Citation
In vitro studies on a putative molecular mechanism of action of valproic acid in treatment of bipolar disorder

Material Information

Title:
In vitro studies on a putative molecular mechanism of action of valproic acid in treatment of bipolar disorder
Creator:
Watterson, Jeannette M, 1974-
Publication Date:
Language:
English
Physical Description:
xi, 139 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Bipolar disorder ( jstor )
Cell growth ( jstor )
Cells ( jstor )
Down regulation ( jstor )
Inositols ( jstor )
Lithium ( jstor )
Medical treatment ( jstor )
Messenger RNA ( jstor )
Neurites ( jstor )
Rats ( jstor )
Bipolar Disorder -- drug therapy ( mesh )
Department of Pharmacology and Therapeutics thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pharmacology and Therapeutics -- UF ( mesh )
Hippocampus -- drug effects ( mesh )
Research ( mesh )
Valproic Acid -- pharmacology ( mesh )
Valproic Acid -- therapeutic use ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1999.
Bibliography:
Bibliography: leaves 124-138.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Jeannette M. Watterson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
020241073 ( ALEPH )
51639014 ( OCLC )

Downloads

This item has the following downloads:


Full Text
















IN VITRO STUDIES ON A PUTATIVE MOLECULAR MECHANISM OF ACTION
OF VALPROIC ACID IN TREATMENT OF BIPOLAR DISORDER










BY


JEANNETTE M. WATTERSON


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


UNIVERSITY OF FLORIDA


1999














ACKNOWLEDGMENTS


I wish to thank Dr. Edwin Meyer first and foremost, for he graciously accepted me
"as is" into his lab, thereby saving me from a decision I did not want to make. I am

grateful that he always saw and helped me to see the positive, and opened me up to new

ideas that I never thought to examine. I am very appreciative of Dr. Robert Lenox's

ability to stimulate me to always look for more, as well as to give more, and to never be

satisfied with less than 110%. 1 also credit him with redefining my conception of
"intensity" and "drive." I am indebted to Dr. David Watson for training me with patience

from start to finish, and for giving me direction and confidence at times when I had little.

I am grateful to my committee members Drs. Stephen Baker, Jeffrey Harrison, and Colin

Sumners for their guidance and approval, which I needed in order to maintain focus and
motivation. I am also thankful for the steadfast support and friendly atmosphere provided

by the department secretaries, the helpful technicians, and the many peers I have worked
with over the past four years, without whose help I could never have reached this point.

Finally, I would like to thank my parents, who instilled in me the ambition to aim high and

the drive to do my best.















TABLE OF CONTENTS


page

A C K N O W LE D G M E N T S ........................................................................................ ii

L IST O F T A B L E S ................................................................................. v

L IS T O F F IG U R E S ........................................................................................ ........ vi

ABBREVIATIONS AND DEFINITIONS ............................................................. ix

A B S T R A C T ............................................................................. ......................... .. x

CHAPTERS

I INTRODUCTION ..................................................... ................... 1

V P A ............................................................................................. 2
Bipolar Disorder .................... ...... 4
Mechanism of Action of Lithium in Treatment of Bipolar Disorder .................... 7
PKC and Its Substrates MARCKS and GAP-43 .......................................... 9
Developmental Effects of VPA .............................................................. 11
S u m m ary ........................................................................................ 1 1
S p ecifi c A im s ............................... ......................... ....................... 13

2 MATERIALS AND METHODS ................................... 20

C h e m ic a ls ....................................................................................... 2 0
H N 33 C ell C ulture ................ .......................................................... 20
P C 12 C ell C ulture ....................................................................... 2 1
N eurite Q uantitation ........................................................................ 22
Protein Quantitation by Western Blotting .................................................. 22
Ribonuclease (Rnase) Protection Assay (RPA) ............................................ 23
P K C In hib ito rs ................................................................................. 24
P K C A ctivity ........................................... .. ................... .. 24
Statistical A nalyses ......................................................................... 25











3 EFFECT OF VPA ON MARCKS AND GAP-43 ............................................ 27

Intro d u ctio n .................................................................................... 2 7
R e su lts ..................................................................................... 3 7
D iscu ssion ............................................................................... 53

4 ROLE OF PKC IN MECHANISM OF ACTION OF VPA .................................. 62

In tro d u ctio n .................................................................................... 6 2
R e su lts .......................................................................................... 6 7
D iscu ssion ............................................................................... .. 78

5 EFFECT OF VPA ON CELL VIABILITY, GROWTH AND MORPHOLOGY .... 88

In tro d u ctio n .................................................................................... 8 8
R e su lts .......................................................................................... 9 1
D isc u ssio n ...................................................................................... 9 8

6 CONCLUSIONS AND FUTURE DIRECTIONS ............................................... 113

R E F E R E N C E S ................................................................ ........................... ....... 124

BIOGRAPHICAL SKETCH .................................................................................. 139















LIST OF TABLES


Table pae

2.1. Properties of HN33 hybrid and parental cells ............................................ 21

2.2. A ntibodies for W estern blots ............................................................. 26

3.1. Addition of carbachol fails to potentiate the VPA-induced
MARCKS down-regulation in HN33 cells..... ......................... 43

3.2. Inositol supplementation fails to attenuate the VPA-induced
MARCKS down-regulation in HN33 cells ............................................. 44

3.3. Drugs, sources, and exposure conditions for HN33 cells ..................... 49

4.1. Antisense oligonucleotide sequences .................................................... 73

5.1. VPA induces an increase in HN33 cell doubling time ................................. 91

5.2. VPA has no effect on cell viability ....................................................... 91

6.1. Effects of two mood-stabilizing drugs, VPA and lithium,
as observed in HN33 cells following chronic exposure:
A su m m ary ................................................................................ 12 3















LIST OF FIGURES


page

1.1. Molecular effects of VPA in brain ......................................................... 17

1.2. Model for lithium regulation of MARCKS ............................................. 18

3.1. Comparison of MARCKS expression in mouse brain vs.
HN33 cells: Representative Western blot ......................................... 37

3.2. Concentration-dependent down-regulation of MARCKS
protein in HN33 cells following I d VPA exposure
a. R epresentative W estern blot ......................................................... 38
b Q u antitatio n ........................................................................... 3 8

3.3. Concentration-dependent down-regulation of MARCKS
protein in HN33 cells following 3 d VPA exposure
a. R epresentative W estern blot ......................................................... 39
b Q u antitatio n ....................... ............................................. ...... 3 9

3.4. Time course of VPA-induced MARCKS down-regulation
in H N 3 3 cells ..... .......................... .............................................. 4 0

3.5. Recovery of MARCKS following VPA-induced
long-term down-regulation
a. Q u antitatio n 4........................................................................... 4 1
b. Representative Western blot .................................................. 41

3.6. Effect of carbachol on the VPA-induced MARCKS down-
regulation in HN33 cells: Representative Western blot .......................... 43

3.7. Effect of myo-inositol on the VPA-induced MARCKS down-
regulation in HN33 cells: Representative Western blot ............................. 44

3.8. Additive effect of combined VPA and lithium exposure
on MARCKS protein
a Q u an titatio n ............................................................................ 4 5
b. Representative W estern blot ....................................................... 45









3.9. Effects of other psychotropic agents and VPA metabolites
on MARCKS protein
a. Q uantitatio n ............................................................................ 4 7
b. MARCKS expression following exposure to
two VPA metabolites: Representative Western blot ............................ 48
c. Structures of VPA and two tested metabolites .................................... 48

3.10. Time course of VPA-induced MARCKS mRNA alterations .......................... 50

3.11. Comparison of GAP-43 expression in rat brain vs.
HN33 cells: Representative Western blot ............................................. 51

3.12. Effect of chronic VPA exposure on GAP-43 expression
in HN33 cells: Representative Western blot ........................................ 51

3.13. Concentration-dependent increase in GAP-43 expression
in HN33 cells following VPA exposure ............................................. 52

3.14. Time course of VPA-induced GAP-43 increase in HN33 cells .............. 52

4.1. Effect of VPA exposure on PKC activity in HN33 cells .............................. 67

4.2. Effect of VPA on soluble PKC-ct expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot ............................. 68
b. Chronic VPA Exposure: Representative Western blot .......................... 68
c Q u antitatio n ........................................................................... 6 8

4.3. Effect of VPA on PKC-6 expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot ............................. 69
b. Chronic VPA Exposure: Representative Western blot .......................... 69
c Q u antitatio n ........................................................................... 69

4.4. Effect of VPA on PKC-& expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot ............................. 70
b. Chronic VPA Exposure: Representative Western blot .......................... 70
c. Q u antitatio n ............................................................................ 70

4.5. Effect of VPA on PKC-C expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot ............................. 72
b. Chronic VPA Exposure: Representative Western blot .......................... 72
c Q u antitatio n ............................................................................ 72

4.6. PKC-c antisense oligonucleotide design ................................ 73

4.7. Effect of PKC-a antisense oligonucleotides on PKC-ct
expression in HN33 cells: Representative Western blot .......................... 75










4.8. Effect of PKC inhibitor LY333531 on VPA-induced
down-regulation of MARCKS
a Q u antitatio n .7.......................................... ............................... 7 6
b. Representative Western blot ................................. 76


4.9. Effect of PKC inhibitor LY333531 on VPA-induced
GAP-43 increase
a. Q u antitatio n ........................................................................... 7 7
b. R epresentative W estern blot ........................................................ 77


5.1. Time-dependent VPA-induced neurite outgrowth
in HN33 cells: Representative photomicrographs
a. U ntreated, control H N 33 cells ...................................................... 92
b. 3 d VPA-exposed HN33 cells ...................................................... 92
c. 14 d VPA-exposed HN33 cells ................................................... 92
d Q u antitatio n ........................................................................... 9 2

5.2. Effect of PKC inhibitor LY333531 on
VPA-induced alterations in HN33 cells
a. E ff ect on cell density ................................................................. 94
b. Effect on neurite outgrowth ....................................................... 94

5.3. Concentration-dependent VPA-induced neurite outgrowth
in P C 12 cells ............................................................. ....... 95

5.4. Effect of PKC inhibitor BIM on VPA-induced
alterations in PC 12 cells
a. Effect of 0.6 mM VPA on PC 12 cell number ................................... 96
b. Effect of 1.8 mM VPA on PC12 cell number ..................................... 96
c. Effect of 0.6 mM VPA on PC12 cell neurite outgrowth ........................ 96
d. Effect of 1.8 mM VPA on PC 12 cell neurite outgrowth ........................ 96

5.5. Neuroprotective potential of VPA
a. Effect of VPA on cell number in
NGF-differentiated PC 12 cells .................................................. 97
b. Effect of VPA on neurite outgrowth in
NGF-differentiated PC 12 cells .................................................. 97














ABBREVIATIONS AND DEFINITIONS

BIM bisindolylmaleimide, an inhibitor of PKC

CMP-PA cytidine monophosphate-phosphatidic acid; also CDP-DAG

CNS central nervous system

DAG diacylglycerol, a stimulator of PKC

FDA U.S. Food and Drug Administration

GABA gamma-aminobutyric acid, an [inhibitory] amino acid neurotransmitter

GAP-43 43 kD growth-associated protein

HN33 immortalized hippocampal cell line derived from fusion of primary mouse
hippocampal and human neuroblastoma cells

IMPase myo-inositol I -monophosphatase

IP inositol [mono]phosphate

LY333531 a PKC inhibitor designed by Eli Lilly and Company

MARCKS myristoylated alanine-rich C kinase substrate

MBP myelin basic protein

NGF nerve growth factor

PC 12 rat pheochromocytoma (adrenal tumor) cell line

PI phosphoinositide

PKC protein kinase C

PDBu phorbol 12,13-dibutyrate

RPA RNase protection assay

SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis

VPA valproic acid or valproate














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

IN VITRO STUDIES ON A PUTATIVE MOLECULAR MECHANISM OF ACTION
OF VALPROIC ACID IN TREATMENT OF BIPOLAR DISORDER

By

Jeannette M. Watterson

December 1999

Chair: Edwin M. Meyer
Major Department: Pharmacology and Therapeutics

Valproic acid (VPA; trade name Depakote) is a broad-spectrum anticonvulsant

which has of late been utilized as a treatment for bipolar disorder and has proven

efficacious in treating a variety of both acute manic and chronic bipolar states. Lithium is

currently the only other FDA-approved drug treatment for this disorder, and although it

has traditionally been the first-line treatment of choice, there is a recognized need for

alternatives, due to the low patient compliance rate resulting from troublesome side

effects, narrow therapeutic index, and less than optimal patient response.

Very little is known about how either drug exerts its mood-stabilizing properties,

and it has been the goal of this project to understand better the actions of VPA from a

mechanistic standpoint.

Initial experiments were conducted utilizing an immortalized hippocampal cell line,

the HN33 cells. Results indicate that at clinically relevant target serum concentrations,

chronic, but not acute, VPA reduces expression of the myristoylated alanine-rich C kinase

substrate (MARCKS), concomitant with increased GAP-43 expression. These effects are

associated with altered activity and expression of protein kinase C (PKC), an important









effector in the CNS, and the effect of VPA on MARCKS appears to be dependent, at least

in part, upon the activation/down-regulation of this enzyme. Furthermore, the observed

effects are accompanied by alterations in cellular morphology and proliferation rate, which
implies that VPA plays a major role in modulating neuronal plasticity.
Additional studies utilizing PC 12 cells showed similar VPA-induced

differentiation, with no evidence for neuroprotection following nerve growth factor-
induced differentiation of cells. These studies bring us closer to elucidating the basis for

the unique therapeutic profile of VPA, which includes both antiepileptic and

antimanic/mood-stabilizing clinical applications, as well as unfortunate teratogenic effects

(such as spina bifida, anencephaly, meningomyelocele, and encephalocele), and may
provide further insight into the pathophysiology underlying bipolar disorder.















CHAPTER 1
INTRODUCTION


VPA is a broad-spectrum anticonvulsant shown in the last two decades to be

efficacious in the treatment of bipolar disorder (manic-depressive illness). These two

actions can be distinguished from each other by their time-courses, with anticonvulsant

activity being observed much more quickly, on a scale of hours to days, in contrast to the

days to weeks treatment required for mood-stabilizing activity. The precise mechanisms

by which VPA exerts either its antiepileptic or antimanic effects are still unclear. VPA has

been found to affect Na+ flux, intracellular Ca++, and brain gamma hydroxybutyric acid

(GABA) levels, these actions probably relate to the anticonvulsant efficacy of this drug.

The basis for its mood-stabilizing efficacy remains a mystery. In addition to its therapeutic

actions, numerous studies have documented the deleterious effects of this drug on the

developing nervous system, especially the embryonic neural tube. It is the goal of this
project to elucidate further the mechanisms of action of VPA in brain, mainly through

studies of its effects on intracellular signaling in one particular cell model, in an effort to

understand better the underlying mechanism for this therapeutic profile. The specific aims

of this project are to: (1) characterize the therapeutically relevant time- and concentration-

dependent effects of VPA on PKC (activity and isozyme expression) and its substrates
(MARCKS and GAP-43), using the hippocampally derived HN33 cells as a model system

for neuronal effects of VPA; and (2) characterize the therapeutically relevant effects of
VPA on the viability, growth, and morphology of both HN33 cells and PC12 cells. Based

on preliminary evidence and previous reports, it is hypothesized that VPA will have a

significant effect on the expression of PKC substrates MARCKS and GAP-43, that PKC









will be integrally involved in any such effects, and that cells will undergo an alteration in

cell growth and morphology indicative of a transition to a more differentiated state.


VPA


VPA is a short-chain carboxylic acid that was first discovered to have antiepileptic

effects in 1963, and has been utilized as an anticonvulsant in the United States for over 20

years (Meunier el al., 1963; Ramsay, 1984). It first became available in the U.S. in 1978

as Depakene, the acid form of the drug. Five years later, the enteric-coated form,
Depakote, was introduced (Penry and Dean, 1989). This particular formulation, the most

popular form in use today, is a stable coordination compound containing equal

proportions of valproic acid and the sodium valproate salt (McElroy el al., 1989). VPA is

striking in its structural difference from other known anticonvulsants, yet it is well known

for its efficacy in treating a wide range of seizure disorders, giving it a broader therapeutic
profile for epilepsy treatment than perhaps any other anticonvulsant yet studied. VPA is

FDA-approved for treatment of simple and complex absence seizures, as well as mixed

seizures which include the absence type. Additionally, VPA is effective in treating such

primary generalized epilepsies as tonic-clonic, clonic, tonic, and myoclonic seizures,
infantile spasms, and photosensitive epilepsy, as well as secondary generalized seizures

and some partial seizures, including complex partial epilepsy (McElroy et al., 1989).

A number of studies have investigated the putative mechanism(s) through which

VPA exerts its anticonvulsant effects. No one signaling effect is likely to be responsible

for the entire panorama of the anticonvulsant efficacy of this drug, and a number of
molecular events appear to contribute to the net effect of reducing neuronal excitability.

The ability of VPA to increase neurotransmission of the inhibitory amino acid GABA has

been well documented. VPA may accomplish this through a variety of mechanisms by

increasing GABA levels through decreased degradation and increased synthesis and









release, or by increasing post-synaptic responsiveness to GABA through effects on the

GABA-receptor complex or altered K' conductance, effectively dampening neuronal

excitability (Chapman et al., 1982; Rimmer and Richards, 1985; McElroy el al., 1989;

Penry and Dean, 1989; Joffe, 1993). Because anticonvulsant effects are often observed

before total brain GABA levels become elevated, the VPA-induced increase in GABA

levels alone may not account for the drug's anticonvulsant effect.

VPA, like other anticonvulsants, is known to block sustained high frequency

repetitive firing of cultured neurons, due to its effect of slowing recovery of the sodium

channel following inactivation (Albus and Williamson, 1998). At high concentrations,

VPA has been shown to alter membrane potassium conductance, and relatively low

concentrations of the drug hyperpolarize neurons, both effects suggestive of direct effects

on neuronal membranes (Porter and Meldrum, 1995). Other studies have provided

evidence for reduced excitatory neurotransmission by the amino acid aspartate (Chapman

el al., 1982). Still other reports have suggested that VPA works through inhibition of

kindling, a model of epilepsy which holds that epileptic activity predisposes the system to

increased frequency of seizures triggered by lower thresholds over time (Porter and

Meldrum, 1995; McElroy el al., 1989).

In summary, the various mechanisms proposed encompass three major net effects

of VPA: (1) increased GABAergic synaptic inhibition post-synaptically; (2) reduction in

the excitatory synaptic transmission ultimately responsible for epileptic kindling; and (3)

reduction in repetitive firing via a direct effect on voltage-sensitive ion channels (Albus

and Williamson, 1998). The broad spectrum of anticonvulsant activity of VPA is probably

related to more than one of the molecular mechanisms demonstrated thus far. However, it

is unclear whether such anticonvulsive effects may also account for the antimanic efficacy

of VPA, and because little is known of the mood-stabilizing mechanism of VPA, much
remains to be explained with regard to what kinds of intracellular changes might underlie

this drug's mood-stabilizing effects.









Bipolar Disorder


In recent years, VPA has emerged as an effective alternative to lithium in the

treatment of acute mania in patients suffering from bipolar disorder (Gerner & Stanton,

1992; McElroy et al., 1992; Bowden el al., 1994), and is also being used for long-term

prophylactic management of this disorder. Bipolar disorder, also known as manic-

depressive illness, is a debilitating psychiatric condition for which there is no known cause
and only a limited range of treatment options available. It is estimated to affect about 1%

of the population in this country (Bowden et al., 1994). Patients afflicted with this

affective disorder experience frequent, recurring episodes of extreme alterations in mood.

Whereas the patient may appear normal between cycles, the episodes of mania and

depression that characterize this disorder can leave the patient functionally incapacitated.
The depressed phase involves motor activity alterations, impaired concentration, impaired

social and occupational functioning, general helplessness, and may result in changes in

sleep, appetite, weight, and energy level (Post, 1989). The manic episodes entail such

disruptive symptoms as hyperactivity, explosive temper, impaired judgement, insomnia,

disorganized behavior, hypersexuality, and grandiosity, any combination of which may

lead to alienation from family and friends, indebtedness, job loss, and other major life

problems (Bowden et al., 1994). If untreated, this disorder has a tendency to worsen,

with increasing severity, duration, and frequency of episodes over time (Post, 1989).

There is a high rate of suicide in patients suffering from this disorder, estimated at about

15% in untreated cases (Bowden el al., 1994).

Lithium has for over 40 years been the first-line drug therapy for both acute and

prophylactic management of this disorder, but due to its narrow therapeutic index and
many undesirable side effects (often resulting in compliance problems), the need for other

viable treatment options has become a major concern (Post, 1989; Bowden, 1996).

Although numerous treatments have been used experimentally in the past, including









antipsychotic drugs, calcium channel blockers (such as verapamil), high doses of thyroxine
(T4), serotonin precursors (such as L-tryptophan), and even electroconvulsive therapy

(Joffe, 1993), results have not been consistently positive, and none is used extensively as a
"mainstream" treatment at the present time. For example, the use of antidepressants is

generally not considered because patients may be brought out of their depressive episodes

only to be immediately submerged into the manic phase; further, certain drugs may

increase the frequency of cycling of mood swings (Post, 1989). The anticonvulsants

carbamazepine, lamotrigine, and gabapentin, as well as other pharmacological agents

(clozapine, clonazepam, verapamil), have shown promise in treating bipolar disorder, but

very few studies have been conducted and FDA approval has yet to be obtained for this

clinical indication (Bowden, 1996; Joffe, 1993; Hollister, 1995). Because lithium is the
approved first-line treatment, and has been used and studied extensively in the past, this

project will be focused in part toward understanding the mechanism of action of bipolar

drugs, based on what is known and has been reported experimentally for lithium.

VPA is emerging as a viable alternative or combinational drug treatment for

bipolar disorder, and in fact, VPA may even have a wider spectrum of efficacy than

lithium. Although VPA seems to be less effective in treating depression than in treating

the manic phase of the disorder, there is accumulating evidence of its use in atypical

(dysphoric/mixed) mania, rapid cycling, and secondary manias, which have poorer
prognoses and are generally less responsive to lithium than the usual forms of the disorder

(Calabrese & Delucchi, 1990; Calabrese et al., 1992, 1993; Bowden el al., 1994; Joffe,

1993; Brown, 1989). VPA has also been used to treat various conditions which are

comorbid with bipolar disorder, such as migraines, panic disorder, bulimia, attention-

deficit hyperactivity disorder (ADHD), and other brain-related illnesses, and VPA appears

to be quite effective in treating lithium non-responders (Bowden, 1995; Joffe, 1993;
Brown, 1989). The therapeutic index of VPA is considerably wider than that of lithium,

thereby allowing for greater freedom in dosage adjustment, with less risk of toxicity.









Furthermore, the side effect profile of VPA is less intimidating than that of lithium.

Whereas the side effects of lithium may be quite severe and disturbing to patients (thyroid,
renal, and CNS effects, dermatologic and cardiovascular complications), VPA typically

exhibits a different, more benign profile of side effects which are generally more

acceptable to patients (gastrointestinal upset, weight gain, hair thinning, increased hepatic

enzyme levels), and many of its bothersome side effects are alleviated over time or with a

decrease in drug dosage (Bowden, 1995; Penry and Dean, 1989). The exception to this is

the idiosyncratic, unpredictable and irreversible hepatic toxicity induced by VPA in

predisposed patients on rare occasion. While this condition may be fatal if not detected in

time, it affects only a very minimal portion of the population the highest incidence
reported was a rate of 0.01%. The majority of reported cases occur within the first three

months of VPA therapy, and because signs common to hepatic failure are clearly present,

patients can be monitored regularly for the first six months of therapy so as to reduce the
risk of fatality and prevent any long-term damage. Despite this risk, the side effect profile

of VPA is still considered to be safer and less bothersome than that of lithium.

Nonetheless, the mechanisms mediating the therapeutic properties of VPA in the

treatment of bipolar disorder are not clearly understood. Whereas the therapeutic actions

of VPA have been reportedly associated with potentiation of GABAergic transmission and
other neurotransmitter effects (Chapman el al., 1982; Post el al., 1992, Loscher, 1993,

Petty, 1995), these effects generally occur at rather high concentrations and may relate

more to the anticonvulsant properties of VPA than to its mood stabilizing effects see

Figure 1. 1 (Waldmeier, 1987, Motohashi, 1990). Further, as is the case for the time

course of lithium, there is a delay of several days in the onset of clinical antimanic action

of VPA, although loading strategies have demonstrated antimanic effects at times as early

as 3 d (McElroy el al., 1996). Thus, any mechanism postulated for the therapeutic action

of these drugs must take this delayed onset into account (Manji el al., 1995). Because
VPA and lithium are currently the only two drugs approved by the FDA for treatment of









acute mania, comparisons of the actions of these agents and efforts to elucidate the
mechanism(s) underlying the mood-stabilizing activity of both drugs is important to

understanding the pathophysiology of this illness. Further, this approach may lead to the

identification of new or existing drugs which may be useful in this application.


Mechanism of Action of Lithium in Treatment of Bipolar Disorder


Accumulating evidence from our laboratory and others strongly implicates
receptor-mediated PI signaling in the mechanism of action of lithium in the brain (Berridge

et al., 1982; Godfrey, 1989; Lenox and Watson, 1994). Numerous studies have

established lithium as an uncompetitive inhibitor of the enzyme inositol monophosphatase

(IMPase), and the resulting accumulation of intermediates in the pathway which are

responsible for additional downstream effects, including intracellular Ca++ release and

PKC activation (see Figure 1.2; Post et al., 1992). It was first suggested by Berridge and

colleagues that depletion of free inositol, resulting from the accumulation of intermediates
in the PI signaling pathway, is a major consequence of long-term lithium exposure

(Berridge et al., 1982; Berridge, 1989). This phenomenon would have greatest

implication in certain areas of the brain which are relatively more inositol-limited by virtue

of the blood-brain barrier as well as less capable of producing inositol de novo. Further,

because the PI signaling cascade is linked to G protein-coupled muscarinic receptor (in,

m3, m5) activation, areas of the brain undergoing the highest rate of receptor activation

may predictably be preferentially affected by lithium. Indeed, studies in our laboratory

have provided evidence for an inositol-driven attenuation of the effects of lithium in HN33

cells, as well as potentiation by a muscarinic receptor agonist (carbachol) and
prevention/reversal by a muscarinic receptor antagonist (atropine) (Watson and Lenox,

1996). Following exposure of HN33 cells to therapeutic concentrations of lithium (1

mM), MARCKS protein expression is significantly down-regulated (Lenox et al., 1996;









Watson and Lenox, 1996). This MARCKS reduction only occurs in the relative absence

of inositol, for supplementation of inositol even at concentrations as low as 5 4tM prevents

the lithium-induced down-regulation of MARCKS (Watson and Lenox, 1996; Watson et

al., 1998; see also Table 3.2). Similarly, the concomitant exposure of HN33 cells to both

lithium and the muscarinic agonist carbachol results in an increase in the level of

MARCKS protein down-regulation observed, and exposure of cells to the muscarinic

antagonist atropine reverses or prevents the lithium-induced reduction in MARCKS

expression (Watson and Lenox, 1996; Watson el al., 1998; see also Table 3.1). These

findings have provided further support for the PI signaling hypothesis.

Studies of the molecular mechanism of action of lithium in brain have focused

largely on its modification of PI signaling, particularly the lithium-induced accumulation of

intermediates such as DAG and the resulting stimulation of PKC. Such a PI-mediated

mechanism may account for the effects of lithium on PKC-regulated events, including

protein down-regulation/degradation and transcriptional and translational events. In

contrast, little evidence has been found to support a similar PI mechanism for VPA.

Vadnal and Parthasarathy (1995) showed that VPA had no effect, either stimulatory or

inhibitory, on IMPase partially purified from bovine brain. Further, lithium and

carbamazepine (a less commonly used antimanic agent) exhibited opposite effects on

IMPase, suggesting that the enzyme is not a common site of action for these two mood-

stabilizing medications. Dixon and Hokin (1997) reported that in cerebral cortical slices,

VPA caused no accumulation of inositol monophosphates, inositol bisphosphates, or

inositol 1,3,4-trisphosphate, all of which accumulated following lithium exposure. Based

on these observations, the depletion of inositol is not universally applicable as the basis for

the antimanic action of mood-stabilizing drugs.

Alternatively, it has become apparent that the action of chronic lithium in the brain,

which follows as a result of its interaction with the PI signaling pathway, is mediated

through subsequent regulation of PKC and the down-stream posttranslational modification









of select protein substrates (Lenox, 1987; Manji and Lenox, 1994; Watson and Lenox,
1996). Further, the clinical effects of lithium may be mediated, at least in part, through

these actions. Although VPA apparently does not share the property of PI signaling

inhibition (Vadnal and Parthasarathy, 1995; Dixon and Hokin, 1997), the question remains

as to whether it affects PKC, as previous studies have suggested (Chen el al., 1994;

O'Brien and Regan, 1998). It is a goal of this project to address further the potential role

of PKC in the action of VPA in the brain, especially as such effects might relate to the

mood-stabilizing properties of this drug.


PKC and its Substrates MARCKS and GAP-43


PKC exists as one of at least II structurally-related isozymes in mammals, many of

which are Cam-activated and/or DAG-dependent, and have been implicated in numerous

cellular responses associated with regulation of signaling and long-term events including

ion channel and gene regulation (Newton, 1995; Nishizuka, 1995). Numerous studies

have suggested a role for PKC in the long-term action of lithium (Lenox, 1987; Manji and

Lenox, 1994), and recent data have provided evidence that chronic VPA exposure alters

PKC activity as well as the expression of a and c isozymes in non-neuronal cells (Chen el

al., 1994; Manji el al., 1996b). Previous studies have demonstrated that chronic (but not

acute) lithium treatment of rats, resulting in clinically relevant brain concentrations,
produces a significant reduction in the PKC substrate MARCKS in the hippocampus,

which persists beyond treatment discontinuation (Lenox el al., 1992; Manji et al., 1996a).

This lithium-induced down-regulation of MARCKS has also been demonstrated in an
immortalized hippocampal cell line (Watson and Lenox, 1996), in which it was previously

demonstrated that phorbol esters, which directly activate PKC, down-regulate MARCKS
protein expression in a PKC-dependent manner (Watson el al., 1994).









MARCKS is a prominent and preferential substrate in the brain for PKC, which by
virtue of phosphorylation, regulates the cellular localization and activity of this protein.

MARCKS binds calmodulin in a calcium-dependent fashion and cross-links filamentous

actin, and has been implicated in cellular processes associated with cytoskeletal
restructuring and neuroplasticity, e.g., transmembrane signaling and neurotransmitter

release (Aderem, 1992; Blackshear, 1993). In fact, previous studies in mutant mice

lacking the MARCKS gene have shown that MARCKS expression is essential to normal

CNS development in animals (Stumpo el al., 1995). Insofar as MARCKS may also

represent a molecular target for mood stabilizers in the brain (Lenox and Watson, 1994,

Watson and Lenox, 1996), it should be of benefit to examine its potential regulation by

VPA as well as by other psychotropic agents.
The growth-associated protein GAP-43 is a second protein thought to be integral
to normal neuronal development and differentiation, as it is expressed at high levels during

nervous system development and is associated with neurite extension (Jap Tjoen San el

al., 1992; Aigner and Caroni, 1993; Strittmatter el al., 1995). GAP-43 shares a number

of properties with MARCKS, including ability to bind actin and calmodulin (albeit in a

Ca++-independent fashion), as well as regulation by PKC (reviewed in Benowitz and

Routtenberg, 1997). Inasmuch as previous studies have provided evidence for an inverse

regulation of MARCKS and GAP-43 in certain regions of rat brain (McNamara and

Lenox, 1998), it may be instructive to assess the relative levels of GAP-43 observed

concomitant with VPA-induced alterations in MARCKS expression. Investigation of

these two related but differentially regulated proteins will likely provide insight into the
molecular mechanism of this drug, as will examination of this drug's effects on PKC, the

primary enzyme responsible for the direct regulation of both of these proteins.









Developmental Effects of VPA


In addition to the therapeutic efficacy of VPA, this drug exhibits its share of

drawbacks, the most severe of which are its effects on the developing embryo. VPA has

been known for years to be a developmental teratogen. In recent years, women who are
pregnant or even of childbearing age are encouraged to supplement their diets with

adequate amounts of folic acid, for supplementation with this nutrient and its metabolites

has been associated with decreased frequency of neural tube defects (including spina

bifida, anencephaly, and encephalocele) in offspring of VPA-treated and nontreated
mothers (Smithells et al., 1981; Rhoads and Mills, 1986; Trotz el al., 1987; Nosel and

Klein, 1992; Ehlers et al., 1996). Numerous studies have investigated this phenomenon,

yet it remains unknown how VPA causes these developmental defects or how folate

supplements are able to prevent such effects. Previous work has demonstrated that VPA

inhibits cell proliferation through arrest of the cell cycle at mid G I phase (Martin and

Regan, 1991), as well as induces differentiation of neuroblastoma and glioma cells (Regan,

1985), which would suggest that VPA is interfering with development through

manipulation of the proliferation rate and differentiation state of cells central to nervous

system structure. Such effects on cell growth and morphology should be considered in

any putative model for the molecular mechanism of this drug, and this study will therefore
investigate the hypothesis that VPA exerts effects on both cell proliferation and cell

morphology which may account for its teratogenicity.


Summary


Experiments designed to elucidate the short- vs. long-term intracellular actions of
VPA will facilitate the development of a model which may explain, in part, the unique

therapeutic profile exhibited by this drug. Initial VPA studies have been modeled after









those conducted previously or in parallel for lithium, so that comparisons may be made

which might shed light on the basis for the common mood-stabilizing therapeutic action of

these two drugs. In summary, thus far, it has been shown that in the HN33 cell model,
lithium elicits a down-regulation in MARCKS primarily through its effect on PI signaling,

as evidenced by the carbachol-driven potentiation and either inositol- or atropine-driven

attenuation of effect (Watson and Lenox, 1996; Watson el al., 1998). This lithium-
induced MARCKS reduction occurs over a chronic time course, requiring days to weeks,

and the effect is greater in the soluble rather than membrane fraction of the cells (Lenox et

al., 1996; Watson and Lenox, 1996). A similar time course of MARCKS mRNA

reduction is also observed (Watson and Lenox, 1997). Further, chronic lithium

administration elicits a significant down-regulation of MARCKS protein in vivo, in a rat

model (Lenox el al., 1992). As for the effects of lithium on cell viability and related

parameters, no alteration in cell growth rate or morphology has been observed following

acute or chronic lithium exposure (Watson and Lenox, 1996), and no alteration in
expression of another prominent PKC substrate, GAP-43, was observed (Watson and

Lenox, 1997). Lithium exposure results in predictable changes in PKC activity (with

initial activation followed by chronic down-regulation), as well as alterations in expression

of the two PKC isoforms c and 6 (Watson and Lenox, 1997), similar to findings

previously reported by Manji el al. (1 996a,b).

Applications of research into the mechanism of action of VPA will be multifold. In
addition to providing a better understanding for the efficacy of VPA in treating bipolar

disorder, a clearer picture of the pathophysiology underlying this illness, and identification

of new or existing drug treatments, investigation of the molecular bases for these effects
may foster a better understanding of how VPA exerts its deleterious developmental effects

and the development of potential therapeutic interventions to prevent or correct such

defects.










Specific Aims


Characterize the Therapeutically Relevant Effects of VPA on PKC and Its

Substrates.

MARCKS and GAP-43 are known to be important in modulating central nervous

system (CNS) plasticity and intracellular signaling, by virtue of the regulatory enzyme

PKC. Using hippocampally derived HN33 cells, I will characterize both acute and chronic

VPA-induced alterations of PKC expression and activity, as well as on levels of expression

of MARCKS and GAP-43.



Effects of VPA on MARCKS protein expression. Previous studies showed that

chronic lithium exposure at therapeutic concentrations resulted in down-regulation of

MARCKS protein expression in HN33 cells, and involved alterations in PI signaling.

After (1) establishing the concentration-response relationship for the effects of VPA on

MARCKS protein, I will (2) establish the time course of action of VPA at therapeutic

concentration, as well as (3) the time course of system recovery after VPA withdrawal. In

addition, I will assess the importance of (4) inositol concentration and (5) muscarinic

receptor/Gq activation, in VPA-induced MARCKS effects. Furthermore, I will (6)

investigate the effects of combined VPA and lithium treatment. Effects will be examined

in both cytosolic and membrane fractions of cells. It is predicted that VPA will have

effects similar to those of lithium, i.e. significant down-regulation of MARCKS protein

expression following chronic (but not acute) exposure to concentrations of the drug within

a therapeutically relevant range, and that MARCKS will return to basal levels after drug

removal. In contrast to lithium, there is no evidence for VPA-induced modulation of the









PI signaling pathway, so no effects by inositol or carbachol (muscarinic receptor agonist)

are expected.

Effects of VPA on MARCKS mRNA. Preliminary studies have provided

evidence for a lithium-induced reduction in MARCKS mRNA. I will establish the time

course of therapeutically relevant VPA-induced effects on MARCKS mRNA levels. It is

predicted that VPA will elicit down-regulation of MARCKS mRNA is a fashion similar to

that of lithium, but following the time course of VPA-induced alterations in MARCKS

protein.

Effects of VPA on GAP-43 expression. I will establish a dose-response and time

course for therapeutically relevant VPA-induced effects on GAP-43 protein levels in both

cytosolic and membrane cellular fractions. Although there are no previous reports for

VPA- or lithium-induced alterations in GAP-43 expression, such effects are a likelihood

given preliminary evidence for VPA-induced morphological alterations, and may partially

account for the plastic changes (such as synaptic remodeling) essential during therapy for

seizure and mood disorders.

Effects of VPA on PKC isozyme expression. I will establish the time course, at

therapeutic concentration, of VPA-induced alterations in four PKC isozymes expressed in

HN33 cells. Based on previous findings with lithium and VPA, which provide preliminary

evidence for effects of each on PKC expression and activity, I predict that VPA will have

significant effects on both PKC activity and isozyme expression, especially if alterations in

MARCKS and GAP-43 are observed following chronic VPA exposure.

Effects of VPA on PKC activity. I will investigate the effects of therapeutic

concentrations of VPA on PKC activity (as measured by phosphorylation of a PKC-

specific substrate) over a broad time course from acute to chronic time points. VPA-









induced alterations in PKC activity are expected for the same reasons as described above

in D.

Role of PKC in mechanism of action of VPA. Through the use of PKC

inhibitors and antisense, which block PKC activity, I plan to further characterize A-E

above; specifically, MARCKS and GAP-43 alterations, as well as morphological effects

(see below), will be investigated following PKC inhibition via enzyme inhibitors and

isozyme-specific antisense, in order to further define the role of PKC in these parameters.



Characterize the Physiologically Relevant Effects of Chronic VPA on Viability of

HN33 Cells and PC12 Cells.



Effects of VPA on HN33 cell growth. I will investigate the effects of chronic

VPA exposure at therapeutic concentrations on cell viability through observation of cell

number and doubling time. The role of PKC in any observed effects will also be assessed.

Based on limited previous studies in other cell lines, it is predicted that VPA will produce

changes in cell growth which may help to explain, in part, the severe developmental effects

associated with prenatal VPA exposure.

Effects of VPA on HN33 cell morphology. I will also investigate the effects of

chronic, therapeutic VPA exposure on cell phenotype, looking specifically for

differentiated morphology and other markers indicative of altered state of maturation. The

role of PKC in any observed effects on cell differentiation will also be assessed. Despite

the lack of evidence for a lithium-induced alteration in cell morphology, VPA-induced

alterations in cell morphology are considered likely, given this drug's therapeutic profile

(anticonvulsant and mood-stabilizing, yet teratogenic). Observed alterations in MARCKS









and GAP-43 will further support this hypothesis, given their integral role in CNS

development and plasticity.

Effects of VPA on PC12 cell viability. Preliminary evidence for effects of VPA

on cell growth and morphology leads me to investigate more in depth the effects of VPA

on cell viability. Whereas the HN33 cell line is a sufficient model for the investigation of

intracellular signaling phenomena as outlined in the studies above, post-mitotic cell lines

may serve as better developmental models for investigation of such phenomena as

neuroprotection or drug toxicity, and these cell lines have been better characterized for

such investigations. Therefore, in order to investigate the potential of VPA to elicit either

toxic or neuroprotective effects, NGF-differentiated and non-differentiated PC 12 cells will

be utilized. In light of the well-established neuronal toxicity of VPA exhibited during

development, toxicity resulting from high levels of the drug would not be a surprise;

however, preliminary studies in our laboratory have shown no evidence for toxic effects of

this drug at therapeutic concentrations. Any cytoprotective effect of VPA has yet to be

established, and there is currently no evidence in support of such an effect.











anticonvulsant effects


VPA

III \ \

glutamate +" !Ii \ \
giulm~~ica \ \ butyric acid
g/utamioacid G / /
decarboxylase GAB V+I '


VPA


PKC


GAP-43


MARCKS


mood-stabilizing effects


Figure 1.1. Molecular Effects of VPA in Brain. Depicted are (1) at right, the effects of
VPA as previously established by other groups, and thought to be responsible for the
anticonvulsant efficacy of the drug, and (2) at left, the long-term effects hypothesized to
contribute to the mood-stabilizing efficacy of VPA. Of the short-term effects at right,
increased GABA neurotransmission and effects on intracellular calcium are generally
thought to contribute less than the effects of VPA on voltage-sensitive ion channels and the
associated decrease in excitatory neurotransmission and neuronal firing. The hypothesized
effects at left will be assessed in detail in the studies which follow.











Figure 1.2. Model for Lithium Regulation of MARCKS. (adapted from Lenox et al., 1996). Following agonist
occupancy of the receptor binding site, G-protein coupling mediates activation of the enzyme phospholipase C (PLC),
which catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) into second messengers diacylglycerol
(DAG) and myo-inositol 1,4,5-triphosphate (Ins 1,4,5 P3). DAG is an endogenous activator of PKC, and Ins 1,4,5 P3
induces release of calcium from intracellular stores such as the endoplasmic reticulum. Lithium inhibits uncompetitively
the enzyme inositol monophosphatase (IMPase), thereby inhibiting the recycling of inositol. This results in accumulation
of metabolites in the DAG pathway during stimulation of receptor signaling. Activation of PKC* through both increased
calcium and DAG initiates posttranslational events such as the translocation of PKC to the membrane where it
phosphorylates the MARCKS protein, which then translocates to the cytosolic fraction. In the phosphorylated form,
MARCKS no longer binds calmodulin or cross-links actin, thereby altering compartmental intracellular calcium signaling
and affecting cytoskeletal restructuring. In addition, chronic lithium and receptor activation down-regulate MARCKS
protein expression, which may result from down-stream transcriptional/translational events.











PK Actin

04
MP-PA CaM ,
In _ositol ransrpin/Translation MARCKS
nslP ns E~ffector(s) mRNA
n P Ins 1,4,5 P3In 3 53
[.I~ns 4PJ .
Ca+Ca+Ca+DNA














CHAPTER 2
MATERIALS AND METHODS


Chemicals


VPA (2-propyl-pentanoic acid) and other chemicals were purchased from Sigma

(St. Louis, MO), unless otherwise noted. For some drug comparison studies, additional

VPA and VPA metabolites were obtained from Abbott Laboratories (Abbott Park, IL),

and fluoxetine and LY333531 (a PKC inhibitor) were provided by Eli Lilly and Company

(Indianapolis, IN). Tissue culture reagents were purchased through Gibco (Gaithersburg,

MD; horse serum, RPMI and DME media), Hyclone (Logan, UT, fetal bovine serum), and

Sigma (trypsin and gentamicin). Electrophoresis reagents were purchased from Bio-Rad

(Hercules, CA), and RPA reagents were purchased from Ambion (Austin, TX).


HN33 Cell Culture


The immortalized hippocampal cell line HN33.dw(9/21) was used to assess effects

of VPA on all parameters, including MARCKS protein and mRNA, GAP-43, PKC activity

and expression, etc. HN33 cells were kindly provided by Dr. Bruce Wainer (Emory

University, Atlanta, GA). These are derived from the fusion of primary neurons from the
hippocampus of postnatal d 21 mice with the N 18TG2 human neuroblastoma cell line, and

exhibit morphological, immunological, and electrophysiological parameters characteristic
of hippocampal neurons in culture (Lee et al., 1990). These properties are described in
Table 2.1. HN33 cells have been used in our laboratory as a model system to address the









mechanism of action of lithium, and we previously established the presence of a functional

second messenger (PI) signaling pathway in these cells (Watson and Lenox, 1997).

Furthermore, HN33 cells have recently been assessed and found to express at least two of

the neurofilament triplet proteins (Watterson, unpublished data), which supports their

consideration as phenotypically neuronal (Debus el al., 1983; Shaw, 1991). Cells were
grown at 37'C and 5% CO2 in Dulbecco's modified Eagle's medium (DMEM) containing

5% defined fetal bovine serum (FBS) and 1 mg/ml gentamicin, with or without the

addition of 0.6 mM VPA. The duration of exposures ranged from 0 to 14 d. VPA was

added to cultures at time of plating, and culture medium was changed or cells were

passaged every 3-4 d, with no replenishment of VPA between platings. Assays were

performed when cells reached 80-90% confluence.


Table 2.1 lists the properties present (+) or absent (-) in HN33 cells and the two parent
cell lines from which the HN33 were derived, primary mouse hippocampal neurons (21 d)
and N18TG2 neuroblastoma cells. This table was adapted from that reported in Lee et al.
(1990), in which the HN33 cell derivation and characteristics were first reported.


Table 2.1. Properties of HN33 Hybrid and Parental Cells

Tissue or Hybrid Neuritic Excitable NGF NGF

Cell Line GPI Isozvme Processes Membrane NF-M Protein mRNA

Hippocampus + + + + +

N18TG2 -

HN33 + + + + + +


PCI2 Cell Culture


PC 12 cells obtained from American Type Culture Collection (Rockville, MD) were

grown in RPMI 1640 containing 10% heat-inactivated horse serum, 5% FBS, and 0.5 mM
l-glutamine. Cultures were maintained at 37C and 5% CO2, and passaged every 3 d. To








assess VPA-induced differentiation, cells were exposed to 0.3 1.8 mM VPA for 2 d, and

then neurite outgrowth was quantitated as described below. The cytoprotective assay was

carried out following differentiation of cells by 7-d exposure to 100 ng/ml NGF, and
subsequent NGF removal and replacement with 0.3, 0.6, and 1.2 mM VPA for 2 d.

Viviana Puig is acknowledged for her extensive assistance in carrying out the technical

aspects of these experiments utilizing the PC 12 cells.


