
which water would not spontaneously form a thin film. This notion was subsequently
shown to be based on artefactual observations“. In contrast, current evidence indicates

that the epithelium is highly wettablezs and the surface tension of intact cells is much closer

to that of water leading to a comea/tear system that is close to a total wetting situation.

Therefore, given the fact that gca is fixed, it is essential to lower gcs and/or gs, in order

to Obtain spontaneous Spreading.

Thermodynamic work of adhesion, Wa

In addition to the spreading coefficient, another important parameter that is
affected by interfacial tensions is the specific work of adhesion. According to the Dupre
equation, the work of adhesion is equal to the sum of the surface tensions of the cornea

and mucoadhesive minus the interfacial tension.
Wa : gsa + gca " gcs (3-4)
Therefore, in order to maximize the Wa, we will have to maximize 859. and

minimize gcs.

38

Given the considerations Of both spreading and adhesion, our goal is to minimize

gcs, and moderately lower gsa, so that both spontaneous spreading and considerable

adhesion strength can be Obtained simultaneously. gcs and gs, are discussed as follows.

2Q, Interfacial tension between the cornea and ophthalmic solution

According to the work by Good and Girifalco”, gcs can be theoretically predicted

according to the following equation:
_ 0.5
gcs — gc + gs ' 2(D(gcgs) (3-5)

(I) is a molecular parameter. They suggested that higher values of (I) and hence,

lower values of gcs are expected to increase the mutual solubility. Generally, a solubility

parameter of a polymer can be estimated by a group contribution method. In this work,
all polymers were based on modified starch and all were based on similar structures
containing grafts of glycopeptide. Therefore, it is reasonable to expect that they are all
mutually soluble as they have similar solubility parameters. Based on their structure, we

can further expect a low interfacial tension between the cornea and the polymer solution.

Given that gcs can be estimated based on the above discussion, in the above

spreading (equation 3.3), and work of adhesion (equation 3.4) equations, two of the three

parameters on the right hand side are known or can be theoretically estimated and the

only unknown key parameter is gsa, the interfacial tension between polymer solutions

and the air. It significantly affects the Spreading and the strength of adhesion.

39

g,_ Interfacial tension between the air and o hthalmic solution
Based on the above discussion, it is essential to moderately reduce gs, to at least

lower than the ST of water, rather than to minimize it. Surface tension reduction in water-
based systems is generally achieved through the addition of surface active agents called
surfactants, which have a characteristic molecular structure consisting of a structural
group that has very little attraction for water, known as hydrophobic group, together with
a group that has a strong attraction to water, called the hydrophilic group. This is known
as amphiphilic structure. When they are dissolved in water the presence of the
hydrophobic group in the interior of the solvent causes a distortion of the water structure,
increasing the fi'ee energy of the system, which requires less work to bring a solute
molecule than a water molecule to the surface. Therefore, the solute molecules
concentrate at the surface. Since less work is needed to bring molecules to the surface,
the presence of surface active molecules decreases the work needed to create unit area of
surface. On the other hand, the presence of the hydrophilic group prevents the solute from
being expelled completely from the solvent as a separate phase, since that would require
desolvation of the hydrophilic group. The amphiphilic structure of surface active
molecules therefore causes, not only concentration of the surfactant at the surface and
reduction of the surface tension of the solution, but also orientation of the molecules at
the surface with its hydrophilic group residing in the aqueous phase and its hydrophobic
group oriented away from it.

The hydrophobic group of surfactants is usually a long-chain hydrocarbon residue
and to a lesser extent a halogenated group, oxygenated hydrocarbon, or a siloxane chain.

The hydrophilic group is usually an ionic or highly polar group. Since both starch and

40

DHS are hydrophilic in nature, it is feasible to make it more amphiphilic by introducing
hydrophobic groups as side chains. Since our goal was to lower the ST moderately, we
needed only to introduce short hydrophobic chains onto the starch backbone. In this work,
polymers with short hydrophobic chains were grafted. The effects of chain length and
degree of substitution on the ST were studied and then correlated with performance as
ophthalmic solutions.

In summary, the interfacial tension between the cornea and air is known. DHS
and their derivatives have similar structures as grafts of mucin, which helps lower
interfacial tension between the drug delivery systems and mucin/comea. It is essential to
control the surface tension of polymer solutions so that both spreading and adhesion
strength can be optimized.
3.1.1.2 Method
The surface tension of polymer solutions at the air-water was measured at different
concentrations. Surface tension was measured by the pendant drop technique with the
Kruss DSA-lO. Polymer solutions were loaded into a syringe, pendant drops were
produced, and their shapes were recorded by a camera. According to the Young-Laplace
equation, the shape of the pendant drop is related with the solutions’ surface tension. The
shape of a drop is determined by its radii of curvature, RI and R2. In the case of a Spherical
drop these are equal. The relationship between interfacial pressure (the pressure across the

interface) and these radii of curvature is called the Young-Laplace equation:

AP = y(l/R1+1/R2) (3.6)

where

AP = interfacial pressure difference

41

y = interfacial tension

R1, R2= surface’s radii of curvature.

3.1.2 Mucoadhesion

Mucoadhesion refers to bonds formed between mucus and polymers that
improves the attachment of drug carriers to mucus for drug delivery purposes. Mucus is a
substance secreted by various tissues in the body made up of water, mucin (a
glycoprotein), salts, and some cells. The mucus glycoproteins consist of hundreds of
short polysaccharide chains, which usually constitute about 70% of the weight of the
molecule, attached to a polypeptide backbone”.

For many years mucoadhesion has attracted the attention of ocular researchers
who have sought to control it and profit from the concepts and techniques of this novel
approach. Since the ocular bioavailability of drugs administrated by conventional eye
drops is low due to the small area for absorption and the short contact time in the eye, any
modification resulting in increased contact time will improve the drug bioavailability.
Therefore, the concept of mucoadhesion was applied in the field of ocular drug delivery
to prolong the residence time in the preocular area, potentially increasing the drug
bioavailability.
3.1.2.1 Literature review

Formation of mucoadhesion involves two steps: (1) the contact stage where an
intimate contact is formed, and (2) the consolidation stage where various interactions
occur to consolidate and strengthen the adhesive joint”. The first step can be explained
by the wetting theory. This theory uses interfacial tensions to predict spreading and in

turn adhesion. The second step has been explained by a few different theories”. That is,

42

the electronic theory relies on the assumption that mucoadhesive and mucus have
different electronic structures. Therefore the mucoadhesive force is originated from the
attractive forces across the interface. The adsorption theory states that the mucoadhesion
is due to the van der Waals interactions, hydrogen bonds, and some other related forces
between mucoadhesive and mucus. It is the most widely accepted theory of adhesion.
The diffusion theory states that mucoadhesion is produced through interpenetration and
entanglement of mucoadhesive polymer chains and mucus polymer chains. Therefore,
according to this theory the bond strength increases in proportion to the degree of
penetration of the polymer chains into the mucus layer. Between all these theories, the
diffusion theory and adsorption theory are of particular interest for this study and they are
discussed in further details below.
Diffusion theogy

The diffusion theory supports that bond strength increases with the degree of
penetration of the polymer chains into the mucous layer”. The penetration depth, 1, can

be estimated with the following relationship:
I = «Dam on

Where t is the time of contact and Dr, is the diffusion coefficient of the bioadhesive
material in mucus. The bond strength for a given polymer is believed to be attained when
the depth of penetration is approximately equal to the end-to-end distance of the polymer
chain3 '. Therefore, to achieve the desired bond strength we will have to either choose an
adhesive polymer with relatively short end-tO-end distance or increase Db.

According to research done by Voyutskii30, for diffusion to occur it requires that

the polymers possess sufficient mobility and are mutually soluble. This latter requirement

43

may be restated by the condition that they possess similar value Of solubility parameter,
which is an index of the compatibility of two components. Thus, the more structurally
similar a bioadhesive is to its target, the greater the mucoadhesive bond will be. In order
to improve diffusion both chain flexibility/mobility and similar structure with mucin are
important.

Adsorption theogy

The adsorption theory of adhesion is the most widely applicable theory to
describe the adhesive joint. It is proposed that given sufficiently intimate molecular
contact at the interface a material will adhere due to inter-atomic and inter-molecular
forces. The adhesion can be divided into two categories: (1) chemical bonds and (2)
secondary bonds. Chemisorption is defined when chemical bonds are formed across the
interface and is beyond this study. We will mainly focus on secondary bonds based on
Van der Waal and hydrogen force. Although these forces are relatively weak compared
with other bonds (Table 6), the sheer number of interactions, as a whole can, produce
intense adhesive strength.

Huntsberger32 has calculated the attractive forces between two planar bulk phases
due to solely dispersion forces. For example, they showed that even at the separation of
one nanometer the attractive force would result in a joint strength in tension of
approximately 100 MPa. Also, the formation of hydrogen bonds across the interface
appears to enhance the intrinsic adhesion and has often been observed. Given the
structure of mucin, it is reasonable to assume it can form hydrogen bonds with the starch-

based polymers.

44

Table 6: Typical bond energy

 

 

 

 

 

 

 

 

 

 

 

Bond Energy
Bond kJ/mol
Primary bonds
Ionic 600~1100
Covalent 60~700
Metallic l 10~350
Secondary bonds
Hydrogen bonds 10~40
Van der Waals bonds 0.08~40

 

A large number of polymers have been studied thoroughly as mucoadhesive drug
delivery systems. It is generally agreed that polymer-related factors influencing
mucoadhesion include hydration or degree of swelling, molecular weight, the nature of
the functional groups, molecular conformation or chain flexibility, and mobility of the
polymer and its concentration. Polymer hydration results in the relaxation of stretched,
entangled, or twisted macromolecules exposing the adhesive sites. Furthermore, chain
interdiffusion is favored by polymer—water interactions dominating the corresponding
polymer—polymer interactions.

A critical chain length is necessary to obtain interpenetration and molecular
entanglement between the polymer and the mucus layer. The threshold required for
successful mucoadhesion is a molecular weight of at least 100,000 Da. Excessive cross-
linking in the polymer, however, decreases the chain length available for interfacial
penetration. Also, excessive formation of interchain physical entanglement and hydrogen
bonding within the polymer itself can lead to conformation that hinders polymer diffusion
into the mucus network. As a result, chain flexibility is critical for interpenetration and

entanglement with the mucus gel. The mobility of the chain segment is directly

45

proportional to the interdiffusion and the interpenetration of the polymer within the
mucus network. Coiling Of polymer chains, due to pH or osmolality Of the medium, can
result in the shielding the active groups necessary for the adhesion process.

In summary, in order to form mucoadhesive bonds the selected polymer(s) should
have at least one of the following characteristics: (a) high molecular weight, (b) high
chain flexibility, (c) surface intention that induces spreading into the mucous layer, ((1)
sufficient quantities of hydrogen-bonding chemical groups, and (e) anionic surface
charges.
3.1.2.2 Method

Determination of mucoadhesive bond strength is important in the development of
ophthalmic solutions, as it can quantitatively compare different mucoadhesive materials.
Hassan and Gallo 3’3 are considered to have pioneered the work on the rheological
assessment of mucin-polymer bioadhesive bond strength. They observed that there was a
synergistic increase in Viscosity when a mucoadhesive polymer and mucin were mixed
together. The Viscosity of a dispersion containing mucin and a bioadhesive polymer is

determined by the contribution of the different components, as in the following equation:
m=nm+np+nb(1&
where

m: Viscosity of the system
11m: individual Viscosity of mucin
rip: individual Viscosity of polymer

11b: Viscosity component due to bioadhesion

46

Therefore, the viscosity component contributed by mucoadhesion can be obtained by

rearranging the last equation,
115 = m — mm 0;.) (3.9)
For these two equations to be valid, 11,, nmand 11,, should be measured at the same
concentration, temperature, time, and shear rate. In this study steady state controlled rate
flow curves, with shear rate varying from 15 to 300 US, of polymer solutions, 5% mucin

dispersion, and the mixture of polymer and mucin were collected. m, was calculated and

used as a direct estimate of the force of mucoadhesion.

3.1.3 Rheology

After a drop of ophthalmic solution is applied into the eye, it experiences two
processes: blinking, and process between blinking. The process between blinking can be
seen as a steady process with low (close to 0) shear rate. The process of blinking can
further be divided into two sub-processes: steady and dynamic processes. Part of the
process during moving between the upper and lower edge of the eye can be seen as a
steady process with a high shear rate. The shear rate associated with the eye has been
estimated to range from 0 to as high as 28500 Us”. When the eyelid is moving close
from/to the upper or lower edge Of the eye, it can be seen as a dynamic process with
changing shear rate. The frequency of the dynamic process can be estimated based on
how often we blink. The muscle that lets our eye blink is the fastest muscle in our body.
It allows us to blink five times a second. However, on average we blink 15,000 times a

day. Women blink twice as much as men”. Given the highest blink rate, five times a

47

second, the highest frequency can be estimated as 5 Hz. In this section, both steady and
dynamic properties were studied.
3.1.3.1 Steady shear rheological properties

Viscosity is a measure of a solution’s resistance to flow and is a function Of the
molecular attraction that resists flow. It is defined as the ratio of the shearing stress
applied to a solution to the velocity gradient in that solution. The relationship between
ophthalmic solutions’ retention time in the eye and their Viscosity is easily understood.
For example, blinking causes high shear rate to the solution applied into the eye. If the
Viscosity of the solution is high, it can lead to high shear stress between the solution and
the conjunctiva around the immediate contact area with the tear film, which is the cause
of irritation and damage. On the other hand, when the Viscosity of the solution is too low
and the Shear rate between blinks is relatively low or even zero, the ophthalmic solution
will not be able to remain on the ocular surface and this will result in increased drainage.
Therefore, to decrease the drainage between blinks the Viscosity at low shear should be
high. For a solution, how the apparent Viscosity changes as a function of shear rate is an
important rheological parameter. In this study, we tried to obtain solutions that had
relatively high Viscosity under low Shear rate and relatively low Viscosity under high
shear rate. Increased viscosity, only when it is not too high and causes irritation, can
increase retention time, reduce the drainage rate, and increase the bioavailability on the
ocular surface. A long contact time, achieved by a high Viscosity of the ophthalmic
solution is usually favorable, since it increases the bioavailability and therefore reduces

the number of applications required to control symptoms and signs.