Neurite Quantitation


HN33 cells were cultured in the presence or absence of 0.6 mM VPA for 3-14 d,

and process outgrowth was assessed using phase-contrast microscopy. A neurite was

scored if its length was at least one respective cell body diameter. Treatments were

compared for percentage of cells with one or more neurites and two or more neurites, and
150-900 cells were scored per treatment group. PC 12 cells were similarly assessed, using

the same criteria, with the aid of NIH Image 1.47 imaging software. Dan O'Donnell,

Jeremy Grimes, Risha King, and Keith Templin are acknowledged for their participation in

quantitation of neurite outgrowth in the PC 12 cells.


Protein Quantitation by Western Blotting


Harvested cells were homogenized in a buffer containing 20 mM HEPES, pH 7.4,

2 mM EGTA, 1 mM PMSF, 2 mM DTE, and 10 .g/ml aprotinin. Cells were sonicated to

disrupt cell membranes, and the soluble and pellet fractions were separated by
centrifugation. The homogenate was centrifuged at 100,000 x g and the soluble fraction

collected. The pellet was resuspended in buffer containing 0.1% Triton X-100 and

solubilized for 30 min. Solubilized fractions were then centrifuged at 50,000 x g and the

supernatant containing the solubilized membrane protein collected. Samples were









adjusted by addition or dilution to 0.05 % Triton X-100. Equal amounts of soluble and
membrane cell protein (50 4g), as determined by the Bradford method, were separated by

SDS-PAGE and transferred to polyvinylidene fluoride membrane (Millipore Corp.,

Bedford, MA) in a Bio-Rad Trans-Blot electrophoresis apparatus at 100 V for 2 h using

Towbin's buffer (25 mM Tris, pH 8.3, 192 mM glycine, and 20% (v/v) methanol). All

antibodies were diluted in TS buffer (20 mM Tris, pH 7.5, and 0.5 M NaCI); see Table 2.2
for dilutions and incubation times. Detection of the immune complex was performed

using HRP-conjugated anti-rabbit or anti-mouse IgG, and the Pierce (Rockford, IL)

enhanced chemiluminescence system. Western blots were quantified using NIH Image

1.47 software for densitometric analysis. Results are expressed as the percentage of

protein present in drug-free controls grown in parallel and assayed on the same Western

blot.


Ribonuclease (RNase) Protection Assay (RPA)


The RPA was performed with the RPAII kit using biotinylated antisense

riboprobes (BIOTINscript and BrightStar system, Ambion). Biotinylated antisense

riboprobes from the Macs (mouse gene coding for MARCKS) and the cyclophilin genes

were generated and gel-purified. A fragment of Macs was cloned by RT-PCR into a TA

cloning vector (InVitrogen, Carlsbad, CA) and restricted at a unique XmaI site. In vitro

transcription using SP6 RNA polymerase and this construct as the template yields a 466 nt

riboprobe and protects a 386 bp fragment of MARCKS mRNA. The cyclophilin probe

(Ambion) protects a 103 bp fragment. For each sample, 15 jLg of RNA was incubated

with 1 ng of each probe at 44C overnight. Non-hybridized RNA was digested with a
mixture of RNases. Protected fragments were separated on a denaturing 5% PAGE and

then transferred onto a nylon membrane to allow detection. Resulting autoradiographs
were then quantified and the amount of Macs expression was normalized to the cyclophilin









internal control. Sharlynn Sweeney is acknowledged for her expertise in development of

the probes, as well as her technical skill in carrying out the RPAs and quantitating results.


PKC Inhibitors


HN33 cells were exposed to the PKC inhibitor LY333531 at 1.0 M (Eli Lilly and

Company) concomitantly with 1 mM VPA for 24 h. Alterations in VPA-induced

MARCKS down-regulation and GAP-43 up-regulation, as well as cell density and

morphology, were measured and compared between control groups and LY33353 1-

treated groups, each in the presence or absence of VPA. PC 12 cells were exposed to the

PKC inhibitor bisindolylmaleimide (BIM; Gibco, Gaithersburg, MD) at 200 nM for 48 h,

and alterations in cell density and morphology were quantitated and compared between

treatment groups.


PKC Activity


PKC activity was measured by the phosphorylation of myelin basic protein (MBP)

4-14 peptide (Upstate Biotechnology, Lake Placid, NY) and MARCKS peptide. Soluble

and membrane cell fractions were prepared as described above. Ten microliters of the cell

protein (1 4g/41) were added to 100 gI1 of a reaction mixture containing 100 pM MBP, 100
W CaC12, 20 mM HEPES (pH 7.5), 0.03% Triton X100, 10 mM MgCI2, 100 pWM ATP,

and 0.0228 p.Ci/J [y-32P]ATP, with or without 100 4.g/ml phosphatidylserine and 20 4g/ml

PMA. The mixture was then incubated at 300 C for 2 min. The reaction was stopped by

the addition of 50 RI of ice-cold 450 mM phosphoric acid. Following a 5 min incubation
on ice, 30 p.1 of each sample was spotted onto Whatman P81 filter paper. Filters were
washed (x4) with a mixture of 150 mM H3PO4 and 10 mM sodium pyrophosphate, rinsed

in ethanol and then acetone, then dried and radioactivity measured. Specific activity for









PKC was measured as nanomoles of 32P/min/mg protein, and expressed as the difference

between activity in the presence and absence of PMA/phosphatidylserine. Dr. David

Watson is acknowledged for his contribution to the development of the PKC activity assay

and his assistance in carrying out the experiments.


Statistical Analyses


Following consultation with a biostatistician, statistical differences between

treatments were analyzed by one-way ANOVA, and post hoc comparisons were

conducted using Fisher's PLSD test. In all cases, minimal acceptable level for significance
was p<0.05. All results are reported as mean percentage of an untreated control group

assayed in parallel, and error bars represent S.E.M.. Drs. Ronald Marks, Robert

McNamara, and Mark Lewis are acknowledged for their assistance in identifying valid

statistical methods for analyzing these data.










Table 2.2. Antibodies for Western Blots. This table lists the primary antibodies utilized in western immunoblots to detect
each of the proteins of interest. For each antibody, the following information is reported: source from which antibody was
obtained, dilution (in tris-saline, or TS, buffer) at which membrane was exposed to antibody, length of exposure of
membrane to antibody, and secondary HRP-conjugated antibody used for amplification and detection of primary antibody,
its source, dilution (in TS buffer + 0.2% Tween-20), and exposure time. The anti-MARCKS polyclonal primary antibody
was raised against a glutathione-S-transferase MARCKS fusion protein designed in our laboratory, as detailed in Watson
and Lenox (1996).


10 ALk


C, ,


I nr.


9o Ah


,~, rr~


n.1 Inr"hnt;nn
I t-k ource, UL U11 "%.,U a U"


;Iulhti;rn


Inn Ih~tlnnl


MARCKS (custom made)

GAP-43 Boehringer-
Mannheim

PKC-ct Transduction Labs

PKC-6 Santa Cruz

PKC-e Santa Cruz

PKC-C Transduction Labs


1:36,000

1:3,000


1:1,000

1: 750

!1,000

1: 1,000


overnight


anti-rabbit Bio-Rad


overnight anti-mouse


anti-mouse

anti-rabbit

anti-rabbit

anti-mouse


Bio-Rad


Bio-Rad

Bio-Rad

Bio-Rad

Bio-Rad


1:24,000

1:20,000


1: 10,000

1:20,000

1:20,000

1:20,000














CHAPTER 3
EFFECTS OF VPA ON MARCKS AND GAP-43


Introduction


VPA has traditionally been used for its antiepileptic properties, which were first

discovered in 1963 by Meunier et al., and encompass a broad range of seizure disorders.

Of late, VPA has been recognized as a suitable alternative to lithium, the traditional first-
line drug, for both the acute and prophylactic treatment of bipolar disorder. VPA and

lithium are currently the only two FDA-approved anti-manic drugs. Although lithium has

been studied more extensively and will serve as the model upon which experiments in this

study are based, the mechanism of anti-manic and mood-stabilizing action of this drug

remains unclear, as does the etiology of the psychiatric disorder targeted by these drugs.

Therefore, investigations into the mechanism(s) underlying the mood-stabilizing activity of

both drugs is an important area of study.


MARCKS


Studies conducted in our laboratory, using both animal and cell models, have

provided evidence for a significant down-regulation of MARCKS following chronic but

not acute exposure to therapeutically relevant concentrations of lithium (Lenox el al,

1992; Watson and Lenox, 1996). In immortalized hippocampal cells, Watson and Lenox
(1996) first demonstrated that long-term (3-10 d) exposure to 1-10 mM LiCl resulted in a

reversible, time- and concentration-dependent down-regulation of MARCKS protein.









Similarly, in vivo studies demonstrated that exposure of rats to chronic lithium (3-4

weeks), resulting in clinically relevant serum concentrations of 0.7-1.2 mEq/kg in brain,

also reduced MARCKS in the hippocampus (Lenox el al., 1992). Additionally, recent

data from our laboratory suggest a time course of MARCKS mRNA regulation by lithium
in HN33 cells similar to the time course of MARCKS protein down-regulation (with
reductions evident by 3 d and persisting beyond 10 d), thereby indicating alterations in

either transcription or post-transcriptional message stability (Watson and Lenox, 1997).

MARCKS expression is widely distributed, with highest levels normally expressed
in brain, spleen, testis, and lung, and lower expression in heart, kidney, liver, and muscle

tissues of mouse (Lobach el al., 1993; Swierczynski el al., 1996). In the brain, the

greatest expression is in the most plastic regions, including the amygdala, striatum,

olfactory bulb and cortex, and septum (Ouimet el al., 1990; McNamara and Lenox, 1997),

and MARCKS is localized to small dendritic branches and axon terminals of neurons and

both cytoplasmic and membranous components of glia (Katz el al., 1985; Patel and

Kligman, 1987; Ouimet el al., 1990; Rosen el al., 1990). MARCKS cross-links actin and

binds calmodulin in a Cal-dependent manner (these events are mutually exclusive; Sheu

et al., 1995; Allen and Aderem, 1995), and as a result, has been implicated in cell motility,

secretion and trafficking, and cell cycle regulation (for reviews, see Aderem, 1992 and

Blackshear, 1993). As is indicated by its name, the activity and expression of MARCKS

are regulated in part through phosphorylation by PKC, though PKC-independent

pathways may also play a role in its post-transcriptional regulation (Brooks et al., 1992).

The inhibition of PKC or mutation of the PKC phosphorylation sites of MARCKS results
in defects in cellular motility (Allen and Aderem, 1995).
MARCKS is a protein for which the precise function is unclear but which is

believed to be essential to normal CNS development, since mice lacking detectable mRNA

and protein, due to gene (Macs) knockout, died perinatally, either before or within just

hours after birth (Stumpo el al., 1995). Developmental defects included defects in midline









events such as neurulation, fusion of the cerebral hemispheres, formation of the great
forebrain commissures, and retinal and cortical lamination (Stumpo el al., 1995).

Ultrastructural examination revealed widespread neuronal ectopia, or abnormal neuronal

pathfinding, during cerebrocortical development (Blackshear et al., 1996). Whereas these
findings do not preclude the possibility of more subtle ultrastructural or immunological

defects in other non-neuronal tissues in which MARCKS is expressed, phenotypic

alterations were not observed outside the CNS, and the data overwhelmingly support the

notion that the mortality is primarily a result of the high frequency of midline defects in

brain resulting from the MARCKS-deficient phenotype. Interestingly, this knockout
phenotype was "rescued" by the expression of a supplemental human MARCKS gene,

MACS, in that the human transgene contained all of the elements necessary for normal

developmental expression of MARCKS, and was able to fully complement the MARCKS-

deficient phenotype of the knockout mice (Swierczynski et al., 1996).

The purpose of the following experiments is to investigate the effects of VPA on

MARCKS expression, since significant effects were reported following chronic lithium

exposure, as noted previously, and any common action of these two mood-stabilizing

agents should be instructive as to their mechanisms of action. If MARCKS alterations are

integral to the mood-stabilizing properties of these and other drugs, one would

hypothesize that VPA should down-regulate MARCKS in a manner similar to that of

lithium. Both concentration-response and time-course curves of VPA-induced effects will

be established. These parameters address the therapeutic relevance of observed effects,
since each drug exhibits its own clinical profile, including such characteristics as

therapeutic serum concentration and latency to response. Previous studies have reported

that the effects of lithium on MARCKS protein expression occur following chronic but not

acute exposure to concentrations within the reported therapeutic serum concentration
range of this drug, and that the observed down-regulation of MARCKS expression is

reversible upon drug withdrawal. These findings are consistent with the known clinical









profile of lithium. Similar results are hypothesized following VPA exposure, since this
drug exhibits similar clinical properties of delayed onset of antimanic drug action and

delayed return to baseline following drug removal, over a therapeutic concentration range

(50-125 p.g/ml). The long-term effects of VPA will be the focus of these studies, for they

may be important to the mood-stabilizing efficacy of this drug, whereas effects observed

more acutely are generally credited to the anticonvulsant properties of VPA.

As mentioned in Chapter 1, accumulating evidence implicates receptor-mediated

PI signaling in the mechanism of action of lithium in the brain (Berridge el al., 1982;

Godfrey, 1989; Lenox and Watson, 1994). Lithium is an uncompetitive inhibitor of the

enzyme IMPase, and as such, it leads to accumulation of intermediates in the pathway

responsible for additional downstream effects, including intracellular Ca++ release and

DAG-mediated PKC activation (refer to Figure 1.2; Post el al., 1992). It is hypothesized

that depletion of free inositol, resulting from the accumulation of intermediates due to
incomplete cycling of the PI signaling pathway, is a major consequence of long-term

lithium exposure (Berridge ei al., 1982; Berridge, 1989). Such a phenomenon would have
its greatest impact on certain areas of the brain which are relatively more inositol-limited,

for example by virtue of the blood-brain barrier or areas less capable of producing inositol
de novo. Further, because the PI signaling cascade is linked to muscarinic receptor/Gq

protein activation, areas of the brain undergoing the highest rate of receptor activation

should be preferentially affected by lithium. Indeed, previous studies in our laboratory
lend support to these hypotheses (Watson and Lenox, 1996). Firstly, in studies of HN33

cells exposed to lithium, inositol supplementation (at concentrations as low as 5 M) was

able to prevent the down-regulation of MARCKS induced by 1 mM lithium, thereby

supporting the inositol-depletion hypothesis. Secondly, addition of the muscarinic
receptor agonist carbachol (1 mM) increased the MARCKS down-regulation observed

(Watson and Lenox, 1996), which supports the notion that the action of lithium is

augmented in cells or tissue regions undergoing the highest rate of receptor activation.









Whereas the mechanism of action of lithium has often been attributed to this

modulation of PI signaling and its downstream effects, there is little evidence in support of

a similar role for VPA; in fact, studies have indicated that VPA is probably not working

through effects on PI signaling. In one study comparing the effects of lithium,

carbamazepine, or VPA on IMPase activity, lithium and carbamazepine were found to

decrease and increase IMPase activity, respectively, and VPA was shown to have no

effect, at concentrations ranging from 2 j.M to 5 mM (Vadnal and Parthasarathy, 1995).

In another study, 2 mM valproate failed to increase accumulation of inositol phosphates,

including the inositol monophosphates, inositol 1,3-bisphosphate, inositol 1,4-

bisphosphate, and inositol 1,3,4-trisphosphate, whereas lithium exposure resulted in

accumulation of all (Dixon and Hokin, 1997). In a study conducted in our laboratory,

VPA failed to result in accumulation of CMP-PA (also known as CDP-DAG; see Figure

1.2), a metabolite of DAG and a sensitive measure of PI signaling, whereas lithium

resulted in significant accumulation of CMP-PA in CHO-KI and HN33 cells (Watson el

al., 1998). These data suggest that inhibitory effects on IMPase may not be the common

site of action for mood-stabilizing agents. However, this does not dismiss the possibility

that such agents may act indirectly or on alternate locations to alter the PI signaling

cascade. Experiments discussed herein have been modeled after those described above for
lithium, and will serve to clarify further the relationship between inositol or carbachol and

VPA-induced effects. Based on data from previous studies showing a lack of influence by

VPA on the PI signaling cascade, it is hypothesized that neither inositol nor carbachol will

alter any observed effect of VPA on MARCKS protein expression.

As mentioned previously, attention will be given to the possible therapeutic

relevance of findings. Inasmuch as lithium and VPA are commonly administered in

combination to patients who fail to respond adequately to either drug alone, often
resulting in amelioration or attenuation of manic symptoms, one study has been designed

to test the hypothesis that there may be additive actions at the cellular level. Following









optimization of exposure conditions for each drug, cells will be exposed to the

combination of VPA and lithium, and effects on MARCKS protein assessed. A final study

involves exposing cells to a series of other psychotropic agents individually, in order to

assess the specificity of action of mood-stabilizing agents and serve as a control for other

studies conducted. If MARCKS mediates any of the mood-stabilizing properties of these

agents, one would hypothesize that a variety of drugs efficacious in the treatment of
bipolar disorder would elicit the MARCKS down-regulation observed following lithium

exposure.

Experiments described thus far have simply focused on alterations in MARCKS

protein expression following exposure to drug and manipulation of various other

conditions in vitro. However, little is known of how these changes come about, or how
MARCKS levels are regulated in vivo. Regulation of MARCKS expression is thought to

occur in several ways: (1) tissue-specific expression, (2) developmentally regulated

expression, (3) differentially regulated transcription rates, and (4) alterations in mRNA

levels in response to PKC (Blackshear, 1993; Lobach et al., 1993). VPA-induced effects

on PKC itself will be discussed in detail in Chapter 5. Accumulating evidence shows that

lithium and VPA have regulatory effects on various transcription factors, thereby

suggesting a role for these mood stabilizers in altering gene expression. Studies in

transfected cultured cells have indicated that VPA enhances expression of genes and

reporters which contain AP-1 elements in their regulatory domains (Simon et al., 1994;

Kuntz-Simon el al., 1995). Manji and colleagues showed further that VPA exposure

resulted in a time- and concentration-dependent increase in DNA binding activity of the

transcription factor AP- 1 in both rat C6 glioma and human neuroblastoma (SH-SY5Y)

cells; lithium reportedly induced similar effects (Chen el al., 1997).
It is unknown how MARCKS expression is regulated beyond transcription, or how

the lithium-induced down-regulation of MARCKS protein occurs. Expression of many
proteins is known to be regulated not only by transcriptionally, but also by post-









transcriptional mechanisms, so that changes in mRNA stability play a crucial role in the

control of GAP-43 expression, for example. Similar post-transcriptional mechanisms
might also be at work in the case of MARCKS, and it is the purpose of these experiments

to assess further this possibility. Preliminary studies in our laboratory have indicated that

following lithium exposure, MARCKS mRNA levels do decline along a time course

similar to that of the protein down-regulation (Watson and Lenox, 1997), and one would

expect the same for VPA-induced alterations in MARCKS mRNA: a down-regulation of

mRNA levels over time, at therapeutically relevant concentrations of VPA, and with a

delayed onset of effect and protracted recovery of expression upon VPA withdrawal.


GAP-43


The growth-associated protein, GAP-43, is another major PKC substrate in brain

which shares many similarities with MARCKS. Like MARCKS, GAP-43 binds both

calmodulin and filamentous actin, albeit in a Ca+-independent manner (reviewed in

Benowitz and Routtenberg, 1997), and may play a role in exocytosis (Dekker et al.,

1989). GAP-43, like MARCKS, appears to be integral to normal nervous system

development, because studies of mice lacking the protein as a result of antisense-mediated

gene knockout show a high frequency of early postnatal death, with few animals surviving

past three weeks of age and even fewer reaching maturity (Strittmatter el al., 1995).
Those mice that do survive exhibit abnormal neuronal pathfinding.

GAP-43 is a rapidly transported, presynaptic protein most highly expressed during

development and post-injury regeneration, and immunohistochemistry and in situ

hybridization studies have shown GAP-43 expression to be almost exclusively neuronal

(Benowitz and Routtenberg, 1997). Neurite outgrowth is generally accompanied by a

large increase in GAP-43 expression, and following differentiation and synaptogenesis,

GAP-43 levels decline sharply and remain low, except in the most plastic regions of the









brain, such as associative and limbic areas, where intense immunostaining persists in adults

throughout life. This selective expression is thought to confer the unique potential for

ongoing structural remodeling (Benowitz and Perrone-Bizzozero, 1991).

GAP-43 is associated with the cytoskeleton and is concentrated in the growth

cones of elongating axons and nerve terminals, thought to help stabilize growth cone
morphology and promote growth cone adhesion (Aigner and Caroni, 1995). Support for

this notion has been provided by a number of studies. Antisense-directed depletion of

GAP-43 in growth cones results in the absence of spreading, branching, and adhesion of
growth cones (and therefore, neurite outgrowth) in primary sensory neurons (Aigner and

Caroni, 1993). Whereas suppression of GAP-43 expression has been found to yield

adverse effects on axon outgrowth both in vivo and in vitro (Meiri et al., 1998; Benowitz

and Routtenberg, 1997; Strittmatter et at., 1995; Jap Tjoen San et al., 1992; Shea et al.,

1991), overexpression of the protein induced nerve sprouting in the adult nervous system

of transgenic mice (Aigner et al., 1995). Even in non-neuronal cells, expression of a

GAP-43 transgene has been found to induce extensive process outgrowth accompanied by
reorganization of the membrane cytoskeleton (Strittmatter el al., 1994; Yankner e al.,

1990; Zuber et al., 1989), yet, a mutant PC 12 cell line which does not express GAP-43

still exhibits neurite outgrowth (Baetge and Hammang, 1991).

Although studies are in disagreement as to the precise role of GAP-43 in these

events, it has been hypothesized that the almost exclusive presence of GAP-43 in neuronal

cells, and the persistence of GAP-43 expression in highly plastic limbic regions, confers

the ability to undergo the structural remodeling necessary for the functional plasticity of

the brain (Benowitz and Perrone-Bizzozero, 1991). Indeed, Skene and Willard (1989)
suggest that it is the induction of GAP expression that is necessary to post-injury
remodeling, and that the inability of many areas of the CNS to induce high levels of GAP

expression may in part underlie the failure of CNS axons to regenerate after axonal injury,

whereas areas of the peripheral nervous system retain the ability to up-regulate GAP-43 as









well as to regenerate axons after injury. However, the presence of GAP-43 is probably

not necessary to neurite outgrowth per se, but rather, plays an important role in regulating
precisely the various parameters involved in the overall event, such as navigation or

pathfinding (Aigner and Caroni, 1995). According to this "GAP hypothesis", nerve
growth is controlled in part by the expression of certain growth associated proteins, such

as GAP-43, which provide a mechanism for regulating neurite outgrowth and structural
plasticity in the nervous system (Skene and Willard, 1989).

Because of the vital role of GAP-43 in effecting plastic changes in brain,

experiments have been designed to address the effect(s) of VPA exposure on GAP-43

expression in HN33 cells. It may be that part of the efficacy of VPA in either bipolar

disorder or epilepsy, or both, lies in its ability to elicit structural modifications of plastic
regions exhibiting the pathology which leads to the manifestation of the particular

disorder. Indeed, Benowitz and Routtenberg (1997) pointed out that in brain of humans
and other primates, the highest levels of GAP-43 expression occur in regions known to be

involved with higher-level associative processes, lending support to the notion that GAP-

43 is important for maintaining plasticity.

Whereas in vilro studies in HN33 cells have shown no evidence for regulation of

GAP-43 by lithium, neither has lithium been shown to alter the morphology of these cells.

Preliminary studies have suggested that chronic VPA exposure induces a morphological

alteration in these cells which might be further characterized as differentiation (see

Chapter 5). Since neurite outgrowth and neuronal differentiation are known to be

associated with up-regulation of GAP-43 protein expression, it will be of interest to

investigate the effect of VPA on this protein.

The results of these studies will be instructive, for in addition to the development

of a model which may help explain the role of MARCKS and other plasticity proteins

(such as GAP-43) in the mechanism of action of both VPA and lithium, findings will

contribute to a better understanding of the pathophysiology involved in bipolar disorder.









The notion of synaptic remodeling as the underlying means for the drug- or therapy-
induced stabilization of mood disorders has been supported by some reports. One study

found increased levels of neural cell adhesion molecule (N-CAM)-immunoreactive

proteins in the cerebrospinal fluid of a population of bipolar patients (Poltorak el al,
1996). Abnormal function or turnover of N-CAM, which plays a role in synaptogenesis,

morphogenesis, and plasticity of the nervous system, may be suggestive of abnormal

synaptic plasticity in affected individuals. With regard to drug therapy for this disorder,

not only do lithium and VPA affect expression of proteins involved in neural plasticity and

cellular response (as previously discussed), but both drugs have also been shown to have

effects on transcriptional factors which regulate gene expression, and it is known that

long-term changes in neuronal synaptic function are dependent upon the selective

regulation of gene expression (Manji and Lenox, 1994). Lamotrigine, another drug used

in treatment of bipolar disorder, was recently shown to alter neuronal hyperexcitability via

inhibition of both Na' and Ca- channels, findings which were interpreted by the authors to

signify modification of synaptic plasticity (Xie and Hagan, 1998). Such evidence is

indirect, but overall suggestive that changes in synaptic plasticity may be a mechanism by

which anti-bipolar (anti-manic and mood-stabilizing) therapy facilitates long-term changes

in affect.









Results













1 2 1 2


Figure 3.1. MARCKS protein. Shown are Western
immunoblots comparing MARCKS protein expression in
(1) mouse brain and (2) HN33 cells. A shows cytosolic
MARCKS expression, and B shows membrane-
associated MARCKS.


Comparative MARCKS expression in HN33 cells (used in this study) vs. mouse

brain is shown in Figure 3.1 following enhanced chemiluminescent (ECL) detection.

Differences in apparent molecular weight may result from differential phosphorylation or

myristoylation of the protein. Note the more pronounced MARCKS protein expression in

mouse brain, even with comparable amounts of total loaded protein.



Effect of VPA Exposure on MARCKS Protein in HN33 Cells

VPA was administered at concentrations ranging from 0.001 to 1.0 mM, for I or 3

d (Figures 3.2 and 3.3, respectively). Representative Western blots are presented, along

with quantitation of mean data. Significant concentration- and time-dependent effects

were observed. Chronic (3 d) VPA exposure results in a concentration-dependent

reduction in MARCKS levels, with a more pronounced reduction in membrane than






38


soluble fraction of these cells. Significant down-regulation of MARCKS is observed in the

membrane fraction at concentrations as low as 0.06 mM (p<0.005), and in the soluble

fraction at 0.1 mM and higher concentrations (p<0.005). These concentrations correlate

well with the target therapeutic serum concentration range used clinically for treatment of

bipolar disorder.


a 2U 3 45 6
1 2 3 4 5 6 7


100-

80-

60-

40-

20 -

A


0.01 0.03 0.06 0.1 0.3 0.6 1


IVPAI, mM


Figure 3.2. Concentration-dependent down-regulation of MARCKS protein in
HN33 cells following I d VPA exposure. (a) Representative Western blot of
membrane-associated MARCKS expression following I d VPA exposure; mean data
are depicted graphically in 3.2b. Lanes 1-7 were treated for I d as follows:
untreated control, 0.01 mrM, 0.03 mM, 0. 1 mM, 0.3 mM, 0.6 mM, and 1.0 mM
VPA, respectively. (b) This figure shows quantitatively the MARCKS protein levels
in HN33 cells following I d exposure to varying concentrations of VPA. Data,
expressed as percent of MARCKS compared to untreated control cells grown in
parallel, were derived from Western immunoblots (e.g., Figure 3.2a) and are the
mean of at least four determinations. Error bars represent S.E.M.


T T T] Soluble
T T-T TTT T E] Membrane
T I i -I- 1

T I TT
T















1 2 3 4 5 6 7 8 9 10 11 12


0.001 0.01 0.03


0.06 0.1 0.3 0.6 1


IVPAI, mM




Figure 3.3. Concentration-dependent down-regulation of MARCKS protein in HN33
cells following 3 d VPA exposure. (a) Representative DAB Western blot of MARCKS
expression following 3 d VPA exposure. Lanes 1-6 show soluble MARCKS following
3 d treatment with 0, 0.03, 0.06, 0.1, 0.3, and 0.6 mM VPA, respectively, and lanes 7-
12 show membrane MARCKS following exposure to the same doses. (b) This figure
shows quantitative data for mean MARCKS levels following 3 d VPA exposure. Data,
expressed as percent of MARCKS compared to untreated control cells grown in
parallel, were derived from Western immunoblots and are the mean of at least four
determinations. Error bars represent S.E.M. *p<0.005


140-

120-

100-

80-

60-

40-

20-


TQ Soluble
I T Q Membrane


T T T
IT

11 T






40


Time Course of VPA-Induced MARCKS Down-Regulation in HN33 Cells

HN33 cells were exposed continuously to 0.6 mM VPA for periods ranging from 0

to 7 d. This concentration is equivalent to 100 ng/ml, which is within the reported

therapeutic range for the mood stabilizing action of VPA (50-125 ng/ml). Cells were

observed to undergo a significant down-regulation of MARCKS protein expression in

both fractions as early as 3 d, with no effect on MARCKS at I d (see Figure 3.4). The

MARCKS reduction in the membrane fraction is more pronounced than that in the

cytosol. Three d, 5 d, and 7 d values are significantly different (p<0.01) from I d

MARCKS values in both soluble and membrane fractions of these cells. These data are

consistent with the clinical profile of this drug, which exhibits a delay in onset of antimanic

action at target therapeutic concentrations.



100-
1 T Soluble ] Membrane

II

60
TI
O 40- T
20-I
S20

1 3 5

Time (Days)



Figure 3.4. Time course of VPA-induced MARCKS down-regulation in
HN33 cells. This figure illustrates the time-dependency of VPA-induced
MARCKS protein down-regulation. HN33 cells were exposed to 0.6 mM
VPA, a clinically relevant therapeutic concentration, for 1, 3, 5, or 7 d, and
MARCKS expression was assessed by quantitative immunoblot analysis.
Results, expressed as percent of untreated controls, represent the mean
S.E.M. of at least four determinations. *p<0.01










a
140 SolubleT
130 Soluble Withdrawal
120 0 Membrane
110 -0 Membrane Withdrawal
U 100 T
90
80 0
70 T
70
S50*
40
30-
S20-
10
0-
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Time (days)


b


1 2 3 4 5


Figure 3.5. Recovery of MARCKS following VPA-induced long-term down-
regulation. (a) This figure shows the time course of VPA-induced MARCKS down-
regulation in HN33 cells, as well as the recovery of MARCKS protein to basal levels
following drug withdrawal. HN33 cells were exposed to 0.6 mM VPA for 3 d, at
which time the drug was either removed or treatment continued. MARCKS levels
were assessed through Western immunoblot (Figure 3.5b is representative), and results
are expressed as percent of untreated control cultures grown in parallel. Results
represent mean S.E.M. of at least three determinations. *p<0.0001 (b) Shown is a
representative Western blot of MARCKS expression in the membrane fraction of HN33
cells following chronic VPA exposure and subsequent withdrawal. Lanes I and 3 show
control levels of MARCKS protein. Lane 2 shows MARCKS expression following 3 d
VPA exposure (compare to Lane 1), and Lane 4 shows MARCKS expression
following 6 d VPA exposure (compare to Lane 3). Lane 5 represents a VPA
withdrawal/recovery group, and shows MARCKS in cells exposed continuously for 3 d
to 0.6 mM VPA, and then for 3 d to media free of drug.


Recovery of MARCKS following VPA-lnduced Long-Term Down-Regulation

As shown in Figure 3.5a, HN33 cells exposed to 0.6 mM VPA for 0 to 14 d

exhibited a significant down-regulation of MARCKS protein expression in both fractions









(membrane > soluble) over time, as previously demonstrated in Figure 3.4. After 3 d of

continuous VPA exposure and subsequent drug washout and withdrawal, MARCKS

protein levels recovered with significant increase (p<0.0001) to control levels within 3-7 d

of culture in drug-free medium (Figure 3.5a). A representative Western blot is shown in

Figure 3.5b. These findings are consistent with clinical evidence for delayed onset of anti-

manic action of VPA, as well as prolonged anti-manic efficacy following drug

discontinuation. It is difficult to discern whether this experimental return of MARCKS to

baseline levels is the result of either a shift in the kinetics of MARCKS

production/degradation, or due to repopulation of the il vitro environment with

MARCKS-replete daughter cells.


Effects of Carbachol and myo-Inositol on the VPA-Induced Down-Regulation of

MARCKS

In order to examine the role of PI signaling in the mechanism of action of this

drug's effects on MARCKS, cells were exposed to VPA alone or in combination with

either carbachol (1 mM) or rnyo-inositol (up to 1 mM). In contrast to previous results

obtained with LiCl (Watson and Lenox, 1996), the VPA-induced down-regulation of

MARCKS protein expression was not altered by addition of carbachol (Table 3.1 and

Figure 3.6) to the culture medium. Similarly, supplementation of myo-inositol (Table 3.2

and Figure 3.7) to the culture medium at concentrations up to 1 mM did not modify the

VPA-induced down-regulation of MARCKS, whereas the LiCI-induced down-regulation

was prevented by the addition of as little as 5.0 iM rnyo-inositol. Representative Western

blots are presented in Figures 3.6 and 3.7. It is difficult to rule out completely the role of

PI signaling based on the data in Table 3.1 alone, due to the maximal level of MARCKS

down-regulation achieved in the case of VPA. However, the data in Table 3.2 are more

conclusive, as the levels of MARCKS reduction elicited by VPA and lithium are more

comparable.










Table 3.1. Addition of Carbachol Fails to Potentiate the VPA-Induced MARCKS Down-
Regulation in HN33 Cells. HN33 cells were exposed to 0.6 mM VPA for 3 d or 1 mM
lithium for 7 d, in the presence or absence of the muscarinic agonist carbachol. Results
are presented as percent of untreated control cultures grown in parallel, and represent the
mean S.E.M. of at least three determinations. *p<0.0001


MARCKS Protein, % Control (Mean S.EM.)


VPA-Exposed


LiCI-Exposed


[Carbachol] Soluble Membrane Soluble Membrane
0mM 41.86.6 17.08.5 81.6+4.5 90.23.1
lmm 46.57.5 10.82.0 66.6+6.1"* 68.75.9**


ens


1 2 3 4


Figure 3.6. Effect of carbachol on VPA-induced MARCKS down-
regulation. Shown is a representative Western blot of MARCKS
expression following VPA exposure with or without the concomitant
addition of carbachol. Lane I shows membrane-associated MARCKS
levels in untreated control HN33 cells, Lane 2 shows MARCKS in 3 d
VPA-treated cells, Lane 3 shows MARCKS in carbachol-treated cells,
and Lane 4 shows MARCKS in cells treated concomitantly with VPA
and carbachol. Mean data are reported in Table 3.1.










Table 3.2. Inositol Supplementation Fails to Attenuate the VPA-Induced MARCKS
Down-Regulation in HN33 Cells. HN33 cells were exposed to 0.6 mM VPA for 3 d or I
mM lithium (+1 mM carbachol) for 7 d, in the presence (or absence) of varying
concentrations of myo-inositol. Results are presented as percent of untreated control
cultures grown in parallel, and represent the mean S.E.M. of at least three
determinations (N.D. = not determined). *p<0.005


MARCKS Protein, % Control (Mean S.E.M.)


VPA-Exposed


LiCl/carbachol-Exposed


[myo-Inositol] Soluble Membrane Soluble Membrane
0 83.0 + 6.0 57.0 5.0 66.6 6.1 68.7 5.9
0.5 IiM N.D. N.D. 71.1 4.8 68.44.4
5 4M N.D. N.D. 97.9 3.6* 104.3 5.3*
40 tM 77.8 2.4 54.8 5.5 98.5 7.5* 103.6 8.2*
1 mM 78.3 9.4 51.3 10.8 N.D. N.D.


1 2 3 4



Figure 3.7. Effect of inositol on VPA-induced MARCKS down-
regulation. Shown is a representative Western blot of MARCKS
expression following VPA exposure with or without concomitant
addition of mnyo-inositol. Lanes 1-4 show membrane-associated
MARCKS in (1) control cells, (2) 3 d 0.6 mM VPA-treated cells,
(3) inositol-treated cells, and (4) combined VPA- and inositol-
treated cells. Mean data are reported in Table 3.2.






45


go-4 100.
a 0o
T -a Soluble
0 80- 1 Membrane
T

TT
W0 40.





VPA LiCl/Carbachol VPA + LiCl/Carbachol
Treatment


b A B


1 2 3 4 5 6 7 I 2 3 4 5 6 7


Figure 3.8. Additive effect of combined VPA and lithium exposure on MARCKS
protein expression. (a) HN33 cells were exposed to 0.6 mM VPA and/or 1 mM
lithium (+ 1 mM carbachol) in inositol-free DMEM for 3 d. MARCKS protein
levels, expressed as percent of untreated control, were assessed by Western
immunoblot analysis (see Figure 3.8b), and values represent mean S.E.M. of at
least three determinations. *p<0.01 (b) Representative Western blot of MARCKS
protein following 3 d exposure of cells to VPA and/or lithium. A shows cytosolic
MARCKS in HN33 cells after the following treatments: no drug (1); 0.3 mM VPA
(2); 0.6 mM VPA (3); 1 mM lithium (4); 1 mM lithium + 1 mM carbachol (5); 1
mM lithium + 1 mM carbachol + 0.3 mM VPA (6); and 1 mM lithium + 1 mM
carbachol + 0.6 mM VPA (7). B shows MARCKS in the membrane fraction of
HN33 cells following the same treatments for 3 d.

Additive Effect of Combined VPA and LiCI Exposure on MARCKS Protein

Expression in HN33 Cells

HN33 cells were exposed to 0.6 mM VPA and/or 1 mM LiCIA mM carbachol for

a period of 3 d, in order to assess the effects of the VPA-lithium combination on

MARCKS protein expression and distribution. Results, presented in Figure 3.8a, show

that a greater reduction of MARCKS protein was observed in both soluble and membrane









fractions of cells exposed to both VPA and lithium/carbachol than was produced by either

VPA or lithium/carbachol alone. The enhanced reduction in MARCKS protein produced

by the combination of VPA and lithium/carbachol at these concentrations appeared to be

additive. Furthermore, the reduction in MARCKS protein produced by the combined

treatment of VPA and lithium/carbachol was greater in the membrane fraction than in the

soluble fraction. A representative Western blot is shown in Figure 3.8b.


Effects of Other Psychotropic Agents on MARCKS Protein Expression

Eight structurally and functionally distinct psychotropic drugs (carbamazepine,

fluoxetine, imipramine, phenytoin, diazepam, haloperidol, morphine, and verapamil) were

administered to HN33 cells for 1, 3, or 7 d, and their effects on MARCKS protein

expression were assessed. Drugs were chosen to represent a wide range of

psychopharmacological actions. Minimal drug concentrations were based on the

therapeutically effective plasma concentration for each drug in humans, as outlined in
Table 3.3. Additionally, cells were exposed to concentrations up to 20 times the highest

reported therapeutic concentration so as not to overlook potential effects at supra-

therapeutic doses. At each concentration tested, cell viability, growth rate, and gross

morphological appearance of HN33 cells exposed to each of the psychotropic drugs were

not notably different than those of cells grown in the absence of drug. Data obtained

following 3 d exposure are shown in Figure 3.9a. For all of the drugs listed, no significant

dose effect on MARCKS protein expression was measured at any of the concentrations or

time points tested.
MARCKS protein levels in HN33 cells were also measured after exposure to each

of two VPA analogues or metabolites, hydroxyvalproic acid (HVPA, or 2-n-propyl-5-
hydroxy-pentanoic acid) and 2-n-propylglutaric acid (2-PGA), for 3 or 7 d (Table 3.3;
Figure 3.9ab) These VPA metabolites, the structures of which are depicted in Figure

3.9c, were provided by Abbott Laboratories (Abbott Park, IL), and have been found to









have little or no anticonvulsant efficacy in vivo (Chapman el aL, 1982; J. Sullivan,

personal communication). Cells were exposed to concentrations of 0.6, 1.5, and 3.0 mM.

Unlike VPA, which produced noticeable changes in cell morphology at 0.6 mM (see

Chapter 6), these analogs produced no such change at this or higher concentrations.

Following HVPA exposure, no significant reduction in MARCKS protein was evident in

the soluble fraction, even at concentrations 5-10 times higher than the minimal effective

concentration of VPA. However, in the membrane fraction, a significant reduction in

MARCKS protein was observed only in cells exposed to 3.0 mM HVPA (p<0.01 at 3 d;

p<0.05 at 7 d). At this highest concentration tested (3.0 mM; 5X the therapeutic level of

VPA), a 30-35% reduction in MARCKS protein was observed following 3-7 d exposure.

Following 2-PGA exposure, no significant reduction in MARCKS protein was apparent in

the soluble fraction, even at concentrations 5-10 times higher than the minimal effective

concentration of VPA A small but insignificant reduction in MARCKS protein in the

membrane fraction was observed following 3.0 mM exposure to 2-PGA for 3-7 d. A

representative Western blot is presented in Figure 3.9b.

0% 140
a I 4 T E] Soluble Membrane
l' 120, TTT
a TT
.10 T I
00T
T100 T T T
0- T F
i 80- .
o 60-

En 40-

1 20-


x Pk


Treatment










A B
X. ...iiiii~ iiiiiii~ iii i .. ...... . ..




1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9


C
HOOC

>OH COO OC


Hydroxyvalproic Acid *2-Propyl Glutaric Acid




COOH

*VaIproic Acid


Figure 3.9. Effects of other psychotropic agents on MARCKS protein expression.
(a) This figure shows the relative lack of effect of various other psychotropic agents
on MARCKS protein expression in HN33 cells. Cells were exposed to drugs other
than VPA or lithium (see Table 3.3 for details), as follows: 3.0 mM hydroxyvalproic
acid, 7 d; 3.0 mM 2-propylglutaric acid, 7 d; 100 p.M carbamazepine, 7 d; I pgg/ml
fluoxetine, 3 d; I pg/ml imipramine, 7 d; 20 pg/ml phenytoin, 3 d; I jig/ml
diazepam, 7 d; 20 ng/ml haloperidol, 7 d; 20 jtg/ml morphine, 7 d; 50 ng/ml
verapamil, 7 d. MARCKS levels were assessed as previously described; data
represent mean S.E.M. of at least three determinations. *p<0.05 (b) MARCKS
expression following exposure to VPA and two metabolites. Shown is a
representative Western blot of MARCKS protein in HN33 cells after 7 d exposure to
the following: no drug (1); 0.6 mM VPA purchased from Sigma Chemical Company
(2); 0.6 mM VPA provided by Abbott Laboratories (for comparison purposes; 3); 0.6
mM (4), 1.5 mM (5), and 3.0 mM (6) 2-propylglutaric acid; 0.6 mM (7), 1.5 mM (8),
and 3.0 mM (9) hydroxyvalproic acid. A shows cytosolic MARCKS expression, and
B shows membrane-associated MARCKS. (c) Structural comparisons of VPA and
metabolites. Shown are the chemical structures of VPA and two of its metabolites,
hydroxyvalproic acid, or 2-n-propyl-5-hydroxy-pentanoic acid, and 2-n-propylglutaric
acid.









Table 3.3. Drugs, Sources, and Exposure Conditions for HN33 Cells.
This table lists the drugs to which HN33 cells were exposed, their sources, target
circulating (plasma) concentrations in humans, concentrations tested in HN33 cells, and
vehicle in which drug was suspended. Drug concentrations tested encompassed a broad
range, including supratherapeutic concentrations, in order to account for discrepancies,
such as those presented by variation in circulating serum vs. brain concentrations of drug.


Drug / Source
Sodium Valproate
Sigma #P-4543
Abbott #A-44089.5


Target Plasma
Conc. in Humans
50-125 ig/ml
= 0.3-0.9 mM
(seizure control )ab


Concentrations Tested
0.01, 0.03, 0.06,
0.1, 0.3, 0.6, 1.0 mM


2-Propylglutaric Acid
Abbott #A-49999.0


Hydroxyvalproate, sodium salt
Abbott #A-49822.5


Carbamazepine
Sigma #C-4024


Fluoxetine-HCl
Lilly #110140L/F08083


Imipramine-HCI
Sigma #1-7379

Phenytoin, sodium salt
Sigma #D-0931

Diazepam
Sigma #D-0899



Haloperidol
Sigma #H- 1512

Morphine Sulfate
Sigma #M-8777

Verapamil-HCI
Sigma #V-4629

aHolford and Benet,
Meldrum, 1995.


8-12 g/ml
= 33-50 4M
(seizure control)b,c


varies


100-300 ng/ml
(antidepressant)a

10-20 .g/ml
(anticonvulsant)d,a

300-400 ng/ml
(anxiolytic)ab
>600 ng/ml
(seizure control)b

4-20 ng/ml
(anti-psychotic)b


5-20 ng/ml


0.6, 1.5, 3.0 mM


0.6, 1.5, 3.0 mM


10, 25, 100 M


50, 200, 1000 ng/ml


50, 200, 1000 ng/ml


1, 10, 20 gtg/ml


50, 200, 1000 ng/ml




5, 20, 100 ng/ml


5, 20, 100 ng/ml


(analgesic in non-tolerant patients)c


varies


50 ng/ml


1995; hBenet and Williams, 1990; cWeiner, 1996; dporter and


Vehicle
dH20


dH20


dH20


ETOH


dH20


dH20


ETOH


ETOH




dH20


dH20


ETOH






50


Effect of VPA on MARCKS mRNA

HN33 cells were exposed to 0.6 mM VPA for 0 to 14 d, and MARCKS mRNA

levels were measured by RPA as described in Materials and Methods. As shown in Figure

3.1 Oa, MARCKS mRNA levels decrease over time with continual VPA exposure

(p<0.05). A representative RPA is shown in Figure 3.10b.


100 -


1 3 5


7 10 14


Time (d)


40 40


:00i


MARCKS



Scyclophilin


$.db
Oo -O A- ....


A B C D E


F G H I J


Figure 3.10. Time course of VPA-induced MARCKS mRNA alterations. (a) This figure
shows the time course of VPA-induced alterations in MARCKS mRNA. HN33 cells
were exposed to 0.6 mM VPA for 0-14 d, and MARCKS mRNA levels were assessed by
RPA, with the aid of a cyclophilin internal control, as described in Materials and
Methods. Results for each treatment are calculated as a ratio of MARCKS to
cyclophilin, and then each value is expressed as a percentage ( S.E.M.) of untreated,
control samples prepared in parallel. *p<0.05 Shown in (b) is a representative RPA.
Upper band reflects MARCKS mRNA, and lower band reflects cyclophilin mRNA.
Treatments are as follows: A, control; B, I d VPA; C, 3 d VPA; D, control; E, 5 d
VPA; F, 7 d VPA; G, control; H, 10 d VPA; I, control; J, 14 d VPA.