48

A solution can have Newtonian or non-Newtonian characteristics, depending
upon how the apparent Viscosity changes with the shear rate. If a solution’s Viscosity is
independent of the shear rate, it is defined as Newtonian fluid and consequently, a non-
Newtonian fluid is defined as a fluid in which the Viscosity changes with the shear rate.
When a fluid’s Viscosity is high under conditions of low shear rate and low under
conditions of high shear rate, it is defined by the term “shear-thinning”. Water and
Silicone oils are Newtonian fluids and have a constant Viscosity regardless of the shear

rate, but normal human tears40 are a non-Newtonian fluid, with viscosity falling from

about 5 mPa.sec at 2 s.I to about 1.5 mPa.sec at 160 s". Shear-thinning behavior is a
typical property of solutions containing long polymeric macromolecules, which are not
strongly bonded into a globular form, such as HPMC (Hydroxypropyl Methyl Cellulose)
and CMC (Carboxyl Methyl Cellulose) solutions. The random orientation and interaction
of these molecules produce high resistance to flow when the shear rate is low. At higher
shear rates, the molecules become aligned in the direction of the shearing force and offer
much less resistance. Thus, solutions become less Viscous when the shear rate increases.
Shear-thinning solutions are more comfortable in the eye than Newtonian solutions of the
same Viscosity since at high Speed blinking shear-thinning solutions offer the advantage
of low Viscosity and have less dragging effect on the ocular surface.

Non-Newtonian behavior is a typical characteristic of polymer solutions. The
solution has a Newtonian Viscosity, which is high at very low rates of shear. However,
over much of the usual accessible shear rate range, the Viscosity decreases nearly linearly
with a shear rate in the log-log plot. Their apparent Viscosity decreases as the rate of

shear increases. In this linear range, the SO-called power law equation holds where:

49

n=KV4

(3.10)

where K and n are constants.
For Newtonian liquids, n=1 and K=Consistency coefficient; the value of n is less than
one for non-Newtonian polymer solutions. At last, when the shear rate is higher than a
certain value, the Viscosity is again independent of the shear rate. Considering the
properties of human tear, this non-Newtonian behavior is of tremendous practical
importance for ophthalmic solution formulation.
3.1.3.1.1 Literature Review

The basic cause for the non-Newtonian behavior of polymer solutions is the
orientation of molecular segments by the flow field. Molecular entanglements with an
appreciable lifetime exist above a critical molecular weight. Entanglements greatly
enhance the possibility of orienting molecular segments in a flow field. The
entanglements act as temporary crosslinks, so that the polymer solutions may have many
of the characteristics of crosslinked rubbers. At very low rates of shear, the
entanglements have time to Slip and become disengaged before enough stress can develop
in them to orient the molecules. At higher shear rates the segments between
entanglements become oriented before the entanglements can disappear. As a load-
bearing entanglement disappears another entanglement, that does not carry any load,
develops somewhere else in the solution. Thus, a steady state condition is developed in
the solution in which the rates of formation and destruction of entanglements is equal.
From the above discussion, a polymer solution at rest Should have a higher concentration

of entanglements than that of a polymer solution that is flowing. The change in

50

concentration of entanglements has two effects: first, once the shear rate is higher than a
certain value, the Viscosity of a polymers’ solution should decrease when the shear rate
increases. Second, at very high rates of shear practically no entanglements can exist. At
this point the Viscosity should reach a relatively small value, which becomes independent
of the shear rate.

There are several theories that are often used to describe the shear rate
dependence on the Viscosity. Besides the power law introduced above, the Cross equation
is another general empirical equation for fitting curves, which have a sigrnoidal shape.
The Cross equation for the effect of shear rate on the apparent Viscosity is:

770 _ "co

= +
77 77.. 1+(C7)’" (3.11)

Where 110 is the Zero Shear Viscosity, the magnitude of the Viscosity at the lower

Newtonian plateau. It is a critical material property. Zero shear Viscosity of a polymer
solution is a function of molecular weight, chemical structure, temperature, and

concentration.

1100 is the Infinite Shear Viscosity. This tells us how our product is likely to

behave in very high Shear processing Situations.

The parameter In is known as the (Cross) Rate Constant. It represents how fast the
Viscosity drops. It is dimensionless and is a measure of the degree of dependence of
viscosity on shear rate in the shear-thinning region. A value of zero for m indicates
Newtonian behaviour with m tending to unify for increasingly shear thinning

behaviour. It is affected by the molecular weight distribution.

51

C is known as the Cross Time Constant (or sometimes the Consistency) and has
dimensions of time. It represents the shifting from zero Viscosity to infinite Viscosity and
is proportional to zero Viscosity.

So, the Cross model not only provides us with a simple way of quantifying the
“full” Viscosity/shear rate profile for a shear thinning fluid, but also it helps us to obtain
the desired steady flow profile by engineering polymers’ structure accordingly.
3.1.3.1.2 Method

The flow properties Of the samples were examined using a Haake RS100
RheoStress equipped with a Haake circulation bath and temperature controller. All
experiments were run at 25 °C. Steady state flow curves were recorded automatically.
Although the shear rate associated with the eye has been estimated to range fiom 0 to as
high as 28500 Us“), we tested the viscosity at shear rates varying from 15 to 300 due to
internal limitations of this rheometer.
3.1.3.2 Viscoelastic behaviors

Viscoelasticity describes materials that exhibit both viscous and elastic
characteristics when undergoing unsteady state deformation. Viscous materials will flow
when a stress is applied and do not recover their original structures after the stress is
removed. Elastic materials strain instantaneously when a stress is applied and quickly
return to their original state once the stress is removed. Viscoelastic materials have
elements of both of these properties. They have both flowing and deforming behaviors
when a stress is applied.

In addition to the flow properties, the Viscoelastic properties Of ophthalmic

solutions are also important, especially at low shear in oscillatory experiments. This is the

52

case since these conditions will impact how well a solution stays in the eye as it most
likely dependent not only on the flow properties but also on the dynamic properties, that
is, how well the carrier, polymers in our case, is held together in situ. AS previously
mentioned, the polymer has to undergo dynamic process because of blinking. How it
behaves under varying frequency is important.
3.1.3.2.1 Literature review

According to Steffe’s work”, all materials are ViscOelastic, but the Viscous or the
elastic character may dominate in certain situations. The Deborah number proposed by
Marcus Reiner37 is a means to distinguish between solids (elastic) and liquids (Viscous).
He recognized that whether a substance is a solid or a liquid depends on the time of the

characterization process. The Deborah number is defined as

NDe = tmatefial/tpmms (3.12)
Where tmatcrial is the characteristic time of the material and tprocess is the characteristic
time of the process.

Pipkin38 suggested that the tmatcria] may be provisionally considered as an order-
of-magnitude estimate for how long it takes the substance to complete a stress relaxation
process. If a material is ideally elastic, tmatcfia] tends to be infinite and no relaxation
occurs. If a material is ideally Viscous, Imateria] equals 0, meaning immediate relaxation

occurs. In our application, the Viscoelastic properties Of systems containing polymer
molecules arose from three factors: (1) the length of the polymer molecules, (2) the

flexibility of the molecular chains, and (3) the interactions of the segments of a polymer

molecule with other segments of the same or different polymer molecules”. The tprocess

53

can be seen as inversely proportional to the frequency in a dynamic test. During the

ocular drug delivery process, the tproccss varies over a large range, depending on the status
of the eye. When the eye stays open or closed tpmcess can be seen as infinite and when the

eye blinks tprocess can be as low as 0.2 second.

Therefore, the Deborah number can be estimated and used as a measure of the
degree of Viscoelasticity. If it is less than 1, the material shows more Viscous character
than elastic character. If it is greater than 1, the material shows more elastic character
than Viscous character.
3.1.3.2.2 Method

Oscillatory shear experiments were performed to evaluate the viscoelastic
properties of the ophthalmic solutions. The shear storage modulus or elastic modulus (G')
and the shear loss modulus or Viscous modulus (G") were evaluated as function of
frequency. G' gives information about the elasticity or the energy stored in the material
during deformation, whereas G" describes the Viscous character or the energy dissipated
as heat.

Strain sweep measurements were made for all samples to determine the maximum
strain amplitude for the solution that would allow all measurements to be made in the
range of linear Viscoelastic behavior. Measurements above this level do not measure the
physical properties relevant for a gel at rest under the lower eyelid. All further
measurements Of rheological properties were made within the linear region, i.e. below
this maximum strain. The oscillation frequency ranged from 0.46 to 100 Hz and the

experiments were run at 25 °C.

54

In oscillation experiments the stress response to a sinusoidally varying strain is
recorded as a function of frequency. The shear strain, the stress, and the phase angle are
determined in the measurement. The parameters obtained are the complex modulus, G“,
and the phase angle, 8. Complex Dynamic modulus G. can be used to represent the

relations between the oscillating stress and strain:

”2

G‘ = (Io/yo = (0’2 + G”2) (3.13)

G'=G.cosb (3.14)
G"=G’sin5 (3.15)
where

0 0: amplitudes of stress

yo: amplitudes of strain

6: phase angle
G': storage modulus

G": loss modulus

3.1.4 In vitro release test

Controlled release is an important step toward improving the delivery of drugs to
the inner part of the eye. In this work, hydrogel was used as a polymer matrix to control
drug release. In the context of this paper, the term hydrogel will be used in its broader
sense. That is, a matrix formed by either a water-swellable material (usually a cross-
linked polymers with limited swelling capacity) or a water-soluble material (usually a
hydrophilic polymer that swells indefinitely and eventually undergoes complete

dissolution)“. Most of the mucoadhesive polymer materials belong to one of these two

55

categories. For the design of new controlled drug release systems, it is essential to know
the exact mass transport mechanism involved in drug release and to predict the resulting
drug release kinetics.

3.1.4.1 Literature review

Based on rate-lirniting step for controlled release, three models are developed and
categorized as follows: (1) Diffusion-controlled, (2) Swelling—controlled, and (3)
Chemically-controlled models. The release of the drug from hydrogel is believed to be
diffusion-controlled.

Fick’s law of diffusion with either constant or variable diffusion coefficient is
commonly used in modeling diffusion-controlled release. Drug diffusivities are generally
determined empirically or estimated a priori using free volume, hydrodynamic, or
obstruction-based theories“. Free volume theory is based on the assumption that the free
volume is the major factor controlling the diffusion rate of molecules. In this theory, the
solute diffuses by jumping into voids formed in the solvent space by the redistribution of
the free volume within the liquid. It is assumed that the free volume can be redistributed
without any energy change. The voids are pictured as being formed by a general
withdrawal of the surrounding liquid molecules due to random thermal motion. These
holes are then filled in by the reverse process. Hydrodynamic theory takes the
hydrodynamic interactions into account. These interactions include frictional interactions
between solute and the polymer, between the solute and the solvent, and also between the
solvent and the polymer. It includes the effect of overlapping of polymer chains, which is
omitted in the obstruction model. According to this model, major factors determining

diffusion coefficients include drug size, molecular weight, and polymer concentration.

56

Obstruction theory is based on the assumption that the polymer chains are regarded as
motionless relative to the diffusing molecules, which leads to an increase in the mean
path length of the diffusing molecules between two points in a system. Therefore, the
diffusion coefficient is a function of polymer fraction.

Baker and Lonsdale model

Peppas‘12 describes the release of drugs from a hydrogel matrix as a three step
process. The first is the initial burst when the liquid dissolves the drug present at the
immediate surfaces of the matrix, creating a small “burst” effect. At this time the water
or biological fluids begin to penetrate the gel at a rate that is dependent on the porosity of
the matrix. The second phase is classified as the stationary phase, where the water
continuously penetrates the matrix at a constant rate. This penetration is accompanied by
an expansion of the gel layer in the direction of the external medium. This phase
accounts for the majority of the drug release. It is generally accepted that the release of
the drug is controlled by diffusion process, not by the rate of drug dissolution or the rate
of penetration of the front for hydrophilic matrices. The third phase is the exhaustion
period which begins when the penetration fiont has reached the center of the matrix and
the drug concentration has dropped below its solubility limit in water. During this stage
the release rate rapidly falls.

These controlled release systems can either be classified as systems with
suspended drugs or systems with dissolved drugs. Mathematical models for both have
been developed and adapted by various authors and a model developed by Baker and
Lonsdale for the case in which the drug is dissolved in the polymer can be used for this

research. This model is presented for the case that the system is homogenous, there is one

57

plane Of diffusion, there is no diffusion boundary layer present, and there are sink

conditions. In this situation, the initial concentration of the drug in the hydrated matrix is
less than the drug solubility in it (C0 < Cm). Setting Co = Moo/V, where M00 is the initial

drug loading (the total amount of drug release at infinite time) and V is the effective

volume of the hydrated matrix, the following expressions are valid for 0 S Mt/ M00 5 0.6:

1/2
M, =2AC0 21 0t”2

72' (3.16)

The rate of release is:

1/2
dM’ = 2AC 2i"— o 1“”2
dt 0 n- (3.17)

 

where:

Mt = the amount of drug released at any time

M00 = the initial drug loading
A= the diffusional area

Co= the initial concentration Of the drug in the system

Dm = the apparent diffusion coefficient

These equations are for the planar case, but can be modified for other shapes. All

of these cases will also Show a to'5 dependency. PeppaS45 brings up faults with this

model due to the following assumptions:

58

1. This model was not developed for systems undergoing dimensional
change.

2. A pseudo steady-state analysis was used that ignores the external mass
transfer resistance and is only valid when the solute loading is in great
excess of it solubility limit.

3. The countercurrent solvent diffusion was not considered

4. Drug diffusion in the gelled matrix was assumed to be the rate limiting
step.

However, the shortcoming of these assumptions are not as much a problem with
the ocular system being described because the gel is already hydrated when applied, will
not undergo dimensional change, and solvent diffusion can be neglected.

Matrix formulations play an important role in the drug release profiles. Release
profiles can usually be modified by type and Viscosity Of the polymers, the polymer
concentration, and the drug particle size. The type of polymer is often determined by the
solubility characteristics of the drug. For example, hydrophilic matrices are generally
used to prolong the release of highly water-soluble drugs. The viscosity of the polymer
appears to play a role in drug release. Polymer concentration follows the general rule that
increasing the proportion of hydrophilic material decreases the rate of release. Drug
particle size affects the dissolution rate of the drug with the smaller particles having
larger surface/volume ratios and therefore, faster dissolutions rates. In summary, we have
to consider the size of drug, the structure of hydrogel, the polymer composition, the water
content, and the size Of the molecules in order to design systems offering desired drug

release.

59

3.1.4.2 Method

A USP dissolution method was used for Obtaining release profiles. A Hanson EZ-
lift dissolution system made with six separate chambers that were all kept at the same
constant temperature water bath was used. Each chamber had a rotating paddle attached
to the same drive motor. One-liter beakers were used to hold the release medium,
simulated tear solution, and they were filled with a specified amount, 600 ml. The
formulated drug was placed in a 5 ml well having a diameter of 5 cm and covered with
the same dialysis membrane. The concentration of the drug in the release medium was
determined by UNICO SQ-2800 Ultraviolet-Visible Spectrophotometer by taking a
sample of 0.10 g at specific time intervals and it to maintain a constant volume. The
composition of the simulated tear solution release medium is given in Table 7.

Table 7: Simulated Tear Solution

 

 

 

 

 

 

 

 

 

 

Chemicals N/mmol Mass/ g
NaHCO3 26. 00 2. 184
NaCl 108. 00 6. 312
IKCI 24. 00 1. 789
CaC12 0. 40 0. 059
M502 2. 50 0. 238
1120 to 1000

 

 

3.1.5 Stability study

The influence of sterilization by autoclaving and storage under various conditions
were studied. In this study, Viscosity was used to assess the solutions’ stability. To test
the effect of autoclaving, Viscosities of the same solutions were recorded before and after
autoclaving and their flow curves were compared. The same method was also utilized to

study the effect of pH and temperature on the storage stability.

60

3.2 Results and discussion

3.2.1 Surface tension

The surface tensions Of DHS and DHS ester solutions are shown in Table 8. They
were all dissolved in pH 7.4 phosphate buffer to maintain the same pH value. Samples
with multiple concentrations were measured until a surface tension plateau was attained.
Among samples tested, no significant effect of the concentration on surface tension was
detected from 20DHS, 60DHS, and lOODHS solutions. This is believed to be because of
all these samples are hydrophilic in nature and only a very small amount of polymers is
absorbed at the solution-air interface. Therefore, when the concentration is higher than a
certain level, which is no greater than the lowest concentration we tested (e.g. 0.5%), the
surface tension does not change as the concentration increases.

In comparison, all DHS esters tended to reduce the surface tension to a various
extent. The ST Of their solutions decreased as the concentrations increased, until a plateau
was reached. These changes are directly related to the hydrophobic nature of methyl and
ethyl groups that were grafted onto the DHS backbones and led to the amphiphilic
structures. When these amphiphilic polymers were dissolved in water the hydrophobic
groups increased the free energy of the system and were more likely to concentrate at the
surface. Since less work was needed to bring these polymers to the surface than water,
their presence decreased the work needed to create unit area of surface, which
consequently reduced the surface tension. After a certain amount of amphiphilic
polymers were adsorbed, the surface was saturated with polymer molecules and could not

adsorb any more giving rise to the observed plateau at high polymer concentrations.

61

In addition, as can be seen from Figure 23, these DHS esters showed different
surface activities, in the following order: DHSP2 > DHSA2 > DHSP1 > DHSA1. That
means that given the same DS, polymers grafted with longer side chains are more active
than those grafted with shorter side chains. Furthermore, given the same length of grafted
side chains, polymers with higher D8 are more active than polymers with lower DS.
These observations can be explained by the difference between methyl and ethyl groups,
as well as the difference between the degrees of substitution. Both grafted —CH2CH3 and
—CH2CH2CH3 are hydrophobic, but an ethyl group is more hydrophobic than a methyl
group. So when methyl and ethyl groups are grafted at the same DS, the polymer with
ethyl is more susceptible to be present on the interface with a lower overall energy. The
same mechanism can be used to explain why polymers with higher DS showed more

activity than polymers with lower DS.

62

Table 8: Surface tension results

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Surface

Tension
Sample dynes/cm
ZODHS
0.50% 72.4
1% 71.7
2% 72.2
GODHS
0.50% 71.8
1% 71.7
2% 71.9
10001-18
0.50% 72.0
1% 71.7
2% 71.6
DHSA1
0.50% 70.5
1% 69.2
2% 67.3
5% 64.8
10% 64.7
DHSA2
0.50% 63.6
1% 63.0
2% 61.6
5% 58.6
10% 58.5
DHSP1
0.50% 68.0
1% 66.7
2% 64.9
5% 61.6
10% 61.6
DHSP2
0.50% 55.5
1% 54.3
2% 52.0
5% 50.9
10% 51.0

 

 

 

 

63

 

Surface Activity in Water

80.0
70.0 a
60.0
50.0
40.0
30.0
20.0
10.0

0.0

 

 

Sufaee Tension, dynadcm

 

 

 

20DHS 60DHS 100DHS DHSA1 DHSA2 DHSP1 DHSP2

 

 

 

Figure 23: Surface tension of solutions

Tiffany reported that the surface tension of intact cells is much closer to that of
water, the comea/tear system is therefore close to a total wetting situation”. However,
when we also take kinetics into account, a solution with the same surface tension as water
may not be able to spontaneously spread. Additional driving forces contributed by lower
surface tension may be required to overcome the retarding forces from viscous resistance.
Therefore, a solution with ST lower than water may be needed if its viscosity is higher
than water. So, although all substances tested could spontaneously spread on the cornea
surface, a low ST is desirable depending on the viscosity of the solution. Meanwhile, a
low ST can result in low work of adhesion and therefore weak adhesion strength. Thus, a
too low ST should be avoided. The effect of ST will be further investigated by in vivo

tests.

64

3.2.2 Mucoadhesion

The retention time of an ophthalmic solution is influenced by its viscosity,
bioadhesion and ST. Although a lower ST indicates a better ability to coat, at present not
all of these factors are known. Bioadhesion, and more specifically mucoadhesion, is
another important polymer property that impacts the adhesion to the glycoproteins on
epithelial surfaces. In this work, the mucoadhesion strength was measured with a
rheological method, which was broadly used in previous studies“. The mucoadhesion
strength of all DHS and DHS ester solutions were tested. Solutions of 20DHS, 6ODHS,
and 100DHS with the same concentration and viscosity were compared. Also, the
mucoadhesion of DHS ester solutions were measured and compared with 100DHS
solution.
3.2.2.1 DHS solutions at the same concentration

The mucoadhesion of 20DHS, 60DHS, and 100DHS solutions at concentration
3% is shown and compared in Figure 24, 25, 26, and 27. The upper dots represent the
viscosity of the mixed dispersion comprised of 3% DHS and 5% mucin. The lower dots
represent the sum of viscosity of 3% DHS and 5% mucin, which were measured

separately.

65

 

 

Mucoadhesion of 3% ZODHS

 

 

 

 

 

 

 

 

 

 

50‘0 _,__ .” ' ‘_ i _ T — ""_““TT “7‘ onixture
93 40.0 ~~- ~ —- - . _.__.._____ ———~--——-—-—-— —-* I Vmucus+Vpolymer
. O O Q . .
_ 30.0 l— _._ ---- __
a 20.0 a” I i —* l~ -——— .-_r'___fi-____._*__- i
5 10.0 «-7 ~ _--_-__fl~_._-. ,__. _-____#___ 4
0'0 T r 1 ‘l I l A
0 50 100 150 200 250 300 350

Shear Rate. 1/s

 

Figure 24: Mucoadhesion of 3% 20DHS

 

 

Mucoadhesion of 3% 60DHS

 

 

 

 

 

 

 

 

300 fl # H" _,______ '___——"4_flq eVmixture
IVmucus-I-V I mer
Q}. 20 0 «-—1— - — ‘_ u... ____ _ __ WY
0 o e e e
. I I . I - - . .
10.0 -2. — ._--. - ‘2. _____
S
0.0 . , , . . 1
O 50 100 150 200 250 300 350

Shear Rate, 1/s

 

Figure 25: Mucoadhesion of 3% 6ODHS

66

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mucoadhesion of 3% 1OODHS
30.0 _#.~___ ~fi ~v —— __,_ ,
o Vmixture
95 - Vmucus+Vpolymer
_ 20.0 -~ +——-—--—-.— — —-——.—— —
O O O O O
'3 I I I I I I I I
10.0 + —- ea
>
0-0 1 T I T l T
0 50 1 00 1 50 200 250 300 350
Shear Rate, 1/s
Figure 26: Mucoadhesion of 3% 100DHS
Mucoadhesion of 3% DHS Solutions
30.0
. e 3% ZODHS
% 20.0 I 3% BODHS J
- ’ . . a 3% 100DHS
5 10.0 4 ______ ——————~— —
' I
‘ ‘ ’ I I = : :
0.0 1 i I Y T I
O 50 100 150 200 250 300 350
Shear rate, 1/s

 

 

 

Figure 27: Mucoadhesion comparison of 3% DHS solutions
It is apparent from these figures that all DHS solutions exhibit mucoadhesion with
positive m, contributed by adhesion. This is because of the existence of hydrogen bonding
formed between DHS and mucin, both of which consist of substantial amount of
hydroxyl groups. The difference in mucoadheison strength between 20DHS, 6ODHS, and

100DHS solutions at the same concentration is caused by their molecular weight

67

difference. At the same DHS concentration, theoretically there should be the same
amount of hydrogen bonding formed. However, another factor affecting the adhesion is
the molecular weight. This explains why 3% 20DHS, 60DHS, and 100DHS gave
different mucoadhesion strengths.
3.2.2.2 DHS solutions with the same viscosity

Since DHS solutions with the same concentration exhibited significantly different
mucoadhesion strength, they were further tested and compared as follows. Herein, 8.5%
60DHS and 10% 100DHS were prepared to have the same viscosity as that of 3% 20DHS
at the second Newtonian region. The mucoadhesion of 3% 20DHS was shown previously,
and the mucoadhesion of 8.5% 6ODHS and 10% 100DHS are shown in Figure 28 and 29,

and their three solutions are compared in Figure 30.

 

 

 

 

 

 

 

 

 

 

Mucoadhesion of 8.5% SODHS
50.0 — —--—- —~--—— ,_,, —— #_____ -__ onixture
1 ’ _ __. _ __~_ ___# _*_ _—... I Vmucus+Vpolymer
95. :3": r . . . . . .
gt . - - # .___ ———~ - -
gzoo~ I _ - u— 4- _.-..__._ ._ .-_._ -4
S 10.0 T ——— — ____ _ _ _____
0.0 f , , fl I I
0 50 1 00 1 50 200 250 300 350
Shear Rate, 1/s

 

 

 

Figure 28: Mucoadhesion of 8.5% 6ODHS

68

 

 

Mucoadhesion of 10% 100DHS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50.0 7— ——- --— —---~ ,
e o lexture
40.0 ~P—— . I Vmucus+Vpolymer
93 . . . 9 O Q
- 30.0
E 20.0 -—n-»-—-—— a r - - Ir — —~r—— ' —~—‘ —. .
> 10.0 l‘“ _ —- s _ 2%—
0.0 W i V T if I l
0 50 1 00 1 50 200 250 300 350
Shear Rate, 1ls
Figure 29: Mucoadhesion of 10% 100DHS
Mucoadhesion of lsoviscous Solutions
30.0
A
95. 20.0 -~-—‘———--————~~— e 3% 20DHS
3‘ . I . ‘
3 0 0 e u 8.5% 60DHS
10.0 ~ A 10%1OODHS
S
0.0 l T T
0 100 200 300 400

Shear Rate, 1/s

 

Figure 30: Mucoadhesion comparison of isoviscous DHS solutions

It should be noted that upon matching the viscosity solution of 20DHS, 6ODHS,

and 100DHS, they exhibited the same strength of mocuadhesion. This further confirmed

our conclusion that the adhesion strength is affected by both the molecular weight and the

amount of hydrogen bonding. Since viscosity of a polymer solution is also affected by its

molecular weight and interaction between chains, it is not surprising to see that solutions

with the same viscosity obtained the same adhesion strength.