Effect of VPA on GAP-43 Expression in HN33 Cells

GAP-43 expression in HN33 cells is compared to its expression in rat brain in

Figure 3.11. Following 3 d, 0.6 mM VPA exposure, GAP-43 levels in HN33 cells

increase, as shown in Figure 3.12. This increase is concentration-dependent, as shown in

Figure 3.13 (p
from 0.01 mM to 1.0 mM, and membrane-associated GAP-43 expression is shown (Figure

3.13).


A B



1 2 1


Figure 3.11. GAP-43. Shown is a Western
immunoblot comparing GAP-43 expression in (1)
soluble and (2) membrane fractions of(A) HN33
cells and (B) rat brain.








1 2


Figure 3.12. Effect of chronic VPA exposure on GAP-43 expression
in HN33 cells. Shown is a representative Western blot of GAP-43
expression in the membrane fraction of HN33 cells following VPA
exposure. Lane 1 shows GAP-43 expression in untreated control
cells, and Lane 2 shows GAP-43 in cells treated for 3 d with 0.6 mM
VPA.












.tVu -

300 -

200 -

100 -

0-


0.01 0.03 0.06 0.1 0.3 0.6


[VPA], mM

Figure 3.13. Concentration-dependent increase in GAP-43 expression in HN33
cells following VPA exposure. HN33 cells were exposed for 3 d to varying
concentrations of VPA, and membrane-associated GAP-43 expression was
assessed by Western immunoblot analysis. Results represent the mean of two
or three determinations, and are expressed as the percent of GAP-43 protein
expressed in untreated control cultures grown in parallel. *p

1 3 5 14
Time (d)

Figure 3.14. Time course of VPA-induced GAP-43 increase in HN33 cells. HN33
cells were exposed to 0.6 mM VPA for times ranging from I d to 14 d, and GAP-
43 levels were assessed by Western immunoblot analysis. Results reflect the mean
of at least three determinations, and are expressed as percent of GAP-43 in an
untreated control. *p








Time Course of VPA-Induced GAP-43 Increase in HN33 Cells

HN33 cells were exposed to 0.6 mM VPA for 0-14 d; results are shown in Figure

3.14. Up-regulation of GAP-43 was observed in membrane and soluble fractions of cells

exposed to VPA for 5 d and 14 d, respectively (p<0.05). Because of the experimental

protocol, which entailed growing cells to confluency and then diluting them every 3-4 d,

the data may not reflect the full potential for GAP-43 induction, since each splitting and

re-seeding procedure involved trypsinization and potential disruption of cellular processes.




Discussion


HN33 cells were utilized as the model for all studies, because their morphological,

immunological, and electrophysiological characteristics resemble those of hippocampal

neurons in culture. HN33 cells were chosen as the in vitro cell model in which to

investigate multiple steps in the PKC/GAP-43/MARCKS pathway, because of their

relative homogeneity and the presence of signal transduction pathways of interest. This is

in contrast to the use of other models, such as primary neuronal cultures, which are more

neurochemically heterogeneous in nature. In primary cultures, for example, it would be

difficult to assess whether PKC and other messenger systems were affected in the same

neuron. The homogeneity of HN33 cells makes signaling research and results less

confounding.


MARCKS


VPA exposure elicits both a time- and concentration-dependent reduction in

MARCKS protein expression. It has been well established that the clinical antimanic

response to VPA requires a lag period of days to weeks following drug administration









before onset of action, although loading strategies have proven useful in achieving faster

response (within 3 d) (Bowden el al., 1994; McElroy el al., 1989). Consistent with such
clinical observations, a significant reduction of MARCKS protein in both soluble and

membrane fractions of HN33 cells was not observed following acute (1 d) VPA

administration, except at levels above the maximal target therapeutic serum concentration
(1 mM). At 0.6 mM VPA, the down-regulation of MARCKS was greatest between d 0

and d 3, and little further reduction was observed upon continued exposure (5-10 d).

With regard to therapeutic concentrations, there have been reports of clinical response at
plasma concentrations of VPA as low as 45 g/ml (Bowden et al., 1994), which is

comparable to the observed effects of VPA on MARCKS down-regulation at

concentrations as low as 0.1 mM in this study. However, VPA is generally thought to be

most effective at therapeutic concentrations of 50-125 g/ml (0.3-0.75 mM), where the
most robust effect of VPA on MARCKS was observed in the HN33 cells. Following

VPA-induced MARCKS down-regulation and subsequent drug withdrawal, MARCKS

levels in these cells recovered to baseline within 3-7 d. This finding is consistent with the

third major parameter discussed with regard to clinical relevance, that of long-term effect

of VPA and delayed return of clinical response to baseline following treatment

discontinuation.

With regard to combined VPA/lithium exposure, a greater reduction of MARCKS
protein was observed in cells following exposure to the combination than was produced by

either VPA or lithium/carbachol alone, and the effects of the two drugs appear to be

approximately additive. Whereas VPA induced a mean reduction in MARCKS expression

of 26% in the soluble fraction and 50% in the membrane, and lithium/carbachol elicited
mean reductions of 14% and 33% in soluble and membrane, respectively, the combination

of VPA plus lithium/carbachol resulted in a 51% drop in soluble MARCKS and a 67%

drop in membrane-associated MARCKS. These data are in agreement with clinical studies

which show VPA/lithium combination therapy to have greater therapeutic efficacy for









treatment of bipolar disorder than either agent administered individually (Calabrese and
Delucchi, 1989; Hayes, 1989). In contrast to lithium, which exerts a preferential down-
regulation of MARCKS in the soluble fraction of HN33 cells (Watson and Lenox, 1996),

the effect of VPA on MARCKS was most pronounced in the membrane fraction of these

cells. While the significance of this finding is unclear, the differences observed in the

patterns of MARCKS down-regulation between these two drugs may be a function of

differences in the pathways of PKC activation, as later discussed. This is of particular
interest in light of the findings that concomitant exposure of cells to both VPA and lithium

appears to have an additive effect on MARCKS down-regulation at therapeutic

concentrations of both drugs, and such findings support the notion that the two drugs

operate through separate mechanisms.

In contrast to lithium, which appears to rely on relative depletion of myo-inositol

and receptor activation of PI signaling to optimize its effect on MARCKS in the

immortalized hippocampal cells, these manipulations had no effect on the VPA-induced

down-regulation of MARCKS in these cells. Addition of carbachol or inositol (even at

mM concentrations) to the culture medium had no effect on the VPA-induced reduction in
MARCKS, and these findings are consistent with the lack of effect of VPA on receptor-

coupled accumulation of CMP-PA in the CHO-KI cells (Watson el al., 1998). CMP-PA,

also known as CDP-DAG, is a byproduct of DAG metabolism and precursor for PI
reformation, and serves as a sensitive measure of cycling through the PI signaling cascade

(refer to Figure 1.2). Thus, in spite of the shared property of lithium and VPA in down-

regulating the expression of MARCKS, these data suggest that the mechanism through

which this occurs may involve different pathways. Nonetheless, there appears to be

evidence for a role of PKC regulation in the action of both lithium and VPA, for previous

studies have demonstrated that phorbol esters down-regulate MARCKS protein in
neuronally derived cell populations in a PKC-dependent manner (Watson el al., 1994).

This will be further discussed in later chapters.









Two structural analogs of VPA (see Figure 3.9c), which are significantly less

potent as anti-convulsants in comparison to VPA (Chapman et al., 1982), were studied.
Both analogs are known metabolites of VPA, although they are produced through minor

metabolic pathways and represent a minimal percentage of total brain plasma

concentration (Chapman ei al., 1982; Eadie, 1991 ). In previous studies, neither
hydroxyvalproic acid nor 2-propylglutaric acid was shown to be effective in preventing

pentylenetetrazol-induced seizures in mice (Chapman et al., 1982; J. Sullivan, personal

communication), and neither agent has been tested clinically for efficacy in the treatment

of bipolar disorder. In our studies, hydroxyvalproic acid produced a statistically

significant effect in down-regulating membrane-associated MARCKS, but at a

concentration well above that observed for VPA (3 .0 mM hydroxyvalproic acid vs. 0.1-

1.0 mM VPA). Comparatively, the 2-propylglutaric acid analog possessed little potency

in down-regulating MARCKS in either the membrane or cytosolic fraction. Future

analysis of structural differences among these VPA compounds may prove useful in

identifying structurally related fatty acids that may be more or less efficacious not only in

down-regulating MARCKS, but also potentially as mood stabilizers.
Exposure of immortalized hippocampal cells to structurally and functionally

diverse psychotropic agents, including carbamazepine, fluoxetine, imipramine, phenytoin,

diazepam, haloperidol, morphine, and verapamil, did not result in a down-regulation of

MARCKS, even at doses that were well above the therapeutic range. It is of particular

note that carbamazepine, another anti-convulsant used less frequently in the treatment of
manic patients, and verapamil, also used unconventionally and perhaps less efficacious

than the more established treatments (Janicak el al., 1998; Walton el al., 1996; Mathis et

al., 1988; Barton and Gitlin, 1987), did not have any effect on MARCKS expression.

Carbamazepine has been shown to have effects within the adenylyl cyclase cascade, a
property shared by lithium at higher concentrations (Manji el al., 1995; Mork et al.,
1992). However, carbamazepine has generally been shown to be less effective overall than









lithium or VPA as a mood stabilizer (Keck el al., 1992; Bowden, 1996), and verapamil, a

calcium channel blocker, has yet to be studied extensively for this application.
Following exposure to a therapeutic serum concentration (0.6 mM) of VPA for up

to 14 d, HN33 cells exhibited decreased MARCKS mRNA levels, along a time course

similar to that of observed MARCKS protein alterations. This finding suggests that the
down-regulation of MARCKS protein might be attributed, at least in part, to a reduction

of its precursor mRNA template. With regard to MARCKS regulation, such evidence
may support more than one hypothesis. The observed reduction in MARCKS protein may

be the result of either altered transcriptional rates, or post-transcriptional regulation, via

decreased mRNA stability, decreased translation, or increased degradation. Yet, the

observed alterations in mRNA expression support the notion that MARCKS is being

regulated at the level of mRNA synthesis or stability. Recent studies in our lab have

shown that lithium, while altering both MARCKS mRNA and protein levels, did not affect

mRNA stability (Watson and Lenox, 1997); this suggests that the lithium-induced

alteration in MARCKS mRNA is primarily attributable to effects on transcription of the

MARCKS gene.

These studies suggest that MARCKS may serve as a target in the brain for the

selective action of mood stabilizing drugs. Although the therapeutic action of these drugs

can probably not be attributed fully to their effects on MARCKS, it is plausible that the

role of MARCKS in modulating plasticity may account for some of the long-term

stabilizing actions exhibited by VPA and lithium in the treatment of bipolar disorder. This
notion is supported by the finding that other drugs were set apart by both their inability to

down-regulate MARCKS as well as their inefficacy in treating bipolar disorder. Further

support for the notion of selective action in the CNS has been provided by data from our

laboratory which shows that the expression of MARCKS protein in two separate murine

macrophage cell lines (IC-21 and MH-S) is unaltered following chronic exposure to

therapeutic levels of VPA (3-7 d at 0.6 mM, data not shown). In addition to effects of









VPA on MARCKS, such alterations in neuronal plasticity may result from actions on

other potential protein substrates, as discussed below.


GAP-43


It is clear from these studies that VPA has a time- and concentration-dependent

effect on GAP-43 expression in HN33 cells. Further, the concentration range over which

this increase in GAP-43 occurred was comparable to the concentration range over which

the down-regulation in MARCKS protein occurred, and this concentration range is well

within the target therapeutic serum concentration. There are several potential implications

of such an increase in GAP-43 expression, since GAP-43 is an early marker of post-

mitotic neurons, and is highly expressed during axonal growth and synapse formation

(Aigner et al., 1995). For example, increased GAP-43 expression might signify synaptic

regrowth and/or remodeling, cytoprotection, or even development of aberrant pathways.

Synaptic remodeling was discussed previously as a potential mechanism for long-term

mood stabilization. A novel finding observed in carrying out these studies was that

following chronic (>1 d) VPA exposure, HN33 cells appeared to become more adherent

to the tissue culture flasks in which they were grown. This is of interest in light of the

finding that chronic VPA exposure also elicited an increase in expression of GAP-43,

which is known to be important for promoting and maintaining growth cone adhesion

during neurite outgrowth (Aigner and Caroni, 1995). Later chapters will include further

studies on the growth and morphological changes induced by VPA.

As previously mentioned, preliminary studies have shown no evidence for lithium-

induced regulation of GAP-43 expression in HN33 cells, and this major difference
between VPA and lithium is intriguing. Since the two drugs share similar effects on

MARCKS, one might hypothesize that this effect provides a basis for their common ability

to treat bipolar disorder. Conversely, the difference in their ability to affect GAP-43 might









relate either to their differing efficacies in treating various types of bipolar disorders (i.e.
VPA is generally more efficacious in treating rapid-cycling bipolar disorder), or,

alternatively, to their very different pharmacological profiles aside from their mood-

stabilizing efficacy (i.e., VPA is also an anti-epileptic, whereas lithium is not).


Conclusions


With regard to alterations in protein expression and their necessity and/or
sufficiency in bringing about the desired end result (that of attaining successful mood-

stabilizing treatment), GAP-43 induction may or may not be either. There is little

evidence to suggest that increased GAP-43 is necessary for the successful amelioration of

bipolar symptoms, for although VPA elicited GAP-43 induction, lithium, the prototypic

mood-stabilizing drug, had no effect on GAP-43 expression in our in vitro cell model.

Nonetheless, this does not preclude the possibility that lithium may exert such an effect on

GAP-43 in other cell model systems or in brain tissue. Whether GAP-43 alterations alone
are sufficient for optimal bipolar therapy is another question entirely, which these data

alone cannot fully address. Many agents may lead to an increase in expression of GAP-43,

if by virtue of nothing other than altering cell growth and differentiation rates, but it does

not necessarily follow, based on what is yet known, that all of these agents would serve as

sufficient mood-stabilizing drugs.

Down-regulation of MARCKS protein may be more consistent with a mood-

stabilization mechanism. The observation that both VPA and lithium, to the exclusion of a

number of drugs used for various other psychotropic applications, elicited a significant

reduction in MARCKS, suggests that alterations in MARCKS may be involved in the
mechanism of action of these agents, and supports the identification of MARCKS as a

potential molecular target which may or may not contribute to the therapeutic effects of

drugs used in the treatment of bipolar disorder. However, the findings that









carbamazepine, another widely used mood-stabilizing drug therapy (although not FDA-

approved for this application), as well as some less commonly used mood-stabilizing

agents, including lamotrigine (data not shown) and verapamil, did not alter MARCKS

expression, point to the possibility that either MARCKS is not integral to mood-stabilizing

efficacy, or that the in vitro exposure conditions and model we employed were not optimal

to allow the down-regulation we may have observed following chronic exposure in vivo.

Another important question concerning the effects observed is that of sufficiency. The

identification of MARCKS as a potential target for the action of drugs used in the

treatment of bipolar disorder suggests that the down-regulation of MARCKS alone, by

any method, would be enough to yield successful amelioration of bipolar symptoms.

However, this diminishes the potential role of other mechanisms, such as the alterations in

PI signaling induced by lithium, or the effects of VPA on GAP-43 expression. It is

conceivable that the alterations in MARCKS expression observed in vitro would lead, in

vivo, to further downstream events important for altering plasticity or other parameters

involved in changing the course of the disorder. In other words, MARCKS may be just

one target of many which may ultimately result in the efficacy of this group of drugs, those

which exhibit anti-manic and/or mood-stabilizing action. This is more compatible with the

notion that other drugs may be able to positively affect the course of bipolar therapy.

Whereas VPA and lithium may operate, at least in part, through their effects on

MARCKS, the involvement of MARCKS in bipolar therapy is indirect, and other drugs

could theoretically elicit the same end result, yet through a separate pathway, perhaps

bypassing MARCKS altogether.

It is interesting that VPA led to opposite effects on two CNS proteins which are

both involved in cellular plasticity and intracellular signaling, and both regulated by PKC.
This is despite the fact that the VPA-induced alterations of these proteins occurred over a
very similar time course and dose range. These data support recent in vivo findings in our

laboratory which imply an apparent inverse regulation of MARCKS and GAP-43 in






61

certain highly plastic brain regions, such as the hippocampus, in that levels of GAP-43 are

high in some regions while MARCKS levels are low, and vice versa (McNamara and

Lenox, 1997). This would suggest that at certain developmental stages or in particular

brain regions, MARCKS and GAP-43 play alternative or opposing roles, even though they

can be regulated similarly. Such comparable pharmacodynamic properties suggest that

both proteins play important roles in the long-term therapeutic action of VPA.














CHAPTER 4
ROLE OF PKC IN MECHANISM OF ACTION OF VPA


Introduction


PKC is an 80-kD protein expressed at high concentrations in neuronal tissues, and

the observation that PKC is the major enzyme responsible for regulation of both GAP-43

and MARCKS make the effects of VPA on PKC an interesting area of investigation. PKC

is a serine/threonine kinase which, by virtue of its effects on intracellular signal

transduction, is thought to play a role in hormone and neurotransmitter secretion and

regulation of cell proliferation and differentiation, through mechanisms which are still not
fully understood (Fujise el al., 1994, Hofmann, 1997). PKC is actually comprised of a

large family of at least 11 mammalian isoforms, classified into three groups based on their

activation profiles. The conventional (cPKC) group includes c, 031/1311, and y, which are
regulated by Ca+, diacylglycerols (DAG), negatively charged phospholipids, and phorbol

esters (potent tumor promoters which mimic DAG). The novel (nPKC) group contains 6,

&, i" (L), and 0, and is characterized by activation requirements similar to those of the

cPKC's, but lack of Ca+' activation. The atypical (aPKC) members are C, t (k), and 4t(D),

and are all independent of Ca++ and DAG or phorbols, requiring only phospholipids for
activation (Fujise el al., 1994, Hofmann, 1997). The PKC isozymes differ in their primary

structure and genomic origin, so that they are not derived from a single mRNA, but rather
from a series of structurally related mRNAs coding for a family of enzymes with very
similar properties. In humans, the genes for the various isozymes are located on different









chromosomes, with the exception of the p1I and p11 isoforms, which are the result of

differential splicing (Stabel and Parker, 1991).
The primary structure of PKC contains four major regions, termed C1 through C4,

which exhibit a high degree of homology between isotypes. These conserved regions are

generally thought to represent areas involved with functions common to all PKC

isozymes, such as the kinase domain. Interspersed between the conserved regions are
variable regions VI through V5, which exhibit a lower degree of homology and probably

allow for properties specific to individual isoforms. The higher level structure of PKC is

comprised of catalytic and regulatory domains. The catalytic domain alone is referred to

as PKM, and has a molecular weight of 45-55 kDa. It is located in the C-terminal region

of the polypeptide, and confers its kinase activity. The regulatory region, located at the

N-terminal and with an approximate molecular weight of 32-36 kDa, contains

phospholipid-and phorbol-binding domains responsible for modulating alterations in the

activity of the enzyme. It is at the regulatory region that diacylglycerol (DAG) and

phorbol esters bind to transiently activate the enzyme, eventually resulting in an increased

rate of degradation (down-regulation). All members of the PKC superfamily carry in the

N-terminal region at least one zinc-finger structure (more commonly two), which is

thought to aid in protein-protein interactions. The N-terminal region also contains a

pseudosubstrate sequence, conferring the enzyme's autoinhibitory activity. This region

serves to regulate activity of the enzyme, and removal of this region results in constitutive

and activator-independent kinase activity. The pseudosubstrate domain resembles the

consensus phosphorylation sequence but contains no phosphorylatable residue; it controls

kinase activity by occupying the catalytic site and thereby blocking access of the substrate

(Stabel and Parker, 1991).

cPKC activation is dependent upon two major events: (1) formation of the ternary

enzyme+Ca2++phospholipid complex (the presence of the phospholipid further enhancing

the affinity of the enzyme for calcium), and (2) binding of DAG (or phorbol), leading to









the conformational changes which result in activation. Down-regulation of the enzyme

follows prolonged activation, and is accomplished through both increased proteolysis and

sequestration (association of the enzyme with cytoskeletal or nuclear components).

Autophosphorylation of the enzyme is not a requirement, but probably plays a role (Stabel

and Parker, 1991).
Data from our laboratory indicate that PKC activity is decreased over time

following chronic lithium exposure, as is expression of PKC isozymes a and e (Watson

and Lenox, 1997). These findings are in agreement with those reported by Manji and

colleagues, which reflected reductions in PKC activity and expression of a and s isoforms

following either lithium or VPA exposure, in either rats or C6 glioma cells (Manji et al.,

1993; Chen et al., 1994; Manji et al., 1996). Tissue distribution of PKC isozyme

expression varies, and the relative expression of each isoform may contribute to its unique
role in signal transduction. Individual PKC isozymes, upon activation, are differentially

compartmentalized, suggesting that they mediate distinct cellular functions. PKC-y
appears to be the only isoform unique to brain and spinal cord, with no expression

elsewhere; its precise role remains unknown. The same is true for most of the other

isoforms although there has been no dearth of studies into roles of the various PKC

isozymes, results have been inconclusive and in some cases contradictory (to be discussed

further in Discussion). Whereas PKC-ox and F have been implicated in tumor growth and

suppression of apoptosis, 8 may induce opposite effects (Hofmann, 1997). Both PKC-cL

and -5 are thought to be important in early brain development (Blackshear et al., 1996),
while PKC-C may play a role in proliferation, and several isoforms have reportedly been

implicated in neurite outgrowth and/or cell differentiation (Ways el al., 1994; Hundle et

al., 1995; Borgatti et al., 1996). We have previously established that the immortalized

hippocampal cell line HN33 expresses PKC isoforms a, 6, F, and C, and our laboratory has

demonstrated in vitro effects of lithium on PKC activity as well as on expression levels of
various PKC isozymes (Watson and Lenox, 1997). In light of previous reports









demonstrating VPA-induced alterations in PKC in a non-neuronal cell line, it will be of

interest to investigate the effects of VPA exposure on these parameters in HN33 cells,

especially given preliminary evidence from our laboratory for effects of VPA on

MARCKS and GAP-43 in these cells. It has been suggested that the colocalization of

some isoforms with substrate(s) may ensure preferential and rapid phosphorylation

following PKC activation. Based on evidence indicating a down-regulation of MARCKS

and increased expression of GAP-43 following chronic VPA exposure, it is predicted that

VPA will activate PKC, at least transiently, so that the downstream effects on MARCKS

and GAP-43 can occur as observed. If VPA does, indeed, activate PKC, both its activity

level and isoform expression should decrease over time, in the characteristic pattern of

down-regulation following long-term activation. This acute activation followed by

chronic down-regulation has been demonstrated previously in the HN33 cells with lithium,

phorbol esters, and retinoic acid.

These and other studies suggest that VPA has significant effects on both PKC and

its substrates, though the connection between such effects is unclear. Although one would

expect PKC alterations to be in large part responsible for any further downstream effects

on its substrates MARCKS and GAP-43, this direct cause-and-effect relationship has yet

to be established. The following studies will therefore help to establish whether or not

PKC is necessary to the downstream effects on MARCKS and GAP-43, and which

isoforms may be most important for each effect. After first establishing the ability of VPA

to effect alterations in both PKC activity and expression of various PKC isozymes in

HN33 cells, the next objective will be to address the role of PKC in altering MARCKS

and GAP-43 levels. VPA-induced alterations in PKC-specific activity will be measured

indirectly by phosphorylation of the substrate MBP, and associated alterations in PKC

isoform expression will be measured using Western immunoblotting techniques. PKC

inhibitors will then be used to assess whether PKC activity is necessary to elicit the

observed VPA-induced alterations in MARCKS and GAP-43, and isoform-specific






66


antisense oligonucleotides will be used to narrow down which isoforms are important for

each parameter. It is hypothesized that PKC activity will be necessary to any observed

effects by VPA on MARCKS or GAP-43, and that inhibition of the enzyme will block
such observed effects.









Results


Effect of VPA Exposure on PKC Activity in HN33 Cells

HN33 cells exposed to 0.6 mM VPA for 15 min 3 d were assayed for effects on

PKC activity, as measured by MBP phosphorylation (Figure 4.1). Acute VPA exposure

(15 min 4 h) resulted in a tendency toward increased membrane-associated activity and a

correlated decrease in PKC activity in the cytosolic fraction. This paradigm most closely

follows a translocation phenomenon, whereby cytosolic PKC is shuttled into the

membrane for activation. By the 3 d time point, activity in the membrane fractions is

down-regulated (p<0.05). Similar results were observed using MARCKS peptide as the

phosphorylation substrate (data not shown).

160
I 175
A 140 150 T T
1- 125
120 1(X, l

100 50
0 05 1 1.5 2 25 3 35 4 45


60-
E
4-"--- Soluble
40-
--- Membrane
20-
0 ... ... .. .. .. .. .
g 0 4 8 12 16 20 24 28 32 3640 44 48 52 5660 64 68 72 76
Time (hours)

Figure 4.1. Effect of VPA exposure on PKC activity in HN33 cells. VPA exposure leads
to apparent PKC activation, followed by down-regulation. HN33 cells exposed to 0.6
mM VPA were processed and soluble and membrane fractions isolated as detailed in
Materials and Methods (Chapter 2). Assay was performed utilizing the PKC activators
PMA (phorbol), CaCl2, and phosphatidylserine. PKC activation was measured by the y-
32p phosphorylation (in pmol/min/mg) of the PKC substrate MBP. Mean activity of
control samples was 102 8.1 pmol/min/mg. The figure shows the overall pattern of
PKC activity from 15 min (acute) to 3 d (chronic) VPA exposure; the inset shows PKC
activation only at acute time points (up to 4 h). Results are mean S.E.M. of at least four
determinations, and are expressed as percent of activity in untreated control samples.
*p<0.05









Effect of VPA on PKC-a

PKC-t is observed entirely in the soluble fraction of HN33 cells. As shown in

Figure 4.2, exposure to 0.6 mM VPA results in an acute (15 min 12 h) increase in PKC-

a expression, followed by down-regulation over time (p<0.05).


a ONO me" AMb 4W#-00 m

1 2 3 4 5 6

160 ,


b 2m
1 2


TIME (hours)
Figure 4.2. Effect of VPA on PKC-ct expression in soluble fraction of HN33 cells. (a)
Representative Western immunoblot of soluble PKC-ax expression following acute (24 h
or less) VPA exposure @ 0.6 mM. Protein samples were loaded into lanes as follows:
Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h VPA; Lane 4, 4 h VPA; Lane
5, 12 h VPA; Lane 6, 24 h VPA. (b) Representative Western immunoblot of cytosolic
PKC-(x expression following chronic (3 d) VPA exposure @ 0.6 mM. Lane 1 was loaded
with protein from an untreated control, and Lane 2 was loaded with protein from a sample
exposed to VPA for 3 d. (c) HN33 cells were exposed to 0.6 mM VPA for 15 min 3 d,
and PKC-at levels (present only in the cytosolic fraction of these cells) were assessed by
Western immunoblot analysis. Results are expressed as percent of untreated control, and
are the mean S.E.M. of four determinations. *p<0.05


Effect of VPA on PKC-6

PKC-6 undergoes a significant increase in expression levels in both soluble

(p<0.001) and membrane (p<0.0001) fractions of cells during chronic VPA exposure (1 &

3 d), and as late as 3 d, there appears to be no down-regulation (Figure 4.3).















7 8 9 10


1 2 1 2


TIME (hours)


Figure 4.3. Effect of VPA on PKC-6 expression in HN33 cells. (a) Representative
Western blot of PKC-6 expression following acute (<24 h) VPA exposure @ 0.6 mM.
The blot at left (Lanes 1-6) shows cytosolic PKC in HN33 cells, while the blot at right
(Lanes 7-12) shows membrane-associated PKC. Protein samples were loaded into
lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h VPA;
Lane 4, 4 h VPA; Lane 5, 12 h VPA; Lane 6, 24 h VPA; Lane 7, untreated control;
Lane 8, 15 min VPA; Lane 9, 1 h VPA; Lane 10, 2 h VPA, Lane 11, 4 h VPA; Lane
12, 12 h VPA. (b) Representative Western of PKC-8 expression following chronic (3
d) VPA exposure @ 0.6 mM. The blot at left shows cytosolic PKC in HN33 cells,
while the blot at right shows membrane-associated PKC. Lane 1 was loaded with
protein from an untreated control, and Lane 2 was loaded with protein from a sample
exposed to VPA for 3 d. (c) HN33 cells were exposed to 0.6 mM VPA for 15 min 3
d, and PKC-6 levels were assessed by Western immunoblot analysis. Results are
expressed as percent of untreated control, and are the mean S.E.M. of four
determinations. *p<0.001, soluble and *p<0.0001, membrane.


54


11 12


a
2 3










a di
1 2 3 4 5 6 1 2 3 4 5 6




b i UFII
I 2 I


TIME (hours)


Figure 4.4. Effect of VPA on PKC-e expression in HN33 cells. (a) Representative
Western immunoblot of PKC-6 expression following acute (24 h or less) VPA
exposure @ 0.6 mM. The six lanes at left show cytosolic (or soluble) PKC in HN33
cells, while the six lanes at right show membrane-associated PKC. Protein samples
were loaded into lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA;
Lane 3, 1 h VPA, Lane 4, 2 h VPA; Lane 5, 4 h VPA; Lane 6, 12 h VPA. (b)
Representative Western immunoblot of PKC-e expression following chronic (3 d)
VPA exposure @ 0.6 mM. The blot at left shows cytosolic (or soluble) PKC in
HN33 cells, while the blot at right shows membrane-associated PKC. Lane I was
loaded with protein from an untreated control, and Lane 2 was loaded with protein
from a sample exposed to VPA for 3 d. (c) Cells were exposed to 0.6 mM VPA for
15 min 3 d, and PKC-e levels were assessed by Western immunoblot analysis.
Results are expressed as percent of untreated control, and are the mean S.E.M. of
four determinations. *p<0.05









Effect of VPA on PKC-P
As shown in Figure 4.4, membrane-associated PKC-s increases acutely (15 min 4
h), and then returns to basal levels by the 3 d time point. Cytosolic PKC-c expression
remains close to baseline acutely, and then increases significantly following prolonged (1-3

d) VPA exposure (p<0.05). This pattern of expression reflects an apparent activation and

translocation of PKC-g, whereby PKC is activated, shuttled from cytosol into the

membrane (where the enzyme is active), and then down-regulated and returned from

membrane to cytosol following prolonged activation.


Effect of VPA on PKC-C
PKC-C shows little change in expression at early time points, and then is up-

regulated sometime between I d and 3 (p
regulation observed in the soluble fraction (Figure 4.5).


Effect of PKC-a Antisense Oligonucleotides on PKC-z Expression in HN33 Cells

Antisense oligonucleotides were designed to bind to various segments of the
mouse and/or human PKC-a mRNA sequence, and block its translation to protein. Figure

4.6 and Table 4.1 list the antisense oligonucleotides and show the regions of mouse PKC-

a gene they were designed to bind. All oligonucleotides were synthesized with
phosphorothioate backbones, in order to increase their stability in culture. A series of

experiments was then conducted to down-regulate PKC-aX expression in HN33 cells, first

using a single antisense oligonucleotide sequence, then using the combination of five

antisense oligonucleotides. A sense oligonucleotide was used as the negative control for

these experiments.












*MAN* M
1 2 3 4 5 6


1 2 1


1 2 3 4 5 6


2


0 10 20 30 40 50 60 70 80


TIME (hours)



Figure 4.5. Effect of VPA on PKC-C expression in HN33 cells. (a) Representative
Western immunoblots of PKC-C expression following acute (24 h or less) VPA
exposure @ 0.6 mM. The blot at left shows cytosolic (or soluble) PKC in HN33 cells,
while the blot at right shows membrane-associated PKC. Protein samples were loaded
into lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h
VPA; Lane 4, 4 h VPA; Lane 5, 12 h VPA; Lane 6, 24 h VPA. (b) Representative
Western immunoblot of PKC-C expression following chronic (3 d) VPA exposure @
0.6 mM. The blot at left shows cytosolic (or soluble) PKC in HN33 cells, while the
blot at right shows membrane-associated PKC. Lane I was loaded with protein from
an untreated control, and Lane 2 was loaded with protein from a sample exposed to
VPA for 3 d. (c) HN33 cells were exposed to 0.6 mM VPA for 15 min 3 d, and
PKC- levels were assessed by Western immunoblot analysis. Results are expressed as
percent of untreated control, and are the mean S.E.M. of four determinations.
*p<0.0001, membrane










25-
24-


26- 27


-28


mouse PKC alpha gene


5 ta
sta rt


Figure 4.6. PKC-ci antisense oligo-nucleotide design. This figure depicts the
antisense oligonucleotides designed to bind to PKC-a mRNA and block its
translation. Shown is the representative PKC-a gene from mouse (as reported in
the NCBI database), and the five oligonucleotides (24, 25, 26, 27, 28) designed to
bind to the nucleotide sequence at the indicated, base-complementary regions.
Start codon is demarcated by the arrow; both oligonucleotides 24 and 25 span this
region. Not shown is #23, the sense oligonucleotide designed as a negative
control, which reads as the sense complement to antisense oligo 24. All
oligonucleotides are 20-base phosphorothioate linked, with the exception of #23
and #24, which are 15-base phosphorothioates, and all were cross-referenced
through GenBank to ensure unique sequence identity with PKC-a.




Table 4.1 lists the 5' to 3'nucleotide sequences of each PKC-a oligonucleotide
(ON) by number, and the species specificity of each.


Table 4.1. Antisense oligonucleotide sequences.

ON # Sequence of Nucleotides Species Specificity

23 ACCATGGCTGACGTT mouse & human

24 AACGTCAGCCATGGT mouse & human

25 CAGCCATGGTTCCCCCCAAC mouse

26 AGAAGGTAGGGCTTCCGTAT mouse & human

27 GTTACCAACAGGACCTAGCT mouse

28 TGAGTTCAGCATGACATTGT mouse









Initial experiments attempted to establish the optimal dose range for

oligonucleotide exposure and protein down-regulation. Oligonucleotide concentrations
ranging from 0.1 to 1.0 p.M (the standard concentrations reported in similar studies) had

no effect on cell viability or PKC-ct expression following 24 h exposure (data not shown).

Increasing oligonucleotide concentration to 10 VM and lengthening exposure time to 48 h

had no effect on PKC-ct expression (data not shown).

Next, methods were employed to increase oligonucleotide uptake and compensate

for potential oligonucleotide degradation. In keeping with previous reports of successful

antisense-mediated protein down-regulation, the cationic carrier lipid LipofectAINE

(Gibco BRL, Gaithersburg, MD) was used at 5 ptg/ml and 15 ptg/ml, and cells were serum-

deprived during oligonucleotide exposure. In addition, medium and its constituents were
replaced fresh every 12 h during exposure. PKC-ct expression remained at basal levels

following 24 or 48 h exposure to oligonucleotide under these conditions (data not shown).

The final approach utilized phorbol to down-regulate PKC expression (non-

selectively), then PKC-a antisense oligonucleotides were added following phorbol

removal, in an effort to selectively block PKC-a recovery. A representative Western is

shown in Figure 4.7. HN33 cells were exposed to 1.0 .M phorbol 12,13-dibutyrate
(PDBu) for 24 h, after which cells were washed with media to remove drug. Next, cells

were exposed to 1.0 p.M sense or antisense oligonucleotides (total of five antisense

oligonucleotides) directed to PKC-ot (Figure 4.6). Cells were exposed to oligonucleotides

for a total of 24 h, but antisense-treated cells showed no difference in PKC-a expression

than sense-treated or untreated control cells. As shown in Figure 4.7, there is clearly no

difference in PKC-ox expression between those PDBu-pre-treated cells which are

antisense-treated or untreated. PDBu exposure resulted in a reversible down-regulation of
PKC-a (Lane 2 vs. 1), and PKC-ct antisense failed to prevent the recovery of PKC-a

expression at rates comparable to control levels (Lanes 7&8 vs. 3-6). Whereas toxicity
commonly arises as a problem in antisense applications in cell and tissue culture, the HN33









cells showed no evidence of cytotoxicity following 24-48 h exposure to any of the
pharmacological agents at any of the concentrations tested.






1 2 3 4 5 6 7 8

Figure 4.7. Effect of PKC-ct antisense oligonucleotides on PKC-a expression in HN33
cells. Shown is a representative Western immunoblot of PKC-X expression in HN33 cells
following antisense treatment. Lane 1 holds an untreated control; Lane 2 shows PKC-ct
expression in cells treated with 1.0 pWM PDBu for 24 h; Lane 3 shows cells treated with
1.0 pM PDBu for 24 h, then no treatment for 24 h; Lane 4 shows cells treated with 1.0
jiM PDBu for 24 h, then LipofectAMINE for 24 h; Lane 5 shows cells treated with 1.0
4M PDBu for 24 h, then sense oligo #23 for 24 h; Lane 6 shows cells treated with 1.0 PM
PDBu for 24 h, then sense oligo #23 and LipofectAMINE for 24 h; Lane 7 shows cells
treated with 1.0 pM PDBu for 24 h, then antisense oligonucleotides #24, 25, 26, 27, and
28, each at 1.0 p.M, for 24 h; Lane 8 shows cells treated with 1.0 JIM PDBu for 24 h, then
antisense oligonucleotides #24, 25, 26, 27, and 28, plus LipofectAMINE, for 24 h.


Effect of PKC Inhibitor LY333531 on VPA-Induced Down-Regulation of MARCKS

LY333531 alone had no effect on MARCKS expression, but when combined with
VPA, LY333531 prevented the VPA-induced MARCKS down-regulation observed in the

membrane, but not soluble, fraction of HN33 cells (Figure 4.8). MARCKS expression in

both cytosol and membrane of VPA-treated cells was significantly different from that of

LY33353 1-treated cells (p<0.005, soluble; p<0.0001, membrane). Cytosolic MARCKS in

combined VPA/LY3 33531-treated cells was significantly different from that of

LY33353 1-treated cells (p<0.005), indicating that, even in the presence of the PKC

inhibitor LY333531, VPA was able to induce down-regulation of cytosolic (but not

membrane-associated) MARCKS expression. In contrast, LY333531 prevented the VPA-

induced down-regulation of MARCKS in the membrane fraction of HN33 cells.










120
110-
100-
90-

70
60
50-
40-
30-
20-
10-
0-


[] Soluble


0 Membrane


Control VPA LY333531 VPA + LY333531
Treatment


2 3 4 5 6 7 8


Figure 4.8. Effect of PKC inhibitor LY333531 on VPA-induced down-regulation
of MARCKS. (a) HN33 cells were exposed for 24 h to 1.0 mM VPA with or
without 1.0 W LY33353 1. MARCKS expression was quantitated by Western
blot analysis, and results (expressed as percent of untreated control) reflect the
mean S.E.M. of at least five determinations. *p<0.005, soluble; *p<0.0001,
membrane (b) Representative Western blot showing MARCKS expression in the
presence of VPA LY33353 1. Lanes 1-4 represent cytosolic MARCKS, and
were treated for 24 h as follows: control, untreated (1); 1.0 mM VPA (2); 1.0 JIM
LY333531 (3); 1.0 mM VPA + 1.0 p.M LY333531 (4). Lanes 5-8 show
membrane-associated MARCKS, and were treated for 24 h as described for 1-4
(respectively) above.

Effect of PKC Inhibitor LY333531 on VPA-Induced GAP-43 Induction

LY33353 1, alone or in combination with VPA, had no effect on GAP-43

expression in HN33 cells (Figure 4.9). GAP-43 levels in both soluble and membrane

fractions are significantly increased (p<0.05) in both VPA-treated groups, compared to







77


controls (untreated or LY33353 1-treated). Therefore, the PKC inhibitor LY333531 was

unable to prevent the VPA-induced GAP-43 increase, for even in the presence of

LY33353 1, VPA increased GAP-43 expression in both soluble and membrane fractions.

These data indicate that PKC inhibition alone is insufficient to prevent the GAP-43

increase observed in HN33 cells following 24 h 1.0 mM VPA exposure.


*


Control VPA


5] Soluble

0 Membrane


Treatment


OM Aw -"0Im


1 2 3 4 5 6 7 8




Figure 4.9. Effect of PKC inhibitor LY333531 on VPA-induced GAP-43 increase. (a)
HN33 cells were exposed for 24 h to 1.0 mM VPA with or without 1.0 M
LY33353 1. GAP-43 expression was quantitated by Western blot analysis, and results
(expressed as percent of untreated control) reflect the mean S.E.M. of three
determinations. *p<0.05 (b) Representative Western blot showing GAP-43 expression
in the presence of VPA LY33353 1. Lanes 1-4 represent membrane-associated GAP-
43, and were treated for 24 h as follows: control, untreated (1); 1.0 mM VPA (2); 1.0
p.M LY333531 (3); 1.0 mM VPA + 1.0 glM LY333531 (4). Lanes 5-8 show cytosolic
GAP-43, and were treated for 24 h as described for 1-4 (respectively) above.


*


240
220 -
200 -
180 -
160-
140 -
120 -
100 -
80-
60-
40-
20-
0-


LY333531 VPA + LY333531









Discussion


Inasmuch as PKC activation has been associated with regulation of expression of

both GAP-43 and MARCKS, the effects of VPA on PKC isozyme levels and PKC activity
were investigated more directly, in an effort to better assess the role of PKC in the

mechanism of action of VPA in the brain. Whereas lithium, the first-line anti-manic/mood-

stabilizer, is believed to operate in part through the PI signaling pathway, our group and

others have provided evidence against such a mechanism for VPA (Vadnal and
Parthasarathy, 1995; Lenox et al., 1996; Dixon and Hokin, 1997; Watson et al, 1998),

although both drugs have similar effects on MARCKS expression.

Our findings indicate that, like lithium, VPA induces alterations in expression of

PKC isoforms cc and (Watson and Lenox, 1997). However, the time course for VPA-

induced changes is earlier than that for lithium, and PKC-s expression actually increases

(acutely in membrane, chronically in cytosol), in contrast to the reduction previously
reported for both lithium and VPA (Chen et al., 1994; Manji et al., 1996; Watson and

Lenox, 1997). Chen et al. (1994) reported that chronic VPA exposure of C6 glioma cells

reduced the expression of PKC isozymes a and & in intact cells, a finding similar to that

observed following chronic lithium administration in the same cell model (Manji et al.,

1993). Subsequent studies revealed PKC-induced alterations in multiple components of
the 13-adrenergic receptor-coupled cAMP-generating system, including J31-AR, Gas, and

adenylyl cyclase, each of which is phosphorylated by PKC (Chen et al., 1996). The

significance of these changes is unclear, but together they suggest a major role for PKC in

the VPA- or lithium-induced molecular effects underlying their efficacy, and may be
instructive when other PKC-related effects are compared between the two drugs.

In this study, VPA exposure resulted in an apparent transient activation of PKC

(within 2 h), followed by a gradual reduction in activity with chronic exposure (4 h and

beyond). PKC activity was significantly down-regulated by the 72 h time point. Similar









results were reported following exposure of HN33 cells to the phorbol PMA, though

activation occurred within 15 minutes and down-regulation of PKC activity was evident

by I h (Watson et al., 1994). This finding is further in agreement with previous reports

which showed that chronic VPA or lithium exposure of C6 glioma cells significantly
reduced PKC activity in both membrane and soluble fractions (Chen et al., 1994; Manji et

al., 1996). The observed patterns of alterations in protein activity and expression suggest

that isoforms ox and F might, under these conditions, be involved in the phosphorylation

and regulation of MARCKS protein expression, since both VPA and lithium elicited

alterations in PKC-a and -e expression in HN33 cells, and the time course of MARCKS

alterations roughly corresponds to that of the changes observed in PKC isoform

expression. Indeed, previous studies have indicated the ability of these two isoforms to
phosphorylate MARCKS, whereas PKC-C is unable to phosphorylate MARCKS in vitro

(Fujise et al., 1994; Herget et al., 1995; Uberall et al., 1997). One could hypothesize a

process in which VPA, directly or indirectly, activates PKC, especially isoforms o and/or

&, at early time points (between 0 and 12 h). By 3 d, there is a significant down-regulation

of activity in the membrane and a comparable drop in the cytosol. Consistent with this
hypothesis are data which show an increase in PKC activity within 15 seconds following

phorbol exposure (Stabel and Parker, 1991), and a time course for MARCKS down-

regulation which closely follows the time course of PKC activation in HN33 cells exposed

to phorbol (Watson et al., 1994).

While VPA and lithium both affect PKC isozyme ax (and perhaps F), the two drugs
have very different effects on PKC-5. VPA induces a chronic up-regulation in PKC-6,

whereas lithium has little or no effect on PKC-5 expression (Watson and Lenox, 1997). In

comparing VPA and lithium, one might hypothesize that such differences in patterns of
PKC activation underlie the differences observed in downstream effects such as protein

expression or cell growth and morphology. For example, the up-regulation of PKC-5
resulting uniquely from VPA exposure is probably not important for the MARCKS down-









regulation observed, but might be involved in the GAP-43 induction observed following

VPA but not lithium exposure; in support of this hypothesis is evidence indicating that

PKC-8 and -& phosphorylate GAP-43 (Oehrlein et al., 1996). Indeed, in our study, the

GAP-43 increase occurred over a time course and of a magnitude similar to those of the

PKC-5 alterations observed. Conversely, PKC isozymes a and/or e may play a role in the

observed VPA- and lithium-induced down-regulation of MARCKS.

Reports are in disagreement as to the role of PKC in neuronal differentiation. A

number of studies have suggested that, rather than being a positive regulator of the

process, PKC may actually be an inhibitory influence, so that suppression of the enzyme

results in increased process outgrowth. For example, use of various PKC inhibitors, and

subsequent reductions in PKC activity, have been associated with increased neurite

outgrowth (Heikkila el al., 1989; Parodi et al., 1990; Tsuneishi, 1992; Wooten, 1992;

Jalava et al., 1993; Ekinci and Shea, 1997). Long-term down-regulation of PKC activity

with phorbols such as PMA has provided similar results (Tsuneishi, 1992; Carlson et al.,

1993; Jalava et al., 1993; Ekinci and Shea, 1997), with one study showing enhanced

neurite outgrowth following combined phorbol and PKC inhibitor treatment (Heikkila et

al., 1989). However, some of these same studies have further shown that down-

regulation of PKC activity alone is insufficient to elicit a differentiated phenotype (Ekinci

and Shea, 1997; Carlson et al., 1993; Wooten, 1992).

On the other hand, numerous reports have presented equally convincing evidence

for a positive role for PKC in regulating neuronal differentiation and a positive role in

maintaining the functionally active state (Coleman and Wooten, 1994; Cabell and

Audesirk, 1993; Parrow et al., 1992; Abraham et al., 1991; Tonini et al., 1991), though

this role is probably not unique to PKC. Previous data have implicated a number of

different PKC isoforms in cellular differentiation, including all four of the isotypes present

in HN33 cells: a, 5, F, and C (Wada et al., 1989; Leli et al., 1992, 1993; Ways et al.,

1994; Hundle et al., 1995; Borgatti et al., 1996). Indeed, the subcellular localization and









activation of PKC isozymes are regulated in an isoform-specific manner during

neurogenesis, suggesting they are involved in control of neural development and in
neuronal differentiation (Oehrlein et al., 1998).

The role of PKC in proliferation and differentiation remains unclear, for reports

differ as to which PKC isoform(s) is/are responsible for various aspects of cellular

differentiation. Some studies have attributed to PKC-at a role as the intermediate in

cellular differentiation, as a result of alterations in PKC-a mRNA and/or protein

expression observed concomitant with differentiation of PC 12 and other cell lines

(Borgatti et al., 1996; Parrow et al., 1995; Murray et al., 1993; Leli et al., 1993; Tonini et

al., 1991; Wada et al., 1989). However, these reports differ in their mechanistic

conclusions, for in three studies, PKC-a expression increased with differentiation

(Borgatti et al., 1996; Parrow et al., 1995; Murray et al., 1993), while in three other

studies, PKC-x and -, levels had decreased following comparable treatment and

neuritogenesis (Leli et al., 1993; Tonini el al., 1991; Wada et al., 1989). Leli's group

further showed that the intracellular delivery of PKC-a and -e isoform-specific antibodies

resulted in a morphologically differentiated phenotype in SH-SY5Y cells (Leli et al.,

1992).

While some groups observed no differentiation-associated alterations in isoforms

other than PKC-ox, a number of reports have indicated that in various cell lines, other

members of the PKC superfamily are involved. Borgatti's group, who showed that PKC-a

was up-regulated during PC 12 cell differentiation, found in the same study that PKC 6, s,

and were down-regulated (Borgatti et al., 1996). A number of studies have provided

evidence for a role of PKC-s in regulating neuronal differentiation, since neurite
outgrowth has been associated with altered PKC-6 levels (Ponzoni et al., 1993; Parrow et
al., 1995; Fagerstrom et al., 1996; Ekinci and Shea, 1997; Hundle et al., 1997; Zeidman et

al., 1999). In support of the notion that PKC-s may be involved in the process, as

suggested by Leli's and Wada's work (Leli et al., 1993; Wada et al., 1989), a separate









study showed that overexpression of PKC-s but not PKC-8 led to an increase in NGF-
induced neurite outgrowth in stably transfected PC 12 cell lines (Hundle et at., 1995). In

stark contrast to these findings, yet another group reported just the opposite in NIH 3T3

cells, in that PKC-8 overexpression induced a morphological change and reduced growth
rate consistent with differentiation that was not observed in cells overexpressing PKC-&

(Mischak et aL, 1993). Further, PKC-C has been similarly examined in transfected
leukemic cells, and results indicate that PKC-C overexpression is able to induce cellular

differentiation (Ways et al., 1994). Another study that utilized antisense techniques also

emphasized the importance of PKC-C in NGF-induced neuronal differentiation (Coleman

and Wooten, 1994).

All of these studies point to different PKC isozymes as mediators of similar effects.

These apparently contrasting results may merely be a function of the varying experimental

methods and conditions used from one study to the next, such as different cell lines,

exposure conditions, and other parameters. These findings may alternatively indicate that

(1) each isozyme plays its own, specialized role in the complex process of cellular
differentiation, and (2) alterations in the level of each isozyme may be compensated for by

alterations in the levels and actions of other isozymes. Further studies will need to be

conducted in order to answer these questions, to confirm more definitively the association

between each PKC isoform and the parameter in question, but these data strongly suggest

that both VPA and lithium may down-regulate MARCKS via a PKC-dependent

mechanism. On the other hand, the unique role of the PI signaling pathway in the action

of lithium, and the differential pattern of activation of the PKC isozymes by VPA and
lithium may indicate that there are multiple pathways to the down-regulation of MARCKS

expression, and may also confer different clinical therapeutic properties on these two
drugs. The findings further suggest that the action of VPA on MARCKS protein

expression could be cell and/or tissue specific, for it may be modulated, in part, by the

expression pattern of the PKC isozymes specific to the cell types. Indeed, Stabel and









Parker (1991) suggest that the distribution of MARCKS closely follows the distribution of
PKC, based on early studies of PKC activity showing wide disparity across regions (Kuo

et al., 1980; Minakuchi et al., 1981).
Having established an apparent causal relationship between VPA and PKC, the

purpose of the next experiments was to establish more precisely the necessity of PKC in
eliciting the obesrved VPA-induced effects on MARCKS and GAP-43. In order to better

establish the role of various PKC isoforms in the MARCKS down-regulation and GAP-43
increase observed, experiments were designed employing antisense technology. Through

this method, antisense oligonucleotides are designed and directed to specifically bind and

"knock down" the nucleic acid precursors of each individual PKC isozyme, resulting in

transient elimination of the protein in question. There are four main mechanisms by which

antisense oligonucleotides are hypothesized to work: (1) by inhibiting the necessary

processing and transport of RNA, (2) by interfering with binding of the ribosome to the

nucleotide sequence, so that translation is not initiated, (3) by inhibiting the completion of

translation after the process has begun, and (4) by binding to the sequence, thereby

characterizing it as abnormal and resulting in degradation by RNase H (Schlingensiepen et

al., 1997).
For this study, a number of trials were performed in an effort to down-regulate

PKC-a in the HN33 cells. One sense and five antisense oligonucleotides were prepared by

Gemini Biotech, Ltd. (The Woodlands, TX) through the DNA Synthesis Core of the UF

Interdisciplinary Center for Biotechnology Research (ICBR). Each oligonucleotide was

designed to bear unique identity to the PKC-ca genome of either mouse or human, as

reported in the National Center for Biotechnology Information database, and screened

through GenBank. Various regions of the cDNA sequence were targeted, including the

start codon (see Figure 4.6). In an effort to maximize stability and binding of

oligonucleotides, all were composed of 15 or 20 bases each, and had phosphorothioate
linkages in place of the endogenous phosphodiester backbone of DNA. Whereas toxicity









commonly arises as a problem in antisense applications in cell and tissue culture, the HN33

cells showed no evidence of cytotoxicity following I to 2 d exposure to I and 10 M
concentrations of PKC-ct oligonucleotides. In initial experiments, cells were exposed to

oligonucleotide(s) for 24-48 h, in an effort to establish a protocol for the consistent
reduction of PKC-a levels prior to VPA exposure. When numerous efforts failed to elicit

any reduction in PKC-ci expression, PKC was first down-regulated through 24 h, 1
PDBu (phorbol) administration prior to exposure of cell cultures to oligonucleotide.

Following thorough phorbol washout, test cells were exposed to all five oligonucleotides

concomitantly for 24 h. In order to increase uptake of oligonucleotides, cells were

deprived of serum nutrients and exposed to LipofectAMINE, an agent which serves as a

cationic carrier lipid to transport oligonucleotide into the cells. Additionally, media and

oligonucleotides were replaced after 12 h to compensate for any degradation which had

taken place. Despite these measures designed to optimize antisense-directed repression of
PKC-a levels following initial down-regulation by phorbol, the protein recovered in

antisense-treated cells at about the same rate as that of drug-free controls (see Figure 4.7).

Common explanations for the difficulty in achieving significant decreases in protein

expression include instability of the antisense oligonucleotide sequence, limited uptake of

the oligonucleotide, suboptimal binding of oligonucleotide to its target sequence, and half-

life of the particular protein. Despite efforts to account for and overcome these issues,

attempts to carry out further studies based on the antisense work were ultimately

abandoned, due to the persistent failure of PKC-ct antisense to down-regulate PKC-a

expression. Perhaps, in this study, the unpredictable tertiary conformation of the antisense

oligonucleotides was such that access and/or binding to the target mRNA sequence was

prevented, or perhaps the oligonucleotides were sequestered and/or degraded before they

even reached their destination. Because antisense technology is still in its developmental

stages, and oligonucleotide design and exposure conditions vary greatly among systems,









significant time and expense may often be invested before down-regulation of the protein

in question is optimized.
Whereas expression of each PKC isoform could not be down-regulated

individually, overall PKC activity inhibition was achieved through the use of the PKC

inhibitor LY333531 (Eli Lilly and Company, Indianapolis, IN), a direct competitor for

ATP binding (Jirousek et al., 1996). LY33353 1 is a PKC-P3-selective inhibitor (at low nm

concentrations) which also inhibits PKC isozymes oX, 6, and e at concentrations in the low

pM range (Ishii et al., 1996; Jirousek et al., 1996). Exposure of HN33 cells to 1.0 W

LY333531 for 24 h resulted in no change in either MARCKS or GAP-43 expression, nor

any evidence of toxicity. When combined with 1.0 mM VPA, LY333531 effectively

prevented the VPA-induced MARCKS down-regulation in the membrane fraction of

HN33 cells, but with no accompanying effect on the VPA-induced GAP-43 increase.
From these studies it is clear that PKC activation plays an integral role in the VPA-

induced MARCKS effects observed, as inhibition of PKC-directed phosphorylation of

MARCKS resulted in the inability of VPA to elicit down-regulation of membrane-

associated MARCKS. It is interesting that this LY33353 1-induced prevention of
MARCKS down-regulation by VPA occurred in membrane, the fraction of HN33 cells in

which VPA exerts its greatest effect. This finding is further consistent with the known

localization of PKC activity in the membrane, and also with a previous report showing that

phosphorylation of MARCKS is necessary for its dissociation and translocation from

membrane (Allen and Aderem, 1995). Whether PKC activation alone is sufficient to

reduce MARCKS has previously been investigated by our laboratory. It was shown that

phorbol esters, potent activators of PKC, yield a significant down-regulation of MARCKS
in the HN33 cells, and that the time course of phorbol-induced MARCKS alterations

follows closely behind the time course of phorbol-induced PKC activation and down-

regulation (Watson et al., 1994). These findings support the notion that activation (and

perhaps also subsequent down-regulation) of PKC, in and of itself, is sufficient to elicit a









reduction in MARCKS, though PKC activation may not be the only route through which

the MARCKS alterations may occur. Indeed, Brooks et al. (1992) reported evidence of

regulation of MARCKS through both PKC-dependent and independent pathways, with
platelet-derived growth factor (PDGF)-induced down-regulation of MARCKS through a

post-transcriptional mechanism.

As for GAP-43, PKC activation is probably not necessary to elicit the alterations in

GAP-43 observed. This is evidenced by the fact that PKC inhibition failed to block the
VPA-induced increase in GAP-43 in the HN33 cells. Even with the strong GAP-43

expression observed in the membrane of these cells, and the significant up-regulation of

this protein to several times basal levels, LY333531 caused no attenuation of the VPA-

induced GAP-43 increase observed. Assuming conditions were optimal for significant

PKC inhibition, we can conclude that PKC activation is not entirely responsible for the

VPA-induced GAP-43 increase observed, nor is PKC activation alone likely to be

sufficient for the effect, since lithium-induced alterations in PKC activity and isoform

expression did not lead to any change in GAP-43 expression.
The findings do little to address the issue of the involvement of PKC in bipolar

disorder and successful mood-stabilizing therapy. However, one can hypothesize that if

MARCKS dysregulation is integral to bipolar disorder, and PKC is the primary modulator

of MARCKS, then perhaps the enzyme is perturbed in some patients exhibiting symptoms

of the disorder. This would be especially applicable to cases in which the patient

demonstrates normal levels of MARCKS, for MARCKS, being downstream from PKC, is

regulated by PKC, and if PKC parameters are abnormal, then PKC-induced MARCKS

regulation will likely also be abnormal. Even if MARCKS itself is not involved in the
manifestation of the disorder, PKC may be a major player. Indeed, previous studies have

provided evidence for alterations in PKC-related signaling in patients with bipolar

disorder. Reports have shown that in both platelets and post-mortem brain samples of

individuals diagnosed with the disease, PKC translocation and activity levels were









significantly increased as compared to controls, as measured by phorbol- or serotonin-

stimulated histone phosphorylation (Friedman et al., 1993; Wang and Friedman, 1996). In

addition, PKC isoform expression was altered in comparison to controls, with higher
levels of PKC-ct, 6, and and lower PKC-e expression in bipolar patients (Wang and

Friedman, 1996). Further, two-week lithium treatment of subjects resulted in a reduction
in both PKC translocation and activity in platelets (Friedman et al., 1993). In a more

recent clinical study, the PKC-selective inhibitor tamoxifen (an estrogen receptor

antagonist widely used in the treatment of breast cancer) has been shown to be effective in

treating acute mania in a small sample of patients (Manji et al., 1999). Inasmuch as PKC

is believed to play a pivotal role in regulating neuronal signal transduction and in

modulating intracellular cross-talk between neurotransmitter systems (Nishizuka, 1992),

these studies are suggestive of the potential for affective (mood) dysfunction as a result of

alterations or dysfunction in PKC and associated neuronal signaling pathways. A

definitive association between PKC alteration and mood stabilization may facilitate the

development of a whole new line of agents useful in the treatment of bipolar disorder.














CHAPTER 5
EFFECT OF VPA ON CELL VIABILITY, GROWTH AND MORPHOLOGY


Introduction


While VPA has proven very useful in its clinical applications to seizures and

bipolar disorders, the detrimental effects of this drug on the developing embryo have been

well documented (Wiger et al., 1988; Gofflot el al., 1996; Menegola et al., 1996). High

frequencies of neural tube defects (NTDs) have been linked to prenatal VPA exposure

(Bjerkedal et al., 1982; Robert and Guibaud, 1982; Lindhout and Schmidt, 1986; Wiger et

al., 1988; Gofflot el al., 1996; Menegola et al., 1996). Such defects may include open

neural tube, spina bifida, anencephaly, meningomyelocele, and encephalocele, and are

thought to be caused by a number of different factors, such as environmental or

pharmacological teratogens, genetic anomalies, and various maternal predisposing factors

(Rhoads and Mills, 1986; Norman et al., 1995). In more recent years, folic acid and/or
methionine supplementation have been shown to reduce the frequency of NTDs, both in

the general population and in offspring of mothers undergoing anticonvulsant therapy

(Laurence et al., 1981; Smithells et al., 1981; Rhoads and Mills, 1986; Trotz el al., 1987;

Mulinare et al., 1988; Nosel and Klein, 1992; Ehlers el a., 1996). Therefore, women are

advised to discontinue VPA therapy or take folic acid supplements during and just before

pregnancy. The mechanism of this protective effect is not understood. Folate is known to

be a precursor of methionine, an important amino acid for cell proliferation, and vitamin
B 12 is an essential cofactor for methionine synthesis. In mothers of NTD-affected

offspring, circulating serum folate levels are low but not clinically deficient; the same is









true for B12 levels (Laurence et al., 1981; Rhoads and Mills, 1986; Kirke et al., 1993). In
addition, homocysteine, the direct precursor for methionine, is found at high levels in the

serum of some mothers in this population (Steegers-Theunissen et al., 1994; Mills et al.,
1995, 1996). This evidence suggests that, in at least a fraction of NTD cases, the mothers

of NTD-affiicted offspring exhibit a metabolic insufficiency, perhaps at the level of the
homocysteine-to-folate conversion step. This "metabolic defect" hypothesis is in contrast

to the simpler notion of a nutritional deficiency (of either folate or methionine), and is

consistent with the possibility that VPA therapy is one factor which may induce or

enhance this metabolic defect. It is of note that, in experimental studies, in vivo

supplementation of either nutrient to pregnant animals resulted in decreased frequencies of

NTDs in offspring, while in vitro studies did not reproduce the protective effect (Trotz et

al., 1987; Hansen and Grafiton, 1991; Nosel and Klein, 1992; Hansen et a., 1995; Ehlers

et al., 1996). Preliminary studies in our laboratory further support the notion that the

effect is of a complicated metabolic nature, for folate or methionine supplementation to

cell growth medium did not prevent the VPA-induced down-regulation of MARCKS

observed in HN33 cells (Watterson unpublished data).
In light of the deleterious developmental effects of this drug, the effects of VPA on

cell growth and differentiation merit in-depth investigation. Previous studies, while few in
number, have provided evidence for a dose-dependent VPA-induced inhibition of cell

proliferation via arrest at mid-GI of the cell cycle (Martin and Regan, 1991), as well as

VPA-induced differentiation of neuroblastoma and glioma cells (Regan, 1985; Cinatl et

al., 1996). The inhibition of cell proliferation, alterations in cellular interactions, and

altered differentiation have all been correlated with prenatal drug teratogenicity, especially
of the developing neural tube (Wilk et al., 1980; Mummery et al., 1984; Martin et al.,

1988). It is plausible that arrest of cell proliferation at a specific phase in the cell cycle,

accompanied by irreversible differentiation, may result in the effects observed to be

consistent with prenatal teratogenicity.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4C4DBCOB_8BDH86 INGEST_TIME 2014-07-17T21:38:21Z PACKAGE AA00022190_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


IN VITRO STUDIES ON A PUTATIVE MOLECULAR MECHANISM OF ACTION
OF VALPROIC ACID IN TREATMENT OF BIPOLAR DISORDER
BY
JEANNETTE M. WATTERSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

ACKNOWLEDGMENTS
I wish to thank Dr. Edwin Meyer first and foremost, for he graciously accepted me
“as is” into his lab, thereby saving me from a decision I did not want to make I am
grateful that he always saw and helped me to see the positive, and opened me up to new
ideas that I never thought to examine. I am very appreciative of Dr Robert Lenox’s
ability to stimulate me to always look for more, as well as to give more, and to never be
satisfied with less than 110% I also credit him with redefining my conception of
“intensity” and “drive.” I am indebted to Dr David Watson for training me with patience
from start to finish, and for giving me direction and confidence at times when I had little.
I am grateful to my committee members - Drs Stephen Baker, Jeffrey Harrison, and Colin
Sumners - for their guidance and approval, which 1 needed in order to maintain focus and
motivation I am also thankful for the steadfast support and friendly atmosphere provided
by the department secretaries, the helpful technicians, and the many peers 1 have worked
with over the past four years, without whose help I could never have reached this point
Finally, I would like to thank my parents, who instilled in me the ambition to aim high and
the drive to do my best.
li

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ii
LIST OF TABLES v
LIST OF FIGURES vi
ABBREVIATIONS AND DEFINITIONS ix
ABSTRACT x
CHAPTERS
1 INTRODUCTION 1
VPA 2
Bipolar Disorder 4
Mechanism of Action of Lithium in Treatment of Bipolar Disorder 7
PKC and Its Substrates MARCKS and GAP-43 9
Developmental Effects of VPA 11
Summary 11
Specific Aims 13
2 MATERIALS AND METHODS 20
Chemicals 20
HN33 Cell Culture 20
PC 12 Cell Culture 21
Neurite Quantitation 22
Protein Quantitation by Western Blotting 22
Ribonuclease (Rnase) Protection Assay (RPA) 23
PKC Inhibitors 24
PKC Activity 24
Statistical Analyses 25
iii

3 EFFECT OF VPA ON MARCKS AND GAP-43 27
Introduction 27
Results 37
Discussion 53
4 ROLE OF PKC IN MECHANISM OF ACTION OF VPA 62
Introduction 62
Results 67
Discussion 78
5EFFECT OF VPA ON CELL VIABILITY, GROWTH AND MORPHOLOGY 88
Introduction 88
Results 91
Discussion 98
6CONCLUSIONS AND FUTURE DIRECTIONS
113
REFERENCES 124
BIOGRAPHICAL SKETCH 139
IV

LIST OF TABLES
Table page
2.1. Properties of HN33 hybrid and parental cells 21
2.2 Antibodies for Western blots 26
3.1. Addition of carbachol fails to potentiate the VPA-induced
MARCKS down-regulation in HN33 cells 43
3.2. Inositol supplementation fails to attenuate the VPA-induced
MARCKS down-regulation in HN33 cells 44
3.3. Drugs, sources, and exposure conditions for HN33 cells 49
4.1. Antisense oligonucleotide sequences 73
5.1. VPA induces an increase in HN33 cell doubling time 91
5.2. VPA has no effect on cell viability 91
6.1. Effects of two mood-stabilizing drugs, VPA and lithium,
as observed in HN33 cells following chronic exposure:
A summary 123
v

LIST OF FIGURES
Figure page
1.1. Molecular effects of VP A in brain 17
1.2. Model for lithium regulation of MARCKS 18
3.1. Comparison of MARCKS expression in mouse brain vs.
HN33 cells: Representative Western blot 37
3.2. Concentration-dependent down-regulation of MARCKS
protein in HN33 cells following 1 d VP A exposure
a. Representative Western blot 38
b. Quantitation 38
3.3. Concentration-dependent down-regulation of MARCKS
protein in HN33 cells following 3 d VPA exposure
a. Representative Western blot 39
b. Quantitation 39
3.4. Time course of VPA-induced MARCKS down-regulation
in HN33 cells 40
3.5. Recovery of MARCKS following VPA-induced
long-term down-regulation
a Quantitation 41
b. Representative Western blot 41
3.6. Effect of carbachol on the VPA-induced MARCKS down-
regulation in HN33 cells: Representative Western blot 43
3.7. Effect of /77>Y>-inositol on the VPA-induced MARCKS down-
regulation in HN33 cells: Representative Western blot 44
3 8. Additive effect of combined VPA and lithium exposure
on MARCKS protein
a Quantitation 45
b. Representative Western blot 45
vi

3.9. Effects of other psychotropic agents and VP A metabolites
on MARCKS protein
a. Quantitation 47
b. MARCKS expression following exposure to
two VPA metabolites: Representative Western blot 48
c. Structures of VPA and two tested metabolites 48
3.10 Time course of VPA-induced MARCKS mRNA alterations 50
3.11. Comparison of GAP-43 expression in rat brain vs.
HN33 cells: Representative Western blot 51
3.12. Effect of chronic VPA exposure on GAP-43 expression
in HN33 cells: Representative Western blot 51
3.13. Concentration-dependent increase in GAP-43 expression
in HN33 cells following VPA exposure 52
3.14 Time course of VPA-induced GAP-43 increase in HN33 cells 52
4.1. Effect of VPA exposure on PKC activity in HN33 cells 67
4.2. Effect of VPA on soluble PKC-a expression in HN33 cells
a.Acute VPA Exposure: Representative Western blot 68
b Chronic VPA Exposure: Representative Western blot 68
c. Quantitation 68
4.3. Effect of VPA on PKC-5 expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot 69
b. Chronic VPA Exposure: Representative Western blot 69
c. Quantitation 69
4 4 Effect of VPA on PKC-e expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot 70
b. Chronic VPA Exposure: Representative Western blot 70
c. Quantitation 70
4.5. Effect of VPA on PKC-£ expression in HN33 cells
a. Acute VPA Exposure: Representative Western blot 72
b. Chronic VPA Exposure: Representative Western blot 72
c. Quantitation 72
4.6. PKC-ot antisense oligonucleotide design 73
4.7. Effect of PKC-a antisense oligonucleotides on PKC-a
expression in HN33 cells: Representative Western blot 75
vii

4.8 Effect of PKC inhibitor LY333531 on VPA-induced
down-regulation of MARCKS
a Quantitation 76
b.Representative Western blot 76
4.9. Effect of PKC inhibitor LY333531 on VPA-induced
GAP-43 increase
a. Quantitation 77
b. Representative Western blot 77
5.1. Time-dependent VPA-induced neurite outgrowth
in HN33 cells: Representative photomicrographs
a. Untreated, control HN33 cells 92
b. 3 d VPA-exposed HN33 cells 92
c. 14 d VPA-exposed HN33 cells 92
d. Quantitation 92
5.2. Effect of PKC inhibitor LY333531 on
VPA-induced alterations in HN33 cells
a. Effect on cell density 94
b. Effect on neurite outgrowth 94
5.3. Concentration-dependent VPA-induced neurite outgrowth
in PC 12 cells 95
5.4. Effect of PKC inhibitor BIM on VPA-induced
alterations in PC 12 cells
a. Effect of 0 6 m\1 VPA on PC 12 cell number 96
b Effect of 1.8 mM VPA on PC 12 cell number 96
c. Effect of 0.6 mM VPA on PC 12 cell neurite outgrowth 96
d Effect of 1.8 mM VPA on PC 12 cell neurite outgrowth 96
5.5. Neuroprotective potential of VPA
a Effect of VPA on cell number in
NGF-differentiated PC 12 cells 97
b. Effect of VPA on neurite outgrowth in
NGF-differentiated PC 12 cells 97

ABBREVIATIONS AND DEFINITIONS
BIM - bisindolylmaleimide, an inhibitor of PKC
CMP-PA - cytidine monophosphate-phosphatidic acid; also CDP-DAG
CNS - central nervous system
DAG - diacylglycerol, a stimulator of PKC
FDA - U S. Food and Drug Administration
GABA - gamma-aminobutyric acid, an [inhibitory] amino acid neurotransmitter
GAP-43 - 43 kD growth-associated protein
HN33 - immortalized hippocampal cell line derived from fusion of primary mouse
hippocampal and human neuroblastoma cells
IMPase - /nyw-inositol 1-monophosphatase
IP - inositol [monojphosphate
LY333531 - a PKC inhibitor designed by Eli Lilly and Company
MARCKS - myristoylated alanine-rich C kinase substrate
MBP - myelin basic protein
NGF - nerve growth factor
PCI2 - rat pheochromocytoma (adrenal tumor) cell line
PI - phosphoinositide
PKC - protein kinase C
PDBu - phorbol 12,13-dibutyrate
RPA - RNase protection assay
SDS-PAGE - sodium dodecyl sulfate - polyacrylamide gel electrophoresis
VPA - valproic acid or valproate
IX

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
IN VITRO STUDIES ON A PUTATIVE MOLECUUAR MECHANISM OF ACTION
OF VALPROIC ACID IN TREATMENT OF BIPOLAR DISORDER
By
Jeannette M. Watterson
December 1999
Chair: Edwin M. Meyer
Major Department: Pharmacology and Therapeutics
Valproic acid (VPA; trade name Depakote) is a broad-spectrum anticonvulsant
which has of late been utilized as a treatment for bipolar disorder and has proven
efficacious in treating a variety of both acute manic and chronic bipolar states. Lithium is
currently the only other FDA-approved drug treatment for this disorder, and although it
has traditionally been the first-line treatment of choice, there is a recognized need for
alternatives, due to the low patient compliance rate resulting from troublesome side
effects, narrow therapeutic index, and less than optimal patient response.
Very little is known about how either drug exerts its mood-stabilizing properties,
and it has been the goal of this project to understand better the actions of VP A from a
mechanistic standpoint.
Initial experiments were conducted utilizing an immortalized hippocampal cell line,
the HN33 cells. Results indicate that at clinically relevant target serum concentrations,
chronic, but not acute, VPA reduces expression of the myristoylated alanine-rich C kinase
substrate (MARCKS), concomitant with increased GAP-43 expression. These effects are
associated with altered activity and expression of protein kinase C (PKC), an important
x

effector in the CNS, and the effect of VP A on MARCKS appears to be dependent, at least
in part, upon the activation/down-regulation of this enzyme. Furthermore, the observed
effects are accompanied by alterations in cellular morphology and proliferation rate, which
implies that VPA plays a major role in modulating neuronal plasticity.
Additional studies utilizing PC 12 cells showed similar VPA-induced
differentiation, with no evidence for neuroprotection following nerve growth factor-
induced differentiation of cells. These studies bring us closer to elucidating the basis for
the unique therapeutic profile of VPA, which includes both antiepileptic and
antimanic/mood-stabilizing clinical applications, as well as unfortunate teratogenic effects
(such as spina bifida, anencephaly, meningomyelocele, and encephalocele), and may
provide further insight into the pathophysiology underlying bipolar disorder.
xi

CHAPTER 1
INTRODUCTION
VPA is a broad-spectrum anticonvulsant shown in the last two decades to be
efficacious in the treatment of bipolar disorder (manic-depressive illness). These two
actions can be distinguished from each other by their time-courses, with anticonvulsant
activity being observed much more quickly, on a scale of hours to days, in contrast to the
days to weeks treatment required for mood-stabilizing activity. The precise mechanisms
by which VPA exerts either its antiepileptic or antimanic effects are still unclear VPA has
been found to affect Na+ flux, intracellular Ca^, and brain gamma hydroxybutyric acid
(GABA) levels; these actions probably relate to the anticonvulsant efficacy of this drug.
The basis for its mood-stabilizing efficacy remains a mystery In addition to its therapeutic
actions, numerous studies have documented the deleterious effects of this drug on the
developing nervous system, especially the embryonic neural tube. It is the goal of this
project to elucidate further the mechanisms of action of VPA in brain, mainly through
studies of its effects on intracellular signaling in one particular cell model, in an effort to
understand better the underlying mechanism for this therapeutic profile. The specific aims
of this project are to: (1) characterize the therapeutically relevant time- and concentration-
dependent effects of VPA on PKC (activity and isozyme expression) and its substrates
(MARCKS and GAP-43), using the hippocampally derived HN33 cells as a model system
for neuronal effects of VPA; and (2) characterize the therapeutically relevant effects of
VPA on the viability, growth, and morphology of both HN33 cells and PCI 2 cells. Based
on preliminary evidence and previous reports, it is hypothesized that VPA will have a
significant effect on the expression of PKC substrates MARCKS and GAP-43, that PKC

will be integrally involved in any such effects, and that cells will undergo an alteration in
cell growth and morphology indicative of a transition to a more differentiated state.
2
VPA
VPA is a short-chain carboxylic acid that was first discovered to have antiepileptic
effects in 1963, and has been utilized as an anticonvulsant in the United States for over 20
years (Meunier el al., 1963; Ramsay, 1984). It first became available in the U S. in 1978
as Depakene, the acid form of the drug. Five years later, the enteric-coated form,
Depakote, was introduced (Penry and Dean, 1989). This particular formulation, the most
popular form in use today, is a stable coordination compound containing equal
proportions of valproic acid and the sodium valproate salt (McElroy el al., 1989). VPA is
striking in its structural difference from other known anticonvulsants, yet it is well known
for its efficacy in treating a wide range of seizure disorders, giving it a broader therapeutic
profile for epilepsy treatment than perhaps any other anticonvulsant yet studied. VPA is
FDA-approved for treatment of simple and complex absence seizures, as well as mixed
seizures which include the absence type. Additionally, VPA is effective in treating such
primary generalized epilepsies as tonic-clonic, clonic, tonic, and myoclonic seizures,
infantile spasms, and photosensitive epilepsy, as well as secondary generalized seizures
and some partial seizures, including complex partial epilepsy (McElroy el al., 1989).
A number of studies have investigated the putative mechanism(s) through which
VPA exerts its anticonvulsant effects. No one signaling effect is likely to be responsible
for the entire panorama of the anticonvulsant efficacy of this drug, and a number of
molecular events appear to contribute to the net effect of reducing neuronal excitability.
The ability of VPA to increase neurotransmission of the inhibitory amino acid GABA has
been well documented. VPA may accomplish this through a variety of mechanisms - by
increasing GABA levels through decreased degradation and increased synthesis and

3
release, or by increasing post-synaptic responsiveness to GABA through effects on the
GABA-receptor complex or altered K' conductance, effectively dampening neuronal
excitability (Chapman el a/., 1982; Rimmer and Richards, 1985; McElroy et a/., 1989;
Penry and Dean, 1989; Joffe, 1993). Because anticonvulsant effects are often observed
before total brain GABA levels become elevated, the VPA-induced increase in GABA
levels alone may not account for the drug’s anticonvulsant effect.
VP A, like other anticonvulsants, is known to block sustained high frequency
repetitive firing of cultured neurons, due to its effect of slowing recovery of the sodium
channel following inactivation (Albus and Williamson, 1998). At high concentrations,
VPA has been shown to alter membrane potassium conductance, and relatively low
concentrations of the drug hyperpolarize neurons, both effects suggestive of direct effects
on neuronal membranes (Porter and Meldrum, 1995). Other studies have provided
evidence for reduced excitatory neurotransmission by the amino acid aspartate (Chapman
el al., 1982). Still other reports have suggested that VPA works through inhibition of
kindling, a model of epilepsy which holds that epileptic activity predisposes the system to
increased frequency of seizures triggered by lower thresholds over time (Porter and
Meldrum, 1995; McElroy et al., 1989).
In summary, the various mechanisms proposed encompass three major net effects
of VPA: (1) increased GABAergic synaptic inhibition post-synaptically; (2) reduction in
the excitatory synaptic transmission ultimately responsible for epileptic kindling; and (3)
reduction in repetitive firing via a direct effect on voltage-sensitive ion channels (Albus
and Williamson, 1998). The broad spectrum of anticonvulsant activity of VPA is probably
related to more than one of the molecular mechanisms demonstrated thus far. However, it
is unclear whether such anticonvulsive effects may also account for the antimanic efficacy
of VPA, and because little is known of the mood-stabilizing mechanism of VPA, much
remains to be explained with regard to what kinds of intracellular changes might underlie
this drug’s mood-stabilizing effects.

4
Bipolar Disorder
In recent years, VPA has emerged as an effective alternative to lithium in the
treatment of acute mania in patients suffering from bipolar disorder (Gerner & Stanton,
1992; McElroy ela!., 1992; Bowden el at., 1994), and is also being used for long-term
prophylactic management of this disorder. Bipolar disorder, also known as manic-
depressive illness, is a debilitating psychiatric condition for which there is no known cause
and only a limited range of treatment options available. It is estimated to affect about 1%
of the population in this country (Bowden el al., 1994). Patients afflicted with this
affective disorder experience frequent, recurring episodes of extreme alterations in mood.
Whereas the patient may appear normal between cycles, the episodes of mania and
depression that characterize this disorder can leave the patient functionally incapacitated.
The depressed phase involves motor activity alterations, impaired concentration, impaired
social and occupational functioning, general helplessness, and may result in changes in
sleep, appetite, weight, and energy level (Post, 1989). The manic episodes entail such
disruptive symptoms as hyperactivity, explosive temper, impaired judgement, insomnia,
disorganized behavior, hypersexuality, and grandiosity, any combination of which may
lead to alienation from family and friends, indebtedness, job loss, and other major life
problems (Bowden el al., 1994). If untreated, this disorder has a tendency to worsen,
with increasing severity, duration, and frequency of episodes over time (Post, 1989).
There is a high rate of suicide in patients suffering from this disorder, estimated at about
15% in untreated cases (Bowden el al., 1994).
Lithium has for over 40 years been the first-line drug therapy for both acute and
prophylactic management of this disorder, but due to its narrow therapeutic index and
many undesirable side effects (often resulting in compliance problems), the need for other
viable treatment options has become a major concern (Post, 1989; Bowden, 1996).
Although numerous treatments have been used experimentally in the past, including

5
antipsychotic drugs, calcium channel blockers (such as verapamil), high doses of thyroxine
(T4), serotonin precursors (such as L-tryptophan), and even electroconvulsive therapy
(Joffe, 1993), results have not been consistently positive, and none is used extensively as a
“mainstream” treatment at the present time. For example, the use of antidepressants is
generally not considered because patients may be brought out of their depressive episodes
only to be immediately submerged into the manic phase, further, certain drugs may
increase the frequency of cycling of mood swings (Post, 1989). The anticonvulsants
carbamazepine, lamotrigine, and gabapentin, as well as other pharmacological agents
(clozapine, clonazepam, verapamil), have shown promise in treating bipolar disorder, but
very few studies have been conducted and FDA approval has yet to be obtained for this
clinical indication (Bowden, 1996; Joffe, 1993; Hollister, 1995). Because lithium is the
approved first-line treatment, and has been used and studied extensively in the past, this
project will be focused in part toward understanding the mechanism of action of bipolar
drugs, based on what is known and has been reported experimentally for lithium.
VPA is emerging as a viable alternative or combinational drug treatment for
bipolar disorder, and in fact, VPA may even have a wider spectrum of efficacy than
lithium. Although VPA seems to be less effective in treating depression than in treating
the manic phase of the disorder, there is accumulating evidence of its use in atypical
(dysphoric/mixed) mania, rapid cycling, and secondary manias, which have poorer
prognoses and are generally less responsive to lithium than the usual forms of the disorder
(Calabrese & Delucchi, 1990; Calabrese el al., 1992, 1993; Bowden el al., 1994; Joffe,
1993; Brown, 1989). VPA has also been used to treat various conditions which are
comorbid with bipolar disorder, such as migraines, panic disorder, bulimia, attention-
deficit hyperactivity disorder (ADHD), and other brain-related illnesses, and VPA appears
to be quite effective in treating lithium non-responders (Bowden, 1995; Joffe, 1993;
Brown, 1989). The therapeutic index of VPA is considerably wider than that of lithium,
thereby allowing for greater freedom in dosage adjustment, with less risk of toxicity.

6
Furthermore, the side effect profile of VP A is less intimidating than that of lithium.
Whereas the side effects of lithium may be quite severe and disturbing to patients (thyroid,
renal, and CNS effects, dermatologic and cardiovascular complications), VP A typically
exhibits a different, more benign profile of side effects which are generally more
acceptable to patients (gastrointestinal upset, weight gain, hair thinning, increased hepatic
enzyme levels), and many of its bothersome side effects are alleviated over time or with a
decrease in drug dosage (Bowden, 1995; Penry and Dean, 1989). The exception to this is
the idiosyncratic, unpredictable and irreversible hepatic toxicity induced by VPA in
predisposed patients on rare occasion. While this condition may be fatal if not detected in
time, it affects only a very minimal portion of the population - the highest incidence
reported was a rate of 0.01%. The majority of reported cases occur within the first three
months of VPA therapy, and because signs common to hepatic failure are clearly present,
patients can be monitored regularly for the first six months of therapy so as to reduce the
risk of fatality and prevent any long-term damage. Despite this risk, the side effect profile
of VPA is still considered to be safer and less bothersome than that of lithium.
Nonetheless, the mechanisms mediating the therapeutic properties of VPA in the
treatment of bipolar disorder are not clearly understood. Whereas the therapeutic actions
of VPA have been reportedly associated with potentiation of GABAergic transmission and
other neurotransmitter effects (Chapman el al., 1982; Post el al., 1992; Loscher, 1993;
Petty, 1995), these effects generally occur at rather high concentrations and may relate
more to the anticonvulsant properties of VPA than to its mood stabilizing effects; see
Figure 1.1 (Waldmeier, 1987; Motohashi, 1990). Further, as is the case for the time
course of lithium, there is a delay of several days in the onset of clinical antimanic action
of VPA, although loading strategies have demonstrated antimanic effects at times as early
as 3 d (McElroy el al., 1996). Thus, any mechanism postulated for the therapeutic action
of these drugs must take this delayed onset into account (Manji el al1995). Because
VPA and lithium are currently the only two drugs approved by the FDA for treatment of

7
acute mania, comparisons of the actions of these agents and efforts to elucidate the
mechanism(s) underlying the mood-stabilizing activity of both drugs is important to
understanding the pathophysiology of this illness. Further, this approach may lead to the
identification of new or existing drugs which may be useful in this application.
Mechanism of Action of Lithium in Treatment of Bipolar Disorder
Accumulating evidence from our laboratory and others strongly implicates
receptor-mediated PI signaling in the mechanism of action of lithium in the brain (Berridge
et al., 1982; Godfrey, 1989; Lenox and Watson, 1994). Numerous studies have
established lithium as an uncompetitive inhibitor of the enzyme inositol monophosphatase
(IMPase), and the resulting accumulation of intermediates in the pathway which are
responsible for additional downstream effects, including intracellular Ca-^ release and
PKC activation (see Figure 1.2; Post et a/., 1992). It was first suggested by Berridge and
colleagues that depletion of free inositol, resulting from the accumulation of intermediates
in the PI signaling pathway, is a major consequence of long-term lithium exposure
(Berridge et al., 1982; Berridge, 1989). This phenomenon would have greatest
implication in certain areas of the brain which are relatively more inositol-limited by virtue
of the blood-brain barrier as well as less capable of producing inositol de novo. Further,
because the PI signaling cascade is linked to G protein-coupled muscarinic receptor (mi,
nv¡, ms) activation, areas of the brain undergoing the highest rate of receptor activation
may predictably be preferentially affected by lithium. Indeed, studies in our laboratory
have provided evidence for an inositol-driven attenuation of the effects of lithium in HN33
cells, as well as potentiation by a muscarinic receptor agonist (carbachol) and
prevention/reversal by a muscarinic receptor antagonist (atropine) (Watson and Lenox,
1996). Following exposure of HN33 cells to therapeutic concentrations of lithium (1
mM), MARCKS protein expression is significantly down-regulated (Lenox et al., 1996;

8
Watson and Lenox, 1996). This MARCKS reduction only occurs in the relative absence
of inositol, for supplementation of inositol even at concentrations as low as 5 (iM prevents
the lithium-induced down-regulation of MARCKS (Watson and Lenox, 1996; Watson el
al., 1998; see also Table 3.2). Similarly, the concomitant exposure of HN33 cells to both
lithium and the muscarinic agonist carbachol results in an increase in the level of
MARCKS protein down-regulation observed, and exposure of cells to the muscarinic
antagonist atropine reverses or prevents the lithium-induced reduction in MARCKS
expression (Watson and Lenox, 1996; Watson el al., 1998; see also Table 3.1). These
findings have provided further support for the PI signaling hypothesis.
Studies of the molecular mechanism of action of lithium in brain have focused
largely on its modification of PI signaling, particularly the lithium-induced accumulation of
intermediates such as DAG and the resulting stimulation of PKC. Such a Pi-mediated
mechanism may account for the effects of lithium on PKC-regulated events, including
protein down-regulation/degradation and transcriptional and translational events. In
contrast, little evidence has been found to support a similar PI mechanism for VPA.
Vadnal and Parthasarathy (1995) showed that VPA had no effect, either stimulatory or
inhibitory, on IMPase partially purified from bovine brain. Further, lithium and
carbamazepine (a less commonly used antimanic agent) exhibited opposite effects on
IMPase, suggesting that the enzyme is not a common site of action for these two mood-
stabilizing medications. Dixon and Hokin (1997) reported that in cerebral cortical slices,
VPA caused no accumulation of inositol monophosphates, inositol bisphosphates, or
inositol 1,3,4-trisphosphate, all of which accumulated following lithium exposure. Based
on these observations, the depletion of inositol is not universally applicable as the basis for
the antimanic action of mood-stabilizing drugs.
Alternatively, it has become apparent that the action of chronic lithium in the brain,
which follows as a result of its interaction with the PI signaling pathway, is mediated
through subsequent regulation of PKC and the down-stream posttranslational modification

9
of select protein substrates (Lenox, 1987; Manji and Lenox, 1994; Watson and Lenox,
1996). Further, the clinical effects of lithium may be mediated, at least in part, through
these actions Although VPA apparently does not share the property of PI signaling
inhibition (Vadnal and Parthasarathy, 1995, Dixon and Hokin, 1997), the question remains
as to whether it affects PKC, as previous studies have suggested (Chen et al., 1994;
O’Brien and Regan, 1998). It is a goal of this project to address further the potential role
of PKC in the action of VPA in the brain, especially as such effects might relate to the
mood-stabilizing properties of this drug.
PKC and its Substrates MARCKS and GAP-43
PKC exists as one of at least 11 structurally-related isozymes in mammals, many of
which are Ca^-activated and/or DAG-dependent, and have been implicated in numerous
cellular responses associated with regulation of signaling and long-term events including
ion channel and gene regulation (Newton, 1995; Nishizuka, 1995). Numerous studies
have suggested a role for PKC in the long-term action of lithium (Lenox, 1987; Manji and
Lenox, 1994), and recent data have provided evidence that chronic VPA exposure alters
PKC activity as well as the expression of a and t; isozymes in non-neuronal cells (Chen et
al., 1994; Manji et al., 1996b). Previous studies have demonstrated that chronic (but not
acute) lithium treatment of rats, resulting in clinically relevant brain concentrations,
produces a significant reduction in the PKC substrate MARCKS in the hippocampus,
which persists beyond treatment discontinuation (Lenox et al., 1992; Manji et al., 1996a).
This lithium-induced down-regulation of MARCKS has also been demonstrated in an
immortalized hippocampal cell line (Watson and Lenox, 1996), in which it was previously
demonstrated that phorbol esters, which directly activate PKC, down-regulate MARCKS
protein expression in a PKC-dependent manner (Watson et al., 1994).

10
MARCKS is a prominent and preferential substrate in the brain for PKC, which by
virtue of phosphorylation, regulates the cellular localization and activity of this protein.
MARCKS binds calmodulin in a calcium-dependent fashion and cross-links filamentous
actin, and has been implicated in cellular processes associated with cytoskeletal
restructuring and neuroplasticity, e g., transmembrane signaling and neurotransmitter
release (Aderem, 1992; Blackshear, 1993). In fact, previous studies in mutant mice
lacking the MARCKS gene have shown that MARCKS expression is essential to normal
CNS development in animals (Stumpo el at., 1995). Insofar as MARCKS may also
represent a molecular target for mood stabilizers in the brain (Lenox and Watson, 1994;
Watson and Lenox, 1996), it should be of benefit to examine its potential regulation by
VPA as well as by other psychotropic agents.
The growth-associated protein GAP-43 is a second protein thought to be integral
to normal neuronal development and differentiation, as it is expressed at high levels during
nervous system development and is associated with neurite extension (Jap Tjoen San el
al., 1992; Aigner and Caroni, 1993; Strittmatter el ai, 1995). GAP-43 shares a number
of properties with MARCKS, including ability to bind actin and calmodulin (albeit in a
Ca^-independent fashion), as well as regulation by PKC (reviewed in Benowitz and
Routtenberg, 1997). Inasmuch as previous studies have provided evidence for an inverse
regulation of MARCKS and GAP-43 in certain regions of rat brain (McNamara and
Lenox, 1998), it may be instructive to assess the relative levels of GAP-43 observed
concomitant with VPA-induced alterations in MARCKS expression. Investigation of
these two related but differentially regulated proteins will likely provide insight into the
molecular mechanism of this drug, as will examination of this drug’s effects on PKC, the
primary enzyme responsible for the direct regulation of both of these proteins.