69

 

3.2.2.3 DHSA Esters vs. 100DHS

Mucoadhesion of DHS esters were measured too and their results were compared

with that of 100DHS, based on which they were synthesized.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mucoadhesion of 10% DHSA1
500 a“ _ g “I m__._ _____ onixture
. - Vmucin+Vpolymer
Q3 40'0 ""'_—— ”V— e_' e "e e e e
. ' 30.0 ~— — -—
g 20.0 ~-— «~— .————.———1———. . '_____
5 10.0 ————
o.o 1 . . . . .
0 5O 1 00 1 50 200 250 300 350
Shear rate, 113
Figure 31: Mucoadhesion of 10% DHSA1
Mucoadhesion of 10% DHSA2
onixture
a. 40'0 TT’" ”TS—— uVmucin+Vpolymer
Q 30.0 *3 ~--—~-+~-»———-e————e—v —o——o——o———e————1
g 20.0 ~~ -~——---—1— .77 --. I I . I
.3 10.0 1F— —
>
0.0 l T l l l l
0 50 100 150 200 250 300 350
Shear Rate, 1/s

 

 

 

Figure 32: Mucoadhesion of 10% DHSA2

70

 

 

Mucoadhesion of 10% DHSP1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60'0 _ onixture
a. 50'0 ‘ i‘ — I Vmucin-I-Vpolymer
: 40.0 -———— -v-—’-—- —+——rh-r“—"T 1* e l
-‘.-; 30.0 ~— ~——-—-— —— I
O
>
10.0 q~——- —-~-— - —-*-— _____.
0.0 17 l l I l I
0 50 100 150 200 250 300 350
Shear Rate, 1Is
Figure 33: Mucoadhesion of 10% DHSP1
Mucoadhesion of 10% DHSP2
40-0 1_“— "“ eVmixture
. I Vmucin+Vpolymer
03 30.0 ~ —— ’ ~—-——¢———.——r—~—.———. ,
.2?
g 20.0 .t- .——— _ _ _ _ _ _ I
5 10.0 -_ ——fl---—-#——
0.0 i l l l T T
0 50 1 00 1 50 200 250 300 350

Shear Rate, 1/s

 

Figure 34: Mucoadhesion of 10% DHSP2

71

 

 

 

Mucoadhesion of DHS Esters

 

 

 

 

 

 

 

 

40.0 —-~- A” _. s__.w-+1_-_-_
. 010% DHS

D.

9 30.0 ___ __,___ T .10% DHSA1
iii . o e . . . 510%DHSA2
g 20.0 ..-.. 2; 3; i g. g ‘— x10%DHSP1
> x10? DHSP2

10.0 ‘ X x x x . x a at 1°

 

Shear Rate, 1/s

 

 

 

Figure 35: Mucoadhesion comparison of 10% DHS ester solutions

All DHS esters showed positive mucoadhesion, however, when they were
compared with 100DHS their mucoadheion was not as high as 100DHS. This lower
mucoadhesion is due to the following two reasons: (1) different amount of hydrogen
bonding and (2) different surface tension and therefore different work of adhesion. After
DHS is esterified, part of hydroxyl groups in the polymer backbone are transferred to
ester groups such that less of the primary hydroxyl groups are available to form hydrogen
bonding with mucin. In addition, after the hydrophobic methyl or ethyl groups are
grafted, the surface tension of the solution is reduced resulting in lower work of adhesion.
In conclusion, all the synthesized polymers showed strengthened mucoadhesion and the

strength of mocuadhesion is affected by the MW, concentration, ST, and functional

groups.

72

3.2.3 Rheological properties

3.2.3.1 Steady shear flow curve
Figure 36-42 show the rheograms of DHS and DHS ester solutions at different
concentrations. Unless noted here, the viscosity in the y-axis of all rheograms refers to

apparent viscosity. As can be seen from these rheograms, all samples are concentration

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

dependent
ZODHS Viscosity Profile
10.0
-. A”- _ _f _ ____ _-__ ____* ___.____..
% 8.0 A ‘ ‘ ‘ ‘ ‘ ‘ . 1%
’ 6.0 + ~~ 7~—--~—- #— —
. I 2%
4.0 “‘ g '— I ‘——'r‘ ' I"—"I —" '—I_—_I ‘ 3%
5 2.0 I_g____ v—* —+--~——+——-—+———o«——+———«————~
0.0 l fir I
0 100 200 300
Shear Rate. 113
Figure 36: Rheogram of 20DHS
SODHS Viscosity Profile
60.0
A
E 30.0 « - — —‘— 7——— —— — —— ———— I 8.5%
17°/
5 15.0 -————— ———-—— - - ‘ o
I I I I I I I I
0 0 o o o, o e . o o ?
0 100 200 300
Shear Rate, 1/s

 

 

 

Figure 37: Rheogram of 60DHS

73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1OODHS Viscosity Profile
40.0
I A ‘ ‘ ‘ ‘ ‘ ‘
Q; 30.0 4m —' — — --—-————~-— —— .
. 94%
20.0 a»-— —— — a 77—— I 10%
0
10.0 _b_____ ____ A20/o
I I I I I I I I
0.0 ’ ’ °. ° ’ . ’ ’ t
O 100 200 300
Shear Rate, 1/s
Figure 38: Rheogram of 100DHS
Viscosity Profile of DHSA1
40.0
9.): 30.0 1r --—‘—- -——‘—— —‘— L ——A-— A— “st—#1— “—~ .4%
20.0 -.--. —* i—— — — —— .3 I 10%
A 20°/
10-0 ‘* I ““1" r I“ — ‘ i—— 3""— I "I I "“‘ °
0.0 . I. .I . I . I . I . ?
0 50 100 150 200 250 300 350
Shear Rate, 1/s
Figure 39: Rheogram of DHSA1
Viscosity Profile of DHSA2
40.0
93. 30'0 ‘”“A‘—"_*—‘~"” ‘ A A A 'T 7‘7 94%
§20.0 a—-—- — —— ~——- —————— — — —— —— I10%
_ M _ _ ____ __ _____ __ _m A20%
5 10°C I I I I I I I I
0.0 ° 3 t ’ . ' . ’ . ' t
0 50 1 00 1 50 200 250 300 350

Shear Rate, 1/s

 

Figure 40: Rheogram of DHSA2

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Viscosity Profile of DHSP1
40.0
QE 30.0 + -‘—— -~--‘-——-—*‘r———*k—'i . ‘ A —a .4%
g 20.0 +— ——-— — —-r~—-»———— .___- I 10%
S 10.0 ‘1*—“ i‘ r“"‘—' I—' "i" ———'- —‘— —‘-- —‘—-——‘*-— —‘—.' "—“—“"* ‘ 20%
0.0 . I. .I . I . I . fl . ?
0 50 1 00 1 50 200 250 300 350
Shear Rate, 1/s
Figure 41: Rheogram of DHSP1
Viscosity Profile of DHSP2
30.0
A
‘ ‘ ‘ A A A A
9". 20.0 — -—- 7 .4%
g I 10%
A 20%
5 10.0 - —-- ——— <~ - -~-—W—
I I I I I I I I
0.0 ' .° °. ° . ° . ° .° 3
0 50 1 00 1 50 200 250 300 350
Shear Rate. 113

 

 

 

Figure 42: Rheogram of DHSP2
The relationship between viscosity and shear rate was studied for all samples to
investigate the non-Newtonian property. The model used to fit shear rate/shear stress
experimental data was the Ostwald model. It relates the apparent viscosity of the
solution and the shear rate applied. The Ostwald model is one of the most used to model

the behavior of non-Newtonian fluids due to its simplicity.

75

— .n
0' " K7 (3.18)

where
o: shear stress
7: shear rate

K: consistency index
n: flow index

When n>l, n=1, and n<1 the flow patterns denote shear-thickening, Newtonian
and shear-thinning patterns, respectively. n, as well as K, were estimated by linear
regression according to the Ostwald equation and are listed in Table 9.

Table 9: Regression of steady shear flow data

 

 

 

 

 

 

 

 

 

Polymer K (Pas) n R2

1% ZODHS 0.0025 0.97 0.99
2% ZODHS 0.0048 0.96 0.99
3% 200145 0.0096 0.95 0.98
4% 6ODHS 0.0029 1.00 0.97
8.5% GODHS 0.0092 0.99 0.98
17% 6ODHS 0.0650 0.93 0.99
4% 100DHS 0.0019 1.00 0.99
10% 1OODHS 0.0073 0.99 0.97
20% 1000115 0.0410 0.96 0.98
4% DHSA1 0.0017 1.00 0.99
10% DHSA1 0.0070 0.99 0.99
20% DHSA1 0.0380 0.94 0.98
4% DHSA2 0.0015 1.00 0.98
10% DHSA2 0.0064 0.99 0.97
20% DHSA2 0.0351 0.95 0.99
4% DHSP1 0.0014 1.00 0.99
10% DHSP1 0.0065 1.00 0.98
20% DHSP1 0.0352 0.96 0.98
4% DHSP2 0.0012 0.99 0.98
10% DHSP2 0.0052 0.99 0.99
20% DHSP2 0.0300 0.97 0.97

 

 

 

 

 

76

It is apparent that the Oswald model is appropriate here since all of the
correlation coefficients (R2) are higher than 0.97, indicating a good fit to the data. In
accordance with the viscosity data shown in above figures, the consistency index also
increases with an increase in the concentration. Also, shear-thinning behaviors are
observed from most solutions. Because of the equipment limitation, we were only able to
measure the viscosity under shear rate no less than 15 US, which might be beyond the
non-Newtonian range of some low viscosity samples. The viscosity drops more quickly
with a dilute solution than that of a solution with higher concentration. This explains
why n decreases with all DHS solutions when concentration is increased.

In addition, DHS and DHS ester solutions at the same concentrations were
compared with each other. Figure 43 shows the rheograms of 3% 20DHS, 60DHS, and
100DHS solutions. Among them, 20DHS showed the highest viscosity, followed by
6ODHS, and 100DHS, respectively. This difference is most likely due to the different
molecular weights of these solutions. Figure 44 shows the rheograms of 10% DHS esters
as well as 100DHS solutions. All DHS ester solutions had lower viscosity than 100DHS;
DHSP solutions had lower viscosity than the DHSA solution with the same degree of
substitution. The viscosity of an ester solution with higher degree of substitution is lower
than that of an ester with lower degree of substitution. These differences are due to the
fact that less hydrogen bonding is formed when hydroxyl groups are substituted by ester

groups during esterification.

77

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DHS Viscosity Comparison

9.0

8.0 ---: — "9.4—;— —4

7_0 --, -,___ -._._ . 9 9 0 4f
‘25 6.0 +—--—
, ' 5.0 _ — e 3%200HS

4,0 4— _ I 3%6ODHS
5 3.0 “1“?” :~ :w—f— —: . ._ A 3%1000Hs

2.0 — A J

1.0 _,_. —1

0.0 I I I

0 100 200 300
Shear Rate, 1/s

 

 

 

Figure 43: Viscosity comparison of DHS solution at the same concentration

 

 

 

 

 

 

 

 

 

 

 

 

DHS Ester Viscosity Comparison
9.0 i
. O
n. 8.0 —~ - 4 1" f r 1 910%1OODHS
°. 1 . _ I10%DHSA1
g 7 0 44 z ‘ I .. l ——-———-—J A10%DHSA2
9 x x A A‘ I i x10%DHSP1
> 6.0 «1 * x -—~X——x——+—X————x———~ x10%DHSP2
5.0 I I I
0 100 200 300
Shear Rate, 1/s

 

 

 

Figure 44: Viscosity comparison of DHS ester solutions at the same concentration
3.2.3.2 Viscoelastic behavior
Storage moduli and loss moduli, G' and G", at different frequencies of DHS and
DHS esters are reported in Tables 10-16. The loss tangent curves as a function of

frequency for the same systems are reported in Figure 45-51.