11
Developmental Effects of VPA
In addition to the therapeutic efficacy of VPA, this drug exhibits its share of
drawbacks, the most severe of which are its effects on the developing embryo. VPA has
been known for years to be a developmental teratogen. In recent years, women who are
pregnant or even of childbearing age are encouraged to supplement their diets with
adequate amounts of folic acid, for supplementation with this nutrient and its metabolites
has been associated with decreased frequency of neural tube defects (including spina
bifida, anencephaly, and encephalocele) in offspring of VP A-treated and nontreated
mothers (Smithells eta!., 1981; Rhoads and Mills, 1986; Trotz el a/., 1987; Nosel and
Klein, 1992; Ehlers el a/., 1996). Numerous studies have investigated this phenomenon,
yet it remains unknown how VPA causes these developmental defects or how folate
supplements are able to prevent such effects. Previous work has demonstrated that VPA
inhibits cell proliferation through arrest of the cell cycle at mid Gl phase (Martin and
Regan, 1991), as well as induces differentiation of neuroblastoma and glioma cells (Regan,
1985), which would suggest that VPA is interfering with development through
manipulation of the proliferation rate and differentiation state of cells central to nervous
system structure. Such effects on cell growth and morphology should be considered in
any putative model for the molecular mechanism of this drug, and this study will therefore
investigate the hypothesis that VPA exerts effects on both cell proliferation and cell
morphology which may account for its teratogenicity.
Summary
Experiments designed to elucidate the short- vs. long-term intracellular actions of
VPA will facilitate the development of a model which may explain, in part, the unique
therapeutic profile exhibited by this drug. Initial VPA studies have been modeled after

12
those conducted previously or in parallel for lithium, so that comparisons may be made
which might shed light on the basis for the common mood-stabilizing therapeutic action of
these two drugs. In summary, thus far, it has been shown that in the HN33 cell model,
lithium elicits a down-regulation in MARCKS primarily through its effect on PI signaling,
as evidenced by the carbachol-driven potentiation and either inositol- or atropine-driven
attenuation of effect (Watson and Lenox, 1996; Watson et al., 1998). This lithium-
induced MARCKS reduction occurs over a chronic time course, requiring days to weeks,
and the effect is greater in the soluble rather than membrane fraction of the cells (Lenox et
al., 1996; Watson and Lenox, 1996). A similar time course of MARCKS mRNA
reduction is also observed (Watson and Lenox, 1997). Further, chronic lithium
administration elicits a significant down-regulation of MARCKS protein in vivo, in a rat
model (Lenox et a/., 1992). As for the effects of lithium on cell viability and related
parameters, no alteration in cell growth rate or morphology has been observed following
acute or chronic lithium exposure (Watson and Lenox, 1996), and no alteration in
expression of another prominent PKC substrate, GAP-43, was observed (Watson and
Lenox, 1997). Lithium exposure results in predictable changes in PKC activity (with
initial activation followed by chronic down-regulation), as well as alterations in expression
of the two PKC isoforms a and s (Watson and Lenox, 1997), similar to findings
previously reported by Manji et al. (1996a,b).
Applications of research into the mechanism of action of VP A will be multifold In
addition to providing a better understanding for the efficacy of VPA in treating bipolar
disorder, a clearer picture of the pathophysiology underlying this illness, and identification
of new or existing drug treatments, investigation of the molecular bases for these effects
may foster a better understanding of how VPA exerts its deleterious developmental effects
and the development of potential therapeutic interventions to prevent or correct such
defects.

13
Specific Aims
Characterize the Therapeutically Relevant Effects of VPA on PKC and Its
Substrates.
MARCKS and GAP-43 are known to be important in modulating central nervous
system (CNS) plasticity and intracellular signaling, by virtue of the regulatory enzyme
PKC. Using hippocampally derived HN33 cells, I will characterize both acute and chronic
VPA-induced alterations of PKC expression and activity, as well as on levels of expression
of MARCKS and GAP-43.
Effects of VPA on MARCKS protein expression. Previous studies showed that
chronic lithium exposure at therapeutic concentrations resulted in down-regulation of
MARCKS protein expression in HN33 cells, and involved alterations in PI signaling.
After (1) establishing the concentration-response relationship for the effects of VPA on
MARCKS protein, I will (2) establish the time course of action of VPA at therapeutic
concentration, as well as (3) the time course of system recovery after VPA withdrawal. In
addition, I will assess the importance of (4) inositol concentration and (5) muscarinic
receptor/Gq activation, in VPA-induced MARCKS effects. Furthermore, I will (6)
investigate the effects of combined VPA and lithium treatment. Effects will be examined
in both cytosolic and membrane fractions of cells. It is predicted that VPA will have
effects similar to those of lithium, i.e. significant down-regulation of MARCKS protein
expression following chronic (but not acute) exposure to concentrations of the drug within
a therapeutically relevant range, and that MARCKS will return to basal levels after drug
removal In contrast to lithium, there is no evidence for VPA-induced modulation of the

14
PI signaling pathway, so no effects by inositol or carbachol (muscarinic receptor agonist)
are expected
Effects of VPA on MARCKS mRNA. Preliminary studies have provided
evidence for a lithium-induced reduction in MARCKS mRNA I will establish the time
course of therapeutically relevant VPA-induced effects on MARCKS mRNA levels. It is
predicted that VPA will elicit down-regulation of MARCKS mRNA is a fashion similar to
that of lithium, but following the time course of VPA-induced alterations in MARCKS
protein
Effects of VPA on GAP-43 expression. 1 will establish a dose-response and time
course for therapeutically relevant VPA-induced effects on GAP-43 protein levels in both
cytosolic and membrane cellular fractions. Although there are no previous reports for
VPA- or lithium-induced alterations in GAP-43 expression, such effects are a likelihood
given preliminary evidence for VPA-induced morphological alterations, and may partially
account for the plastic changes (such as synaptic remodeling) essential during therapy for
seizure and mood disorders.
Effects of VPA on PKC isozyme expression. 1 will establish the time course, at
therapeutic concentration, of VPA-induced alterations in four PKC isozymes expressed in
HN33 cells. Based on previous findings with lithium and VPA, which provide preliminary
evidence for effects of each on PKC expression and activity, I predict that VPA will have
significant effects on both PKC activity and isozyme expression, especially if alterations in
MARCKS and GAP-43 are observed following chronic VPA exposure.
Effects of VPA on PKC activity. I will investigate the effects of therapeutic
concentrations of VPA on PKC activity (as measured by phosphorylation of a PKC-
specific substrate) over a broad time course from acute to chronic time points. VPA-

15
induced alterations in PKC activity are expected for the same reasons as described above
in D.
Role of PKC in mechanism of action of VPA. Through the use of PKC
inhibitors and antisense, which block PKC activity, I plan to further characterize A-E
above; specifically, MARCKS and GAP-43 alterations, as well as morphological effects
(see below), will be investigated following PKC inhibition via enzyme inhibitors and
isozyme-specific antisense, in order to further define the role of PKC in these parameters.
Characterize the Physiologically Relevant Effects of Chronic VPA on Viability of
HN33 Cells and PC12 Cells.
Effects of VPA on HN33 cell growth. I will investigate the effects of chronic
VPA exposure at therapeutic concentrations on cell viability through observation of cell
number and doubling time The role of PKC in any observed effects will also be assessed.
Based on limited previous studies in other cell lines, it is predicted that VPA will produce
changes in cell growth which may help to explain, in part, the severe developmental effects
associated with prenatal VPA exposure
Effects of VPA on HN33 cell morphology. I will also investigate the effects of
chronic, therapeutic VPA exposure on cell phenotype, looking specifically for
differentiated morphology and other markers indicative of altered state of maturation. The
role of PKC in any observed effects on cell differentiation will also be assessed Despite
the lack of evidence for a lithium-induced alteration in cell morphology, VPA-induced
alterations in cell morphology are considered likely, given this drug’s therapeutic profile
(anticonvulsant and mood-stabilizing, yet teratogenic). Observed alterations in MARCKS

16
and GAP-43 will further support this hypothesis, given their integral role in CNS
development and plasticity.
Effects of VPA on PC 12 cell viability. Preliminary evidence for effects of VP A
on cell growth and morphology leads me to investigate more in depth the effects of VPA
on cell viability. Whereas the HN33 cell line is a sufficient model for the investigation of
intracellular signaling phenomena as outlined in the studies above, post-mitotic cell lines
may serve as better developmental models for investigation of such phenomena as
neuroprotection or drug toxicity, and these cell lines have been better characterized for
such investigations. Therefore, in order to investigate the potential of VPA to elicit either
toxic or neuroprotective effects, NGF-differentiated and non-differentiated PC 12 cells will
be utilized. In light of the well-established neuronal toxicity of VPA exhibited during
development, toxicity resulting from high levels of the drug would not be a surprise;
however, preliminary studies in our laboratory have shown no evidence for toxic effects of
this drug at therapeutic concentrations. Any cytoprotective effect of VPA has yet to be
established, and there is currently no evidence in support of such an effect.

17
Figure 1.1. Molecular Effects of VPA in Brain. Depicted are (1) at right, the effects of
VPA as previously established by other groups, and thought to be responsible for the
anticonvulsant efficacy of the drug, and (2) at left, the long-term effects hypothesized to
contribute to the mood-stabilizing efficacy of VPA. Of the short-term effects at right,
increased GABA neurotransmission and effects on intracellular calcium are generally
thought to contribute less than the effects of VPA on voltage-sensitive ion channels and the
associated decrease in excitatory neurotransmission and neuronal firing. The hypothesized
effects at left will be assessed in detail in the studies which follow.

Figure 1.2. Model for Lithium Regulation of MARCKS. (adapted from Lenox et ai, 1996). Following agonist
occupancy of the receptor binding site, G-protein coupling mediates activation of the enzyme phospholipase C (PLC),
which catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2) into second messengers diacylglycerol
(DAG) and myo-inositol 1,4,5-triphosphate (Ins 1,4,5 P3). DAG is an endogenous activator of PKC, and Ins 1,4,5 P3
induces release of calcium from intracellular stores such as the endoplasmic reticulum. Lithium inhibits uncompetitively
the enzyme inositol monophosphatase (IMPase), thereby inhibiting the recycling of inositol. This results in accumulation
of metabolites in the DAG pathway during stimulation of receptor signaling. Activation of PKC* through both increased
calcium and DAG initiates posttranslational events such as the translocation of PKC to the membrane where it
phosphorylates the MARCKS protein, which then translocates to the cytosolic fraction. In the phosphorylated form,
MARCKS no longer binds calmodulin or cross-links acón, thereby altering compartmental intracellular calcium signaling
and affecting cytoskeletal restructuring. In addition, chronic lithium and receptor activation down-regulate MARCKS
protein expression, which may result from down-stream transcriptional/translational events.

DNA

CHAPTER 2
MATERIALS AND METHODS
Chemicals
VPA (2-propyl-pentanoic acid) and other chemicals were purchased from Sigma
(St. Louis, MO), unless otherwise noted. For some drug comparison studies, additional
VPA and VPA metabolites were obtained from Abbott Laboratories (Abbott Park, IL),
and fluoxetine and LY333531 (a PKC inhibitor) were provided by Eli Lilly and Company
(Indianapolis, IN). Tissue culture reagents were purchased through Gibco (Gaithersburg,
MD; horse serum, RPMI and DME media), Hyclone (Logan, UT; fetal bovine serum), and
Sigma (trypsin and gentamicin). Electrophoresis reagents were purchased from Bio-Rad
(Hercules, CA), and RPA reagents were purchased from Ambion (Austin, TX).
HN33 Cell Culture
The immortalized hippocampal cell line HN33 .dw(9/21) was used to assess effects
of VPA on all parameters, including MARCKS protein and mRNA, GAP-43, PKC activity
and expression, etc. HN33 cells were kindly provided by Dr. Bruce Wainer (Emory
University, Atlanta, GA). These are derived from the fusion of primary neurons from the
hippocampus of postnatal d 21 mice with the N18TG2 human neuroblastoma cell line, and
exhibit morphological, immunological, and electrophysiological parameters characteristic
of hippocampal neurons in culture (Lee et a/., 1990). These properties are described in
Table 2.1. HN33 cells have been used in our laboratory as a model system to address the
20

21
mechanism of action of lithium, and we previously established the presence of a functional
second messenger (PI) signaling pathway in these cells (Watson and Lenox, 1997).
Furthermore, HN33 cells have recently been assessed and found to express at least two of
the neurofilament triplet proteins (Watterson, unpublished data), which supports their
consideration as phenotypically neuronal (Debus et a/., 1983; Shaw, 1991). Cells were
grown at 37°C and 5% CO2 in Dulbecco's modified Eagle's medium (DMEM) containing
5% defined fetal bovine serum (FBS) and 1 mg/ml gentamicin, with or without the
addition of 0.6 mM VPA. The duration of exposures ranged from 0 to 14 d. VPA was
added to cultures at time of plating, and culture medium was changed or cells were
passaged every 3-4 d, with no replenishment of VPA between platings. Assays were
performed when cells reached 80-90% confluence.
Table 2.1 lists the properties present (+) or absent (-) in HN33 cells and the two parent
cell lines from which the HN33 were derived, primary mouse hippocampal neurons (21 d)
and N18TG2 neuroblastoma cells. This table was adapted from that reported in Lee et al.
(1990), in which the HN33 cell derivation and characteristics were first reported.
Table 2.1. Properties of
-IN33 Hybrid and Parental Cells
Tissue or
Hybrid
Neurilic
Excitable
NGF
NGF
Cell Line
GPI Isozyme
Processes
Membrane
NF-M
Protein
mRNA
Hippocampus
+
+
+
+
+
N18TG2
HN33
+
+
+
+
+
+
PCI2 Cell Culture
PC 12 cells obtained from American Type Culture Collection (Rockville, MD) were
grown in RPMI 1640 containing 10% heat-inactivated horse serum, 5% FBS, and 0.5 mM
1-glutamine. Cultures were maintained at 37°C and 5% CO2, and passaged every 3 d. To

22
assess VPA-induced differentiation, cells were exposed to 0.3 - 1.8 mM VPA for 2 d, and
then neurite outgrowth was quantitated as described below. The cytoprotective assay was
carried out following differentiation of cells by 7-d exposure to 100 ng/ml NGF, and
subsequent NGF removal and replacement with 0.3, 0.6, and 1.2 mM VPA for 2 d.
Viviana Puig is acknowledged for her extensive assistance in carrying out the technical
aspects of these experiments utilizing the PC 12 cells.
Neurite Quantitation
HN33 cells were cultured in the presence or absence of 0.6 mM VPA for 3-14 d,
and process outgrowth was assessed using phase-contrast microscopy. A neurite was
scored if its length was at least one respective cell body diameter. Treatments were
compared for percentage of cells with one or more neurites and two or more neurites, and
150-900 cells were scored per treatment group. PC 12 cells were similarly assessed, using
the same criteria, with the aid of NIH Image 1.47 imaging software. Dan O’Donnell,
Jeremy Grimes, Risha King, and Keith Templin are acknowledged for their participation in
quantitation of neurite outgrowth in the PC 12 cells.
Protein Quantitation by Western Blotting
Harvested cells were homogenized in a buffer containing 20 mM HEPES, pH 7.4,
2 mM EGTA, 1 mM PMSF, 2 mM DTE, and 10 ng/ml aprotinin. Cells were sonicated to
disrupt cell membranes, and the soluble and pellet fractions were separated by
centrifugation. The homogenate was centrifuged at 100,000 x g and the soluble fraction
collected. The pellet was resuspended in buffer containing 0.1% Triton X-100 and
solubilized for 30 min. Solubilized fractions were then centrifuged at 50,000 x g and the
supernatant containing the solubilized membrane protein collected. Samples were

23
adjusted by addition or dilution to 0.05 % Triton X-100. Equal amounts of soluble and
membrane cell protein (50 gg), as determined by the Bradford method, were separated by
SDS-PAGE and transferred to polyvinylidene fluoride membrane (Millipore Corp.,
Bedford, MA) in a Bio-Rad Trans-Blot electrophoresis apparatus at 100 V for 2 h using
Towbin's buffer (25 mM Tris, pH 8.3, 192 mM glycine, and 20% (v/v) methanol). All
antibodies were diluted in TS buffer (20 mM Tris, pH 7.5, and 0.5 M NaCl); see Table 2.2
for dilutions and incubation times. Detection of the immune complex was performed
using HRP-conjugated anti-rabbit or anti-mouse IgG, and the Pierce (Rockford, IL)
enhanced chemiluminescence system Western blots were quantified using NIH Image
1.47 software for densitometric analysis. Results are expressed as the percentage of
protein present in drug-free controls grown in parallel and assayed on the same Western
blot.
Ribonuclease (RNase) Protection Assay (RPA)
The RPA was performed with the RPAII kit using biotinylated antisense
riboprobes (BIOTINscript and BrightStar system, Ambion). Biotinylated antisense
riboprobes from the Macs (mouse gene coding for MARCKS) and the cyclophilin genes
were generated and gel-purified A fragment of Macs was cloned by RT-PCR into a TA
cloning vector (InVitrogen, Carlsbad, CA) and restricted at a unique Xmal site. In vitro
transcription using SP6 RNA polymerase and this construct as the template yields a 466 nt
riboprobe and protects a 386 bp fragment of MARCKS mRNA. The cyclophilin probe
(Ambion) protects a 103 bp fragment. For each sample, 15 ng of RNA was incubated
with 1 ng of each probe at 44°C overnight Non-hybridized RNA was digested with a
mixture of RNases. Protected fragments were separated on a denaturing 5% PAGE and
then transferred onto a nylon membrane to allow detection. Resulting autoradiographs
were then quantified and the amount of Macs expression was normalized to the cyclophilin

24
internal control. Sharlynn Sweeney is acknowledged for her expertise in development of
the probes, as well as her technical skill in carrying out the RPAs and quantitating results.
PKC Inhibitors
HN33 cells were exposed to the PKC inhibitor LY333531 at 1.0 nM (Eli Lilly and
Company) concomitantly with 1 mM VPA for 24 h. Alterations in VPA-induced
MARCKS down-regulation and GAP-43 up-regulation, as well as cell density and
morphology, were measured and compared between control groups and LY333531-
treated groups, each in the presence or absence of VPA. PC 12 cells were exposed to the
PKC inhibitor bisindolylmaleimide (BIM; Gibco, Gaithersburg, MD) at 200 nM for 48 h,
and alterations in cell density and morphology were quantitated and compared between
treatment groups.
PKC Activity
PKC activity was measured by the phosphorylation of myelin basic protein (MBP)
4-14 peptide (Upstate Biotechnology, Lake Placid, NY) and MARCKS peptide. Soluble
and membrane cell fractions were prepared as described above. Ten microliters of the cell
protein (1 ng/|d) were added to 100 gl of a reaction mixture containing 100 MBP, 100
nM CaCl2, 20 mM HEPES (pH 7.5), 0.03% Triton XI00, 10 mM MgCl2, 100 ^M ATP,
and 0.0228 |iCi/|d [y-32P]ATP, with or without 100 |ig/ml phosphatidylserine and 20 gg/ml
PMA. The mixture was then incubated at 30° C for 2 min. The reaction was stopped by
the addition of 50 ^1 of ice-cold 450 mM phosphoric acid. Following a 5 min incubation
on ice, 30 |d of each sample was spotted onto Whatman P81 filter paper. Filters were
washed (x4) with a mixture of 150 mM H3PO4 and 10 mM sodium pyrophosphate, rinsed
in ethanol and then acetone, then dried and radioactivity measured. Specific activity for

25
PKC was measured as nanomoles of’2P/min/mg protein, and expressed as the difference
between activity in the presence and absence of PMA/phosphatidylserine. Dr David
Watson is acknowledged for his contribution to the development of the PKC activity assay
and his assistance in carrying out the experiments.
Statistical Analyses
Following consultation with a biostatistician, statistical differences between
treatments were analyzed by one-way ANOVA, and post hoc comparisons were
conducted using Fisher’s PLSD test In all cases, minimal acceptable level for significance
was p<0.05. All results are reported as mean percentage of an untreated control group
assayed in parallel, and error bars represent S.E.M Drs. Ronald Marks, Robert
McNamara, and Mark Lewis are acknowledged for their assistance in identifying valid
statistical methods for analyzing these data.

Table 2.2. Antibodies for Western Blots This table lists the primary antibodies utilized in western immunoblots to detect
each of the proteins of interest For each antibody, the following information is reported: source from which antibody was
obtained, dilution (in tris-saline, or TS, buffer) at which membrane was exposed to antibody, length of exposure of
membrane to antibody, and secondary HRP-conjugated antibody used for amplification and detection of primary antibody,
its source, dilution (in TS buffer + 0 2% Tween-20), and exposure time The anti-MARCKS polyclonal primary antibody
was raised against a glutathione-S-transferase MARCKS fusion protein designed in our laboratory, as detailed in Watson
and Lenox (1996).
1° Ab
Source
Dilution
Incubation
2° Ab
Source
Dilution
Incubation
MARCKS
(custom made)
1:36,000
overnight
anti-rabbit
Bio-Rad
1:24,000
2 h
GAP-43
Boehringer-
Mannheim
1:3,000
overnight
anti-mouse
Bio-Rad
1:20,000
2 h
PKC-a
Transduction Labs
1:1,000
2 h
anti-mouse
Bio-Rad
1:10,000
1 h
PKC-8
Santa Cruz
1: 750
2 h
anti-rabbit
Bio-Rad
1:20,000
1 h
PKC-e
Santa Cruz
1:1,000
2 h
anti-rabbit
Bio-Rad
1:20,000
1 h
PKC<
Transduction Labs
1:1,000
2 h
anti-mouse
Bio-Rad
1:20,000
1 h

CHAPTER 3
EFFECTS OF VPA ON MARCKS AND GAP-43
Introduction
VPA has traditionally been used for its antiepileptic properties, which were first
discovered in 1963 by Meunier el al., and encompass a broad range of seizure disorders.
Of late, VPA has been recognized as a suitable alternative to lithium, the traditional first-
line drug, for both the acute and prophylactic treatment of bipolar disorder. VPA and
lithium are currently the only two FDA-approved anti-manic drugs. Although lithium has
been studied more extensively and will serve as the model upon which experiments in this
study are based, the mechanism of anti-manic and mood-stabilizing action of this drug
remains unclear, as does the etiology of the psychiatric disorder targeted by these drugs.
Therefore, investigations into the mechanism(s) underlying the mood-stabilizing activity of
both drugs is an important area of study.
MARCKS
Studies conducted in our laboratory, using both animal and cell models, have
provided evidence for a significant down-regulation of MARCKS following chronic but
not acute exposure to therapeutically relevant concentrations of lithium (Lenox el a!.,
1992; Watson and Lenox, 1996). In immortalized hippocampal cells, Watson and Lenox
(1996) first demonstrated that long-term (3-10 d) exposure to 1-10 mM LiCl resulted in a
reversible, time- and concentration-dependent down-regulation of MARCKS protein.
27

28
Similarly, in vivo studies demonstrated that exposure of rats to chronic lithium (3-4
weeks), resulting in clinically relevant serum concentrations of 0.7-1.2 mEq/kg in brain,
also reduced MARCKS in the hippocampus (Lenox el al., 1992). Additionally, recent
data from our laboratory suggest a time course of MARCKS mRNA regulation by lithium
in HN33 cells similar to the time course of MARCKS protein down-regulation (with
reductions evident by 3 d and persisting beyond 10 d), thereby indicating alterations in
either transcription or post-transcriptional message stability (Watson and Lenox, 1997).
MARCKS expression is widely distributed, with highest levels normally expressed
in brain, spleen, testis, and lung, and lower expression in heart, kidney, liver, and muscle
tissues of mouse (Lobach el al., 1993; Swierczynski el al., 1996). In the brain, the
greatest expression is in the most plastic regions, including the amygdala, striatum,
olfactory bulb and cortex, and septum (Ouimet el al., 1990; McNamara and Lenox, 1997),
and MARCKS is localized to small dendritic branches and axon terminals of neurons and
both cytoplasmic and membranous components of glia (Katz el al., 1985; Patel and
Kligman, 1987; Ouimet el al., 1990; Rosen el al., 1990). MARCKS cross-links actin and
binds calmodulin in a Ca^-dependent manner (these events are mutually exclusive; Sheu
el al., 1995; Allen and Aderem, 1995), and as a result, has been implicated in cell motility,
secretion and trafficking, and cell cycle regulation (for reviews, see Aderem, 1992 and
Blackshear, 1993). As is indicated by its name, the activity and expression of MARCKS
are regulated in part through phosphorylation by PKC, though PKC-independent
pathways may also play a role in its post-transcriptional regulation (Brooks et al., 1992).
The inhibition of PKC or mutation of the PKC phosphorylation sites of MARCKS results
in defects in cellular motility (Allen and Aderem, 1995).
MARCKS is a protein for which the precise function is unclear but which is
believed to be essential to normal CNS development, since mice lacking detectable mRNA
and protein, due to gene (Macs) knockout, died perinatally, either before or within just
hours after birth (Stumpo et a!., 1995). Developmental defects included defects in midline

29
events such as neurulation, fusion of the cerebral hemispheres, formation of the great
forebrain commissures, and retinal and cortical lamination (Stumpo el al., 1995).
Ultrastructural examination revealed widespread neuronal ectopia, or abnormal neuronal
pathfinding, during cerebrocortical development (Blackshear et al., 1996). Whereas these
findings do not preclude the possibility of more subtle ultrastructural or immunological
defects in other non-neuronal tissues in which MARCKS is expressed, phenotypic
alterations were not observed outside the CNS, and the data overwhelmingly support the
notion that the mortality is primarily a result of the high frequency of midline defects in
brain resulting from the MARCKS-deficient phenotype. Interestingly, this knockout
phenotype was "rescued" by the expression of a supplemental human MARCKS gene,
MACS, in that the human transgene contained all of the elements necessary for normal
developmental expression of MARCKS, and was able to fully complement the MARCKS-
deficient phenotype of the knockout mice (Swierczynski et a/., 1996).
The purpose of the following experiments is to investigate the effects of VP A on
MARCKS expression, since significant effects were reported following chronic lithium
exposure, as noted previously, and any common action of these two mood-stabilizing
agents should be instructive as to their mechanisms of action. If MARCKS alterations are
integral to the mood-stabilizing properties of these and other drugs, one would
hypothesize that VPA should down-regulate MARCKS in a manner similar to that of
lithium. Both concentration-response and time-course curves of VPA-induced effects will
be established. These parameters address the therapeutic relevance of observed effects,
since each drug exhibits its own clinical profile, including such characteristics as
therapeutic serum concentration and latency to response. Previous studies have reported
that the effects of lithium on MARCKS protein expression occur following chronic but not
acute exposure to concentrations within the reported therapeutic serum concentration
range of this drug, and that the observed down-regulation of MARCKS expression is
reversible upon drug withdrawal These findings are consistent with the known clinical

30
profile of lithium Similar results are hypothesized following VPA exposure, since this
drug exhibits similar clinical properties of delayed onset of antimanic drug action and
delayed return to baseline following drug removal, over a therapeutic concentration range
(50-125 ng/ml). The long-term effects of VPA will be the focus of these studies, for they
may be important to the mood-stabilizing efficacy of this drug, whereas effects observed
more acutely are generally credited to the anticonvulsant properties of VPA.
As mentioned in Chapter 1, accumulating evidence implicates receptor-mediated
PI signaling in the mechanism of action of lithium in the brain (Berridge el a/., 1982;
Godfrey, 1989; Lenox and Watson, 1994). Lithium is an uncompetitive inhibitor of the
enzyme IMPase, and as such, it leads to accumulation of intermediates in the pathway
responsible for additional downstream effects, including intracellular Ca-^ release and
DAG-mediated PKC activation (refer to Figure 1.2; Post el al., 1992). It is hypothesized
that depletion of free inositol, resulting from the accumulation of intermediates due to
incomplete cycling of the PI signaling pathway, is a major consequence of long-term
lithium exposure (Berridge el a/., 1982; Berridge, 1989). Such a phenomenon would have
its greatest impact on certain areas of the brain which are relatively more inositol-limited,
for example by virtue of the blood-brain barrier or areas less capable of producing inositol
de novo. Further, because the PI signaling cascade is linked to muscarinic receptor/Gq
protein activation, areas of the brain undergoing the highest rate of receptor activation
should be preferentially affected by lithium. Indeed, previous studies in our laboratory
lend support to these hypotheses (Watson and Lenox, 1996). Firstly, in studies of FIN33
cells exposed to lithium, inositol supplementation (at concentrations as low as 5 ^M) was
able to prevent the down-regulation of MARCKS induced by 1 mM lithium, thereby
supporting the inositol-depletion hypothesis. Secondly, addition of the muscarinic
receptor agonist carbachol (1 mM) increased the MARCKS down-regulation observed
(Watson and Lenox, 1996), which supports the notion that the action of lithium is
augmented in cells or tissue regions undergoing the highest rate of receptor activation.

31
Whereas the mechanism of action of lithium has often been attributed to this
modulation of PI signaling and its downstream effects, there is little evidence in support of
a similar role for VP A; in fact, studies have indicated that VPA is probably not working
through effects on PI signaling. In one study comparing the effects of lithium,
carbamazepine, or VPA on IMPase activity, lithium and carbamazepine were found to
decrease and increase IMPase activity, respectively, and VPA was shown to have no
effect, at concentrations ranging from 2 jjM to 5 mM (Vadnal and Parthasarathy, 1995).
In another study, 2 mM valproate failed to increase accumulation of inositol phosphates,
including the inositol monophosphates, inositol 1,3-bisphosphate, inositol 1,4-
bisphosphate, and inositol 1,3,4-trisphosphate, whereas lithium exposure resulted in
accumulation of all (Dixon and Hokin, 1997). In a study conducted in our laboratory,
VPA failed to result in accumulation of CMP-PA (also known as CDP-DAG; see Figure
1.2), a metabolite of DAG and a sensitive measure of PI signaling, whereas lithium
resulted in significant accumulation of CMP-PA in CHO-K1 and HN33 cells (Watson et
a/., 1998). These data suggest that inhibitory effects on IMPase may not be the common
site of action for mood-stabilizing agents. However, this does not dismiss the possibility
that such agents may act indirectly or on alternate locations to alter the PI signaling
cascade. Experiments discussed herein have been modeled after those described above for
lithium, and will serve to clarify further the relationship between inositol or carbachol and
VPA-induced effects. Based on data from previous studies showing a lack of influence by
VPA on the PI signaling cascade, it is hypothesized that neither inositol nor carbachol will
alter any observed effect of VPA on MARCKS protein expression.
As mentioned previously, attention will be given to the possible therapeutic
relevance of findings. Inasmuch as lithium and VPA are commonly administered in
combination to patients who fail to respond adequately to either drug alone, often
resulting in amelioration or attenuation of manic symptoms, one study has been designed
to test the hypothesis that there may be additive actions at the cellular level. Following

32
optimization of exposure conditions for each drug, cells will be exposed to the
combination of VP A and lithium, and effects on MARCKS protein assessed. A final study
involves exposing cells to a series of other psychotropic agents individually, in order to
assess the specificity of action of mood-stabilizing agents and serve as a control for other
studies conducted. If MARCKS mediates any of the mood-stabilizing properties of these
agents, one would hypothesize that a variety of drugs efficacious in the treatment of
bipolar disorder would elicit the MARCKS down-regulation observed following lithium
exposure.
Experiments described thus far have simply focused on alterations in MARCKS
protein expression following exposure to drug and manipulation of various other
conditions in vitro However, little is known of how these changes come about, or how
MARCKS levels are regulated in vivo. Regulation of MARCKS expression is thought to
occur in several ways: (1) tissue-specific expression, (2) developmentally regulated
expression, (3) differentially regulated transcription rates, and (4) alterations in mRNA
levels in response to PKC (Blackshear, 1993; Lobach el a/., 1993). VPA-induced effects
on PKC itself will be discussed in detail in Chapter 5. Accumulating evidence shows that
lithium and VPA have regulatory effects on various transcription factors, thereby
suggesting a role for these mood stabilizers in altering gene expression. Studies in
transfected cultured cells have indicated that VPA enhances expression of genes and
reporters which contain AP-1 elements in their regulatory domains (Simon el al., 1994;
Kuntz-Simon el al., 1995). Manji and colleagues showed further that VPA exposure
resulted in a time- and concentration-dependent increase in DNA binding activity of the
transcription factor AP-1 in both rat C6 glioma and human neuroblastoma (SH-SY5Y)
cells; lithium reportedly induced similar effects (Chen et al., 1997).
It is unknown how MARCKS expression is regulated beyond transcription, or how
the lithium-induced down-regulation of MARCKS protein occurs. Expression of many
proteins is known to be regulated not only by transcriptionally, but also by post-

33
transcriptional mechanisms, so that changes in mRNA stability play a crucial role in the
control of GAP-43 expression, for example Similar post-transcriptional mechanisms
might also be at work in the case of MARCKS, and it is the purpose of these experiments
to assess further this possibility. Preliminary studies in our laboratory have indicated that
following lithium exposure, MARCKS mRNA levels do decline along a time course
similar to that of the protein down-regulation (Watson and Lenox, 1997), and one would
expect the same for VPA-induced alterations in MARCKS mRNA: a down-regulation of
mRNA levels over time, at therapeutically relevant concentrations of VP A, and with a
delayed onset of effect and protracted recovery of expression upon VPA withdrawal.
GAP-43
The growth-associated protein, GAP-43, is another major PKC substrate in brain
which shares many similarities with MARCKS. Like MARCKS, GAP-43 binds both
calmodulin and filamentous actin, albeit in a Ca^-independent manner (reviewed in
Benowitz and Routtenberg, 1997), and may play a role in exocytosis (Dekker et al.,
1989). GAP-43, like MARCKS, appears to be integral to normal nervous system
development, because studies of mice lacking the protein as a result of antisense-mediated
gene knockout show a high frequency of early postnatal death, with few animals surviving
past three weeks of age and even fewer reaching maturity (Strittmatter el al., 1995).
Those mice that do survive exhibit abnormal neuronal pathfinding.
GAP-43 is a rapidly transported, presynaptic protein most highly expressed during
development and post-injury regeneration, and immunohistochemistry and in situ
hybridization studies have shown GAP-43 expression to be almost exclusively neuronal
(Benowitz and Routtenberg, 1997). Neurite outgrowth is generally accompanied by a
large increase in GAP-43 expression, and following differentiation and synaptogenesis,
GAP-43 levels decline sharply and remain low, except in the most plastic regions of the

34
brain, such as associative and limbic areas, where intense immunostaining persists in adults
throughout life. This selective expression is thought to confer the unique potential for
ongoing structural remodeling (Benowitz and Perrone-Bizzozero, 1991).
GAP-43 is associated with the cytoskeleton and is concentrated in the growth
cones of elongating axons and nerve terminals, thought to help stabilize growth cone
morphology and promote growth cone adhesion (Aigner and Caroni, 1995). Support for
this notion has been provided by a number of studies. Antisense-directed depletion of
GAP-43 in growth cones results in the absence of spreading, branching, and adhesion of
growth cones (and therefore, neurite outgrowth) in primary sensory neurons (Aigner and
Caroni, 1993). Whereas suppression of GAP-43 expression has been found to yield
adverse effects on axon outgrowth both in vivo and in vitro (Meiri et a/., 1998; Benowitz
and Routtenberg, 1997; Strittmatter et al., 1995; Jap Tjoen San et al., 1992; Shea et al.,
1991), overexpression of the protein induced nerve sprouting in the adult nervous system
of transgenic mice (Aigner et a/., 1995). Even in non-neuronal cells, expression of a
GAP-43 transgene has been found to induce extensive process outgrowth accompanied by
reorganization of the membrane cytoskeleton (Strittmatter et al., 1994; Yankner et a!.,
1990; Zuber et al., 1989), yet, a mutant PC 12 cell line which does not express GAP-43
still exhibits neurite outgrowth (Baetge and Hammang, 1991).
Although studies are in disagreement as to the precise role of GAP-43 in these
events, it has been hypothesized that the almost exclusive presence of GAP-43 in neuronal
cells, and the persistence of GAP-43 expression in highly plastic limbic regions, confers
the ability to undergo the structural remodeling necessary for the functional plasticity of
the brain (Benowitz and Perrone-Bizzozero, 1991). Indeed, Skene and Willard (1989)
suggest that it is the induction of GAP expression that is necessary to post-injury
remodeling, and that the inability of many areas of the CNS to induce high levels of GAP
expression may in part underlie the failure of CNS axons to regenerate after axonal injury,
whereas areas of the peripheral nervous system retain the ability to up-regulate GAP-43 as

35
well as to regenerate axons after injury. However, the presence of GAP-43 is probably
not necessary to neurite outgrowth per se, but rather, plays an important role in regulating
precisely the various parameters involved in the overall event, such as navigation or
pathfinding (Aigner and Caroni, 1995). According to this “GAP hypothesis”, nerve
growth is controlled in part by the expression of certain growth associated proteins, such
as GAP-43, which provide a mechanism for regulating neurite outgrowth and structural
plasticity in the nervous system (Skene and Willard, 1989).
Because of the vital role of GAP-43 in effecting plastic changes in brain,
experiments have been designed to address the efifect(s) of VP A exposure on GAP-43
expression in HN33 cells. It may be that part of the efficacy of VP A in either bipolar
disorder or epilepsy, or both, lies in its ability to elicit structural modifications of plastic
regions exhibiting the pathology which leads to the manifestation of the particular
disorder. Indeed, Benowitz and Routtenberg (1997) pointed out that in brain of humans
and other primates, the highest levels of GAP-43 expression occur in regions known to be
involved with higher-level associative processes, lending support to the notion that GAP-
43 is important for maintaining plasticity.
Whereas in vitro studies in HN33 cells have shown no evidence for regulation of
GAP-43 by lithium, neither has lithium been shown to alter the morphology of these cells.
Preliminary studies have suggested that chronic VPA exposure induces a morphological
alteration in these cells which might be further characterized as differentiation (see
Chapter 5). Since neurite outgrowth and neuronal differentiation are known to be
associated with up-regulation of GAP-43 protein expression, it will be of interest to
investigate the effect of VPA on this protein.
The results of these studies will be instructive, for in addition to the development
of a model which may help explain the role of MARCKS and other plasticity proteins
(such as GAP-43) in the mechanism of action of both VPA and lithium, findings will
contribute to a better understanding of the pathophysiology involved in bipolar disorder.

36
The notion of synaptic remodeling as the underlying means for the drug- or therapy-
induced stabilization of mood disorders has been supported by some reports. One study
found increased levels of neural cell adhesion molecule (N-CAM)-immunoreactive
proteins in the cerebrospinal fluid of a population of bipolar patients (Poltorak et al.,
1996). Abnormal function or turnover of N-CAM, which plays a role in synaptogenesis,
morphogenesis, and plasticity of the nervous system, may be suggestive of abnormal
synaptic plasticity in affected individuals. With regard to drug therapy for this disorder,
not only do lithium and VPA affect expression of proteins involved in neural plasticity and
cellular response (as previously discussed), but both drugs have also been shown to have
effects on transcriptional factors which regulate gene expression, and it is known that
long-term changes in neuronal synaptic function are dependent upon the selective
regulation of gene expression (Manji and Lenox, 1994). Lamotrigine, another drug used
in treatment of bipolar disorder, was recently shown to alter neuronal hyperexcitability via
inhibition of both Na and Ca channels, findings which were interpreted by the authors to
signify modification of synaptic plasticity (Xie and Hagan, 1998). Such evidence is
indirect, but overall suggestive that changes in synaptic plasticity may be a mechanism by
which anti-bipolar (anti-manic and mood-stabilizing) therapy facilitates long-term changes
in affect.
Á

37
Results
A
*>**&#•
12 12
Figure 3.1. MARCKS protein. Shown are Western
immunoblots comparing MARCKS protein expression in
(1) mouse brain and (2) HN33 cells. A shows cytosolic
MARCKS expression, and B shows membrane-
associated MARCKS
Comparative MARCKS expression in HN33 cells (used in this study) vs. mouse
brain is shown in Figure 3.1 following enhanced chemiluminescent (ECL) detection.
Differences in apparent molecular weight may result from differential phosphorylation or
myristoylation of the protein Note the more pronounced MARCKS protein expression in
mouse brain, even with comparable amounts of total loaded protein.
Effect of VPA Exposure on MARCKS Protein in HN33 Cells
VPA was administered at concentrations ranging from 0.001 to 1.0 mM, for 1 or 3
d (Figures 3.2 and 3.3, respectively). Representative Western blots are presented, along
with quantitation of mean data. Significant concentration- and time-dependent effects
were observed. Chronic (3 d) VPA exposure results in a concentration-dependent
reduction in MARCKS levels, with a more pronounced reduction in membrane than

38
soluble fraction of these cells. Significant down-regulation of MARCKS is observed in the
membrane fraction at concentrations as low as 0.06 mM (p<0.005), and in the soluble
fraction at 0.1 mM and higher concentrations (p<0.005). These concentrations correlate
well with the target therapeutic serum concentration range used clinically for treatment of
bipolar disorder.
a
0 000 00
|VPA|, mM
Figure 3.2. Concentration-dependent down-regulation of MARCKS protein in
HN33 cells following 1 d VPA exposure, (a) Representative Western blot of
membrane-associated MARCKS expression following 1 d VPA exposure; mean data
are depicted graphically in 3.2b. Lanes 1-7 were treated for 1 d as follows:
untreated control, 0.01 mM, 0.03 mM, 0.1 mM, 0.3 mM, 0.6 mM, and 1.0 mM
VPA, respectively, (b) This figure shows quantitatively the MARCKS protein levels
in HN33 cells following 1 d exposure to varying concentrations of VPA. Data,
expressed as percent of MARCKS compared to untreated control cells grown in
parallel, were derived from Western immunoblots (e.g., Figure 3.2a) and are the
mean of at least four determinations. Error bars represent S .E.M.

39
©
140
120 -
$
o
100 -
£
a
80 -
3
©
l-c
60 -
Cm
in
w
40 -
u
06
<
20 -
T
1
I
T
I
1
i
I
in
1
I
1
T
1
I
1
T
i
â–¡ Soluble
Membrane
* *
1
T
i
0.001 0.01
0.03 0.06
0.1
0.3
0.6
[VPA|, mM
Figure 3.3. Concentration-dependent down-regulation of MARCKS protein in HN33
cells following 3 d VPA exposure (a) Representative DAB Western blot of MARCKS
expression following 3 d VPA exposure. Lanes 1-6 show soluble MARCKS following
3 d treatment with 0, 0.03, 0.06, 0.1, 0.3, and 0.6 mM VPA, respectively, and lanes 7-
12 show membrane MARCKS following exposure to the same doses (b) This figure
shows quantitative data for mean MARCKS levels following 3 d VPA exposure. Data,
expressed as percent of MARCKS compared to untreated control cells grown in
parallel, were derived from Western immunoblots and are the mean of at least four
determinations. Error bars represent S.E.M. *p<0.005

40
Time Course of VPA-Induced MARCKS Down-Regulation in HN33 Cells
HN33 cells were exposed continuously to 0.6 mM VPA for periods ranging from 0
to 7 d. This concentration is equivalent to 100 ng/ml, which is within the reported
therapeutic range for the mood stabilizing action of VPA (50-125 ng/ml). Cells were
observed to undergo a significant down-regulation of MARCKS protein expression in
both fractions as early as 3 d, with no effect on MARCKS at 1 d (see Figure 3.4). The
MARCKS reduction in the membrane fraction is more pronounced than that in the
cytosol. Three d, 5 d, and 7 d values are significantly different (p<0.01) from 1 d
MARCKS values in both soluble and membrane fractions of these cells. These data are
consistent with the clinical profile of this drug, which exhibits a delay in onset of antimanic
action at target therapeutic concentrations.
Time (Days)
Figure 3.4. Time course of VPA-induced MARCKS down-regulation in
FTN33 cells. This figure illustrates the time-dependency of VPA-induced
MARCKS protein down-regulation. HN33 cells were exposed to 0.6 mM
VPA, a clinically relevant therapeutic concentration, for 1, 3, 5, or 7 d, and
MARCKS expression was assessed by quantitative immunoblot analysis.
Results, expressed as percent of untreated controls, represent the mean
± S.E.M. of at least four determinations. *p<0.01

41
b
2 3 4 5
Figure 3.5. Recovery of MARCKS following VPA-induced long-term down-
regulation. (a) This figure shows the time course of VPA-induced MARCKS down-
regulation in HN33 cells, as well as the recovery of MARCKS protein to basal levels
following drug withdrawal. HN33 cells were exposed to 0.6 mM VPA for 3 d, at
which time the drug was either removed or treatment continued. MARCKS levels
were assessed through Western immunoblot (Figure 3.5b is representative), and results
are expressed as percent of untreated control cultures grown in parallel. Results
represent mean ± S.E.M. of at least three determinations. *p<0.0001 (b) Shown is a
representative Western blot of MARCKS expression in the membrane fraction of HN33
cells following chronic VPA exposure and subsequent withdrawal. Lanes 1 and 3 show
control levels of MARCKS protein Lane 2 shows MARCKS expression following 3 d
VPA exposure (compare to Lane 1), and Lane 4 shows MARCKS expression
following 6 d VPA exposure (compare to Lane 3). Lane 5 represents a VPA
withdrawal/recovery group, and shows MARCKS in cells exposed continuously for 3 d
to 0.6 mM VPA, and then for 3 d to media free of drug.
Recovery of MARCKS following VPA-induced Long-Term Down-Regulation
As shown in Figure 3.5a, HN33 cells exposed to 0.6 mM VPA for 0 to 14 d
exhibited a significant down-regulation of MARCKS protein expression in both fractions

42
(membrane > soluble) over time, as previously demonstrated in Figure 3.4. After 3 d of
continuous VPA exposure and subsequent drug washout and withdrawal, MARCKS
protein levels recovered with significant increase (p<0.0001) to control levels within 3-7 d
of culture in drug-free medium (Figure 3.5a). A representative Western blot is shown in
Figure 3.5b. These findings are consistent with clinical evidence for delayed onset of anti-
manic action of VPA, as well as prolonged anti-manic efficacy following drug
discontinuation. It is difficult to discern whether this experimental return of MARCKS to
baseline levels is the result of either a shift in the kinetics of MARCKS
production/degradation, or due to repopulation of the in vitro environment with
MARCKS-replete daughter cells.
Effects of Carbachol and myoinositol on the VPA-Induced Down-Regulation of
MARCKS
In order to examine the role of PI signaling in the mechanism of action of this
drug's effects on MARCKS, cells were exposed to VPA alone or in combination with
either carbachol (1 mM) or myoinositol (up to 1 mM). In contrast to previous results
obtained with LiCl (Watson and Lenox, 1996), the VPA-induced down-regulation of
MARCKS protein expression was not altered by addition of carbachol (Table 3.1 and
Figure 3.6) to the culture medium. Similarly, supplementation of ffryo-inositol (Table 3.2
and Figure 3.7) to the culture medium at concentrations up to 1 mM did not modify the
VPA-induced down-regulation of MARCKS, whereas the LiCl-induced down-regulation
was prevented by the addition of as little as 5.0 pM myoinositol. Representative Western
blots are presented in Figures 3.6 and 3.7. It is difficult to rule out completely the role of
PI signaling based on the data in Table 3.1 alone, due to the maximal level of MARCKS
down-regulation achieved in the case of VPA. However, the data in Table 3.2 are more
conclusive, as the levels of MARCKS reduction elicited by VPA and lithium are more
comparable.