78

Table 10: Viscoelastic property of 20DHS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Frequency, It

200Hs 1% 3% 5%
f. Hz G'. Pa G", Pa G'. Pa G". Pa 6'. Pa G", Pa
0.46 0.00 0.01 0.00 0.04 0.00 0.11
0.68 0.00 0.02 0.00 0.04 0.01 0.16
1.00 0.00 0.03 0.00 0.06 0.04 0.24
1.47 0.00 0.04 0.00 0.09 0.15 0.35
2.15 0.00 0.05 0.00 0.15 0.36 0.66
3.16 0.00 0.07 0.00 0.28 0.82 0.74
4.64 0.00 0.12 0.00 0.21 1.75 1.64
6.81 0.00 0.34 0.30 1.37 5.34 3.34
10.00 0.00 0.69 0.45 1.73 7.38 4.95
14.70 0.55 1.49 2.34 7.00 18.68 12.15
25.10 6.35 11.49 37.63 35.88 141.13 56.22
39.00 46.47 68.64 158.00 110.00 332.29 155.25
61.00 467.22 448.12 600.00 460.00 865.64 487.49
100.00 4852.44 1883.55 6000.00 2401.25 8654.15 3265.15
ZODHS
100.00 +1% .4
5 +3%
l; 1.00 . . . . Hf.
0.10
98899985829688

 

Figure 45: Loss tangent of 20DHS

79

 

 

 

Table 11: Viscoelastic property of 6ODHS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60DHS 4% 8.5% 17%
1, Hz 6', Pa G", Pa G'. Pa G", Pa G'. Pa G", Pa
0.46 0.00 0.02 0.00 0.04 0.00 0.06
0.68 0.00 0.02 0.00 0.04 0.00 0.09
1.00 0.00 0.04 0.00 0.06 0.04 0.18
1.47 0.00 0.06 0.00 0.09 0.11 0.27
2.15 0.00 0.09 0.00 0.13 0.66 0.52
3.16 0.00 0.12 0.00 0.19 1.52 1.02
4.64 0.00 0.16 0.00 0.22 2.54 2.16
6.81 0.00 0.55 0.00 0.99 4.32 3.34
10.00 0.00 1.05 0.42 1.76 7.16 5.27
14.70 0.42 1.64 1.94 5.32 18.53 12.22
25.10 7.86 18.22 64.75 41.79 78.24 51.32
39.00 36.17 36.24 147.75 61.50 152.22 96.15
61.00 2877.35 1524.21 5238.00 2388.75 5328.19 2732.65
100.00 9537.48 2634.16 13938.50 3693.50 13925.22 5124.26
60DHS
10.00

‘2

-e—4%
g 1.00 1 1 1 1 1 1 1 1 1 1 +8.59%
.— +17%
5 0.10

0

Frequency, Hz

 

Figure 46: Loss tangent of 60DHS

80

 

Table 12: Viscoelastic property of 100DHS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Frequency, Hz

1OODHS 4% 10% 20%

f, Hz 6'. Pa G", Pa G', Pa G", Pa G'. Pa G", Pa

0.46 0.00 0.03 0.00 0.04 0.00 0.05

0.68 0.00 0.03 0.00 0.04 0.00 0.09

1.00 0.00 0.05 0.00 0.06 0.01 0.19

1.47 0.00 0.08 0.00 0.09 0.13 0.30

2.15 0.00 0.15 0.00 0.15 0.55 0.45

3.16 0.00 0.28 0.00 0.28 1.25 0.75

4.64 0.00 0.29 0.02 0.21 2.84 1.81

6.81 0.12 1.78 0.15 1.37 3.46 2.50

10.00 0.21 1.86 0.30 1.73 6.75 4.64

14.70 0.54 4.59 2.34 12.00 17.55 10.58

25.10 4.47 26.48 37.63 38.65 73.46 44.45

39.00 39.48 113.18 107.50 100.25 133.17 87.35

61.00 3090.59 982.55 2600.25 531.25 4568.16 2875.84

100.00 11536.22 2685.47 15069.75 2401 .25 11253.47 4458.46

1OODHS

100.00
L9
19

10.00 - +4%
g +10%
I- 1.00 1 1 4. 20%
0.10
9.) 0

 

Figure 47: Loss tangent of 100DHS

81

 

Table 13: Viscoelastic property of DHSA1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DHSA1 4% 10% 20%

f, Hz G', Pa G", Pa G', Pa G", Pa G', Pa G", Pa
0.46 0.00 0.03 0.00 0.04 0.00 0.05
0.68 0.00 0.03 0.00 0.04 0.00 0.08
1.00 0.00 0.04 0.00 0.06 0.01 0.19
1.47 0.00 0.07 0.00 0.09 0.08 0.22
2.15 0.00 0.12 0.00 0.14 0.44 0.41
3.16 0.00 0.25 0.00 0.26 1.21 0.82
4.64 0.00 0.26 0.00 0.19 2.16 1.59
6.81 0.00 1.35 0.04 1.21 3.20 2.62
10.00 0.00 1.78 0.25 1.53 5.73 4.05
14.70 0.17 12.69 2.04 11.19 15.19 9.53
25.10 3.25 34.58 33.18 41.45 62.20 40.25
39.00 34.18 121.23 95.35 112.39 117.67 79.65
61 .00 2758.15 786.54 2273.64 464.42 4540.45 2993.45
100.00 10054.46 2308.29 14352.47 2118.70 11563.54 4002.45

 

 

 

 

 

 

 

 

 

 

Loos Tangent, G"IG'

 

 

 

 

DHSA1
100.00
10.00 1 +4%
+10%
1.00 I T I I ‘ 7‘ 20%
0.10

0
I
‘3

13’ <9 5" ,9“ I989 65° 9°

Frequency, Hz

 

Figure 48: Loss tangent of DHSA1

82

 

Table 14: Viscoelastic property of DHSA2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ Frequency, Hz

DHSA2 4% 10% 20%

f. Hz G', Pa G", Pa 6'. Pa G", Pa G', Pa G", Pa

0.46 0.00 0.02 0.00 0.04 0.00 0.03

0.68 0.00 0.03 0.00 0.04 0.00 0.05

1.00 0.00 0.04 0.00 0.06 0.02 0.18

1.47 0.00 0.07 0.00 0.08 0.08 0.25

2.15 0.00 0.11 0.00 0.13 0.40 0.38

3.16 0.00 0.23 0.00 0.24 0.98 0.84

4.64 0.00 0.17 0.00 0.29 2.16 1.51

6.81 0.00 1.32 0.00 1.15 3.17 2.35

10.00 0.00 1.62 0.23 1.43 5.64 3.98

14.70 0.59 10.17 2.24 10.62 15.17 9.01

25.10 6.17 23.22 31.49 30.13 58.38 36.16

39.00 44.35 113.07 88.17 59.50 111.49 74.88

61 .00 2618.55 684.55 2005.22 433.17 3902.56 3305.36

100.00 8400.49 1795.57 10254.19 2014.55 11353.46 3821.13

DHSA2

100.00
{2
b

10.00 3 +4%
E \ + 10%
l- 1.00 1 1 1 1 .L 20%
g 0.10
‘3 o

 

Figure 49: Loss tangent of DHSA2

83

 

 

Table 15: Viscoelastic property of DHSP1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DHSP1 4% 10% 20%

f, Hz 6', Pa G", Pa G'. Pa G", Pa G'. Pa G". Pa

0.46 0.00 0.04 0.00 0.03 0.00 0.02

0.68 0.00 0.04 0.00 0.04 0.00 0.06

1.00 0.00 0.05 0.00 0.05 0.03 0.16

1.47 0.00 0.09 0.00 0.06 0.07 0.23

2.15 0.00 0.17 0.00 0.12 0.44 0.33

3.16 0.00 0.23 0.00 0.22 1.05 0.69

4.64 0.00 0.26 0.00 0.15 2.35 1.52

6.81 0.00 1.32 0.00 1.30 3.07 2.32

10.00 0.00 1.63 0.24 3.69 5.33 3.85

14.70 0.63 12.05 1.86 11.02 14.02 8.90

25.10 3.54 32.14 27.63 30.24 60.31 37.25

39.00 32.06 114.17 112.35 105.36 111.53 75.86

61 .00 2568.46 695.06 2036.95 443.65 3778.29 2865.03

100.00 8512.43 2144.86 13151.24 2015.42 10862.22 3814.33

DHSP1

100.00
52
1.9

10.00 ~ +41%
E + 10%
l- 1.00 1 1 20%
g 0.10
<0 0
055’ .99 ,5 8? I989 .95 69°

Freque ncy, Hz

 

84

Figure 50: Loss tangent of DHSP1

 

Table I6: Viscoelastic property of DHSP2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Frequency, Hz

 

 

DHSP2 4% 10% 20%

1, Hz 6'. Pa G", Pa G', Pa G", Pa G'. Pa G", Pa

0.46 0.00 0.03 0.00 0.03 0.00 0.03

0.68 0.00 0.04 0.00 0.04 0.00 0.06

1.00 0.00 0.04 0.00 0.04 0.01 0.13

1.47 0.00 0.06 0.00 0.08 0.07 0.19

2.15 0.00 0.13 0.00 0.13 0.39 0.35

3.16 0.00 0.25 0.00 0.24 0.87 0.66

4.64 0.00 0.19 0.00 0.25 1.98 1.33

6.81 0.00 1.12 0.21 1.04 2.56 2.10

10.00 0.00 1.38 0.60 1.31 4.77 3.41

14.70 0.49 10.55 5.12 10.12 12.35 8.30

25.10 2.77 28.69 14.98 29.58 56.28 34.23

39.00 28.36 104.37 79.35 73.53 96.68 70.12

61.00 2318.42 598.65 1954.24 401.65 3658.49 2193.53
100.00 7763.64 1954.63 10982.33 1796.35 9534.51 3215.25

DHSP2
100.00
9
b
10.00 1 +4%
g \+\ +10%
1- 1.00 1 a 1 1 1 1 1 1 1 t 20%
g 0.10
0
015° .59 .9" ,8? ~08? If; 6.5.9

 

All the systems display elastic and viscous moduli that are frequency dependent.

Figure 5 1: Loss tangent of DHSP2

85

Furthermore, in all cases both modulus increased as the frequency increased. At low
frequency all the formulations show a predominant viscous character (G" > 6') but as the

frequency is increased, the elastic modulus increases faster than the viscous modulus so

 

 

that G' curve crosses G" curve at a certain frequency called the cross-over frequency.
Even with higher fiequencies than the cross-over frequency the elastic modulus remains
high and the systems show a predominant elastic characteristic. This rheological behavior
is a feature of an entangled network that can finther be seen in the loss tangent curves.
For the solutions at the cross-over frequency the loss tangent is equal to 1, while it is
greater than 1 at low frequency and lower than I at high frequency. What should be
particularly pointed out is the cross-over point, which means the material demonstrates
equal viscous and elastic characters. At frequencies lower than the cross-over point, the
relaxation time of the material is shorter than the process time and the material has
sufficient time to relax. Thus, it shows more viscous behavior. When the frequency is
higher than the cross-over point, the process time is shorter than the material relaxation
time and there is not sufficient time for the material to relax, resulting in a more elastic
behavior.

The cross-over points of all the solutions we tested are dependent on
concentration. Namely, the concentration is inversely proportional to the cross-over
frequency. This is due to the fact that when the concentration is high there is more
entanglement and interactions between polymer chains and longer relaxation time is
needed when a stress is applied.

One of the key objectives in this study was to obtain ophthalmic solutions that
show an entangled solution behavior. Since such solutions are characterized by the
important feature of flowing as viscous fluid (G" > 6') when stressed at low frequency

therefore adapting to the ocular surface just like a natural tear and behaving elastically

86

when stressed quickly, therefore the entangled DHS should be able to stay together in

presence of quick stress change.

3.2.4 In vitro release profile of Ofloxacin

 

Figure 52: Structure of ofloxacin

Ofloxacin is an antibacterial agent belonging to the fluoroquinolone family with a
molecular weight of 361.37. Of the available fluoroquinolones, ofloxacin is usually
administered as a single agent and has been shown to have the best aqueous humor

penetration. As an ophthalmic formulation, ofloxacin is formulated as a 0.3% solution

and is known by the trade name OCUFLOX®. According to Allergan’s prescribing

information packet, OCUFLOX solution is unbuffered and formulated with a pH of 6.4
(range - 6.0 to 6.8). Ofloxacin is a fluorinated 4-quinolone which differs from other
fluorinated 4—quinolones in that there is a six member pyridobenzoxazine ring from
positions I to 8 of the basic ring structure.

The drug release profiles were studied using a UNICO SQ-2800 Ultraviolet-
Visible Spectrophotometer. The absorbance spectrum for the drug ofloxacin is shown in

Figure 53.

87

 

 

I I I I I I l I I
h' ....... t ------- ‘ ------- t ------- l ....... t ....... ' ....... t ------- ' ....... |
I I l I I I I I I

24ppm Drug

 

 

 

-1 : : L : n : 1 :
200 300 400 500 600 700 800 900 1000 1100
Wavelength (nm)

 

 

 

 

Figure 53: UVN is Spectrum of ofloxacin
The strongest peak, at 288 nm, was used to determine the concentration of
ofloxacin as compared to a calibration curve (Figure 54). The absorbance at
concentrations of 0.0036% w/v to 0.00075%w/v was found linearly dependent and
measurable using the parameters described in the analytical technique chapter. A fixed-
point measurement method was used so that only the absorbance at 288 nm was recorded.

This eliminated the need to develop a full spectra for all of the samples.

88

 

 

UV calibration! Conc vs Abs

 

 

    

 

 

50
4o _ y = 15.699X - 0.5251
30 - R2 = 0.9947
20 —
10 —
0
0 0.5 1 1.5
Abs at288 nm

 

’ Figure 54: Calibration curve for Ofloxacin

3.2.4.1 Release from isoviscous solutions

Figure 55 shows the release profiles from seven isoviscous DHS and DHS ester

solutions and phosphate buffer as control at pH 7.4 using phosphate buffer.