43
Table 3.1. Addition of Carbachol Fails to Potentiate the VPA-Induced MARCKS Down-
Regulation in HN33 Cells. HN33 cells were exposed to 0.6 mM VPA for 3 d or 1 mM
lithium for 7 d, in the presence or absence of the muscarinic agonist carbachol. Results
are presented as percent of untreated control cultures grown in parallel, and represent the
mean ± S.E.M. of at least three determinations. *p<0.0001
MARCKS Protein, % Control (Mean ± S.E.M.)
VPA-Exposed
LiCl-Exposed
[Carbachol]
Soluble
Membrane
Soluble
Membrane
0 mM
41.8 ± 6.6
17.0 ± 8.5
81.6 ± 4.5
90.2 ±3.1
1 mM
46.5 ±7.5
10.8 ± 2.0
66.6 ±6.1**
68.7 ± 5.9**
12 3 4
Figure 3.6. Effect of carbachol on VPA-induced MARCKS down-
regulation. Shown is a representative Western blot of MARCKS
expression following VPA exposure with or without the concomitant
addition of carbachol. Lane 1 shows membrane-associated MARCKS
levels in untreated control HN33 cells, Lane 2 shows MARCKS in 3 d
VPA-treated cells, Lane 3 shows MARCKS in carbachol-treated cells,
and Lane 4 shows MARCKS in cells treated concomitantly with VPA
and carbachol. Mean data are reported in Table 3.1.

44
Table 3.2. Inositol Supplementation Fails to Attenuate the VPA-Induced MARCKS
Down-Regulation in HN33 Cells. HN33 cells were exposed to 0.6 mM VPA for 3 d or 1
mM lithium (+1 mM carbachol) for 7 d, in the presence (or absence) of varying
concentrations of myo-inositol. Results are presented as percent of untreated control
cultures grown in parallel, and represent the mean ± S.E.M. of at least three
determinations (N.D. = not determined). *p<0.005
MARCKS Protein, % Control (Mean ± S.E.M.)
VPA-Exposed
LiCl/carbachol-Exposed
[myoinositol]
Soluble
Membrane
Soluble
Membrane
0
83.0 ±6.0
57.0 ±5.0
66.6 ±6.1
68.7 ±5.9
0.5 pM
N D
N.D.
71.1 ±4.8
68.4 ±4.4
5 pM
N.D.
N.D
97.9 ±3.6*
104.3 ±5.3*
40 pM
77.8 ±2.4
54.8 ±5.5
98.5 ±7.5*
103.6 ±8.2*
1 mM
78.3 ±9.4
51.3 ± 10.8
N.D.
N.D.
12 3 4
Figure 3.7. Effect of inositol on VPA-induced MARCKS down-
regulation. Shown is a representative Western blot of MARCKS
expression following VPA exposure with or without concomitant
addition of myoinositol. Lanes 1-4 show membrane-associated
MARCKS in (1) control cells, (2) 3 d 0.6 mM VPA-treated cells,
(3) inositol-treated cells, and (4) combined VPA- and inositol-
treated cells. Mean data are reported in Table 3.2.

45
VPA LiCl/Carbachol VPA + LiCl/Carbachol
Treatment
b
A
B
1 2 3 4 5 6 7
mm
1 2 3 4 5 6 7
Figure 3.8. Additive effect of combined VPA and lithium exposure on MARCKS
protein expression, (a) HN33 cells were exposed to 0.6 mM VPA and/or 1 mM
lithium (+ 1 mM carbachol) in inositol-free DMEM for 3 d MARCKS protein
levels, expressed as percent of untreated control, were assessed by Western
immunoblot analysis (see Figure 3.8b), and values represent mean ± S .E .M. of at
least three determinations. *p<0.01 (b) Representative Western blot of MARCKS
protein following 3 d exposure of cells to VPA and/or lithium. A shows cytosolic
MARCKS in HN33 cells after the following treatments: no drug (1); 0.3 mM VPA
(2); 0.6 mM VPA (3); 1 mM lithium (4); 1 mM lithium + 1 mM carbachol (5); 1
mM lithium + 1 mM carbachol + 0.3 mM VPA (6); and 1 mM lithium + 1 mM
carbachol + 0.6 mM VPA (7). B shows MARCKS in the membrane fraction of
HN33 cells following the same treatments for 3 d.
Additive Effect of Combined VPA and LiCI Exposure on MARCKS Protein
Expression in HN33 Cells
HN33 cells were exposed to 0.6 mM VPA and/or 1 mM LiCl/1 mM carbachol for
a period of 3 d, in order to assess the effects of the VP A-lithium combination on
MARCKS protein expression and distribution. Results, presented in Figure 3 8a, show
that a greater reduction of MARCKS protein was observed in both soluble and membrane

46
fractions of cells exposed to both VPA and lithium/carbachol than was produced by either
VPA or lithium/carbachol alone. The enhanced reduction in MARCKS protein produced
by the combination of VPA and lithium/carbachol at these concentrations appeared to be
additive. Furthermore, the reduction in MARCKS protein produced by the combined
treatment of VPA and lithium/carbachol was greater in the membrane fraction than in the
soluble fraction. A representative Western blot is shown in Figure 3.8b.
Effects of Other Psychotropic Agents on MARCKS Protein Expression
Eight structurally and functionally distinct psychotropic drugs (carbamazepine,
fluoxetine, imipramine, phenytoin, diazepam, haloperidol, morphine, and verapamil) were
administered to HN33 cells for 1, 3, or 7 d, and their effects on MARCKS protein
expression were assessed. Drugs were chosen to represent a wide range of
psychopharmacological actions. Minimal drug concentrations were based on the
therapeutically effective plasma concentration for each drug in humans, as outlined in
Table 3.3. Additionally, cells were exposed to concentrations up to 20 times the highest
reported therapeutic concentration so as not to overlook potential effects at supra-
therapeutic doses. At each concentration tested, cell viability, growth rate, and gross
morphological appearance of HN33 cells exposed to each of the psychotropic drugs were
not notably different than those of cells grown in the absence of drug. Data obtained
following 3 d exposure are shown in Figure 3.9a. For all of the drugs listed, no significant
dose effect on MARCKS protein expression was measured at any of the concentrations or
time points tested.
MARCKS protein levels in HN33 cells were also measured after exposure to each
of two VPA analogues or metabolites, hydroxyvalproic acid (HVPA, or 2-n-propyl-5-
hydroxy-pentanoic acid) and 2-n-propylglutaric acid (2-PGA), for 3 or 7 d (Table 3.3;
Figure 3.9a,b) These VPA metabolites, the structures of which are depicted in Figure
3.9c, were provided by Abbott Laboratories (Abbott Park, IL), and have been found to

47
have little or no anticonvulsant efficacy in vivo (Chapman et al., 1982; J. Sullivan,
personal communication). Cells were exposed to concentrations of 0.6, 1.5, and 3.0 mM.
Unlike VP A, which produced noticeable changes in cell morphology at 0.6 mM (see
Chapter 6), these analogs produced no such change at this or higher concentrations.
Following HVPA exposure, no significant reduction in MARCKS protein was evident in
the soluble fraction, even at concentrations 5-10 times higher than the minimal effective
concentration of VPA. However, in the membrane fraction, a significant reduction in
MARCKS protein was observed only in cells exposed to 3.0 mM HVPA (p<0.01 at 3 d;
p<0.05 at 7 d). At this highest concentration tested (3.0 mM; 5X the therapeutic level of
VPA), a 30-35% reduction in MARCKS protein was observed following 3-7 d exposure.
Following 2-PGA exposure, no significant reduction in MARCKS protein was apparent in
the soluble fraction, even at concentrations 5-10 times higher than the minimal effective
concentration of VPA. A small but insignificant reduction in MARCKS protein in the
membrane fraction was observed following 3.0 mM exposure to 2-PGA for 3-7 d. A
representative Western blot is presented in Figure 3.9b.
140
T reatment

48
123456789 123456789
C
Figure 3.9. Effects of other psychotropic agents on MARCKS protein expression.
(a) This figure shows the relative lack of effect of various other psychotropic agents
on MARCKS protein expression in HN33 cells. Cells were exposed to drugs other
than VPA or lithium (see Table 3.3 for details), as follows: 3.0 mM hydroxyvalproic
acid, 7 d; 3.0 mM 2-propylglutaric acid, 7 d; 100 pM carbamazepine, 7 d; 1 pg/ml
fluoxetine, 3 d; 1 pg/ml imipramine, 7 d; 20 pg/ml phenytoin, 3 d; 1 pg/ml
diazepam, 7 d; 20 ng/ml haloperidol, 7 d; 20 pg/ml morphine, 7 d; 50 ng/ml
verapamil, 7 d MARCKS levels were assessed as previously described; data
represent mean ± S.E.M. of at least three determinations. *p<0.05 (b) MARCKS
expression following exposure to VPA and two metabolites. Shown is a
representative Western blot of MARCKS protein in HN33 cells after 7 d exposure to
the following: no drug (1); 0.6 mM VPA purchased from Sigma Chemical Company
(2); 0.6 mM VPA provided by Abbott Laboratories (for comparison purposes; 3); 0.6
mM (4), 1.5 mM (5), and 3.0 mM (6) 2-propylglutaric acid; 0.6 mM (7), 1.5 mM (8),
and 3.0 mM (9) hydroxyvalproic acid. A shows cytosolic MARCKS expression, and
B shows membrane-associated MARCKS (c) Structural comparisons of VPA and
metabolites. Shown are the chemical structures of VPA and two of its metabolites,
hydroxyvalproic acid, or 2-n-propyl-5-hydroxy-pentanoic acid, and 2-n-propylglutaric
acid.

49
Table 3.3. Drugs, Sources, and Exposure Conditions for HN33 Cells.
This table lists the drugs to which HN33 cells were exposed, their sources, target
circulating (plasma) concentrations in humans, concentrations tested in HN33 cells, and
vehicle in which drug was suspended. Drug concentrations tested encompassed a broad
range, including supratherapeutic concentrations, in order to account for discrepancies,
such as those presented by variation in circulating serum vs. brain concentrations of drug
Drug / Source
Target Plasma
Cone, in Humans
Concentrations Tested
Vehicle
Sodium Valproate
Sigma #P-4543
Abbott #A-44089.5
50-125 ng/ml
= 0.3-0 9 mM
(seizure control)3’11
0.01, 0.03, 0.06,
0.1,0.3,0 6, 1.0 mM
dH20
2-Propylglutaric Acid
Abbott #A-49999.0
0.6, 1.5, 3.0 mM
dH20
Hydroxyvalproate, sodium salt
Abbott #A-49822.5
0.6, 1.5, 3.0 mM
dH20
Carbamazepine
Sigma #C-4024
8-12 ng/ml
= 33-50 nM
(seizure control)b’c
10, 25, 100 [M
ETOH
Fluoxetine-HCl
Lilly #110140L/F08083
varies
50, 200, 1000 ng/ml
dH20
Imipramine-HCl
Sigma #1-7379
100-300 ng/ml
(antidepressant)3
50, 200, 1000 ng/ml
dH20
Phenytoin, sodium salt
Sigma #D-0931
10-20 ng/ml
(anticonvulsant)11’3
1, 10, 20 ng/ml
ETOH
Diazepam
Sigma #D-0899
300-400 ng/ml
(anxiolytic)3’11
>600 ng/ml
(seizure control)11
50, 200, 1000 ng/ml
ETOH
Haloperidol
Sigma #H-1512
4-20 ng/ml
(anti-psychotic)b
5, 20, 100 ng/ml
dH20
Morphine Sulfate
Sigma #M-8777
5-20 ng/ml 5, 20, 100 ng/ml
(analgesic in non-tolerant patients)c
dH20
Verapamil-HCl
Sigma #V-4629
varies
50 ng/ml
ETOH
aHolford and Benet, 1995; bBenet and Williams, 1990; cWeiner, 1996; dPorter and
Meldrum, 1995.

50
Effect of VPA on MARCKS mRNA
HN33 cells were exposed to 0.6 mM VPA for 0 to 14 d, and MARCKS mRNA
levels were measured by RPA as described in Materials and Methods. As shown in Figure
3.10a, MARCKS mRNA levels decrease over time with continual VPA exposure
(p<0.05). A representative RPA is shown in Figure 3.10b.
a
Time (d)
b
m m
MARCKS
cyclophilin
A B C D E
F G H
J
Figure 3.10. Time course of VP A-induced MARCKS mRNA alterations, (a) This figure
shows the time course of VP A-induced alterations in MARCKS mRNA. HN33 cells
were exposed to 0.6 mM VPA for 0-14 d, and MARCKS mRNA levels were assessed by
RPA, with the aid of a cyclophilin internal control, as described in Materials and
Methods. Results for each treatment are calculated as a ratio of MARCKS to
cyclophilin, and then each value is expressed as a percentage (± S .E.M .) of untreated,
control samples prepared in parallel. *p<0.05 Shown in (b) is a representative RPA.
Upper band reflects MARCKS mRNA, and lower band reflects cyclophilin mRNA.
Treatments are as follows: A, control; B, 1 d VPA; C, 3 d VPA; D, control; E, 5 d
VPA; F, 7 d VPA; G, control; H, 10 d VPA; I, control; J, 14 d VPA.

51
Effect of VPA on GAP-43 Expression in HN33 Cells
GAP-43 expression in HN33 cells is compared to its expression in rat brain in
Figure 3.11. Following 3 d, 0.6 mM VPA exposure, GAP-43 levels in HN33 cells
increase, as shown in Figure 3.12. This increase is concentration-dependent, as shown in
Figure 3.13 (p<0.01). HN33 cells were exposed for 3 d to concentrations of VPA ranging
from 0.01 mM to 1.0 mM, and membrane-associated GAP-43 expression is shown (Figure
3.13).
A
B
Figure 3.11. GAP-43. Shown is a Western
immunoblot comparing GAP-43 expression in (1)
soluble and (2) membrane fractions of (A) HN33
cells and (B) rat brain.
1 2
Figure 3.12. Effect of chronic VPA exposure on GAP-43 expression
in HN33 cells. Shown is a representative Western blot of GAP-43
expression in the membrane fraction of HN33 cells following VPA
exposure. Lane 1 shows GAP-43 expression in untreated control
cells, and Lane 2 shows GAP-43 in cells treated for 3 d with 0.6 mM
VPA

52
[VPA], mM
Figure 3.13. Concentration-dependent increase in GAP-43 expression in HN33
cells following VPA exposure. HN33 cells were exposed for 3 d to varying
concentrations of VPA, and membrane-associated GAP-43 expression was
assessed by Western immunoblot analysis. Results represent the mean of two
or three determinations, and are expressed as the percent of GAP-43 protein
expressed in untreated control cultures grown in parallel. *p<0.01
Time (d)
Figure 3.14 Time course of VPA-induced GAP-43 increase in HN33 cells. HN33
cells were exposed to 0.6 mM VPA for times ranging from 1 d to 14 d, and GAP-
43 levels were assessed by Western immunoblot analysis. Results reflect the mean
of at least three determinations, and are expressed as percent of GAP-43 in an
untreated control. *p<0.05

53
Time Course of VPA-Induced GAP-43 Increase in HN33 Cells
HN33 cells were exposed to 0.6 mM VPA for 0-14 d; results are shown in Figure
3.14. Up-regulation of GAP-43 was observed in membrane and soluble fractions of cells
exposed to VPA for 5 d and 14 d, respectively (p<0.05). Because of the experimental
protocol, which entailed growing cells to confluency and then diluting them every 3-4 d,
the data may not reflect the full potential for GAP-43 induction, since each splitting and
re-seeding procedure involved trypsinization and potential disruption of cellular processes.
Discussion
HN33 cells were utilized as the model for all studies, because their morphological,
immunological, and electrophysiological characteristics resemble those of hippocampal
neurons in culture. HN33 cells were chosen as the in vitro cell model in which to
investigate multiple steps in the PKC/GAP-43/MARCKS pathway, because of their
relative homogeneity and the presence of signal transduction pathways of interest. This is
in contrast to the use of other models, such as primary neuronal cultures, which are more
neurochemically heterogeneous in nature. In primary cultures, for example, it would be
difficult to assess whether PKC and other messenger systems were affected in the same
neuron. The homogeneity of HN33 cells makes signaling research and results less
confounding
MARCKS
VPA exposure elicits both a time- and concentration-dependent reduction in
MARCKS protein expression. It has been well established that the clinical antimanic
response to VPA requires a lag period of days to weeks following drug administration

54
before onset of action, although loading strategies have proven useful in achieving faster
response (within 3 d) (Bowden et al., 1994; McElroy et al., 1989). Consistent with such
clinical observations, a significant reduction of MARCKS protein in both soluble and
membrane fractions of HN33 cells was not observed following acute (1 d) VP A
administration, except at levels above the maximal target therapeutic serum concentration
(1 mM). At 0.6 mM VP A, the down-regulation of MARCKS was greatest between d 0
and d 3, and little further reduction was observed upon continued exposure (5-10 d).
With regard to therapeutic concentrations, there have been reports of clinical response at
plasma concentrations of VPA as low as 45 |ig/ml (Bowden et al., 1994), which is
comparable to the observed effects of VPA on MARCKS down-regulation at
concentrations as low as 0.1 mM in this study. However, VPA is generally thought to be
most effective at therapeutic concentrations of 50-125 (ig/ml (0.3-0.75 mM), where the
most robust effect of VPA on MARCKS was observed in the HN33 cells Following
VPA-induced MARCKS down-regulation and subsequent drug withdrawal, MARCKS
levels in these cells recovered to baseline within 3-7 d This finding is consistent with the
third major parameter discussed with regard to clinical relevance, that of long-term effect
of VPA and delayed return of clinical response to baseline following treatment
discontinuation.
With regard to combined VP A/lithium exposure, a greater reduction of MARCKS
protein was observed in cells following exposure to the combination than was produced by
either VPA or lithium/carbachol alone, and the effects of the two drugs appear to be
approximately additive Whereas VPA induced a mean reduction in MARCKS expression
of 26% in the soluble fraction and 50% in the membrane, and lithium/carbachol elicited
mean reductions of 14% and 33% in soluble and membrane, respectively, the combination
of VPA plus lithium/carbachol resulted in a 51% drop in soluble MARCKS and a 67%
drop in membrane-associated MARCKS. These data are in agreement with clinical studies
which show VP A/lithium combination therapy to have greater therapeutic efficacy for

55
treatment of bipolar disorder than either agent administered individually (Calabrese and
Delucchi, 1989; Hayes, 1989). In contrast to lithium, which exerts a preferential down-
regulation of MARCKS in the soluble fraction of HN33 cells (Watson and Lenox, 1996),
the effect of VP A on MARCKS was most pronounced in the membrane fraction of these
cells. While the significance of this finding is unclear, the differences observed in the
patterns of MARCKS down-regulation between these two drugs may be a function of
differences in the pathways of PKC activation, as later discussed This is of particular
interest in light of the findings that concomitant exposure of cells to both VPA and lithium
appears to have an additive effect on MARCKS down-regulation at therapeutic
concentrations of both drugs, and such findings support the notion that the two drugs
operate through separate mechanisms.
In contrast to lithium, which appears to rely on relative depletion of /wyo-inositol
and receptor activation of PI signaling to optimize its effect on MARCKS in the
immortalized hippocampal cells, these manipulations had no effect on the VPA-induced
down-regulation of MARCKS in these cells. Addition of carbachol or inositol (even at
mM concentrations) to the culture medium had no effect on the VPA-induced reduction in
MARCKS, and these findings are consistent with the lack of effect of VPA on receptor-
coupled accumulation of CMP-PA in the CHO-K1 cells (Watson et al., 1998). CMP-PA,
also known as CDP-DAG, is a byproduct of DAG metabolism and precursor for PI
reformation, and serves as a sensitive measure of cycling through the PI signaling cascade
(refer to Figure 1.2). Thus, in spite of the shared property of lithium and VPA in down-
regulating the expression of MARCKS, these data suggest that the mechanism through
which this occurs may involve different pathways. Nonetheless, there appears to be
evidence for a role of PKC regulation in the action of both lithium and VPA, for previous
studies have demonstrated that phorbol esters down-regulate MARCKS protein in
neuronally derived cell populations in a PKC-dependent manner (Watson et al., 1994).
This will be further discussed in later chapters.

56
Two structural analogs of VPA (see Figure 3.9c), which are significantly less
potent as anti-convulsants in comparison to VPA (Chapman et a!., 1982), were studied.
Both analogs are known metabolites of VPA, although they are produced through minor
metabolic pathways and represent a minimal percentage of total brain plasma
concentration (Chapman et«/., 1982; Eadie, 1991). In previous studies, neither
hydroxyvalproic acid nor 2-propylglutaric acid was shown to be effective in preventing
pentylenetetrazol-induced seizures in mice (Chapman et a!., 1982; J. Sullivan, personal
communication), and neither agent has been tested clinically for efficacy in the treatment
of bipolar disorder. In our studies, hydroxyvalproic acid produced a statistically
significant effect in down-regulating membrane-associated MARCKS, but at a
concentration well above that observed for VPA (3.0 mM hydroxyvalproic acid vs. 0.1-
1.0 mM VPA). Comparatively, the 2-propylglutaric acid analog possessed little potency
in down-regulating MARCKS in either the membrane or cytosolic fraction. Future
analysis of structural differences among these VPA compounds may prove useful in
identifying structurally related fatty acids that may be more or less efficacious not only in
down-regulating MARCKS, but also potentially as mood stabilizers.
Exposure of immortalized hippocampal cells to structurally and functionally
diverse psychotropic agents, including carbamazepine, fluoxetine, imipramine, phenytoin,
diazepam, haloperidol, morphine, and verapamil, did not result in a down-regulation of
MARCKS, even at doses that were well above the therapeutic range. It is of particular
note that carbamazepine, another anti-convulsant used less frequently in the treatment of
manic patients, and verapamil, also used unconventionally and perhaps less efficacious
than the more established treatments (Janicak et a/., 1998; Walton et a/., 1996; Mathis et
a1988; Barton and Gitlin, 1987), did not have any effect on MARCKS expression.
Carbamazepine has been shown to have effects within the adenylyl cyclase cascade, a
property shared by lithium at higher concentrations (Manji et a/., 1995; Mork et a/.,
1992). However, carbamazepine has generally been shown to be less effective overall than

57
lithium or VP A as a mood stabilizer (Keck el«/., 1992; Bowden, 1996), and verapamil, a
calcium channel blocker, has yet to be studied extensively for this application.
Following exposure to a therapeutic serum concentration (0.6 mM) of VP A for up
to 14 d, HN33 cells exhibited decreased MARCKS mRNA levels, along a time course
similar to that of observed MARCKS protein alterations. This finding suggests that the
down-regulation of MARCKS protein might be attributed, at least in part, to a reduction
of its precursor mRNA template. With regard to MARCKS regulation, such evidence
may support more than one hypothesis. The observed reduction in MARCKS protein may
be the result of either altered transcriptional rates, or post-transcriptional regulation, via
decreased mRNA stability, decreased translation, or increased degradation. Yet, the
observed alterations in mRNA expression support the notion that MARCKS is being
regulated at the level of mRNA synthesis or stability. Recent studies in our lab have
shown that lithium, while altering both MARCKS mRNA and protein levels, did not affect
mRNA stability (Watson and Lenox, 1997); this suggests that the lithium-induced
alteration in MARCKS mRNA is primarily attributable to effects on transcription of the
MARCKS gene
These studies suggest that MARCKS may serve as a target in the brain for the
selective action of mood stabilizing drugs. Although the therapeutic action of these drugs
can probably not be attributed fully to their effects on MARCKS, it is plausible that the
role of MARCKS in modulating plasticity may account for some of the long-term
stabilizing actions exhibited by VPA and lithium in the treatment of bipolar disorder. This
notion is supported by the finding that other drugs were set apart by both their inability to
down-regulate MARCKS as well as their inefficacy in treating bipolar disorder. Further
support for the notion of selective action in the CNS has been provided by data from our
laboratory which shows that the expression of MARCKS protein in two separate murine
macrophage cell lines (IC-21 and MH-S) is unaltered following chronic exposure to
therapeutic levels of VPA (3-7 d at 0.6 mM, data not shown). In addition to effects of

58
VPA on MARCKS, such alterations in neuronal plasticity may result from actions on
other potential protein substrates, as discussed below.
GAP-43
It is clear from these studies that VPA has a time- and concentration-dependent
effect on GAP-43 expression in HN33 cells. Further, the concentration range over which
this increase in GAP-43 occurred was comparable to the concentration range over which
the down-regulation in MARCKS protein occurred, and this concentration range is well
within the target therapeutic serum concentration. There are several potential implications
of such an increase in GAP-43 expression, since GAP-43 is an early marker of post¬
mitotic neurons, and is highly expressed during axonal growth and synapse formation
(Aigner et a!., 1995). For example, increased GAP-43 expression might signify synaptic
regrowth and/or remodeling, cytoprotection, or even development of aberrant pathways.
Synaptic remodeling was discussed previously as a potential mechanism for long-term
mood stabilization. A novel finding observed in carrying out these studies was that
following chronic (>1 d) VPA exposure, HN33 cells appeared to become more adherent
to the tissue culture flasks in which they were grown. This is of interest in light of the
finding that chronic VPA exposure also elicited an increase in expression of GAP-43,
which is known to be important for promoting and maintaining growth cone adhesion
during neurite outgrowth (Aigner and Caroni, 1995). Later chapters will include further
studies on the growth and morphological changes induced by VPA.
As previously mentioned, preliminary studies have shown no evidence for lithium-
induced regulation of GAP-43 expression in HN33 cells, and this major difference
between VPA and lithium is intriguing. Since the two drugs share similar effects on
MARCKS, one might hypothesize that this effect provides a basis for their common ability
to treat bipolar disorder. Conversely, the difference in their ability to affect GAP-43 might

59
relate either to their differing efficacies in treating various types of bipolar disorders (i.e.
VPA is generally more efficacious in treating rapid-cycling bipolar disorder), or,
alternatively, to their very different pharmacological profiles aside from their mood-
stabilizing efficacy (i.e., VPA is also an anti-epileptic, whereas lithium is not).
Conclusions
With regard to alterations in protein expression and their necessity and/or
sufficiency in bringing about the desired end result (that of attaining successful mood-
stabilizing treatment), GAP-43 induction may or may not be either. There is little
evidence to suggest that increased GAP-43 is necessary for the successful amelioration of
bipolar symptoms, for although VPA elicited GAP-43 induction, lithium, the prototypic
mood-stabilizing drug, had no effect on GAP-43 expression in our in vitro cell model.
Nonetheless, this does not preclude the possibility that lithium may exert such an effect on
GAP-43 in other cell model systems or in brain tissue. Whether GAP-43 alterations alone
are sufficient for optimal bipolar therapy is another question entirely, which these data
alone cannot fully address. Many agents may lead to an increase in expression of GAP-43,
if by virtue of nothing other than altering cell growth and differentiation rates, but it does
not necessarily follow, based on what is yet known, that all of these agents would serve as
sufficient mood-stabilizing drugs.
Down-regulation of MARCKS protein may be more consistent with a mood-
stabilization mechanism. The observation that both VPA and lithium, to the exclusion of a
number of drugs used for various other psychotropic applications, elicited a significant
reduction in MARCKS, suggests that alterations in MARCKS may be involved in the
mechanism of action of these agents, and supports the identification of MARCKS as a
potential molecular target which may or may not contribute to the therapeutic effects of
drugs used in the treatment of bipolar disorder. However, the findings that

60
carbamazepine, another widely used mood-stabilizing drug therapy (although not FDA-
approved for this application), as well as some less commonly used mood-stabilizing
agents, including lamotrigine (data not shown) and verapamil, did not alter MARCKS
expression, point to the possibility that either MARCKS is not integral to mood-stabilizing
efficacy, or that the in vitro exposure conditions and model we employed were not optimal
to allow the down-regulation we may have observed following chronic exposure in vivo.
Another important question concerning the effects observed is that of sufficiency. The
identification of MARCKS as a potential target for the action of drugs used in the
treatment of bipolar disorder suggests that the down-regulation of MARCKS alone, by
any method, would be enough to yield successful amelioration of bipolar symptoms.
However, this diminishes the potential role of other mechanisms, such as the alterations in
PI signaling induced by lithium, or the effects of VPA on GAP-43 expression. It is
conceivable that the alterations in MARCKS expression observed in vitro would lead, in
vivo, to further downstream events important for altering plasticity or other parameters
involved in changing the course of the disorder. In other words, MARCKS may be just
one target of many which may ultimately result in the efficacy of this group of drugs, those
which exhibit anti-manic and/or mood-stabilizing action. This is more compatible with the
notion that other drugs may be able to positively affect the course of bipolar therapy.
Whereas VPA and lithium may operate, at least in part, through their effects on
MARCKS, the involvement of MARCKS in bipolar therapy is indirect, and other drugs
could theoretically elicit the same end result, yet through a separate pathway, perhaps
bypassing MARCKS altogether
It is interesting that VPA led to opposite effects on two CNS proteins which are
both involved in cellular plasticity and intracellular signaling, and both regulated by PKC.
This is despite the fact that the VPA-induced alterations of these proteins occurred over a
very similar time course and dose range. These data support recent in vivo findings in our
laboratory which imply an apparent inverse regulation of MARCKS and GAP-43 in

61
certain highly plastic brain regions, such as the hippocampus, in that levels of GAP-43 are
high in some regions while MARCKS levels are low, and vice versa (McNamara and
Lenox, 1997). This would suggest that at certain developmental stages or in particular
brain regions, MARCKS and GAP-43 play alternative or opposing roles, even though they
can be regulated similarly. Such comparable pharmacodynamic properties suggest that
both proteins play important roles in the long-term therapeutic action of VPA.

CHAPTER 4
ROLE OF PKC IN MECHANISM OF ACTION OF VPA
Introduction
PKC is an 80-kD protein expressed at high concentrations in neuronal tissues, and
the observation that PKC is the major enzyme responsible for regulation of both GAP-43
and MARCKS make the effects of VPA on PKC an interesting area of investigation PKC
is a serine/threonine kinase which, by virtue of its effects on intracellular signal
transduction, is thought to play a role in hormone and neurotransmitter secretion and
regulation of cell proliferation and differentiation, through mechanisms which are still not
fully understood (Fujise el a/., 1994; Hofmann, 1997). PKC is actually comprised of a
large family of at least 11 mammalian isoforms, classified into three groups based on their
activation profiles. The conventional (cPKC) group includes a, pi/pil, and y, which are
regulated by Ca++, diacylglycerols (DAG), negatively charged phospholipids, and phorbol
esters (potent tumor promoters which mimic DAG). The novel (nPKC) group contains 5,
e, r| (L), and 0, and is characterized by activation requirements similar to those of the
cPKC’s, but lack of Ca++ activation. The atypical (aPKC) members are £, t (>.), and |i(D),
and are all independent of Ca++ and DAG or phorbols, requiring only phospholipids for
activation (Fujise el a/., 1994; Hofmann, 1997). The PKC isozymes differ in their primary
structure and genomic origin, so that they are not derived from a single mRNA, but rather
from a series of structurally related mRNAs coding for a family of enzymes with very
similar properties. In humans, the genes for the various isozymes are located on different
62

63
chromosomes, with the exception of the pi and pH isoforms, which are the result of
differential splicing (Stabel and Parker, 1991).
The primary structure of PKC contains four major regions, termed Cj through C4,
which exhibit a high degree of homology between isotypes. These conserved regions are
generally thought to represent areas involved with functions common to all PKC
isozymes, such as the kinase domain. Interspersed between the conserved regions are
variable regions Vj through V5, which exhibit a lower degree of homology and probably
allow for properties specific to individual isoforms. The higher level structure of PKC is
comprised of catalytic and regulatory domains. The catalytic domain alone is referred to
as PKM, and has a molecular weight of 45-55 kDa. It is located in the C-terminal region
of the polypeptide, and confers its kinase activity. The regulatory region, located at the
N-terminal and with an approximate molecular weight of 32-36 kDa, contains
phospholipid-and phorbol-binding domains responsible for modulating alterations in the
activity of the enzyme. It is at the regulatory region that diacylglycerol (DAG) and
phorbol esters bind to transiently activate the enzyme, eventually resulting in an increased
rate of degradation (down-regulation). All members of the PKC superfamily carry in the
N-terminal region at least one zinc-finger structure (more commonly two), which is
thought to aid in protein-protein interactions. The N-terminal region also contains a
pseudosubstrate sequence, conferring the enzyme’s autoinhibitory activity. This region
serves to regulate activity of the enzyme, and removal of this region results in constitutive
and activator-independent kinase activity. The pseudosubstrate domain resembles the
consensus phosphorylation sequence but contains no phosphorylatable residue; it controls
kinase activity by occupying the catalytic site and thereby blocking access of the substrate
(Stabel and Parker, 1991).
cPKC activation is dependent upon two major events: (1) formation of the ternary
enzyme+Ca2++phospholipid complex (the presence of the phospholipid further enhancing
the affinity of the enzyme for calcium), and (2) binding of DAG (or phorbol), leading to

64
the conformational changes which result in activation. Down-regulation of the enzyme
follows prolonged activation, and is accomplished through both increased proteolysis and
sequestration (association of the enzyme with cytoskeletal or nuclear components).
Autophosphorylation of the enzyme is not a requirement, but probably plays a role (Stabel
and Parker, 1991).
Data from our laboratory indicate that PKC activity is decreased over time
following chronic lithium exposure, as is expression of PKC isozymes a and s (Watson
and Lenox, 1997). These findings are in agreement with those reported by Manji and
colleagues, which reflected reductions in PKC activity and expression of a and s isoforms
following either lithium or VP A exposure, in either rats or C6 glioma cells (Manji et al.,
1993; Chen et al., 1994; Manji et al., 1996). Tissue distribution of PKC isozyme
expression varies, and the relative expression of each isoform may contribute to its unique
role in signal transduction. Individual PKC isozymes, upon activation, are differentially
compartmentalized, suggesting that they mediate distinct cellular functions. PKC-y
appears to be the only isoform unique to brain and spinal cord, with no expression
elsewhere; its precise role remains unknown. The same is true for most of the other
isoforms - although there has been no dearth of studies into roles of the various PKC
isozymes, results have been inconclusive and in some cases contradictory (to be discussed
further in Discussion). Whereas PKC-a and s have been implicated in tumor growth and
suppression of apoptosis, 8 may induce opposite effects (Hofmann, 1997). Both PKC-a
and -6 are thought to be important in early brain development (Blackshear et al., 1996),
while PKC-C may play a role in proliferation, and several isoforms have reportedly been
implicated in neurite outgrowth and/or cell differentiation (Ways et al., 1994; Hundle et
al., 1995; Borgatti et al., 1996). We have previously established that the immortalized
hippocampal cell line HN33 expresses PKC isoforms a, 5, s, and and our laboratory has
demonstrated in vitro effects of lithium on PKC activity as well as on expression levels of
various PKC isozymes (Watson and Lenox, 1997). In light of previous reports

65
demonstrating VPA-induced alterations in PKC in a non-neuronal cell line, it will be of
interest to investigate the effects of VP A exposure on these parameters in HN33 cells,
especially given preliminary evidence from our laboratory for effects of VP A on
MARCKS and GAP-43 in these cells. It has been suggested that the colocalization of
some isoforms with substrate(s) may ensure preferential and rapid phosphorylation
following PKC activation. Based on evidence indicating a down-regulation of MARCKS
and increased expression of GAP-43 following chronic VPA exposure, it is predicted that
VPA will activate PKC, at least transiently, so that the downstream effects on MARCKS
and GAP-43 can occur as observed. If VPA does, indeed, activate PKC, both its activity
level and isoform expression should decrease over time, in the characteristic pattern of
down-regulation following long-term activation. This acute activation followed by
chronic down-regulation has been demonstrated previously in the HN33 cells with lithium,
phorbol esters, and retinoic acid.
These and other studies suggest that VPA has significant effects on both PKC and
its substrates, though the connection between such effects is unclear. Although one would
expect PKC alterations to be in large part responsible for any further downstream effects
on its substrates MARCKS and GAP-43, this direct cause-and-effect relationship has yet
to be established. The following studies will therefore help to establish whether or not
PKC is necessary to the downstream effects on MARCKS and GAP-43, and which
isoforms may be most important for each effect. After first establishing the ability of VPA
to effect alterations in both PKC activity and expression of various PKC isozymes in
HN33 cells, the next objective will be to address the role of PKC in altering MARCKS
and GAP-43 levels. VPA-induced alterations in PKC-specific activity will be measured
indirectly by phosphorylation of the substrate MBP, and associated alterations in PKC
isoform expression will be measured using Western immunoblotting techniques. PKC
inhibitors will then be used to assess whether PKC activity is necessary to elicit the
observed VPA-induced alterations in MARCKS and GAP-43, and isoform-specific

66
antisense oligonucleotides will be used to narrow down which isoforms are important for
each parameter. It is hypothesized that PKC activity will be necessary to any observed
effects by VPA on MARCKS or GAP-43, and that inhibition of the enzyme will block
such observed effects.

67
Results
Effect of VPA Exposure on PKC Activity in HN33 Cells
HN33 cells exposed to 0.6 mM VPA for 15 min -3d were assayed for effects on
PKC activity, as measured by MBP phosphorylation (Figure 4.1). Acute VPA exposure
(15 min - 4 h) resulted in a tendency toward increased membrane-associated activity and a
correlated decrease in PKC activity in the cytosolic fraction. This paradigm most closely
follows a translocation phenomenon, whereby cytosolic PKC is shuttled into the
membrane for activation. By the 3 d time point, activity in the membrane fractions is
down-regulated (p<0.05). Similar results were observed using MARCKS peptide as the
phosphorylation substrate (data not shown).
Time (hours)
Figure 4.1. Effect of VPA exposure on PKC activity in HN33 cells. VPA exposure leads
to apparent PKC activation, followed by down-regulation. HN33 cells exposed to 0.6
mM VPA were processed and soluble and membrane fractions isolated as detailed in
Materials and Methods (Chapter 2). Assay was performed utilizing the PKC activators
PMA (phorbol), CaCl2, and phosphatidylserine. PKC activation was measured by the y-
32P phosphorylation (in pmol/min/mg) of the PKC substrate MBP. Mean activity of
control samples was 102 ±8.1 pmol/min/mg. The figure shows the overall pattern of
PKC activity from 15 min (acute) to 3 d (chronic) VPA exposure; the inset shows PKC
activation only at acute time points (up to 4 h). Results are mean ± S.E.M. of at least four
determinations, and are expressed as percent of activity in untreated control samples.
*p<0.05

68
Effect of VPA on PKC-a
PKC-a is observed entirely in the soluble fraction of FTN33 cells. As shown in
Figure 4.2, exposure to 0.6 mM VPA results in an acute (15 min - 12 h) increase in PKC-
a expression, followed by down-regulation over time (p<0.05).
H .'â– â– â– 'â– â– -amm.
ifVCfUMKOn
1 2 3 4 5 6
TIME (hours)
Figure 4.2. Effect of VPA on PKC-a expression in soluble fraction of HN33 cells, (a)
Representative Western immunoblot of soluble PKC-a expression following acute (24 h
or less) VPA exposure @0.6 mM. Protein samples were loaded into lanes as follows:
Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h VPA; Lane 4, 4 h VPA; Lane
5, 12 h VPA; Lane 6, 24 h VPA. (b) Representative Western immunoblot of cytosolic
PKC-a expression following chronic (3 d) VPA exposure @ 0.6 mM. Lane 1 was loaded
with protein from an untreated control, and Lane 2 was loaded with protein from a sample
exposed to VPA for 3 d. (c) HN33 cells were exposed to 0.6 mM VPA for 15 min - 3d,
and PKC-a levels (present only in the cytosolic fraction of these cells) were assessed by
Western immunoblot analysis. Results are expressed as percent of untreated control, and
are the mean ± S.E.M of four determinations. *p<0.05
Effect of VPA on PKC-S
PKC-6 undergoes a significant increase in expression levels in both soluble
(p<0.001) and membrane (p<0.0001) fractions of cells during chronic VPA exposure (1 &
3 d), and as late as 3 d, there appears to be no down-regulation (Figure 4.3).

69
1 2 1 2
Figure 4.3. Effect of VP A on PKC-5 expression in HN33 cells, (a) Representative
Western blot of PKC-8 expression following acute (<24 h) VPA exposure @ 0.6 mM.
The blot at left (Lanes 1-6) shows cytosolic PKC in HN33 cells, while the blot at right
(Lanes 7-12) shows membrane-associated PKC. Protein samples were loaded into
lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h VPA;
Lane 4, 4 h VPA; Lane 5, 12 h VPA; Lane 6, 24 h VPA; Lane 7, untreated control;
Lane 8, 15 min VPA; Lane 9, 1 h VPA; Lane 10, 2 h VPA; Lane 11, 4 h VPA; Lane
12, 12 h VPA. (b) Representative Western of PKC-6 expression following chronic (3
d) VPA exposure @ 0.6 mM. The blot at left shows cytosolic PKC in HN33 cells,
while the blot at right shows membrane-associated PKC. Lane 1 was loaded with
protein from an untreated control, and Lane 2 was loaded with protein from a sample
exposed to VPA for 3 d. (c) HN33 cells were exposed to 0.6 mM VPA for 15 min - 3
d, and PKC-8 levels were assessed by Western immunoblot analysis. Results are
expressed as percent of untreated control, and are the mean ± S .E.M. of four
determinations. *p<0.001, soluble and *p<0.0001, membrane.

70
a
1 234561 23456
b
1
Figure 4.4. Effect of VP A on PKC-e expression in HN33 cells, (a) Representative
Western immunoblot of PKC-e expression following acute (24 h or less) VPA
exposure @ 0.6 mM. The six lanes at left show cytosolic (or soluble) PKC in HN33
cells, while the six lanes at right show membrane-associated PKC Protein samples
were loaded into lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA;
Lane 3, 1 h VPA; Lane 4, 2 h VPA; Lane 5, 4 h VPA; Lane 6, 12 h VPA. (b)
Representative Western immunoblot of PKC-e expression following chronic (3 d)
VPA exposure @ 0.6 mM. The blot at left shows cytosolic (or soluble) PKC in
HN33 cells, while the blot at right shows membrane-associated PKC. Lane 1 was
loaded with protein from an untreated control, and Lane 2 was loaded with protein
from a sample exposed to VPA for 3 d (c) Cells were exposed to 0.6 mM VPA for
15 min - 3d, and PKC-e levels were assessed by Western immunoblot analysis
Results are expressed as percent of untreated control, and are the mean ± S.E.M. of
four determinations. *p<0.05

71
Effect of VPA on PKC-e
As shown in Figure 4.4, membrane-associated PKC-s increases acutely (15 min - 4
h), and then returns to basal levels by the 3 d time point. Cytosolic PKC-s expression
remains close to baseline acutely, and then increases significantly following prolonged (1-3
d) VPA exposure (p<0.05). This pattern of expression reflects an apparent activation and
translocation of PKC-e, whereby PKC is activated, shuttled from cytosol into the
membrane (where the enzyme is active), and then down-regulated and returned from
membrane to cytosol following prolonged activation.
Effect of VPA on PKC<
PKC-C shows little change in expression at early time points, and then is up-
regulated sometime between 1 d and 3 (p<0.0001), with a similar tendency toward up-
regulation observed in the soluble fraction (Figure 4.5).
Effect of PKC-a Antisense Oligonucleotides on PKC-a Expression in HN33 Cells
Antisense oligonucleotides were designed to bind to various segments of the
mouse and/or human PKC-a mRNA sequence, and block its translation to protein. Figure
4.6 and Table 4.1 list the antisense oligonucleotides and show the regions of mouse PKC-
a gene they were designed to bind. All oligonucleotides were synthesized with
phosphorothioate backbones, in order to increase their stability in culture. A series of
experiments was then conducted to down-regulate PKC-a expression in HN33 cells, first
using a single antisense oligonucleotide sequence, then using the combination of five
antisense oligonucleotides. A sense oligonucleotide was used as the negative control for
these experiments.