 

Release Percentage, %

 

0.3% Ofloxacin Release Profiles from lsoViscous Systems and Buffer Solution

120.00

 

100.00 J

.388
888

O
l
20.00 - °

3

0.00F T 7 I 1 1 T T T I

 

+5"

 

 

e 3% ZODHS

I 8.5% 6ODHS
A10% 100DHS
x 11%DHSA1
x 11.5%DHSP1
e Buffer

+ 11.5%DHSA2
-12%DHSP2

 

 

 

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0

Time, min

f

r

 

Figure 55: Release profiles of ofloxacin from isoviscous solutions

89

 

 

 

 

 

 

 

 

 

O 3% ZODHS

I 8.5% SODHS

A 10% 1OODHS
x 11%DHSA1

X 11.5%DHSP1
e Bufkr

+ 11.5%DHSA2
- 12%DHSP2

 

 

 

Regression Data
60.00
32. 50.00 . l-
8
53 40.00 4 o
8 to
g 30.00 4 n
a.
3 20.00 - '
a I;
a 10.00 J
m
0.00 k: l T l I I
0.00 1.00 2.00 3.00 4.00 5.00 6.00
Square root of time, min"0.5

 

Figure 56: Release Percentage from Isoviscous Solutions vs. Square Root of Release

Time

Table 17: Regression Data of Ofloxacin Release fi'om Isovisous Solutions

 

2

 

 

 

 

 

 

 

 

 

 

 

 

 

Solution . Slope value m (10"(-4)cm"%)
% ZODHS 9.54 0.9915 0.027 1.95
.5% 60DHS 9.17 0.9639 0.010 1.87

10% 100DHS 9.36 0.9813 0.011 1.91

11% DHSA1 9.20 0.9772 0.034 1.87

11.5% DHSA2 9.46 0.9957 0.016 1.93

11.5% DHSP1 9.29 0.9975 0.021 1.89

12% DHSP2 9.33 0.9936 0.013 1.90

Buffer solution 11.35 0.9978 1.000 2.31

 

 

The release percentage is also plotted in Figure 56 as a function of the square root

of the release time to fit the Baker and Lonsdale model and it appears to yield a linear

correlation. The slope, R2 values, p value, and diffusion coefficients, Dm are listed in

Table 17.

It is apparent that all R2 values are greater than 0.95, confirming the fit of the

release profiles from DHS and DHS ester solutions to Baker and Lonsdale model.

Furthermore, the slopes, which are proportional to diffusion coefficients of DHS as well

90

 

DHS esters, were compared with that of the blank buffer solution by T test. The p values
shown in Table 17 are all less than 0.05, which indicates that the release profiles of all
isoviscous samples are significantly different fi'om that of buffer solution.

Also, an ANOVA test was done to study whether there was a significant
difference between the release profiles of these isoviscous samples. The p value is
0.78473, which is greater than 0.05, indicating that there is no significant difference of
release profiles between these isoviscous solutions.
3.2.4.2 Release from solutions with different viscosity

Figure 57 shows the release from 100DHS at different concentrations (e.g.

different Viscosities).

 

 

 

 

 

 

 

 

0.3% Olloxacin Release Profiles from 1OODHS Solutions at
120.00 Different Concentrations
°\° 100.00 4
80.00 4 I
o
I A
600° ‘ .- ‘ 010% 1OODHS
A
40.00 _ I 4% 1OODHS
" A 20% 1OODHS
20.00 - ’
0.00 I Y I I I f
0.00 10.00 20.00 30.00 40.00 50.00 60.00
Time. min

 

 

 

Figure 57: Release Profiles of Ofloxacin from 100DHS at Different Concentrations
These data are also plotted as before as a function of the square root of the release
time yielding a linear correlation. The linear regression parameters (e.g. slope and R2

values) are listed in Table 18.

91

 

Regession Data

70.00

 

.88
88
D

 

40.00 4

30.00 1 o 010% 1OODHS

20.00 - I 4% 1OODHS

10.00 _ " A20% 1OODHS

 

Release Percentage. %

 

 

 

 

0cm l f I I I I I I I]
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00

Square root of time, min“0.5

 

 

 

 

Figure 58: Non Isoviscous Release Percentage vs. Square Root of Release Time

Table 18: Regression Data of Ofloxacin Release from Solutions with Different Viscosity

 

 

 

 

 

 

 

 

 

Solution slope R42 Dm (10"(-4)cm"2l92
4% 1OODHS 9.94 0.981 2.03
10% 100DHS 9.36 0.981 1.91
20% 100DHS 8.91 0.999 1.82

 

All R2 values are greater than 0.95, indicating a good fit to Baker and Lonsdale
model. T tests were examined between solutions at concentrations of 4% and 10%, 10%
and 20%, as well as 4% and 20% (Table 12). They all have p values less than 0.05,
indicating that they are significantly different from each other statistically. As shown in
Table 11, 20% 100DHS has the lowest diffusion coefficient and 4% 100DHS has the
highest diffusion coefficient.

Table 19: T Test Results of Solutions with Different Viscosities

 

 

 

 

test between P value
4%, 10% 0.034
10%, 20% 0.044
4%. 20% 0.020

 

 

 

 

92

Based on the in vitro release tests, we found that viscosity was a factor that
significantly affected the incorporated drug release profile. The drug diffused faster in
low viscosity solutions than in high viscosity solutions. However, this does not
necessarily agree with their in viva release profile. In this test the effects of drug-eye
contact time and interaction with mucus were not included, while they can significantly

affect in vivo drug release as well.

3.2.5 Stability Data

3.2.5.1 Storage stability

In this project, the effects of pH and temperature on the polymers’ storage
stability were studied. Samples 20DHS, 60DHS, 100DHS, DHSA1, DHSA2, DHSP1,
and DHSP2 were dissolved in pH 4, 7, and 9 buffer solutions. Six samples of each
solution were made. Three of them were stored at 4 °C and the other three were stored at
room temperature (22 +/- 2 °C). Their apparent viscosity was recorded at a shear rate of
300 US immediately after they were prepared and then compared with the viscosity
measured 30 days later. The results are shown in the Table 20-26.

Table 20: Storage stability of 3% 20DHS

cPs °C
1--1.4 .1--1.2

.0--7.5 .0-7.5
.0--7.5 .0--7.5

 

Table 21: Storage stability of 8.5% 60DHS

cPs °C
.8--1 .4

.7—7.7
.5--7.4

 

93

Table 22: Storage stability of 10% 100DHS

cPs °C

 

Table 23: Storage stability of 10% DHSA1

cPs °C
.3--2.0
.2—7.0
.4—7.3

 

Table 24: Storage stability of 10% DHSA2

cPs °C
.9--2.1
.7—6.8
.2—7.3

 

Table 25: Storage stability of 10% DHSP1

cPs °C

 

Table 26: Storage stability of 10% DHSP2

 

 

 

 

 

 

Viscosity, cPs 4 °C Room temp._
H 4 3.6—-2.1 3.6—20
H 7 6.3«6.4 6.3--6.3
H 9 7.2—7.2 7.2—7.1

 

 

 

No difference was observed between the data collected at 4 °C and room

temperature, indicating these solutions can be stored at room temperature. However, the

pH had a significant effect on the conditions these solutions were dissolved and stored.

At pH 4 the viscosity of DHS solutions was much lower than that of the solutions at pH 7

94

and 9. This agrees with Sihtola’s conclusion22 that DHS hydrolyzes at acidic
environment. At pH 7 and 9 solutions of samples 60DHS, 100DHS, and DHS esters
showed excellent stability with no viscosity change. However, we did observed a
viscosity drop of 20DHS solutions. Apparently, the structure of 20DHS is very close to
that of starch and retrogression takes place as in a starch solution.
3.2.5.2 Effect of autoclaving

The purpose of this test was to study the effect of autoclaving (121 °C, 20 min) on
the polymer molecular weight. In this test the viscosity was measured as an indicator of
the molecular weight change. DHS and DHS ester solutions were tested at pH 7. The
viscosity of these samples was measured with shear rate set at 300 US by a Haake
rheometer. The viscosity measurement was repeated twice before and after autoclaving
and the results are shown as follows.

Table 27: Viscosity drop after autoclaving

 

 

 

 

 

 

 

 

 

Before After %(viscosity
Viscosity, cPs autoclaving autoclavinL drop)
3% ZODHS 8.0 4.5 43.8
8.5% 600HS 7.7 5.8 24.7
10% 1OODHS 7.3 5.9 19.2
10% DHSA1 7.1 5.1 28.2
10% DHSA2 6.9 4.7 31.9
10% DHSP1 6.8 4.6 32.4
10% DHSP2 6.2 5.0 19.4

 

 

 

 

 

A drop in the viscosity was observed with all DHS solutions, with 20DHS having
the largest drop indicating that hydrolysis occurred under the high temperature and
pressure conditions. However, this hydrolysis should not affect the performance of

ophthalmic solutions too much, as long as autoclaving does not impact their rheological

95

properties too much. We also observed that the viscosity drop takes place in CMC
solutions, which is one of the most commonly used polymers in ophthalmic solutions.

In summary, based on the property characterizations described in this chapter, the
synthesized polymers showed very good properties. The variable studied allowed us to
control the surface activities, mucoadhesnion, rheological behaviors, sustained drug
release, and storage stability. In vivo tests would be conducted to further investigate their

potential as ophthalmic applications.

96

Chapter 4 Performance Characterization

Based on the property characterizations described in the previous chapter, the
modified starch products were shown to be good candidates for ocular drug delivery
applications. In this chapter, we report the results of our in vitro performance tests and
the results of in vivo irritancy tests that were run to confirm safety issues. Specifically,
the following tests will be discussed: in vitro EpiOcular test, in vivo irritancy test, Tear
Break-up Time (TBUT) test, and Timolol release test. All these tests were conducted in
cooperation with Dr. Wendy Townsend.
4.1 Experiments
4.1.1 EpiOcular in vitro test
The EpiOcularTM model provides a predictive, morphologically relevant in vitro means to
assess ocular irritancy. In this study, the Epi Ocular in vitro test was run with the

MatTek's EpiOcularTM corneal model in Dr. Wendy Townsend’s lab using the following

the SOP:

I. Pre-warm the MatTek assay medium to 37 0C by placing it in an incubator for
one hour

2. Use sterile technique pipet 0.9 ml of assay medium into each well of sterile
six well plates

3. Transfer EpiOcular samples using sterile forceps into the six well plates
containing the prewarmed assay medium

4. Place at 37 °C and 5% C02, humidified chamber for one hour prior to dosing.

5. Remove the assay media and replace with 0.9 ml/well of fresh media

97

10.

11.

12.

13.

14.

Pipet 100 ul of test material into cell culture insert atop the EpiOcular sample.
For negative controls dose with 100 ul of deionized water. Perform in
triplicate for each material. Use Triton X as positive control at same time
exposures.

Exposure time is 16 minutes initially, if viability > 90% then 64 minutes and
256 minutes

Before the end of the exposure period, thaw the MTT concentrate and place 2
ml of concentrate in with 8 ml of MTT diluent. Filter to remove any
precipitate. Store remaining MTT in the dark at 4 °C

Prepare a 24 well place by pipetting 300 ul of the MTT solution into each well
needed for an insert and label the 24 well plate top.

Discard liquid atop EpiOcular tissues. Submerge each insert to fill with PBS
and decant. Repeat three times. Submerge cell culture, insert in a well of a 12
well plate containing 5 ml of assay media for 10 minutes at 37 °C and 5% C02
Decant and shake off assay medium. Place EpiOcular sample in the MTT
containing 24 well plate. Return to the incubator at 37°C and 5% CO2 for 3
hours

Remove each insert and gently rinse with PBS. Shake the insert and blot with
a Kimwipe. Place in pre-labeled 24 well extraction plate.

If test article is colored then pipet 1 ml into the well and place the insert in
each well (If it is not colored pipet 2 ml into the cell culture insert itself).
Place cover and wrap in aluminum foil. Place on counter overnight in a

sealed plastic bag.

98

15. If placed test material in each well then remove insert and add 1 ml of
extractant. If placed test material in the cell culture insert then decant into the
well and discard the insert.

16. Pipet extractant solution up and down three times to mix well

17. Pipet 200 ul to a 96 well plate, perform in triplicate for each well

18. Measure optical density (OD) at 540 nm. Blank is 200 U] of extractant.

Calculate % viability = 100 x [OD sample / OD negative control] (4.1)

4.1.2 In vivo irritation test

In vivo irritation tests were done before any other in vivo tests to ensure the test
substances were accepted by the test animals. Both short-term and long-term irritation
tests were carried on as described below.
4.1.2.1 Assessment of ocular irritation after a single administration of the test
formulations

Eighteen rabbits were divided into six groups evenly. The groups were composed
of rabbits receiving either a saline control, Gellan gumTM control, or the DHS
formulations. One drop (50 ul) was administered to one eye of each rabbit. The rabbits
were examined via slit-lamp biomicroscopy and fluorescein staining (as needed) pre-
administration and at 15 minutes, 30 minutes, 1 hour, 2 hours, 3 hours, and 5 hours post-
administration. Scoring of ocular signs was performed according to the Hackett—
McDonald Ocular Scoring System 44 . Any complications or irritation from the
preparations was treated as required [either topical antimicrobial ophthalmic medication
(neomycin/polymyxin/bacitracin ophthalmic ointment '/4 inch strip q4-8 hours) if corneal

or conjunctival ulceration developed or topical anti-inflammatory ophthalmic

99

medications (1% prednisolone acetate ophthalmic suspension 1 drop q4-12 hours) if
inflammatory responses developed]. Any pain associated with corneal ulceration or
severe inflammation was treated with ketoprofen administered subcutaneously at 3 mg/kg
q24h and butorphanol tartrate administered at 0.25 mg/kg subcutaneously every four
hours until resolved. When any systemic signs of toxicity or disease should arise they
were addressed by the attending laboratory animal veterinarian.
4.1.2.2 Assessment of ocular irritation after multiple doses of the test formulations

In this long-term irritation test we followed a similar procedure as the short-term
irritation test. The differences were that the doses were administrated every 2 hours from
7 AM until 7 PM for a period of 7 days and the rabbits were examined at pre-

administration, day 2, and day 7.