72
b
Figure 4.5. Effect of VP A on PKC-£ expression in HN33 cells, (a) Representative
Western immunoblots of PKC-£ expression following acute (24 h or less) VPA
exposure @ 0.6 mM The blot at left shows cytosolic (or soluble) PKC in HN33 cells,
while the blot at right shows membrane-associated PKC. Protein samples were loaded
into lanes as follows: Lane 1, untreated control; Lane 2, 15 min VPA; Lane 3, 1 h
VPA; Lane 4, 4 h VPA; Lane 5, 12 h VPA, Lane 6, 24 h VPA. (b) Representative
Western immunoblot of PKC-(^ expression following chronic (3 d) VPA exposure @
0.6 mM The blot at left shows cytosolic (or soluble) PKC in HN33 cells, while the
blot at right shows membrane-associated PKC. Lane 1 was loaded with protein from
an untreated control, and Lane 2 was loaded with protein from a sample exposed to
VPA for 3 d (c) HN33 cells were exposed to 0.6 mM VPA for 15 min - 3d, and
PKC-£ levels were assessed by Western immunoblot analysis. Results are expressed as
percent of untreated control, and are the mean ± S.E.M of four determinations
*p<0.0001, membrane

73
25
24
26.
27
28
5'
t
start
mouse PKC alpha gene
Figure 4.6. PKC-a antisense oligo-nucleotide design. This figure depicts the
antisense oligonucleotides designed to bind to PKC-a mRNA and block its
translation. Shown is the representative PKC-a gene from mouse (as reported in
the NCBI database), and the five oligonucleotides (24, 25, 26, 27, 28) designed to
bind to the nucleotide sequence at the indicated, base-complementary regions.
Start codon is demarcated by the arrow; both oligonucleotides 24 and 25 span this
region. Not shown is #23, the sense oligonucleotide designed as a negative
control, which reads as the sense complement to antisense oligo 24. All
oligonucleotides are 20-base phosphorothioate linked, with the exception of #23
and #24, which are 15-base phosphorothioates, and all were cross-referenced
through GenBank to ensure unique sequence identity with PKC-a.
Table 4.1 lists the 5’ to 3’nucleotide sequences of each PKC-a oligonucleotide
(ON) by number, and the species specificity of each.
Table 4.1 Antisense oligonucleotide sequences.
ON#
Sequence of Nucleotides
Species Specificity
23
ACCATGGCTGACGTT
mouse & human
24
AACGTCAGCCATGGT
mouse & human
25
CAGCCATGGTTCCCCCCAAC
mouse
26
AGAAGGTAGGGCTTCCGTAT
mouse & human
27
GTT ACC AAC AGGACCT AGCT
mouse
28
T GAGTT C AGC AT GAC ATTGT
mouse

74
Initial experiments attempted to establish the optimal dose range for
oligonucleotide exposure and protein down-regulation. Oligonucleotide concentrations
ranging from 0.1 to 1.0 pM (the standard concentrations reported in similar studies) had
no effect on cell viability or PKC-a expression following 24 h exposure (data not shown).
Increasing oligonucleotide concentration to 10 pM and lengthening exposure time to 48 h
had no effect on PKC-a expression (data not shown).
Next, methods were employed to increase oligonucleotide uptake and compensate
for potential oligonucleotide degradation In keeping with previous reports of successful
antisense-mediated protein down-regulation, the cationic carrier lipid LipofectAMINE
(Gibco BRL, Gaithersburg, MD) was used at 5 pg/ml and 15 pg/ml, and cells were serum-
deprived during oligonucleotide exposure. In addition, medium and its constituents were
replaced fresh every 12 h during exposure. PKC-a expression remained at basal levels
following 24 or 48 h exposure to oligonucleotide under these conditions (data not shown).
The final approach utilized phorbol to down-regulate PKC expression (non-
selectively), then PKC-a antisense oligonucleotides were added following phorbol
removal, in an effort to selectively block PKC-a recovery. A representative Western is
shown in Figure 4.7. HN33 cells were exposed to 1.0 pM phorbol 12,13-dibutyrate
(PDBu) for 24 h, after which cells were washed with media to remove drug. Next, cells
were exposed to 1.0 pM sense or antisense oligonucleotides (total of five antisense
oligonucleotides) directed to PKC-a (Figure 4.6). Cells were exposed to oligonucleotides
for a total of 24 h, but antisense-treated cells showed no difference in PKC-a expression
than sense-treated or untreated control cells. As shown in Figure 4.7, there is clearly no
difference in PKC-a expression between those PDBu-pre-treated cells which are
antisense-treated or untreated. PDBu exposure resulted in a reversible down-regulation of
PKC-a (Lane 2 vs. 1), and PKC-a antisense failed to prevent the recovery of PKC-a
expression at rates comparable to control levels (Lanes 7&8 vs. 3-6). Whereas toxicity
commonly arises as a problem in antisense applications in cell and tissue culture, the HN33

cells showed no evidence of cytotoxicity following 24-48 h exposure to any of the
pharmacological agents at any of the concentrations tested.
75
1 2 3 4 5 6 7 8
Figure 4.7. Effect of PKC-a antisense oligonucleotides on PKC-a expression in HN33
cells Shown is a representative Western immunoblot of PKC-a expression in HN33 cells
following antisense treatment. Lane 1 holds an untreated control; Lane 2 shows PKC-a
expression in cells treated with 1.0 pM PDBu for 24 h; Lane 3 shows cells treated with
1.0 pM PDBu for 24 h, then no treatment for 24 h; Lane 4 shows cells treated with 1.0
pM PDBu for 24 h, then LipofectAMINE for 24 h; Lane 5 shows cells treated with 1.0
pM PDBu for 24 h, then sense oligo #23 for 24 h; Lane 6 shows cells treated with 1.0 pM
PDBu for 24 h, then sense oligo #23 and LipofectAMINE for 24 h; Lane 7 shows cells
treated with 1.0 pM PDBu for 24 h, then antisense oligonucleotides #24, 25, 26, 27, and
28, each at 1.0 pM, for 24 h; Lane 8 shows cells treated with 1.0 pM PDBu for 24 h, then
antisense oligonucleotides #24, 25, 26, 27, and 28, plus LipofectAMINE, for 24 h.
Effect of PKC Inhibitor LY333531 on VPA-Induced Down-Regulation of MARCKS
LY333531 alone had no effect on MARCKS expression, but when combined with
VP A, LY333531 prevented the VPA-induced MARCKS down-regulation observed in the
membrane, but not soluble, fraction of HN33 cells (Figure 4.8). MARCKS expression in
both cytosol and membrane of VPA-treated cells was significantly different from that of
LY333531-treated cells (p<0.005, soluble; p<0.0001, membrane). Cytosolic MARCKS in
combined VPA/LY333531 -treated cells was significantly different from that of
LY3 3 3 531 -treated cells (p<0.005), indicating that, even in the presence of the PKC
inhibitor LY333531, VPA was able to induce down-regulation of cytosolic (but not
membrane-associated) MARCKS expression. In contrast, LY333531 prevented the VPA-
induced down-regulation of MARCKS in the membrane fraction of HN33 cells.

76
120-r
"o
s-
+*
110-
100-
s
o
90-
U
80-
70-
e
• ■M
60-
©
50-
—
40-
cn
30-
U
20-
<
10-
0-
y Soluble
Membrane
Control
VPA LY333531 VPA + LY333531
Treatment
b
Figure 4.8. Effect of PKC inhibitor LY333531 on VPA-induced down-regulation
of MARCKS. (a) HN33 cells were exposed for 24 h to 1.0 mM VPA with or
without 1.0 (iM LY333531. MARCKS expression was quantitated by Western
blot analysis, and results (expressed as percent of untreated control) reflect the
mean ± S.E.M of at least five determinations. *p<0.005, soluble; *p<0.0001,
membrane (b) Representative Western blot showing MARCKS expression in the
presence of VPA ± LY333531. Lanes 1-4 represent cytosolic MARCKS, and
were treated for 24 h as follows: control, untreated (1); 1.0 mM VPA (2); 1.0 pM
LY333531 (3); 1.0 mM VPA + 1.0 pM LY333531 (4). Lanes 5-8 show
membrane-associated MARCKS, and were treated for 24 h as described for 1 -4
(respectively) above.
Effect of PKC Inhibitor LY333531 on VPA-induced GAP-43 Induction
LY333531, alone or in combination with VPA, had no effect on GAP-43
expression in HN33 cells (Figure 4.9). GAP-43 levels in both soluble and membrane
fractions are significantly increased (p<0.05) in both VPA-treated groups, compared to

77
controls (untreated or LY333531-treated). Therefore, the PKC inhibitor LY333531 was
unable to prevent the VPA-induced GAP-43 increase, for even in the presence of
LY333531, VPA increased GAP-43 expression in both soluble and membrane fractions.
These data indicate that PKC inhibition alone is insufficient to prevent the GAP-43
increase observed in HN33 cells following 24 h 1.0 mM VPA exposure.
a
Treatment
b
Figure 4.9. Effect of PKC inhibitor LY333 531 on VPA-induced GAP-43 increase (a)
HN33 cells were exposed for 24 h to 1.0 mM VPA with or without 1.0 (iM
LY333531. GAP-43 expression was quantitated by Western blot analysis, and results
(expressed as percent of untreated control) reflect the mean ± S .E.M. of three
determinations. *p<0.05 (b) Representative Western blot showing GAP-43 expression
in the presence of VPA ± LY333531. Lanes 1-4 represent membrane-associated GAP-
43, and were treated for 24 h as follows: control, untreated (1); 1.0 mM VPA (2); 1.0
pM LY333531 (3); 1.0 mM VPA + 1.0 pM LY333531 (4). Lanes 5-8 show cytosolic
GAP-43, and were treated for 24 h as described for 1-4 (respectively) above.

78
Discussion
Inasmuch as PKC activation has been associated with regulation of expression of
both GAP-43 and MARCKS, the effects of VP A on PKC isozyme levels and PKC activity
were investigated more directly, in an effort to better assess the role of PKC in the
mechanism of action of VP A in the brain. Whereas lithium, the first-line anti-manic/mood-
stabilizer, is believed to operate in part through the PI signaling pathway, our group and
others have provided evidence against such a mechanism for VPA (Vadnal and
Parthasarathy, 1995; Lenoxetal, 1996; Dixon and Hokin, 1997; Watson et al., 1998),
although both drugs have similar effects on MARCKS expression.
Our findings indicate that, like lithium, VPA induces alterations in expression of
PKC isoforms a and s (Watson and Lenox, 1997). However, the time course for VPA-
induced changes is earlier than that for lithium, and PKC-e expression actually increases
(acutely in membrane, chronically in cytosol), in contrast to the reduction previously
reported for both lithium and VPA (Chen et al., 1994; Manji et al., 1996; Watson and
Lenox, 1997). Chen et al. (1994) reported that chronic VPA exposure of C6 glioma cells
reduced the expression of PKC isozymes a and e in intact cells, a finding similar to that
observed following chronic lithium administration in the same cell model (Manji et al.,
1993). Subsequent studies revealed PKC-induced alterations in multiple components of
the (3-adrenergic receptor-coupled cAMP-generating system, including Pi-AR, Gas, and
adenylyl cyclase, each of which is phosphorylated by PKC (Chen et al, 1996). The
significance of these changes is unclear, but together they suggest a major role for PKC in
the VPA- or lithium-induced molecular effects underlying their efficacy, and may be
instructive when other PKC-related effects are compared between the two drugs.
In this study, VPA exposure resulted in an apparent transient activation of PKC
(within 2 h), followed by a gradual reduction in activity with chronic exposure (4 h and
beyond). PKC activity was significantly down-regulated by the 72 h time point. Similar

79
results were reported following exposure of HN33 cells to the phorbol PMA, though
activation occurred within 15 minutes and down-regulation of PKC activity was evident
by 1 h (Watson et al., 1994). This finding is further in agreement with previous reports
which showed that chronic VPA or lithium exposure of C6 glioma cells significantly
reduced PKC activity in both membrane and soluble fractions (Chen et al., 1994; Manji et
al., 1996). The observed patterns of alterations in protein activity and expression suggest
that isoforms a and e might, under these conditions, be involved in the phosphorylation
and regulation of MARCKS protein expression, since both VPA and lithium elicited
alterations in PKC-a and -e expression in HN33 cells, and the time course of MARCKS
alterations roughly corresponds to that of the changes observed in PKC isoform
expression Indeed, previous studies have indicated the ability of these two isoforms to
phosphorylate MARCKS, whereas PKC-£ is unable to phosphorylate MARCKS in vitro
(Fujise etal, 1994; Herget et al., 1995; Uberall et al., 1997). One could hypothesize a
process in which VPA, directly or indirectly, activates PKC, especially isoforms a and/or
e, at early time points (between 0 and 12 h). By 3 d, there is a significant down-regulation
of activity in the membrane and a comparable drop in the cytosol. Consistent with this
hypothesis are data which show an increase in PKC activity within 15 seconds following
phorbol exposure (Stabel and Parker, 1991), and a time course for MARCKS down-
regulation which closely follows the time course of PKC activation in HN33 cells exposed
to phorbol (Watson et al., 1994).
While VPA and lithium both affect PKC isozyme a (and perhaps e), the two drugs
have very different effects on PKC-5. VPA induces a chronic up-regulation in PKC-5,
whereas lithium has little or no effect on PKC-8 expression (Watson and Lenox, 1997). In
comparing VPA and lithium, one might hypothesize that such differences in patterns of
PKC activation underlie the differences observed in downstream effects such as protein
expression or cell growth and morphology. For example, the up-regulation of PKC-5
resulting uniquely from VPA exposure is probably not important for the MARCKS down-

80
regulation observed, but might be involved in the GAP-43 induction observed following
VPA but not lithium exposure; in support of this hypothesis is evidence indicating that
PKC-5 and -e phosphorylate GAP-43 (Oehrlein et al., 1996). Indeed, in our study, the
GAP-43 increase occurred over a time course and of a magnitude similar to those of the
PKC-8 alterations observed Conversely, PKC isozymes a and/or s may play a role in the
observed VPA- and lithium-induced down-regulation of MARCKS.
Reports are in disagreement as to the role of PKC in neuronal differentiation. A
number of studies have suggested that, rather than being a positive regulator of the
process, PKC may actually be an inhibitory influence, so that suppression of the enzyme
results in increased process outgrowth. For example, use of various PKC inhibitors, and
subsequent reductions in PKC activity, have been associated with increased neurite
outgrowth (Heikkila et al., 1989; Parodie/a/., 1990; Tsuneishi, 1992; Wooten, 1992;
Jalava et al., 1993; Ekinci and Shea, 1997). Long-term down-regulation of PKC activity
with phorbols such as PMA has provided similar results (Tsuneishi, 1992; Carlson et al.,
1993; Jalava et al., 1993; Ekinci and Shea, 1997), with one study showing enhanced
neurite outgrowth following combined phorbol and PKC inhibitor treatment (Heikkila et
al., 1989). However, some of these same studies have further shown that down-
regulation of PKC activity alone is insufficient to elicit a differentiated phenotype (Ekinci
and Shea, 1997; Carlson et al., 1993; Wooten, 1992).
On the other hand, numerous reports have presented equally convincing evidence
for a positive role for PKC in regulating neuronal differentiation and a positive role in
maintaining the functionally active state (Coleman and Wooten, 1994; Cabell and
Audesirk, 1993; Parrow et al., 1992; Abraham et al., 1991; Tonini et al., 1991), though
this role is probably not unique to PKC. Previous data have implicated a number of
different PKC isoforms in cellular differentiation, including all four of the isotypes present
in HN33 cells: a, 6, e, and C, (Wada et al., 1989; Leli et al., 1992, 1993; Ways et al.,
1994; Hundle et al., 1995; Borgatti et al, 1996). Indeed, the subcellular localization and

81
activation of PKC isozymes are regulated in an isoform-specific manner during
neurogenesis, suggesting they are involved in control of neural development and in
neuronal differentiation (Oehrlein et al., 1998).
The role of PKC in proliferation and differentiation remains unclear, for reports
differ as to which PKC isoform(s) is/are responsible for various aspects of cellular
differentiation. Some studies have attributed to PKC-a a role as the intermediate in
cellular differentiation, as a result of alterations in PKC-a mRNA and/or protein
expression observed concomitant with differentiation of PC 12 and other cell lines
(Borgatti et al., 1996; Parrow et al., 1995; Murray et al., 1993; Leli et al., 1993; Tonini et
al., 1991; Wada et al., 1989). However, these reports differ in their mechanistic
conclusions, for in three studies, PKC-a expression increased with differentiation
(Borgatti et al., 1996; Parrow et al., 1995; Murray et al., 1993), while in three other
studies, PKC-a and -e levels had decreased following comparable treatment and
neuritogenesis (Leli et al., 1993; Tonini et al., 1991; Wada et al., 1989). Leli's group
further showed that the intracellular delivery of PKC-a and -e isoform-specific antibodies
resulted in a morphologically differentiated phenotype in SH-SY5Y cells (Leli et al.,
1992).
While some groups observed no differentiation-associated alterations in isoforms
other than PKC-a, a number of reports have indicated that in various cell lines, other
members of the PKC superfamily are involved. Borgatti's group, who showed that PKC-a
was up-regulated during PC 12 cell differentiation, found in the same study that PKC 6, s,
and C, were down-regulated (Borgatti et al., 1996). A number of studies have provided
evidence for a role of PKC-e in regulating neuronal differentiation, since neurite
outgrowth has been associated with altered PKC-e levels (Ponzoni et al., 1993; Parrow et
al., 1995; Fagerstrom et al., 1996; Ekinci and Shea, 1997; Hundle et al., 1997; Zeidman et
al., 1999). In support of the notion that PKC-e may be involved in the process, as
suggested by Leli's and Wada's work (Leli et al., 1993; Wada et al., 1989), a separate

82
study showed that overexpression of PKC-e but not PKC-5 led to an increase in NGF-
induced neurite outgrowth in stably transfected PC 12 cell lines (Hundle et al., 1995). In
stark contrast to these findings, yet another group reported just the opposite in NIH 3T3
cells, in that PKC-5 overexpression induced a morphological change and reduced growth
rate consistent with differentiation that was not observed in cells overexpressing PKC-s
(Mischak et al., 1993). Further, PKC-£ has been similarly examined in transfected
leukemic cells, and results indicate that PKC-C overexpression is able to induce cellular
differentiation (Ways et al., 1994). Another study that utilized antisense techniques also
emphasized the importance of PKC-C in NGF-induced neuronal differentiation (Coleman
and Wooten, 1994).
All of these studies point to different PKC isozymes as mediators of similar effects.
These apparently contrasting results may merely be a function of the varying experimental
methods and conditions used from one study to the next, such as different cell lines,
exposure conditions, and other parameters. These findings may alternatively indicate that
(1) each isozyme plays its own, specialized role in the complex process of cellular
differentiation, and (2) alterations in the level of each isozyme may be compensated for by
alterations in the levels and actions of other isozymes. Further studies will need to be
conducted in order to answer these questions, to confirm more definitively the association
between each PKC isoform and the parameter in question, but these data strongly suggest
that both VPA and lithium may down-regulate MARCKS via a PKC-dependent
mechanism. On the other hand, the unique role of the PI signaling pathway in the action
of lithium, and the differential pattern of activation of the PKC isozymes by VPA and
lithium may indicate that there are multiple pathways to the down-regulation of MARCKS
expression, and may also confer different clinical therapeutic properties on these two
drugs. The findings further suggest that the action of VPA on MARCKS protein
expression could be cell and/or tissue specific, for it may be modulated, in part, by the
expression pattern of the PKC isozymes specific to the cell types. Indeed, Stabel and

83
Parker (1991) suggest that the distribution of MARCKS closely follows the distribution of
PKC, based on early studies of PKC activity showing wide disparity across regions (Kuo
et al., 1980; Minakuchi etal., 1981).
Having established an apparent causal relationship between VPA and PKC, the
purpose of the next experiments was to establish more precisely the necessity of PKC in
eliciting the obesrved VPA-induced effects on MARCKS and GAP-43. In order to better
establish the role of various PKC isoforms in the MARCKS down-regulation and GAP-43
increase observed, experiments were designed employing antisense technology. Through
this method, antisense oligonucleotides are designed and directed to specifically bind and
“knock down” the nucleic acid precursors of each individual PKC isozyme, resulting in
transient elimination of the protein in question. There are four main mechanisms by which
antisense oligonucleotides are hypothesized to work: (1) by inhibiting the necessary
processing and transport of RNA, (2) by interfering with binding of the ribosome to the
nucleotide sequence, so that translation is not initiated, (3) by inhibiting the completion of
translation after the process has begun, and (4) by binding to the sequence, thereby
characterizing it as abnormal and resulting in degradation by RNase H (Schlingensiepen et
al., 1997).
For this study, a number of trials were performed in an effort to down-regulate
PKC-a in the HN33 cells One sense and five antisense oligonucleotides were prepared by
Gemini Biotech, Ltd. (The Woodlands, TX) through the DNA Synthesis Core of the UF
Interdisciplinary Center for Biotechnology Research (ICBR). Each oligonucleotide was
designed to bear unique identity to the PKC-a genome of either mouse or human, as
reported in the National Center for Biotechnology Information database, and screened
through GenBank Various regions of the cDNA sequence were targeted, including the
start codon (see Figure 4.6). In an effort to maximize stability and binding of
oligonucleotides, all were composed of 15 or 20 bases each, and had phosphorothioate
linkages in place of the endogenous phosphodiester backbone of DNA. Whereas toxicity

84
commonly arises as a problem in antisense applications in cell and tissue culture, the HN33
cells showed no evidence of cytotoxicity following 1 to 2 d exposure to 1 and 10 |¿M
concentrations of PKC-a oligonucleotides. In initial experiments, cells were exposed to
oligonucleotide(s) for 24-48 h, in an effort to establish a protocol for the consistent
reduction of PKC-a levels prior to VP A exposure. When numerous efforts failed to elicit
any reduction in PKC-a expression, PKC was first down-regulated through 24 h, 1 pM
PDBu (phorbol) administration prior to exposure of cell cultures to oligonucleotide.
Following thorough phorbol washout, test cells were exposed to all five oligonucleotides
concomitantly for 24 h. In order to increase uptake of oligonucleotides, cells were
deprived of serum nutrients and exposed to LipofectAMINE, an agent which serves as a
cationic carrier lipid to transport oligonucleotide into the cells. Additionally, media and
oligonucleotides were replaced after 12 h to compensate for any degradation which had
taken place. Despite these measures designed to optimize antisense-directed repression of
PKC-a levels following initial down-regulation by phorbol, the protein recovered in
antisense-treated cells at about the same rate as that of drug-free controls (see Figure 4.7).
Common explanations for the difficulty in achieving significant decreases in protein
expression include instability of the antisense oligonucleotide sequence, limited uptake of
the oligonucleotide, suboptimal binding of oligonucleotide to its target sequence, and half-
life of the particular protein. Despite efforts to account for and overcome these issues,
attempts to carry out further studies based on the antisense work were ultimately
abandoned, due to the persistent failure of PKC-a antisense to down-regulate PKC-a
expression. Perhaps, in this study, the unpredictable tertiary conformation of the antisense
oligonucleotides was such that access and/or binding to the target mRNA sequence was
prevented, or perhaps the oligonucleotides were sequestered and/or degraded before they
even reached their destination. Because antisense technology is still in its developmental
stages, and oligonucleotide design and exposure conditions vary greatly among systems,

85
significant time and expense may often be invested before down-regulation of the protein
in question is optimized
Whereas expression of each PKC isoform could not be down-regulated
individually, overall PKC activity inhibition was achieved through the use of the PKC
inhibitor LY333531 (Eli Lilly and Company, Indianapolis, IN), a direct competitor for
ATP binding (Jirousek et al., 1996). LY333531 is a PKC-P-selective inhibitor (at low nm
concentrations) which also inhibits PKC isozymes a, 5, and e at concentrations in the low
nM range (Ishii et al., 1996; Jirousek et al., 1996). Exposure of HN33 cells to 1.0 |iM
LY333531 for 24 h resulted in no change in either MARCKS or GAP-43 expression, nor
any evidence of toxicity. When combined with 1.0 mM VP A, LY333531 effectively
prevented the VPA-induced MARCKS down-regulation in the membrane fraction of
HN33 cells, but with no accompanying effect on the VPA-induced GAP-43 increase.
From these studies it is clear that PKC activation plays an integral role in the VPA-
induced MARCKS effects observed, as inhibition of PKC-directed phosphorylation of
MARCKS resulted in the inability of VP A to elicit down-regulation of membrane-
associated MARCKS. It is interesting that this LY333531-induced prevention of
MARCKS down-regulation by VPA occurred in membrane, the fraction of HN33 cells in
which VPA exerts its greatest effect. This finding is further consistent with the known
localization of PKC activity in the membrane, and also with a previous report showing that
phosphorylation of MARCKS is necessary for its dissociation and translocation from
membrane (Allen and Aderem, 1995). Whether PKC activation alone is sufficient to
reduce MARCKS has previously been investigated by our laboratory. It was shown that
phorbol esters, potent activators of PKC, yield a significant down-regulation of MARCKS
in the HN33 cells, and that the time course of phorbol-induced MARCKS alterations
follows closely behind the time course of phorbol-induced PKC activation and down-
regulation (Watson et al., 1994). These findings support the notion that activation (and
perhaps also subsequent down-regulation) of PKC, in and of itself, is sufficient to elicit a

86
reduction in MARCKS, though PKC activation may not be the only route through which
the MARCKS alterations may occur. Indeed, Brooks et al. (1992) reported evidence of
regulation of MARCKS through both PKC-dependent and independent pathways, with
platelet-derived growth factor (PDGF)-induced down-regulation of MARCKS through a
post-transcriptional mechanism.
As for GAP-43, PKC activation is probably not necessary to elicit the alterations in
GAP-43 observed This is evidenced by the fact that PKC inhibition failed to block the
VPA-induced increase in GAP-43 in the HN33 cells. Even with the strong GAP-43
expression observed in the membrane of these cells, and the significant up-regulation of
this protein to several times basal levels, LY333531 caused no attenuation of the VPA-
induced GAP-43 increase observed Assuming conditions were optimal for significant
PKC inhibition, we can conclude that PKC activation is not entirely responsible for the
VPA-induced GAP-43 increase observed, nor is PKC activation alone likely to be
sufficient for the effect, since lithium-induced alterations in PKC activity and isoform
expression did not lead to any change in GAP-43 expression.
The findings do little to address the issue of the involvement of PKC in bipolar
disorder and successful mood-stabilizing therapy. However, one can hypothesize that if
MARCKS dysregulation is integral to bipolar disorder, and PKC is the primary modulator
of MARCKS, then perhaps the enzyme is perturbed in some patients exhibiting symptoms
of the disorder. This would be especially applicable to cases in which the patient
demonstrates normal levels of MARCKS, for MARCKS, being downstream from PKC, is
regulated by PKC, and if PKC parameters are abnormal, then PKC-induced MARCKS
regulation will likely also be abnormal. Even if MARCKS itself is not involved in the
manifestation of the disorder, PKC may be a major player Indeed, previous studies have
provided evidence for alterations in PKC-related signaling in patients with bipolar
disorder Reports have shown that in both platelets and post-mortem brain samples of
individuals diagnosed with the disease, PKC translocation and activity levels were

87
significantly increased as compared to controls, as measured by phorbol- or serotonin-
stimulated histone phosphorylation (Friedman et al., 1993; Wang and Friedman, 1996). In
addition, PKC isoform expression was altered in comparison to controls, with higher
levels of PKC-a, 5, and C, and lower PKC-s expression in bipolar patients (Wang and
Friedman, 1996). Further, two-week lithium treatment of subjects resulted in a reduction
in both PKC translocation and activity in platelets (Friedman et al., 1993). In a more
recent clinical study, the PKC-selective inhibitor tamoxifen (an estrogen receptor
antagonist widely used in the treatment of breast cancer) has been shown to be effective in
treating acute mania in a small sample of patients (Manji et al., 1999). Inasmuch as PKC
is believed to play a pivotal role in regulating neuronal signal transduction and in
modulating intracellular cross-talk between neurotransmitter systems (Nishizuka, 1992),
these studies are suggestive of the potential for affective (mood) dysfunction as a result of
alterations or dysfunction in PKC and associated neuronal signaling pathways. A
definitive association between PKC alteration and mood stabilization may facilitate the
development of a whole new line of agents useful in the treatment of bipolar disorder.

CHAPTER 5
EFFECT OF VPA ON CELL VIABILITY, GROWTH AND MORPHOLOGY
Introduction
While VPA has proven very useful in its clinical applications to seizures and
bipolar disorders, the detrimental effects of this drug on the developing embryo have been
well documented (Wiger et al., 1988; Gofflot el al., 1996; Menegola el al., 1996). High
frequencies of neural tube defects (NTDs) have been linked to prenatal VPA exposure
(Bjerkedal et a!., 1982; Robert and Guibaud, 1982; Lindhout and Schmidt, 1986; Wiger et
al., 1988; Gofflot et al., 1996; Menegola et ai, 1996). Such defects may include open
neural tube, spina bifida, anencephaly, meningomyelocele, and encephalocele, and are
thought to be caused by a number of different factors, such as environmental or
pharmacological teratogens, genetic anomalies, and various maternal predisposing factors
(Rhoads and Mills, 1986; Norman et ai, 1995). In more recent years, folic acid and/or
methionine supplementation have been shown to reduce the frequency of NTDs, both in
the general population and in offspring of mothers undergoing anticonvulsant therapy
(Laurence et al., 1981; Smithells et ai, 1981; Rhoads and Mills, 1986; Trotz et al, 1987;
Mulinare et ai, 1988; Nosel and Klein, 1992; Ehlers et al., 1996). Therefore, women are
advised to discontinue VPA therapy or take folic acid supplements during and just before
pregnancy. The mechanism of this protective effect is not understood. Folate is known to
be a precursor of methionine, an important amino acid for cell proliferation, and vitamin
B12 is an essential cofactor for methionine synthesis. In mothers of NTD-affected
offspring, circulating serum folate levels are low but not clinically deficient; the same is
88

89
true for B12 levels (Laurence et al., 1981; Rhoads and Mills, 1986; Kirke et al., 1993). In
addition, homocysteine, the direct precursor for methionine, is found at high levels in the
serum of some mothers in this population (Steegers-Theunissen et al., 1994; Mills et al.,
1995, 1996). This evidence suggests that, in at least a fraction of NTD cases, the mothers
of NTD-afflicted offspring exhibit a metabolic insufficiency, perhaps at the level of the
homocysteine-to-folate conversion step. This "metabolic defect" hypothesis is in contrast
to the simpler notion of a nutritional deficiency (of either folate or methionine), and is
consistent with the possibility that VPA therapy is one factor which may induce or
enhance this metabolic defect. It is of note that, in experimental studies, in vivo
supplementation of either nutrient to pregnant animals resulted in decreased frequencies of
NTDs in offspring, while in vitro studies did not reproduce the protective effect (Trotz et
al., 1987; Hansen and Grafton, 1991; Nosel and Klein, 1992; Hansen et al., 1995; Ehlers
et al., 1996). Preliminary studies in our laboratory further support the notion that the
effect is of a complicated metabolic nature, for folate or methionine supplementation to
cell growth medium did not prevent the VPA-induced down-regulation of MARCKS
observed in HN33 cells (Watterson unpublished data).
In light of the deleterious developmental effects of this drug, the effects of VPA on
cell growth and differentiation merit in-depth investigation. Previous studies, while few in
number, have provided evidence for a dose-dependent VPA-induced inhibition of cell
proliferation via arrest at mid-Gl of the cell cycle (Martin and Regan, 1991), as well as
VPA-induced differentiation of neuroblastoma and glioma cells (Regan, 1985; Cinatl et
al., 1996). The inhibition of cell proliferation, alterations in cellular interactions, and
altered differentiation have all been correlated with prenatal drug teratogenicity, especially
of the developing neural tube (Wilk et al., 1980; Mummery et al., 1984; Martin et al.,
1988) It is plausible that arrest of cell proliferation at a specific phase in the cell cycle,
accompanied by irreversible differentiation, may result in the effects observed to be
consistent with prenatal teratogenicity.

90
The experiments of this chapter have been designed to address further the issues of
cell growth and morphology, and the effects VPA has on these parameters The purpose
of the following experiments is to ascertain whether VPA induces alterations associated
with differentiation in the same cell model in which altered MARCKS and GAP-43
expression were observed Based on the significantly increased GAP-43 levels discussed
previously (in Chapter 4), it is hypothesized that VPA will increase neurite outgrowth,
since both processes (increased GAP-43 and outgrowth of neurites) generally accompany
cell differentiation, as described in depth in Chapter 3. Alterations in cell growth rate
and/or cellular morphology would be consistent with the hypothesis that VPA is capable
of modifying cytoplasticity. This is in contrast to previous observations showing a lack of
effect by lithium on HN33 cell growth or morphology. It has already been demonstrated,
through these studies and others, that these two drugs operate through different
mechanisms, and have different profiles clinically and at a molecular level. In addition,
previous reports reflect VPA-induced modifications of cell growth and differentiation state
(Martin and Regan, 1991; Regan, 1985). If VPA were indeed found to elicit alterations in
cell growth and morphology, the next step would be to address the mechanism of these
effects by examining the role of PKC. As discussed in Chapter 4, PKC-specific inhibitors
(LY333531 and bisindolylmaleimide, or BIM, Calbiochem, La Jolla, CA) will be used in
an attempt to block the VPA effect, to test the hypothesis that PKC activity is necessary
for the morphological changes.
In addition to the HN33 cells, PC 12 cells will be employed to assess the potential
of VPA to serve as a differentiating and neuroprotective agent. PC 12 cells are a well-
characterized cell line widely used for studying various aspects of cellular differentiation.
PC 12 cells exhibit NGF sensitivity while HN33 cells do not, a characteristic which will be
brought to bear upon these studies. In this apoptotic cell death model, NGF-differentiated
PC 12 cells will be used to assess the ability of VPA to protect against cell death and/or
neurite retraction induced by removal of NGF.

91
These experiments will further address the question of the effects of VP A on cell
plasticity. The ability of VP A to alter cell plasticity is relevant to its unique
pharmacological profile, which includes both anticonvulsant and long-term mood-
stabilizing effects in the adult brain, as well as teratogenic effects on the developing
nervous system. All of these major effects would seem to suggest altered plasticity (either
therapeutic or abnormal) as a contributing factor.
Results
Effect of VPA on HN33 Cell Division
HN33 cells were exposed to 0.6 mM VPA for up to 7 d. At this therapeutically
relevant concentration, an inhibition of cell growth was observed, evident in the increased
cell doubling time (p<0.01; Table 5.1). Despite this effect on cell number, no effect on
cell viability was observed when cells were plated at high confluence, even at a higher
concentration of VPA (1.0 mM), as shown in Table 5.2 (no significance). This supports
the notion that the reduction in cell number is a result of decreased growth rate and not
due to cell death, since it was only observed under conditions in which cell division was
possible.
Table 5.1 shows the mean (± S.E.M) time required for the HN33 cell population to
double in number when exposed to either no drug (control) or chronic 0.6 mM VPA.
*p<0.05 (t-test)
Table 5.1. Cell Doubling Time
Control
19 ± 0.1 h
0.6 mM VPA
26 ± 1.4 h *
Table 5.2 shows the mean (± S.E.M.) number of cells per viewing field in cultures plated
too densely to allow for neurite outgrowth. No significant effect was observed
Table 5.2. # Cells/Field
Control
2,970 ± 280
1.0 mM VPA
3,583 ± 103

92
Figure 5.1. VPA-induced neurite outgrowth in HN33 cells. These photomicrographs
show (a) untreated HN33 cells; (b) HN33 cells exposed to 0.6 mM VPA for 3 d; and (c)
HN33 cells exposed to 0.6 mM VPA for 14 d The figure (d) summarizes quantitatively
the effect of VPA on H1M33 cell morphology. Cells were exposed to 0.6 mM VPA for
varied lengths of time, and then cells were photographed and analyzed. Data are
presented as mean percent (± S.E.M.) of cells with one neurite or more, and mean percent
(± S.E.M.) of cells with two neurites or more; a neurite was defined as having length
equaling at least the diameter of the respective cell body. *p<0.0001

93
Effect of VPA on HN33 Cell Morphology
In comparison to control cells grown in parallel in the absence of drug, an increase in
neurite outgrowth was observed in VPA-exposed cells as early as 3 d in culture, and
persisted for the duration of the exposure (Figure 5.1). The percentage of cells with one
neurite or more increased significantly (p<0.0001) between 0 and 3 d, and remained at
about 71-76% throughout the duration of VPA exposure, from 3 to 14 d. The percentage
of cells with two neurites or more also increased (p<0.0001) between 0 and 3 d (from
11.7% in controls to 36.6% in VPA-treated cells), and increased again (p<0.0001)
between 5 and 8 d (to 52%). It was also noted, during cell collection and harvesting, that
HN33 cells exposed to VPA concentrations higher than 0.3 mM for at least 1 d exhibited
an increased adherence to culture flasks, as evidenced by the greater effort required to
remove cells by sloughing with versene (data not shown). This greater adherence to
substrate is consistent with the increased outgrowth of processes and increased expression
of GAP-43, and together are suggestive of differentiation to a more mature state. It was
also observed that cell density significantly influenced neurite outgrowth of HN33 cells.
Even in the absence of drug treatment, there was an inverse correlation between number
of cells in the viewing field and percent of these cells exhibiting neurites (p<0.0005; data
not shown).
Effects of the PKC Inhibitor LY333531 on VPA-Induced Alterations in HN33 Cell
Number and Neurite Outgrowth
HN33 cells were exposed to 1.0 mM VPA and/or 1.0 |iM LY333531 for 48 h. As
depicted in Figure 5.2a, cell number was decreased significantly only in cultures treated
with the combination of VPA and LY333531 (p<0.05), though cultures treated with either
drug alone exhibited a tendency toward decreased growth. Figure 5.2b shows the
significant effect of either VPA or LY333531 on neurite outgrowth in HN33 cells. Either
agent increased neurite outgrowth as compared to controls (p<0.01), and the two agents

94
combined yielded a greater effect than either agent alone (p<0.05). These results suggest
that PKC exerts a negative influence on cell differentiation, so that addition of a PKC
inhibitor allows for increased neurite outgrowth. Further, the mechanism of VPA-induced
increase in neurites is perhaps not activation of PKC, but either long-term down-
regulation of PKC, or some alternate mechansim altogether. The observations that VP A
and LY333531 exhibited additive effects on both neurite outgrowth and cell number are
suggestive of different mechanisms.
a
Treatment
b
o
Z
01
U
4
3
2
1
0
Control VPA LY333531 VPA + LY333531
Treatment
Figure 5.2. Effects of VPA and LY333531 on HN33 cell density and neurite outgrowth.
HN33 cells were exposed to either 1.0 mM VPA or 1.0 mM LY333531, or the
combination, for 48 h. Cells were then photographed and analyzed to determine (a) cell
number and (b) percent of cell population with neurites. Data are presented as (a) mean
number of cells in the viewing field (± S.E.M.; *p<0.05) and (b) mean percent of cell
population with neurites (± S.E.M.; *p<0.01; **p<0.05).

95
Effect of VP A on Neurite Outgrowth in PC12 Cells
Similar to the effects of VP A observed in HN33 cells, a concentration-dependent
increase in neurite outgrowth in PC 12 cells occurred following 2 d exposure to
therapeutic concentrations of VP A (p<0.05), as shown in Figure 5.3. In cells exposed to a
higher concentration of VP A (1.8 mM), neurite outgrowth appeared to be no different
from that of controls. It is of note that, similar to effects observed in HN33 cells, the
PC 12 cells exhibited an inverse correlation between cell density and neurite outgrowth
(p<0.005; data not shown).
Figure 5.3. VPA-induced neurite outgrowth in PC 12 cells. Cells were exposed to
VPA for 2 d, and then photographs were taken for comparison. Data are presented
as mean percent (± S.E.M.) of cells displaying one neurite or more. *p<0.05
Effects of VPA and the PKC Inhibitor BIM on PC12 Cell Number and Neurite
Outgrowth
Non-differentiated PC 12 cells exposed to either 0.6 mM or 1.8 mM VPA exhibited
no alteration in growth rate compared to controls, as evidenced by lack of difference in
cell density (Figure 5.4 a and b). Similarly, 200 nM BIM, either alone or in combination
with VPA, elicited no alteration in overall cell density. Following exposure to 0.6 mM

96
VP A, PC 12 cells exhibited increased neurite outgrowth, which was unaffected by the
addition of 200 nM BIM (p<0.05; Figure 5.4c). Neither BIM nor 1.8 mM VP A, alone or
in combination, had any effect on neurite outgrowth in PC 12 cells (Figure 5.4d). These
results suggest that, at concentrations of VP A therapeutic for mood-stabilization (e.g., 0.6
mM), PC 12 cells exhibit increased neurite outgrowth, and that this effect is independent of
PKC.
a
b
Control 1.8 mM VPA 200 nM BIM VPA + BIM
Treatment
Control 0.6 mM VPA 200 nM BIM VPA + BIM
Treatment
Figure 5.4 Effects of VPA and BIM, a PKC Inhibitor, on PC 12 Cell Density and
Neurite Outgrowth. Non-differentiated PC 12 cells were exposed to 0.6 mM
VPA, 1.8 mM VPA, 200 nM BIM, or some combination thereof for 24-48 h.
Shown in (a) and (b) are mean cell density (± S.E.M.) following exposure to 0.6
mM VPA and 1.8 mM VPA, respectively. Shown in (c) and (d) are percent of
total population exhibiting neurites (± S.E.M.) following exposure to 0.6 mM
VPA, 1.8 mM VPA, 200 nM BIM, or a combination. *p<0.05

97
Post-NGF T reatment
Figure 5.5. Effect of VP A on NGF-Differentiated PC12 Cells. Differentiation of PC12
cells was induced by 4-7 d exposure to 100 ng/ml NGF, and then NGF was either
removed or replaced, and VPA was added at various concentrations (0.3, 0.6, 1.2
mM). NGF-positive control is represented as baseline (0 on y-axis). (a) Cell Density.
Shown is mean (± S.E.M.) number of cells, expressed as a percent of NGF-positive
control values. No protective effect was observed (b) Neurite Outgrowth. Shown is
mean percent (± S.E.M.) of cell population exhibiting neurites in treatment groups
compared to NGF-positive controls. Protection against loss of neurites was observed
*p<0.05

98
Neuroprotective Effect of VP A on NGF-Differentiated PC12 Cells
Following NGF-induced differentiation of PC 12 cells, varying concentrations of
VPA (0.3, 0.6, 1.2 mM) were added, alone or in combination with NGF (100 ng/ml).
NGF-deprived cultures experienced a loss of neurites and decreased cell number
(attributable to cell death), as shown in Figure 5.5. VPA-treated cultures (alone or in
combination with NGF) exhibited a decreased rate of neurite loss, though this effect was
no greater than that elicited by NGF alone (p<0.05; Figure 5.5b). Exposure to the
combination of VPA and NGF resulted in a tendency, albeit insignificant, toward
decreased cell death, while VPA alone had no apparent effect (Figure 5.5a). These results
suggest that VPA is protective against neurite loss in NGF-differentiated PC 12 cells, but
not protective against cell loss associated with trophic factor deprivation.
Discussion
HN33 Cells
Exposure of HN33 cells to therapeutic concentrations (up to 1.0 mM) of VPA for
up to 14 d produced no apparent cytotoxicity, and cells appeared healthy and viable when
assessed by Trypan blue dye exclusion assay (data not shown). However, a concentration-
dependent increase in cell doubling time was noted. Following long-term exposure,
cellular growth rates were notably slower for HN33 cells exposed to 0.6 mM VPA. This
effect was reversible upon removal of the drug, and could be avoided altogether by plating
cells more densely (Table 5.2). These findings are similar to reported effects of VPA on
glioma (C6) and neuroblastoma (neuro 2A, UKF-NB-2, UKF-NB-3) cells, in which VPA
exhibited a reversible anti-proliferative effect which was not attributable to cytotoxicity

99
(Martin and Regan, 1991; Cinatl et al., 1996). In one particular study, VP A was found to
arrest cells at the post-mitotic mid-Gl phase of the cell cycle (Martin and Regan, 1991).
The next parameter assessed in the HN33 cells was that of differentiation.
Following chronic (>1 d) exposure to therapeutic concentrations of VP A, cells exhibited a
significant increase in neurite outgrowth consistent with previous reports (Martin and
Regan, 1991; Cinatl et al., 1996). The increase in neurite outgrowth observed in this
study occurred over a time course similar to that of the MARCKS down-regulation
observed (see Chapter 3). While no significant alterations in either parameter were
observed at 1 d following 0.6 mM VP A exposure, both cell morphology (neurite
outgrowth) and MARCKS expression had undergone significant changes by 3 d.
Following continued VPA exposure, little or no change was evident in either MARCKS or
cell morphology between 3 d and 5 or 6 d, but a significant increase in neurite outgrowth
was observed by the 8 d time point, which corresponds roughly to the further decrease in
MARCKS protein observed between 6 and 10 d (see Figure 3.5).
Although these similarities in overall time course are interesting, they may simply
point to the notion that VPA-induced alterations in both cell morphology and MARCKS
expression are effects observed only following chronic, but not acute, exposure to this
drug. Previous studies would suggest that these two parameters are unrelated, for in this
particular cell model, chronic lithium exposure also resulted in significant reductions in
MARCKS expression, but with no accompanying change in cell morphology (Watson and
Lenox, 1996). Further, exposure of HN33 cells to either of the VPA metabolites
hydroxyvalproic acid or 2-propylglutaric acid, whose structures are analagous to that of
VPA, produced no change in cell morphology, and only one metabolite, at high
concentrations, had a notable effect on MARCKS.
On the other hand, retinoic acid, a known differentiating agent, has been shown to
elicit both MARCKS down-regulation and neurite outgrowth in these cells (Watson et al.,
1994). Further, the effect of lithium on MARCKS differs from that of VPA in the HN33

100
cells. Whereas lithium appears to exert its greatest effect on MARCKS in the soluble
fraction of these cells, and requires a longer exposure, VPA exposure results in a greater
reduction in MARCKS expression, at a much earlier time point (3 d for VPA vs. 10 d for
lithium), and the effect is most pronounced in the membrane fraction of these cells. As has
been suggested previously, VPA and lithium likely operate through two different or only
partially overlapping mechanisms. While lithium works in part through its effect on the PI
signaling system, VPA does not. Similarly, VPA exposure leads to effects on cell growth
and morphology which may or may not be related to its therapeutic efficacy. Indeed, the
observed VPA-induced effects more than likely contribute to this drug’s unique
therapeutic profile, as will be further discussed in Chapter 6.
It is clear from previous studies that morphological alterations are not a necessary
prerequisite for down-regulation of MARCKS protein expression, for lithium exposure
results in MARCKS alterations without accompanying changes in HN33 cell growth or
morphology. Whether MARCKS down-regulation is necessary for the kinds of changes in
cell growth and morphology necessary to cell differentiation is a more difficult question to
answer based on the data presented, but will be further discussed below. Future studies
might further address this phenomenon in a variety of cell lines simply by assessing
MARCKS expression levels following differentiation.
As for GAP-43 and its potential involvement in the morphological alterations
observed, there is extensive evidence in the literature linking GAP-43 with neurite
outgrowth, as discussed in Chapter 3. In this study, both parameters (GAP-43 and neurite
outgrowth) continued to increase throughout the length of VPA exposure examined.
This, along with the fact that lithium had no effect on either GAP-43 expression or neurite
outgrowth (data not shown), suggests that the effects are somehow related. Although
necessity and/or sufficiency have not been addressed sufficiently by the data presented,
there is a wealth of evidence which supports the notion that increased GAP-43 levels
greatly contribute to the overall process of neurite outgrowth and cell differentiation. As