4.1.3 Tear break-up time test

The normal tear film is continuous. Blinking maintains the tear film continuity. If
the eye is kept open long enough, without blinking, the tear film will start breaking up.
The eye will feel uncomfortable and be forced to blink. In subjects with dry eyes the tear
film is unstable and breaks up faster. Therefore the tear breakup time in subjects who
have dry eyes is shorter. However, an effective artificial tear can extend the tear breakup
time. In this study tear break-up test was used to assess the feasibility of DHS and DHS
esters for use of dry eye solutions.

There were 10 rabbits in a group. The group was composed of rabbits receiving
initially balanced salt solution (B88) and then the polymer solutions. Each rabbit was
anesthetized by mask administration of isoflurane and oxygen. One drop (SOul) of

balanced salt solution containing 1.25mg/ml of fluorescein was administered to one eye

100

of each rabbit once. The lid was manually closed twice to distribute the fluorescein stain.
The lids were then manually kept open for measurement. Immediately afier the second
blink a timing instrument was started. The tear film was scanned using a broad beam slit
lamp (Kowa SL-IS) in a darkened room using a cobalt blue filter. The timer was stopped
when the film disruption occured first The ocular surface was flushed with BSS. The
polymer formulations were then tested by an identical procedure. If any post-operative
discomfort was noted as detected by the presence of blepharospasm, rubbing at the eye,
decreased activity, or inappetance, butorphanol tartrate was administered at 0.25 mg/kg
subcutaneously (can be repeated every 4 hours if needed). Any complications or
irritation from the preparations was treated as required [either topical antimicrobial
ophthalmic medication (neomycin/polymyxin/bacitracin ophthalmic ointment '/4 inch
strip q4-8 hours) if corneal or conjunctival ulceration developed or topical anti-
inflammatory ophthalmic medications (1% prednisolone acetate ophthalmic suspension 1
drop q4-12 hours) if inflammatory responses developed]. If any systemic signs of toxicity

or disease should arose was addressed by the attending laboratory animal veterinarian.

4.1.4 In viva Timolol release test

4.1.4.1 In viva drug release profiles

Timolol is a drug used to lower pressure in the eye for glaucoma patients by
reducing aqueous production. In viva Timolol release tests were used to evaluate the
performance of synthesized polymers as ocular drug delivery systems. Sixty rabbits were
divided into three groups evenly. The groups were composed of rabbits receiving either a
timolol 0.5% commercial ophthalmic solution (topical glaucoma agent) or timolol 0.5%

in our polymer solutions. One drop (50 ul) was administered to one eye of each rabbit

101

once. At each time point (15, 30, 60, 90, and 120 minutes) post-administration, four
rabbits from each group were anesthetized by mask administration of isoflurane and
oxygen. Topical proparacaine (a topical anesthetic agent) was applied to the cornea. A
sample of 60-80ul of aqueous humor was obtained from the anterior chamber. The
sample was stored at —20°C. The concentration of Timolol present within the sample was
then determined using high pressure liquid chromatography (I-IPLC). Prior to recovery,
ketoprofen was administered subcutaneously at 3 mg/kg. Also, 250 ml of warmed
subcutaneous fluids was administered while anesthetized to decrease the risk of GI stasis.
If any post-operative discomfort was noted as detected by the presence of blepharospasm,
rubbing at the eye, decreased activity, or inappetance butorphanol tartrate was
administered at 0.25 mg/kg subcutaneously (can be repeated every 4 hours if needed). If
signs of ocular inflammation resulted from the aqueouscentesis either topical anti-
inflammatory agents (1% prednisolone acetate ophthalmic suspension 1 drop q4-12
hours) was administered or systemic anti-inflammatory agents (ketoprofen
subcutaneously at 3 mg/kg q24h) was administered to alleviate the inflammation. If any
signs of intra-ocular infection were noted, systemic antibiotics (enrofloxacin 10 mg/kg
per as q 12 hours) was administered after consultation with the attending lab animal
veterinarian.
4.1.4.2 Intra-Ocular Pressure test

Timolol is the drug being delivered to decrease in intra-ocular pressure (IOP).
Therefore by serially measuring the IOP one should have a measure of whether the drug
is being delivered for a sustained period of time. Measurement of IOP is a very non-

invasive procedure and can be completed with the use of a topical anesthetic. The IOP of

102

each eye of ten rabbits will be evaluated at time 0 and then 2 hours, 6 hours, and 8 hours
after the administration of one drop (50 ul) of solution. One drop of proparacine
hydrochloride will be administered before the IOP is measured utilizing the TonoPenVet
applanation tonometer. A one day recovery period will be utilized after the application.
The [OP of each eye of ten rabbits will then be evaluated as above afier the
administration of one drop (50 ul) of either commercial Timolol eye drop or polymer
solution containing 0.5% timolol maleate in both eyes. A one week washout period will
occur between measurement series. A sample size of 10 animals was selected based on
the following calculations. If one assumes a 0.9 mmHg difference between the BSS
control and the trial solution, with a standard deviation of 2.5 mmHg and a p<0.05 then to
achieve a power of 80%, a sample size of 10 is required“.

4.2 Results and Discussion

4.2.1 EpiOcular

The viability data, with 256 minutes exposure time, of 20DHS and 100DHS at
concentrations 1%, 3%, 5%, and 10% were collected and are shown in Figure 59. They
all had satisfactory results with viabilities higher than 80%, which indicate minimal
irritancy. The same tests were also done with 10% DHS esters and they showed

viabilities close to 100%.

103

 

 

EpiOcular Viability Results

 

 

 

 

 

 

r?
E
(I
.5
i=2
8
8
0.
Figure 59: EpiOcular results of 20DHS and 100DHS
EplOcular Viability Results
.3
8
E I 10% solutions

 

Test Substance

 

Figure 60: EpiOcular results of DHS Esters

104

 

4.2.2 In viva irritation test

There were two irritation tests performed on the rabbits. For the short-term
irritancy the drops were applied once and then the animals were evaluated with the slit
lamps at 15 min, 30 min, 60 min, 2 hours, 3 hours, and 5 hours post-application. Samples
tested included solutions of DHS, DHSA2, and DHSP2. The scores were recorded as 0
(normal) or 1-4. The scores were all 0 at all time points, which meant everyone was as
normal as the control eye (the other eye). The same polymers were also tested with the
long-tenn method. For the long-term irritancy the drops were applied six times daily for
seven days and evaluated pre and at days 2 and 7. The scores were the same as

before. All had a total score of 0, which meant normal.

4.2.3 Tear Break-up time

In this section the solutions of DHS and DHS esters were tested. According to the
previous discussion, a prolonged TBUT is helpful for stabilizing the tear film. For
comparison some commercial products (Advance Eye Relief from Bausch & Lomb and
Systane from Alcon) and solutions of commercial polymers, which were commonly used
in ophthalmic solutions, (carboxymethylcellulose, hydroxylpropyl methylcellulose, and

sodium hyaluronate) were also tested. The results are summarized as following.

105

 

Comparison between DHS Solutions with Different Concentrations

80.00
70.00
60.00
50.00
40.00
30.00
20.00
10.00

0.00

 

TBUTofSoluion-TBUTd'BSS. Sec.

0.05% 20 DHS 0.25% 100 DHS 3% ZODHS 10% 100DHS

 

 

 

Figure 61: TBUT comparison of DHS solutions with different concentrations
In Figure 61 20DHS and 100DHS were tested under different concentrations. The
solutions with high concentrations had much longer TBUT than those with low
concentrations. This can be contributed to multiple reasons. Comparing with low
concentration solutions, high concentration solutions have higher viscosity, lower
relaxation time, and stronger mucoadhesion therefore, they are able to resist shear caused

by blinking and stay in the eye for a prolonged time as an unbroken surface.

 

TBUT of Isoviscous Solutions

 

80.00
70.00
60.00
50.00
40.00
30.00
20.00
1 0.00

0.00

 

 

 

TBUTofSolution—TBUTOIBSS, Sec.

 

§§§w§§§
3%
0°.- 5

 

 

Figure 62: TBUT of Isoviscous solutions

106

In Figure 62, the isoviscous solutions made of DHS and DHS esters are
compared. Among the three DHS solutions, 3% 20DHS had the longest TBUT, and
60DHS and 100DHS had similar TBUT, which was lower than that of 20DHS. Given
that they have similar surface tension, rheological properties, and mucoadhesion strength
it is not clear why the performance of 3% 20DHS is different. The performance of DHS
ester solutions was compared with that of 100DHS. Among them, DHSA1 had the best
performance with a TBUT at 69 seconds. It is followed by DHSA2, which had a similar
TBUT as 100DHS. Both DHSP1 and DHSP2 had shorter TBUT than 100DHS and
DHSP2 had the shortest one at only 27 seconds. The most important reason for these
differences is most likely related to the difference in different surface tension between
these samples. The results confirm our prediction that a surface tension lower than that of
water should help an ocular drug delivery system to remain in the eye for a prolonged
time. These results further explain the reason DHSA1 had much longer TBUT than
100DHS. In addition, we also predicted previously that if the surface tension was lower
than a certain level, it would not help any more and may even turn to be a negative factor.
This explains why the TBUT of DHSA2, DHSP1, and DHSP1 is shorter than TBUT of
100DHS and their TBUT is in the order of DHSA2 > DHSP1 > DHSP2. This means
when the surface tension decreases, although it helps spontaneous spreading on the
cornea, it also lowered the work of adhesion, and therefore, the surface tension of an

ophthalmic solution should be only slightly lower than that of water.

107

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TBUT of Commercial Prodcuts and Polymers

70.00
3 60.00
a 50.00
‘8
E 40.00
' 30.00 «——
8
3
3 20.00
‘8

0.00 i . . .
Advanced Eye Systane 1% HPMC 1.3% CMC 0.3% Na hyal
Relief

 

 

 

Figure 63: TBUT of commercial products and polymers

In addition, in order to examine the performance of our polymers when
comparing with available products, we tested two commercial products and three
commercial polymers that are commonly used in ophthalmic solutions. It was exciting to
see that several of our polymers had a better performance than commercial products.

Based on the above TBUT results, we found that in order to obtain improved tear
film stability a tear substitute formulation should have the following characteristics: (1)
optimized viscosity, if the viscosity is too low, the solution is not able to stay on the
cornea, if the viscosity is too high irritancy can be caused, which thereafter results in
increased blinking frequency; (2) optimized surface tension, if the surface tension is too
high, it prevents spontaneous spreading on the cornea, if the surface tension is too low, it
will result in low adhesion strength; and (3) mucoadhesion, which keeps the polymer on

the cornea for prolonged time.

108

4.2.3 In vivo Timolol release results

Timolol release profiles were measured with selected polymers to make sure that
Timolol entered the inside of the eye. Afterwards, IOP tests were done to determine their
performance. The results are as follows.
4.2.3.1 In viva Timolol release profiles

The in vivo Timolol release profiles of 20DHS and 100DHS solutions were
collected. 3% 20DHS, 3% 100DHS, and 10% 100DHS were tested, along with a
commercial Timolol eye drop as a control. In this test, 3% 20DHS and 10% 100DHS had
much higher concentrations at all of the time points, this means they were able to stay in
the eye for a longer time compared with the commercial control. Their performance is
believed to be contributed by its (1) stronger mocuadhesion, which helps drug stay in the
eye with longer retention time in the eye; (2) relatively higher viscosity, which means
slower drug release compared with 3% 100DHS and the control; and (3) lower relaxation
time, which means more elastic character during blinking. In contrast, 3% 100DHS did
not help because of its low viscosity and mucoadhesion strength. The results are shown in
Figure 64. Based on these results, we confirmed that our polymers had the ability to
deliver Timolol into the eye. However, although the drug was detected in the eye, it did
not mean our goal (to decrease the ocular pressure) was achieved. In the next section IOP
results are presented to show the performance of our polymers in terms of decreasing

ocular pressure.

109

 

In vivo Timolol Release

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

160
140 .- + Control Timolol
+3% 100DHS
c 120 -- , A 3% 20DHS
% " I 10%1OODHS
g 100 ‘——‘ "'
'8 80 4e .._ ___ - M-_ _ , #iw ‘_-_
E II I
75: 30 l ._-. - _.-_ _a._-— .___.-___~_: .---~__ -_~
3 i
0 I
0 40 .5. _fi _ _I _ _._ ;~._ .___¥ 7 _. ;m
.5 ‘L T
20-4.-..5 v . i
I I i
o T I I I ‘ fir .
0 20 40 60 80 100 120 140
Time (Minutes)

 

Figure 64: First set of Timolol release test

4.2.3.2 IOP results

Change in IOP (AIOP) for each eye is expressed as follows:

MOP = IOPzero time - 101)time, t (42)

In this test, besides our selected solutions (4% 20DHS, 5% 20DHS, 20% 100DHS,
and 20% DHSA1), two commercial products were also tested as controls plain Timolol
and Gel Forming Solution (GFS). According to their Indications and Dosage information,
the plain Timolol should be applied twice a day and the GFS should be applied once a
day. In this test, all solutions with 0.5% Timolol were applied once and the IOP were
measured at 2 hours, 6 hours, and 8 hours. Their IOP change calculated according to

Equation (4.2) and the standard deviations were listed in the following table. AIOP is

110

plotted as a function of time. The results can be seen in Figure 65. The IOP measurement
of plain Timolol showed a drop at 2 hours but did not sustain and diminished afterwards.
This is because there was no carrier in the plain Timolol solution and only a small
fraction of drug diffused into the eye at the beginning due to the high drug concentration
in the precomeal region right after the solution was applied. The measurement of GFS
did not show any IOP drop at any time point. This might be caused by its short retention
time in the eye. When the GFS was dropped into the eye the aqueous solution of xanthan
gum in the presence of tear protein (lysozyme) formed a gel with elastic rather than
viscoelastic character, which might cause irritancy and the irritancy further caused fast
and frequent blinking, therefore resulted in a short retention time in the eye. When
compared with either of the commercial products, our polymers showed sustained effect.
At 2 hours, 4% 20DHS had significantly better performance than the GFS and
comparable performance as plain Timolol. At hour 8, 4% 20DHS showed significantly
greater IOP drop from both of the commercial products. At 6 hours, it is not clear what
caused the poor performance. The performance of 5% 20DHS was not as good as that of
4% DHS. This might be due to its higher viscosity and more elastic character when
comparing with 4% 20DHS. Although, the average lOP drop of 5% 20DHS showed a
sustained trend. Similar as 5% 20DHS, 20% 100DHS and DHSA1 also showed slight
IOP drop at all times. What needs to be particularly pointed out is the performance of
DHSA1. It did not show better performance than 100DHS, as in the TBUT test. This
observation means that surface tension is a more important factor for tear film stability
than Timolol delivery. This is because that surface activity is a key factor for spreading

and spreading determines how the ophthalmic solution covers on the cornea. However,

111

instead of targeting the outer surface of the cornea the polymers stays in the corner of eye

for Timolol delivery, therefore spreading is not as important as for TBUT.

Table 28: IOP change

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IOP Change
Hour 2 Hour 6 Hour 8
Samfle Average, mmHg Std. Dev. Average. mmHg Std. Dev. Average, mmH Std. Dev.
Plain Timolol 0.9 1.3 -0.4 1.6 -0.1 1.4
GFS -1.6 1.6 0 1.4 0 1.1
% 20DHS 0.7 1.2 -0.2 1.4 1.4 2.2
5% ZODHS 0.5 1.6 0.4 1.5 0.4 1.3
20% 1000HS 0.3 1.3 0.2 1.3 0.2 1.4
20% 1OODHSA1 0.2 1.6 0.6 1.3 0.3 1.2
IOP Change vs. Time
2
1.5 .
1 .
i? +P|ain Timolol
E 0-5 —I—GFS
o“ 0 +4% 20DHS
g —I—5% 20DHS
o -0.5 1 +20%100DHS
5 +20% 100DHSA1
-1 l
-1.5 ..
-2
Time. hour

 

 

Figure 65: IOP change vs. Time

Based on the above Timolol release test results, we can draw the following

conclusions: (1) in order to obtain effective treatment and minimized instillation

frequency drug delivery systems with controlled drug release are needed. (2) Comparing

with drug delivery systems that forms gel in the eye a system with optimized viscoelastic

 

character has more advantage by offering low irritancy under low frequency of blinking
and high stability under high frequency of blinking, therefore maximized effective
treatment and patient acceptance. (3) Comparing with TBUT, surface tension is not a
significant factor for Timolol delivery.

In summary, given the prolonged TBUT and sustained drug release from in viva
tests, our polymers demonstrated a good potential for ophthalmic solutions, either as drug
delivery systems or dry eye solutions. Therefore, commercialization of these polymers
should be further pursued. In addition, the data confirm our prediction that the ocular
drug delivery process should be improved though a multidisciplinary approach. All
factors including surface tension, mucoadhesion, rheology, and controlled drug release
work together to determine the performance as an ocular drug delivery system. Such
approach will help others and us in this field to design effective ocular drug delivery

systems.

113

Chapter 5 Conclusions and Recommendations
5.1 Conclusions

0 The synthesized polymers, DHS and DHS esters, can be characterized by FTIR,
NMR, and chemical method to analyze the structure and degree of modification,
qualitatively and/or quantitatively.

o The degree of modification is controllable by varying reaction conditions.

0 The polymers are water dispersible or soluble, depending on the controlled degree
of modification.

0 The DHS ester solutions can reduce the surface tension of an ophthalmic solution,
which helps them spreading on the corneal surface.

0 The polymers show strengthened adhesion with mucin, which can offer the drug
delivery system prolonged retention time when in contact with mucin.

0 The polymer solutions have desired flow and viscoelastic properties for
application of ophthalmic formulations.

o The polymer solutions have controlled drug release profiles consistent of hydrogel
matrices.

o The products have acceptable irritation test results from both in vitro and in viva
tests.

0 In viva Tear Break-Up Time Tests show some candidates have prolonged TBUT,
which indicates that they can be used to stabilize the tear film.

0 In viva Timolol release tests provide promising results for sustained drug release,

which offers the potential for effective ocular drug delivery.

114

5.2 Fundamental Contribution
Our fundamental contribution is the building of a structure-property-performance

relationship for polymer-based ocular drug delivery systems. According to our work, the
performance of an ocular drug delivery system is determined by properties such as
surface tension, adhesion, and rheology. The surface tension is important for spreading
and strength of adhesion, mocuadhesion is essential for the retention of prolonged
retention time in the eye, and rheological properties regulate resistance to the shear stress,
which further affects the retention time, and drug release profile. In addition, all these
properties can be optimized according to specific needs by engineering polymers with
various structures. By understanding the structure-property-performance relationship, it
will help designing of effective polymer-based ocular drug delivery systems.
5.3 Recommendations

0 Study the temperature’s effect on polymers’ rheological properties.

0 IOP tests with prolonged test time.

0 Further investigation for dry eye application with corneal topographic

modeling system.

115

References

lKaur, I. P. and Kanwar, M. (2002). "Ocular preparations: The formulation approach."
Drug Development and Industrial Pharmacy 28(5): 473-493.

2 Mindel, J. S. (I 982) Bioavailability. Biomedical Foundations of Ophthalmology:
Chapter 22.

3 Burrows, J ., Tsibouklis, J ., and Smart, J. (2002) “Drug delivery to the eye.” The Drug
Delivery Companies Report 2002 (Pharmaventures)

4 Robinson JC. (1993) “Ocular anatomy and physiology relevant to ocular drug delivery.”
Ophthalmic Drug Delivery Systems, Marcel Dekker: New York, US, 29-57

5 Davies, N.M. (2000). "Biopharmaceutical considerations in topical ocular drug
delivery." Clinical and Experimental Pharmacology and Physiology 27:55 8-562.

6 Brewitt, H. and Sistani, F. (2001) “Dry Eye Disease: The Scale of the Problem”
SURVEY OF OPHTHALMOLOGY VOLUME 45 SUPPLEMENT 2 MARCH 2001

7 Albietz, J. M. (2001). "Dry eye: an update on clinical diagnosis, management and
promising new treatments." Clin Exp Optom 84(1): 4-18.

8 Geerling, G., and Liu, 08. et al. (2002). "Costs and gains of complex procedures to
rehabilitate end stage ocular surface disease." Br J Ophthalmol 86(11): 1220-1.

9 Baudouin, C. (2001). "The pathology of dry eye." Surv Ophthalmol 45 Suppl 2: 8211-
20.

'0 Calonge, M. (2001). "The treatment of dry eye." Surv Ophthalmol 45 Suppl 2: S227-
39.

” Bach, F. C., J. B. Adam, et al. (1972). "Ocular retention of artificial tear solutions.
Comparison of hydroxypropyl methylcellulose and polyvinyl alcohol vehicles using an
argyrol marker." Ann Ophthalmol 4(2): 116-9.

'2 Sugiyama, T., S. Miyauchi, et al. (1991). "The effect of sodium hyaluronate on the
migration of rabbit corneal epithelium. II. The effect of topical administration." J Ocul
Pharmacol 7(1): 53-64.

'3 Sullivan, D. A., H. Ariga, et al. (1994). "Androgen-induced suppression of autoimmune
disease in lacrimal glands of mouse models of Sjogren's syndrome." Adv Exp Med Biol
350: 683-90.

116

'4 Rieger, G. (1990). "Lipid-containing eye drops: a step closer to natural tears."
Ophthalmologica 201(4): 206-] 2.

'5 Urtti, A. (1995). “Kinetic Aspects in the Design of prolonged action ocular drug
delivery systems.” Advances in Pharmaceutical Sciences.

'6 Rolando M, Zierhut M,‘ (2001). The Ocular Surface and Tear Film and Their
Dysfunction in Dry Eye Disease, Survey of Ophthalmology Volume 45 , Pages 8203 -
8210.

'7 Lindberg, B.; Misiomy, A. Quantitative analysis of reducible carbohydrates by means
of sodium borohydride. Svensk. Papperstidn. 1952, 55, 13-14.

'8 Miladinov,V.D. and Hanna,M.A. (2000) Starch esterification by reactive extrusion,
Industrial Crops and Products 11 (2000) 51—57

1")Solomon, 0. F. and Ciuta,I. Z. (1962) Determination of the intrinsic viscosity of
polymer solutions by a simple determination of viscosity , J. Appl. Polym. Sci, 6,683.

20 Sihtola, H. Chemical properties of modified cellulose. 2003'Die Makromolekulare
Chemie Volume 35, Issue 1. Pages 250 — 265

2' Narayan, R. cellulose modification Engineering, L. o. R. R. (1983). Conversion of
cellulose and xylan into glycols. NSF Grant No. 8022426AI.

22 Billmers, R. L. and Tessler, M. M. Method of preparing intermediate DS starch esters
in aqueous solution, United States Patent 5321 I32

23 Holly, F. J. and M. A. Lemp (1971). "Wettability and wetting of corneal epithelium." Exp
Eye Res 11: 239-250.

24 Cape, C., Dilly, P. N. et al. (1986). "Wettability of the corneal surface: a reappraisal." Curr
Eye Res 5(10): 777-85.

25 Tiffany, J. M. (1990). "Measurement of wettability of the corneal epithelium. I. Particle
attachment method." Acta Ophthalmol (Copenh) 68(2): 175-81.

26 Girifalco, L. A. and Good, R J. (1957) A Theory for the Estimation of Surface and
Interfacial Energies. I. Derivation and Application to Interfacial Tension, R. J. , J. Phys.
Chem, 61, 904.

27 Rose, M.C. (1992) Mucins: structure, function, and role in pulmonary diseases, Am J
Physiol Lung Cell Mal Physiol 263: L413-L429,

ll7

 

28 Smart, J .D. (1999) “The role of water movement and polymer hydration in
mucoadhesion” Bioadhesive Drug Delivery Systems: Chapter 2

29 Checkering, DE. and Mathiowitz E. (1999) “Definitions, mechamisms, and theories of
bioadhesion” Bioadhesive Drug Delivery Systems: Chapter 1

3° Voyutskii, S.S. Autohesio'n and adhesion of high polymers. lSt edition. 1963 P272

3' Mikos, AG, and Peppas, N.A. Systems for controlled release of drugs. V. Bioadhesive
systems. S T P Pharmacol 2:705-716, 1986

32 Huntsberger, J. R.. Locus of adhesive failure. Journal of Polymer Science, Part A:
General Papers (1963), 1 1339-44.

33 Hassan, EB. and Gallo, J.M., 1990. A simple rheological method for the in vitro
assessment of mucin—polymer bioadhesive bond strength. Pharm. Res. 7, pp. 491—495.

3'4 Carpenter RHS. Movements of the eyes. 2nd ed. London: Pion, 1988: 55-131

31‘ *3sz 1L4».- h- _-¢um‘wv-.h

35 Ravn K., A better faux teardrop, Los Angeles Times;Health,"Febni’ary 12,2007

36 Steffe, J. Rheological methods in food process engineering, 2"d Edition. Chapter 5.
37 Reiner, M. 1964, The Deborah number, Physics today, Jan.: 62.
38 Pipkin, AC. 1986, Lectures on Viscoelasticity theory, 2“d Edition

39 Tobolsky, A.V. , Dunell, B.A., and Andrews R.D. Relaxation and Dynamic Properties
of Polymers Textile Research Journal, Vol. 21, No. 6, 404-411 (1951)

40 Ciramella, GMT," R‘ossi, S., and Bonferoni M.C. .(1999) “A rheological approach to
explain the mucoadhesive behavior of polymer hydrogels” Bioadhesive Drug Delivery
Systems: Chapter 3

4' Amsden, B., Solute Diffusion within Hydrogels. Mechanisms and Models
Macromolecules, 31 (23), 8382 -8395, 1998

42 Peppas, NA. and Lustig S.R., Solute diffusion in hydrophilic network structures. In:
N.A. Peppas, Editor, Hydrogels in Medicine and Pharmacy 1, CRC Press, Boca Raton
(1986), pp. 57—84.

43 Allen, A., Foster SNE, Person JP. 1986. Br J Pharmacol 87: 126p

, ’, _.-'f"-pr.' , .

44 Munger, RJ. “Veterinary ophthalmology in laboratory animal studies” Veterinary
Ophthalmology. 2002. 5 (3): 167-175

118

 

45 Cao y, Zhang C, et al " Poly (N-isopropylacrylamide)-chitosan as thermosensitive in
situ gel-forming system for ocular drug delivery" J. of Controlled Release, 2007, vol.
120, p. 186-194

119

YLBR

4542

T ARIES

Mlclmllllllllllll lililllllll
3 1 2 9 3 0 3 6 2

0