101
discussed in Chapter 3, neurite outgrowth is accompanied by a large increase in GAP-43
expression, and is thought to help stabilize growth cone morphology and promote growth
cone adhesion (Aigner and Caroni, 1995). Antisense-directed depletion of GAP-43 in
growth cones results in the absence of spreading, branching, and adhesion of growth cones
(and therefore, neurite outgrowth) in primary sensory neurons (Aigner and Caroni, 1993).
Suppression of GAP-43 expression has been found to yield adverse effects on axon
outgrowth both in vivo and in vitro (Meiri et al., 1998; Benowitz and Routtenberg, 1997;
Strittmatter et al., 1995; Jap Tjoen San et al., 1992; Shea et al., 1991), and
overexpression of GAP-43 induced nerve sprouting in the adult nervous system of
transgenic mice (Aigner et al., 1995). Even in non-neuronal cells, expression of a GAP-43
transgene has been found to induce extensive process outgrowth accompanied by
reorganization of the membrane cytoskeleton (Strittmatter et al., 1994; Yankner et al.,
1990; Zuber et al., 1989). Skene and Willard (1989) have suggested that the induction of
GAP (GAP-43 and other growth-associated proteins) expression is necessary to post-
injury remodeling, and that the inability of many areas of the CNS to induce high levels of
GAP expression may in part underlie the failure of CNS axons to regenerate after axonal
injury, whereas areas of the peripheral nervous system retain the ability to up-regulate
GAP-43 as well as to regenerate axons after injury. The presence of GAP-43 is probably
not necessary to neurite outgrowth per se, but rather, plays an important role in regulating
precisely the various parameters involved in the overall event, such as navigation or
pathfmding (Aigner and Caroni, 1995). According to this “GAP hypothesis”, certain
growth associated proteins, such as GAP-43, provide a mechanism for regulating neurite
outgrowth and structural plasticity in the nervous system (Skene and Willard, 1989).
With regard to PKC activity and isoform expression, and their association with the
VPA-induced neurite outgrowth observed in HN33 cells, all effects occurred at the same
therapeutically relevant concentration of VP A (0.6 mM). The alterations in PKC activity
appear to precede the morphological effects, for membrane-associated PKC activation,

102
followed by down-regulation, occurred within hours following addition of VP A to the
culture medium, whereas neurite outgrowth was observed only following chronic VPA
exposure. Similarly, alterations in PKC isoform expression were observed acutely, within
1-2 h following VPA addition, as well as following longer-term VPA exposure (1-3 d).
Although reports are inconclusive and even contradictory as to the either positive
(Abraham et al., 1991; Parrow et al., 1992; Cabel and Audesirk, 1993; Coleman and
Wooten, 1994) or negative (Heikkila et al, 1989; Parodie/ al, 1990; Tsuneishi, 1992;
Carlson et al., 1993; Jalava et al., 1993; Chakravarthy et al., 1995; Ekinci and Shea,
1997) regulatory role of PKC in neuronal differentiation, each of the four PKC isozymes
present in HN33 cells has, in one report or another, been implicated in neurite outgrowth
and/or cellular differentiation, as discussed at length in Chapter 4. The present study
supports the association of neurite outgrowth with either one or any combination of the
following: (a) acute induction followed by down-regulation of PKC-a expression, (b)
chronic up-regulation of PKC-5, (c) apparent translocation (consistent with activation) of
PKC-e from cytosol to membrane, followed by down-regulation, and/or (d) chronic up-
regulation of PKCThe findings discussed herein are consistent with previous studies
showing alterations in PKC-a mRNA and/or protein expression observed concomitant
with differentiation of PC 12 and other cell lines (Borgatti et al., 1996; Parrow et al., 1995;
Murray et al., 1993; Leli etal., 1993; Tonini etal., 1991; Wada et al., 1989). However,
these reports differ in their mechanistic conclusions, for in three studies, PKC-a
expression increased with differentiation (Borgatti et al., 1996; Parrow et al., 1995;
Murray et al., 1993), while in three other studies, PKC-a levels had decreased following
comparable treatment and neuritogenesis (Leli et al., 1993; Tonini et al., 1991; Wada et
al., 1989). The findings presented herein are also consistent with reports which have
provided evidence for a role of PKC-s in regulating neuronal differentiation, since neurite
outgrowth has been associated with altered PKC-e levels (Ponzoni et al, 1993; Parrow et
al., 1995; Fagerstrom et al., 1996; Ekinci and Shea, 1997; Hundle et al, 1997; Zeidman et

103
al., 1999). Leli and Wada observed increased expression of PKC-e concomitant with
neurite outgrowth (Leli et ai, 1993; Wada et ai, 1989), and another report indicated that
overexpression of PKC-e led to an increase in NGF-induced neurite outgrowth in stably
transfected PC 12 cell lines (Hundle et ai, 1995). Still other reports emphasized the
importance of PKC-£ in neuronal differentiation, showing that PKC-£ overexpression is
able to induce cellular differentiation in transfected leukemic cells (Ways et al., 1994), and
antisense-mediated down-regulation of PKC-£ inhibits NGF-induced neurite outgrowth in
PC 12 cells (Coleman and Wooten, 1994).
In this study, inhibition of PKC by LY333531 alone increased neurite outgrowth in
HN33 cells compared to that of untreated controls. LY333531 also enhanced the VPA-
induced increase in neurite outgrowth, and decreased HN33 cell density, when the two
drugs were combined. These data suggest that, rather than serving as the necessary
intermediate for the effects of VP A on HN33 cell growth and morphology, as initially
hypothesized, one or more isoforms of PKC may actually be inhibitory toward cell
differentiation. PKC inhibition not only elicited effects on its own, but enhanced the
observed effects of VP A on these parameters. Such findings suggest that VPA may
operate through inactivation of PKC to elicit observed effects on cell growth and
morphology, or perhaps through a separate mechanism altogether. The finding that PKC
was not a necessary intermediate in the VPA-induced GAP-43 induction observed
(Chapter 4) is consistent with these data, since increased GAP-43 expression is associated
with cell differentiation, as discussed in Chapter 3.
A novel finding observed, but not quantified, in this study was the increased
adherence of VPA-treated cells to the tissue culture flasks in which they were grown.
Another group reported a similar effect in C6 glioma and primary astrocytes (Martin et al.,
1988). Further investigation revealed that VPA induced a significant increase in the
expression of the collagen type IV receptor, a heat-sensitive protein thought to be
responsible for the increased adherence to substrate (Martin and Regan, 1988). The heat

104
shock proteins specifically respond to environmental stress, including that of teratogenic
burden. This reversible, stress-induced effect is suggestive of a mechanism for the
teratogenic effects of VPA.
PC12 Cells
Insofar as VPA induced morphological alterations in HN33 cells, we next
investigated the effect of VPA on PC 12 cells, a model in which neurite outgrowth and
differentiation have been well characterized. Indeed, exposure of PC 12 cells for 2 d to
therapeutic concentrations of VPA yielded a significant and dose-dependent increase in
neurite outgrowth. Future studies may examine alterations in PKC substrates in these
cells following chronic VPA exposure, in order to strengthen the link already proposed for
protein expression during such cellular events.
The PKC inhibitor BIM was used to assess the involvement of PKC in the
observed VPA-induced increase in neurite outgrowth in PC 12 cells. It was found that
BIM alone had no effect on either PC 12 cell density or neurite outgrowth. Similarly, a
high concentration of VPA (1.8 mM, well above therapeutic) had no effect on either
parameter At a therapeutic concentration of VPA (0.6 mM), however, neurite outgrowth
increased with or without addition of BIM, suggesting that (1) VPA, at concentrations
therapeutic for mood stabilization, may serve as a differentiating agent in PC 12 cells, and
(2) VPA-induced differentiation of these cells does not involve PKC, as inhibition of the
enzyme had neither a positive nor negative impact on the effect of VPA.
These findings are consistent with those observed following VPA exposure in the
HN33 cells, insofar as therapeutic concentrations of VPA elicited increased neurite
outgrowth, a process which did not appear to rely on PKC. However, the findings in
PC 12 cells differ from those of the HN33 cells, in that PKC inhibition did not enhance the

105
VPA-induced neurite outgrowth, nor did it have any effect when added alone. Further,
PC 12 cell growth was unaffected by either VPA or BIM, or the combination.
A number of recent studies have investigated the role of PKC in neuroprotection.
Results are conflicting, but often support such a role for PKC. In a model of ischemic
preconditioning-induced neuroprotection (utilizing a two-vessel occlusion protocol),
altered distribution and levels of PKC-y were observed in rat hippocampal CA1 and
neocortex regions, suggesting an association between PKC and neuroprotection (Shamloo
and Wieloch, 1999). Yet, in either El8 rat cortical cultures or hippocampal slices,
preconditioning studies utilizing oxygen-glucose deprivation or anoxic depolarization
paradigms, respectively, showed no evidence for a role of PKC, either stimulatory or
inhibitory, in these mechanism of anoxic preconditioning-induced neuroprotection (Perez-
Pinzon and Born, 1999; Tauseka et al., 1999). Other studies have also suggested a
positive role for PKC in neuroprotection. Cerebral ischemia and glutamate- or NMDA-
induced cell death have been shown to follow PKC inactivation (Pizzi et al., 1996;
Tremblay et al., 1999). Further, neuroprotective agents including low-dose amyloid
precursor protein, which increases uptake of the excitatory amino acids aspartate and
glutamate (Masliah etal., 1998), thyrotropin-releasing hormone, which protects against
NMD A toxicity (Pizzi et al., 1999), and vasoactive intestinal peptide, which was found to
be neuroprotective against ibotenate-induced white matter lesions (Gressens et al., 1998),
all were dependent upon PKC activation for their overall neuroprotective effect. In
contrast, still other reports suggest that PKC activity actually contributes to neurotoxicity.
In toxicity models involving anoxia, glucose deprivation, or nitric oxide toxicity, growth
factors were found to protect against cytotoxicity and increase cell survival, independent
of PKC (Boniece and Wagner, 1993; Maiese and Boccone, 1995). Down-regulation or
inhibition of PKC alone prior to insult was found to be cytoprotective, supporting the
notion that PKC is not neuroprotective, but rather, a contributor to toxicity in certain
models.

106
One series of experiments was designed to test the hypothesis that VPA had
neuroprotective activity in addition to its neurite-stabilizing role. Inasmuch as various
animal models have provided evidence for progressive seizure-induced neuronal damage,
VPA has been investigated as a potential neuroprotective agent (Jallon, 1997).
Furthermore, lithium was shown in a recent report to be neuroprotective against NMDA-
mediated glutamate-induced excitotoxicity (apoptosis) in cortical, hippocampal, and
cerebellar granule cells. This effect, which apparently involved modulation of NMD A-
mediated Ca++ influx, was both time- and concentration-dependent, and independent of
inositol concentration (therefore independent of IMPase) and not a result of alterations in
either NMDA receptor composition or number (Nonaka et al., 1998). Other studies, in
cultured cerebellar granule cells as well as rat frontal cortex, hippocampus, and striatum,
showed chronic lithium-induced up-regulation of BCL-2 mRNA and protein levels, as well
as protection against glutamate-induced cytotoxicity (Chen and Chuang, 1999; Manji et
al., 1999). Bcl-2 is a proto-oncogene thought to be protective against cell death arising
from a number of different pathways, including apoptosis, and may also play a role in the
regulation of neural cell differentiation (Vaux, 1993; Naumovski and Cleary, 1994; Yang
and Korsmeyer, 1996; Zhang et al., 1996).
NGF-differentiated PC 12 cells were used as a model for neuroprotection, as NGF
deprivation following cell differentiation leads to apoptotic cell death. In this study, VPA
was added to NGF-differentiated cells, either alone or in combination with NGF, and both
cell number and neurite outgrowth were measured. No effect of VPA on cell density was
observed; VPA-treated cultures exhibited a loss in cell number similar to that observed in
NGF-deprived cultures. With regard to neurite outgrowth, VPA, alone and in
combination with NGF, appeared to protect differentiated PC 12 cells against loss of
neurites following NGF withdrawal. These findings are consistent with those observed in
non-differentiated PC 12 and HN33 cells, in that VPA induced neurite outgrowth and did
not stimulate (at times, even inhibited) cell growth. VPA appears to have selective effects

107
on different cell populations. With regard to the neuroprotective potential of VP A against
apoptosis, findings are inconclusive: VPA exhibited protective tendencies toward neurites,
but did not protect against cell loss, following NGF-induced differentiation and subsequent
withdrawal in PC 12 cells. These findings agree with recent preliminary observations by
Watson (personal communication), which show no evidence for neuroprotection by acute
(8 h) VPA exposure against glutamate-induced cytotoxicity in hippocampally-derived HT-
22 cells. The ability of anticonvulsants such as VPA to decrease the frequency or severity
of epileptic episodes does not necessarily confer or reflect neuroprotective properties.
Such drugs are prophylactic against further seizure-induced damage, but these drugs are
not generally considered to possess the power to correct the pathology underlying
epilepsy (Jallon, 1997). Instead, these studies suggest that perhaps VPA is able to
stabilize existing synaptic connections. Alternatively, VPA may have protected and
induced differentiation in cells which had not already differentiated in response to NGF,
while cells which were NGF-differentiated were not protected by VPA and subsequently
died.
Additional Comments
The ability of VPA to effect changes in plasticity correlates well with previous
evidence presented by other groups showing significant developmental effects of this drug.
It has been known for some time that epileptic women have a greater incidence of birth
defects in their offspring than do nonepileptics, and that certain anticonvulsant drugs used
to treat these women are largely responsible for this teratogenicity (Golbus, 1980; Nakane
et al, 1980). VPA has been the focus of many studies in this regard, as its prenatal use
has been associated with a high incidence of offspring bom with various dysmorphic birth
defects, collectively referred to as the Fetal Valproate Syndrome (Daston et al, 1995;
Ardinger et al, 1988; Wiger et al, 1988; Winter et al, 1987; Chessa and Iannetti, 1986;

108
Tein and MacGregor, 1985; DiLiberti et al., 1984; Clay et al., 1981; Nau et al., 1981;
Dalens et al., 1980). In addition to a number of somatic defects consistently observed,
infants exposed prenatally to VPA also exhibited a strikingly higher incidence of neural
tube defects than the general population (Bjerkedal et a/., 1982; Robert and Guibaud,
1982; Lindhout and Schmidt, 1986).
A number of animal studies have shown that prenatal exposure to VPA at
physiologically relevant plasma levels results in a tendency toward decreased size and
weight of offspring (Ehlers et al., 1996; Hansen and Grafton, 1991), as well as increased
incidence of embryolethality (Hansen et al., 1995b; Ehlers et al., 1996). There is also
overwhelming evidence for the teratogenic effects of VPA on the rodent neural tube: data
indicate that even a single treatment with VPA at critical developmental times can result in
such neural tube defects as exencephaly, spina bifida aperta, and spina bifida occulta in
offspring (Hansen et al., 1995b; Trotz et al., 1987). Hansen's group showed that
offspring of mice exposed in vivo on gd 8 to one subcutaneous injection of VPA exhibited
a significant increase in the rate of exencephaly as compared to controls (Hansen et al.,
1995b); in a separate study, in vitro rat embryos exposed to VPA for 44 h beginning on gd
9 exhibited a higher incidence of open neural tubes (Hansen and Grafton, 1991). Other
studies have shown similar teratogenic abnormalities, as well as defects in cranial nerve
development, of in vitro cultured rat, mouse, or chick embryos exposed to VPA for just
24-48 h (Chatot et al., 1984; Kelly and Regan, 1992; Nosel and Klein, 1992; Gofflot et
al., 1996). Reports reflect comparable increases in exencephaly and other neural tube
defects in mouse fetuses briefly exposed to VPA in vivo during early gestation (Trotz et
al., 1987; Ehlers et al., 1996). These effects are indicative of a disruption in formation of
the neural tube, rather than simply a neurodevelopmental delay (Kelly and Regan, 1992).
Both retrospective and prospective studies have attempted to examine this
phenomenon, but the pattern of malformations caused by VPA appears to vary greatly
across species (Ehlers et al., 1996), and findings differ from one report to the next. One

109
striking consistency among all studies is the selectivity of VP A for defects of the midline.
Perhaps the specificity of action of VP A on G1 phase of the cell cycle may help explain
the drug's specificity in inducing neural tube defects (Martin and Regan, 1991). Some
studies suggest peak plasma levels of VP A itself, and not its metabolites, are responsible
for the drug's teratogenicity; it has also been suggested that the drug accumulates to a
greater degree in embryonic target tissues relative to maternal plasma concentrations
(Menegola et al., 1996; Kelly and Regan, 1992; Martin and Regan, 1991). Thus far, no
consensus has been reached as to the putative mechanism of this drug's teratogenic effects,
and data are equally inconclusive with regard to preventive regimens. Whereas folic acid
supplementation has been reported to prevent anticonvulsant-induced neural tube defects
in rats and mice in vivo (Ehlers et al., 1996; Nosel and Klein, 1992; Trotz et al., 1987),
the in vitro evidence is not as compelling. In vitro reports in general have shown a failure
of folic acid or its metabolites to prevent VPA-induced neural tube defects in mouse or rat
embryos (Hansen and Grafton, 1991; Hansen et al., 1995). Additionally, preliminary
findings from our laboratory, in which we used the immortalized hippocampal cell model
to investigate the preventive effect of folic acid or methionine (a folic acid metabolite) on
VPA-induced MARCKS down-regulation, showed no prevention or reversal of the
MARCKS effect (data not shown). These data suggest that folic acid alone is not
responsible for prevention of the effects elicited by VP A, but rather, that a more complex,
unknown, metabolic phenomenon is somehow altered by folic acid supplementation, which
in turn allows modulation of VPA-induced effects.
These findings suggest an effect of VP A on cellular differentiation during early
developmental stages, and our data are consistent with previous findings. Such dramatic
and selective effects on central nervous system development suggest that VPA may be
altering the cellular developmental timing program at a molecular level. In accordance
with this hypothesis, cells would experience a shift in their developmental timing program,
from the growth phase, in which they divide and multiply, to the differentiating phase,

110
during which growth stops and specialized functional differentiation begins. The key is
that this differentiation commences at a time prior to growth completion, and therefore
before the cells are organized at a tissue level; as a result, the whole organism exhibits
such gross abnormalities as open neural tube. Our findings and those of others (Martin et
al., 1988), in which chronic VPA exposure resulted in decreased growth rates and
increased morphological differentiation, are consistent with this hypothesis. Regan’s
group further suggests that agents such as VPA, which significantly alter mitosis and cell-
substratum adhesivity, can be predicted to induce neural tube defects, for cell proliferation
and related functions are known to be integral to neural tube formation (Martin and
Regan, 1988).
It is also of note that, while prenatal VPA exposure often leads to neural tube
defects, lack of MARCKS protein results in similar effects. In a transgenic mouse model
homozygous null for the MARCKS gene Macs, all pups died perinatally and exhibited a
number of neural tube and other midline defects (Stumpo et a/., 1995; Blackshear et al.,
1996). Swierczynski et al. (1996) subsequently showed that transgenic expression of the
human MARCKS gene MACS in these knockout mice effectively rescued the MARCKS-
deficient phenotype and allowed for survival of the transgenic mice. These studies suggest
that MARCKS is essential for normal CNS development, and that lack of MARCKS
results in abnormal development of various midline structures, and even death. These
findings are of interest in light of the present study, in which we have presented evidence
for VPA-induced down-regulation of MARCKS and associated morphological alterations.
Our findings suggest a putative mechanism for the developmental abnormalities associated
with prenatal VPA exposure That there is extensive evidence of VPA-induced neural
tube defects, and that VPA elicits a down-regulation in MARCKS, a protein without
which normal neural tube development cannot proceed, is intriguing. While our findings
are merely suggestive, and more extensive studies will need to be done, this is perhaps the

Ill
first attempt to attribute MARCKS (or lack thereof) with a significant role in the
mechanism of VPA-induced neural tube defects.
Beyond the developing CNS, such VPA-induced effects are likely to be relevant to
plastic or regenerating areas of the adult brain, as well. Alterations in growth,
morphology, and PKC substrate expression have implications for mood stabilization as
well as for seizure therapy. Indeed, studies have shown that seizures induce structural and
functional alterations in the brain, at cellular and molecular levels, which lead to further
downstream changes in gene expression, cell morphology and function (Jallon, 1997). The
notion of synaptic remodeling as the underlying means for the drug- or therapy-induced
stabilization of mood disorders has also been supported by some reports, as previously
discussed. One study found increased levels of N-CAM-immunoreactive proteins in the
cerebrospinal fluid of a population of bipolar patients (Poltorak el a!., 1996), which may
be suggestive of abnormal synaptic plasticity in affected individuals. Lithium and VPA
affect transcriptional factors which regulate gene expression, and it is known that long¬
term changes in neuronal synaptic function are dependent upon the selective regulation of
gene expression (Manji and Lenox, 1994). Lamotrigine was recently shown to alter
neuronal hyperexcitability via inhibition of both Na and Ca channels, perhaps signifying
modification of synaptic plasticity, as the authors suggest (Xie and Hagan, 1998). These
findings are suggestive of changes in synaptic plasticity as a putative mechanism by which
anti-bipolar therapy facilitates long-term changes in affect.
Perhaps the dramatic effects observed in this study on cellular protein expression
and growth/morphology, in some combination, result in alterations in cellular signaling
which are responsible for the therapeutic effects of VPA In other words, VPA-induced
molecular alterations may underlie some functional change, as yet undefined, which is
responsible for modifying CNS plasticity at a cellular level This may, in turn, allow for
remodeling of the tissues in select areas of the brain, such as the limbic system, to better
accommodate appropriate cell signaling events, such as those responsible for mood As

we have shown, these molecular events occur following chronic but not acute VP A
exposure, and this speaks to the therapeutic relevance of our findings: the functional
changes necessary for long-term mood stabilization must be such that they lag in onset
are not easily reversed.

CHAPTER 6
CONCLUSIONS AND FUTURE DIRECTIONS
VPA has proven to be very useful in its application to both seizure disorders and
bipolar disorders Because of the relative dearth of mood-stabilizing therapies available
today, it is the latter application on which these research efforts have been focused. Over
the past two decades, a number of reviews have been published on the efficacy of VPA in
treating bipolar affective disorder (Keck and McElroy, 1996; McElroy and Keck, 1993;
McElroy, 1989; Hayes, 1989; Brown, 1989; Fawcett, 1989). From various studies, it has
become clear that VPA elicits a significant antimanic response rate, even in lithium-
refractory patients and others who display more resistant forms of bipolar disorder, such
as the rapid cycling variant (Calabrese and Delucchi, 1989; Brown, 1989). Further, VPA
has proven to be quite useful and safe in combination with lithium or as a long-term
prophylactic drug therapy (Granneman et al., 1996; Hayes, 1989). Because of its broad
therapeutic range (actually thought to be even broader than the formerly established 50-
100 |ig/ml range), VPA can be prescribed with less consideration of risk and less close
monitoring. In fact, VPA has received more attention as a first-line mood stabilizer
(before lithium or carbamazepine), because of its efficacy, relatively rapid onset of action,
favorable side effect profile, and relative ease of administration (McElroy and Keck,
1993).
The findings of our current and previous studies are summarized in Table 6.1.
Both chronic VPA and lithium share the property of regulating MARCKS expression, at
clinically relevant concentrations (0 6 - 1 0 mM VPA was used for all experiments in this
study), to the exclusion of all other psychotropic agents examined to date. Comparisons
between VPA and lithium are instructive, for these two FDA-approved mood-stabilizing
113

114
agents share a striking number of differences, and further investigation may begin to
answer some very important questions regarding pathophysiology as well as drug therapy.
Indeed, the mood-stabilizing properties of these two drugs may be partially attributable to
the similarities observed on a cellular level (such as the similar effects on PKC-a), whereas
their differences, including morphological and cell growth effects of VP A, may contribute
to the unique anticonvulsant properties of VPA as well as its severe developmental effects.
The fact that only VPA and lithium elicit a significant reduction in MARCKS suggests that
alterations in MARCKS may be involved in the mechanism of action of these agents, and
supports the identification of MARCKS as a potential molecular target for the action of
drugs used in the treatment of bipolar disorder Based on previous reports, MARCKS is
likely to play an important role in modulating CNS plasticity. Although the precise role of
MARCKS alterations remains unknown, the identification of common targets in the brain
for these drugs under conditions that reflect both effective therapeutic concentrations and
the time course of clinical action may provide a unique window into the pathophysiology
of bipolar disorder. It may be that MARCKS is an intermediate, and that something
downstream of MARCKS is responsible for the anti-bipolar effect of these drugs. Even if
the regulation of MARCKS is a clinically irrelevant "side effect" of mood-stabilizing drug
therapy, it may be a useful marker nonetheless. Such findings may offer us opportunities
for the design and development of the next generation of mood stabilizers, compounds
which will ideally possess broad spectrum of efficacy, minimal side effects, and widened
therapeutic index, all of which will contribute to increased patient compliance.
On the other hand, the finding that carbamazepine, another widely used mood-
stabilizing drug therapy (although not FDA-approved for this application), as well as some
less commonly used mood-stabilizing agents, including lamotrigine (data not shown) and
verapamil, did not result in alterations in MARCKS expression, points to the possibility
that either (1) MARCKS down-regulation itself is not integral to mood-stabilizing
efficacy, or (2) the in vitro exposure conditions and model employed were not optimal to

115
allow the down-regulation which may have been observed following chronic exposure in
vivo. Alternatively, MARCKS may activate some protein or process downstream which is
responsible for the anti-manic, mood-stabilizing efficacy of these drugs; while lithium and
VPA arrive at this downstream signal via MARCKS, carbamazepine and other anti-bipolar
drugs may arrive at the same signal through some alternate mechanism, so that all anti-
bipolar drugs do actually converge upon the same pathway ultimately, albeit through
different routes.
Whereas MARCKS down-regulation may not be necessary to the successful
amelioration of bipolar symptoms, the question remains as to whether this effect, alone, is
sufficient for bipolar therapy. Since other drugs are known to be effective in treatment of
bipolar disorder without affecting MARCKS expression, and since non-therapeutic
differentiating agents such as phorbols and retinoic acid are capable of eliciting MARCKS
alterations, one may conclude that MARCKS down-regulation alone is not likely to lead
to successful bipolar treatment in all cases. Indeed, lithium and VPA elicit other effects
aside from MARCKS alterations, and such effects may also contribute to the overall
mood-stabilizing action of these drugs.
As for the VPA-induced alterations in HN33 cell morphology and GAP-43
expression, such effects are not likely to be necessary or sufficient for the successful
amelioration of bipolar symptoms. This is evidenced by the fact that lithium, the
prototypic mood-stabilizing drug, had no effect on either parameter in our in vitro cell
model, and that many agents may elicit GAP-43 alterations without necessarily exhibiting
mood-stabilizing efficacy. One must question, then, the significance of the observed
effects of VPA on these parameters, which seem to be related, based on previous findings
Numerous reports have emphasized GAP-43 as a protein integral to normal neuronal
differentiation. Although the sequence of events is unknown, it is generally believed that
cell differentiation is accompanied by a reduction in cell growth rates, the outgrowth of
neuritic processes, and an increase in GAP-43 in neurons. The findings discussed herein

116
implicate VPA in invoking cellular differentiation, for all three of these parameters were
observed. Perhaps VPA, by altering GAP-43 expression and cell growth/differentiative
state, in turn alters the plasticity and function of areas in the brain that are integral to
mood. According to such a scenario, the observed alterations in both MARCKS and
GAP-43 may go hand-in-hand to elicit the plastic changes underlying clinical antimanic or
mood-stabilizing efficacy. Indeed, animal models of epileptic kindling have shown a
progressive and permanent alteration in neuronal structure and function. Perhaps a similar
change underlies the pathology involved in bipolar disorder, so that for both diseases,
VPA helps to bring about plasticity changes important in stabilizing the course of illness
over time.
These findings are also suggestive of a mechanism to explain the developmental
effects of VPA. It has been well established that prenatal VPA exposure results in a high
frequency of neural tube defects, although the mechanistic basis for such effects is unclear.
These data suggest that VPA elicits a change in the developmental program whereby cell
growth is prematurely halted, prior to growth completion, and specialized functional
differentiation begins before the tissue or organ has adequately prepared macroscopically
for advanced developmental stages. As a result of such dramatic effects on functions (cell
proliferation, differentiation, and substrate adhesivity) known to be integral to neural tube
formation, the whole organism exhibits such gross abnormalities as open neural tube.
Other reports are consistent with this hypothesis and the findings discussed herein (Wilk et
al., 1980; Mummery et al., 1984; Martin and Regan, 1988; Martin et al., 1988). In fact,
the VPA-induced down-regulation of MARCKS also fits well and may contribute further
to this model As mentioned previously, mice lacking MARCKS due to antisense-
mediated gene knockout die perinatally, exhibiting a number of gross abnormalities in
midline structure formation (Stumpo et a/., 1995; Blackshear et al., 1996).
Perhaps the ratio of MARCKS to GAP-43 is an important signal which, when out
of balance, leads to a pathological, even fatal, phenotype. According to this hypothesis,

117
relatively higher MARCKS expression, concomitant with low GAP-43 expression, would
signal for growth (cell division), and low MARCKS / high GAP-43 would signal for
growth arrest and induction of differentiation. The data are thus far consistent with such a
model, and future studies might further address the issue by examining and comparing
relative MARCKS and GAP-43 levels in both control (unaffected) animals and those
exhibiting abnormalities of the neural tube and other CNS regions. Such studies would be
of great benefit medically, for they would help identify a mechanism for the development
of neural tube defects in human offspring, and may even lead to diagnostic means and
preventive measures to deal with such effects.
Expression of both GAP-43 and MARCKS is directly regulated by the enzyme
PKC. If we understand better the role of PKC, we may also understand better the
relevance of the observed VPA-induced effects. These studies have shown that PKC is
clearly involved in the intracellular effects exhibited by VPA. However, PKC is apparently
not central to the entire panorama of effects observed, for exposure of HN33 cells to
inhibitors of PKC failed to halt the observed VPA-induced alterations in GAP-43
expression, and prevented the VPA-induced down-regulation of MARCKS only in the
membrane, not cytosolic, fraction of cells. This raises the issue of how such effects could
occur, especially in the case of GAP-43, if PKC is the primary regulator of both proteins.
Previous studies have indicated that phosphorylation of GAP-43 is spatially and
temporally restricted in vivo (Meiri et al., 1991), and is implicated in LTP, a model for
memory and learning (Stabel and Parker, 1991; Benowitz and Routtenberg, 1997; Son et
al., 1997), in part due to its association with state of growth and sprouting or neurite
extension (Benowitz and Routtenberg, 1997). Meiri et al. (1998) showed that neurite
outgrowth stimulated by neural cell adhesion molecules required GAP-43
phosphorylation Yet another group showed that modulation of membrane-associated
PKC activity played a necessary role in cell differentiation (Chakravarthy et al., 1995).
Such findings would suggest that PKC activity (specifically, phosphorylation of GAP-43)

118
is integral to neurite outgrowth. However, the role of PKC in neuronal differentiation
remains in question. While there is substantial evidence for a positive role of PKC in
regulating neuronal differentiation and maintaining the functionally active state (Coleman
and Wooten, 1994; Cabell and Audesirk, 1993; Parrow et al., 1992; Abraham et al., 1991;
Tonini et al., 1991), and previous data have implicated a number of different PKC
isoforms in cellular differentiation (Wada el al., 1989; Leli etal., 1992, 1993; Wayse/
a!., 1994; Hundle et al., 1995; Borgatti et al., 1996), other studies have suggested that,
rather that being a positive regulator of the process, PKC may actually be an inhibitory
influence, so that suppression of the enzyme results in increased process outgrowth
(Heikkila et al., 1989; Parodi et al., 1990; Tsuneishi, 1992; Wooten, 1992; Carlson et al,
1993; Jalava et al., 1993; Ekinci and Shea, 1997). The results of the present experiments
further argue against the notion that PKC is integral to cell differentiation, for addition of
a PKC inhibitor to VPA-exposed HN33 cell cultures enhanced the neurite outgrowth and
cell growth inhibition observed, and failed to reverse similar VPA-induced alterations in
PC 12 cells. Perhaps the cell differentiation parameters observed in this study, those of
increased GAP-43 expression, decreased cell growth, and increased neurite outgrowth,
occur concomitantly, and somehow signal to one another when the time is right to begin.
In the case of the HN33 cells, VP A exposure may be leading merely to a slow in growth
through altered expression of various cell cycling proteins, which in turn induces
accompanying changes in cell morphology and GAP-43 expression. Or perhaps increased
GAP-43 leads to neurite outgrowth, which then results in slowed division rate This,
again, is consistent with findings from studies of lithium, which had no effect on cell
growth rates, and consequently had no effects on either cell morphology or GAP-43
expression. On the other hand, these VPA-induced effects on GAP-43 may be important
for the kinds of plastic changes required for optimal anticonvulsant or mood-stabilizing
therapy The differences observed between lithium and VPA may explain their differing

119
abilities to treat various types of bipolar disorders, or the ability of VP A to effect partial
amelioration of symptoms more quickly than lithium.
In PC 12 cells, we investigated the potential of VP A to serve as a neuroprotective
agent, since VPA exposure leads to cell differentiation, and has clearly been shown to be
an efficacious drug for the treatment of either epilepsy or bipolar disorder. These two
disorders are characterized by dysfunction of thought or electrical impulse, and one could
hypothesize that VPA treatment brings about plastic changes important for either
antiepileptic or mood-stabilizing therapy. In addition, lithium has been shown to have
neuroprotective effects in various in vitro models. In this study, VPA exerted a protective
influence against neurite retraction following trophic factor deprivation, while it had no
protective effect on loss of cell number. Therefore, these data, while consistent with those
discussed earlier (VPA decreased cell growth and increased neurite outgrowth), are
inconclusive as to the potential for neuroprotection by VPA. Future studies may utilize
other protection paradigms to address this issue.
In an effort to further the findings presented herein, one could investigate more in-
depth the effects of VPA on PKC. It is yet unknown precisely how VPA interacts with
this enzyme, or whether it is direct or indirect. As discussed previously, lithium affects
PKC indirectly through its more direct inhibitory effects on PI signaling, which in turn lead
to increased D AG-driven activation of PKC, as well as Ca++ release and further
potentiation of PKC activation. VPA, on the other hand, is structurally very different
from lithium, and it has been shown by these studies and others that VPA does not affect
PI signaling. It would be useful to determine more definitively the mechanism by which
VPA modulates PKC. Potential targets may be Ca++ release, which VPA has already
been shown to affect, and which would in turn activate PKC; activation of an intermediate
protein responsible for direct PKC activation; or simply via the enzyme itself, since VPA is
a fatty acid, and some fatty acids are known to be activators of PKC (though one group
has provided evidence against this hypothesis; Chen et a!., 1994).

120
Another important area of study related to differentiation is the investigation of the
effects of VP A on gene regulation, for in addition to the role of PKC in altering protein
expression by virtue of phosphorylation, PKC and other protein kinases (such as protein
kinase A and tyrosine kinase), as well as protein phosphatases, are known to be involved
in activation and inactivation of transcription factors AP-1 and CREB (Jackson, 1992;
Karin and Smeal, 1992). AP-1 and CREB are known to modulate expression of key
proteins which are involved in neuronal excitability, such as receptors, neuropeptides, and
key enzymes in neurotransmitter biosynthesis, and these parameters are certainly of
interest with regard to either epilepsy or bipolar disorder. Manji’s group showed that in
C6 glioma and human neuroblastoma (SH-SY5Y) cells, as well as in rat brain, VPA
increased AP-1 DNA binding activity in both a time- and concentration-dependent
manner, effects partially dependent upon PKC activity (Chen et a/., 1997). We have
further shown evidence for VPA-induced alterations in MARCKS mRNA levels, also
suggestive of effects on gene expression. This may be an important effect of VPA, or of
mood-stabilizing agents in general. It may be that the efficacy of these anti-manic, mood-
stabilizing drugs lies, in part, in their ability to influence and regulate expression of various
genes. Retinoic acid, a fatty acid known to arrest growth and induce differentiation of a
number of cell lines, is also known to affect gene expression through intracellular retinoic
acid receptors which bind to gene promoter regions and modulate transcription (Watson et
al., 1994). It is of interest that retinoic acid has been found to elicit down-regulation of
MARCKS expression in HN33 cells, which correlates well with findings reported herein
for VPA: decreased MARCKS concomitant with decreased growth and increased
differentiation. Furthermore, retinoic acid appears to elicit this effect without intermediary
involvement of PKC (Watson et al., 1994).
Yet another area of interest is the mechanistic basis for the developmental (in
útero) effects of VPA, which, unfortunately, occur following therapeutic, not toxic,
maternal plasma concentrations. In addition to the considerations mentioned above, the

121
metabolic nature of the VPA-induced effect should be examined more in depth. Previous
studies suggest that while folic acid may prevent neural tube defects in vivo, this
protective effect is not reproducible in vitro. Our preliminary studies in HN33 cells
indicate that the protective effect of folic acid and its metabolites, such as methionine, are
more complicated than merely inhibiting VPA-induced MARCKS down-regulation.
Nonetheless, these findings are not inconsistent with the notion that VPA induces a
significant alteration in the MARCKS/GAP-43 ratio which may underlie a pathological
alteration in the developmental program of cells, leading to premature growth arrest and
differentiation.
On the other hand, such modulation of cell growth may be considered therapeutic,
rather than pathological, in other settings. For example, this anti-proliferative effect of
VPA may prove useful in the area of cancer therapy. Other groups have already
recognized this potential, and one report suggests an antitumor effect of VPA. In human
neuroblastoma cells (UKF-NB-2 and UKF-NB-3), at nontoxic concentrations, VPA not
only arrested cell growth and induced phenotypic differentiation, but also increased the
immunogenicity of cells. This occurred through increased expression of surface neural cell
adhesion molecules (NCAMs), and increased sensitivity to natural killer/lymphokine-
activated killer (NK/LAK) cytolysis, through non-toxic mechanisms (Cinatl et al., 1996).
Therefore, VPA shows promise for treatment of cancer, and at concentrations already
known to be well tolerated by patients.
With regard to experimental models, future studies might utilize primary cells
derived from regions of brain thought to be important in affective disorders, such as the
hippocampus. Even more importantly, in vivo animal experiments will be required before
all of the effects reported thus far can be put into proper perspective with regard to their
roles in the course and treatment of bipolar disorder. Animal studies, while quite
instructive, are a considerably more difficult model in which to investigate the mechanism
of action of VPA. Previous efforts in our laboratory have failed due to the difficulty

122
experienced in attempting to attain therapeutic serum levels of VP A, for the rat’s
metabolism is such that the drug is rapidly broken down and cleared following
administration, with a half-life of about 1 h. This pharmacokinetic parameter is quite in
contrast to the long half-life of VP A in humans, about 12 h.
VPA remains an invaluable drug for both the acute and prophylactic treatment of
bipolar disorder, as well as for various seizure disorders This drug has also proven quite
useful in studying multiple aspects of the dysregulation involved in these diseases, and may
one day lead to a clearer understanding of the pathophysiology which leads to bipolar
disorder, as well as aid in the development of new and better drug therapies for the
disorder.

123
Table 6.1. Effects of two mood-stabilizing drugs, VPA and lithium, as observed in
HN33 cells following chronic exposure: A Summary. The presence of the
described effect is indicated by “+”, and indicates the absence of the effect.
Effect VPA Lithium
•iMARCKS Protein
+
+ a
¿MARCKS mRNA
+
+ b
iPI Signaling
-
+ c
TGAP-43
+
_ b
¿PKC Activity
+
+ b
¿PKC-a
+
+ b
TPKC-5
+
_ b
iPKC-e
-
+ b
Tpkc<
+
?
^Cell Growth
+
_ d
TNeurite Outgrowth
+
_ d
a Watson et ai, 1994
b Watson and Lenox, 1997
c Lenox et ai, 1996
d Watson and Lenox, 1996

REFERENCES
Abraham, I., Sampson, K.E., Powers, E A., Mayo, J K., Ruff, V.A., and Leach, K.L.
(1991). Increased PKA and PKC activities accompany neuronal differentiation of
NT2/D1 cells. J. Neurosci. Res. 28 (1): 29-39.
Aderem, A. (1992). The MARCKS brothers: A family of protein kinase C substrates.
Cell 71: 713-716.
Aigner, L., Arber, S., Kapfhammer, J.P., Laux, T , Schneider, C., Botteri, F., Brenner, H-
R, and Caroni, P. (1995). Overexpression of the neural growth-associated protein
GAP-43 induces nerve sprouting in the adult nervous system of transgenic mice. Cell
83: 269-278.
Aigner, L. and Caroni, P (1993). Depletion of 43-kD growth-associated protein in
primary sensory neurons leads to diminished formation and spreading of growth cones.
J. Cell Biol. 123: 417-429.
Aigner, L. and Caroni, P. (1995). Absence of persistent spreading, branching, and
adhesion in GAP-43-depleted growth cones J Cell Biol. 128 (4): 647-660.
Albus, H. and Williamson, R (1998). Electrophysiologic analysis of the actions of
valproate on pyramidal neurons in the rat hippocampal slice. Epilepsia 39 (2): 124-
139.
Allen, L-A.H and Aderem, A. (1995). A role for MARCKS, the a isozyme of protein
kinase C and myosin I in zymosan phagocytosis by macrophages. J. Exp. Med. 182:
829-840.
Baetge, E.E. and Hammang, J.P (1991). Neurite outgrowth in PC 12 cells deficient in
GAP-43. Neuron 6: 21-30.
Barton, B.M and Gitlin, M.J (1987). Verapamil in treatment-resistant mania: an open
trial. J. Clin. Psychopharmacol. 7 (2): 101-103.
Benet, L.Z and Williams, R.L (1990). Pharmacokinetic data. ///Goodman and
Gilman’s The Pharmacological Basis of Therapeutics, 8th ed., ed. by A.G. Gilman,
T W. Rail, A S. Nies and P Taylor, pp 1655-1715, Pergamon Press, Elmsford, NY.
Benowitz, L.I. and Perrone-Bizzozero, N.I. (1991). The expression of GAP-43 in
relation to neuronal growth and plasticity: when, where, how, and why9 Progress in
Brain Research 89: 69-87.
124

125
Benowitz, L.I. and Routtenberg, A. (1997). GAP-43: An intrinsic determinant of
neuronal development and plasticity. TINS 20: 84-91.
Berridge, M.J (1989). Inositol trisphosphate, calcium, lithium, and cell signaling. JAMA
262: 1834-1841.
Berridge, M.J., Downes, CP, and Hanley, M R. (1982). Lithium amplifies agonist-
dependent phosphatidylinositol responses in brain and salivary glands. Biochem. J.
206: 587-595.
Blackshear, P J (1993). The MARCKS family of cellular protein kinase C substrates. J.
Biol. Chem. 268: 1501-1504.
Boniece, I.R and Wagner, J.A. (1993). Growth factors protect PC12 cells against
ischemia by a mechanism that is independent of PKA, PKC, and protein synthesis. J.
Neurosci. 13 (10): 4220-4228.
Borgatti, P., Mazzoni, M., Carini, C , Neri, L.M., Marchisio, M, Bertolaso, L., Previati,
M, Zauli, G, and Capitani, S. (1996). Changes of nuclear protein kinase C activity
and isotype composition in PC 12 cell proliferation and differentiation. Experimental
Cell Research 224: 72-78.
Bowden, C.L. (1995). Predictors of response to divalproex and lithium. J. Clin.
Psychiatry 56 [Suppl 3]: 25-30
Bowden, C.L. (1996). Role of newer medications for bipolar disorder J. Clin.
Psychopharmacology 16 (Suppl. 1): 48S-55S.
Bowden, CL, Brugger, A M., Swann, AC, Calabrese, JR, Janicak, P.G., Petty, F.,
Dilsaver, S.C., Davis, J.M., Rush, A.J., Small, J.G., Garza-Trevino, E.S., Risch, SC.,
Goodnick, P.J., and Morris, D.D (1994). Efficacy of divalproex vs. lithium and
placebo in the treatment of mania. JAMA 271: 918-924.
Brooks, S.F., Herget, T., Broad, S., and Rozengurt, E. (1992). The expression of
80K/MARCKS, a major substrate of protein kinase C (PKC), is down-regulated
through both PKC-dependent and -independent pathways. JBC 267 (20): 14212-
14218.
Brown, R. (1989). U S. experience with valproate in manic depressive illness: a
multicenter trial. J. Clin. Psychiatry 50 [3, Suppl]: 13-16.
Cabell, L. and Audesirk, G. (1993). Effects of selective inhibition of protein kinase C,
cyclic AMP-dependent protein kinase, and Ca(2+)-calmodulin-dependent protein kinase
on neurite development in cultured rat hippocampal neurons. Int. J Dev Neurosci. 11
(3): 357-368.

126
Calabrese, JR and Delucchi, G.A. (1989). Phenomenology of rapid cycling manic
depression and its treatment with valproate. J. Clin. Psychiatry 50 [3, SuppI]: 30-34.
Calabrese, J.R. and Delucchi, G.A. (1990). Spectrum of efficacy of valproate in 55
patients with rapid-cycling bipolar disorder. Am. J. Psychiatry 147: 431-434.
Calabrese, J.R., Markovitz, P J., Kimmel, S.E., and Wagner, S.C. (1992). Spectrum of
efficacy of valproate in 78 rapid-cycling bipolar patients. J. Clin. Psychopharmacol.
12: 53S-56S.
Calabrese, J.R., Rapport, D.J., Kimmel, S.E., Reece, B., and Woyshville, M.J. (1993).
Rapid-cycling biolar disorder and its treatment with valproate Can. J. Psychiatry 38
(suppl. 2): S57-S61.
Carlson, R.O., Martin, D M, Feldman, E.L., and Agranoff, B.W. (1993). PKC activity
and PKC-alpha mRNA content are reduced in serum-derived human neuroblastoma
cells without concomitant induction of differentiation. Exp. Cell Res. 207 (2): 340-
347.
Chakravarthy, B.R., Wong, J., and Durkin, J.P (1995). Evidence that the modulation of
membrane-associated protein kinase C activity by an endogenous inhibitor plays a role
in N1E-115 murine neuroblastoma cell differentiation. J. Neurochem. 65: 1569-1579.
Chapman, A., Keane, P.E., Meldrum, B.S., Simiand, J., and Vemieres, J.C. (1982).
Mechanism of anticonvulsant action of valproate Progress in Neurobiol. 19: 315-359.
Chen, G., Manji, H.K., Hawver, D.B., Wright, C.B., and Potter, W.Z.: (1994). Chronic
sodium valproate selectively decreases protein kinase C a and e in vitro. J.
Neurochem. 63: 2361-2364.
Chen, G., Yuan, P , Hawver, D.B., Potter, W.Z., and Manji, H.K. (1997). Increase in
AP-1 transcription factor DNA binding activity by valproic acid. Neuropsychopharm.
16 (3): 238-245.
Chen, R.W. and Chuang, D M. (1999). Long term lithium treatment suppresses p53 and
Bax expression but increases Bcl-2 expression. A prominent role in neuroprotection
against excitotoxicity. J. Biol. Chem. 274 (10): 6039-6042.
Cinatl, J., Cinatl, J., Scholz, M., Driever, P.H., Henrich, D., Kabickova, H., Vogel, J.-U.,
Doerr, H.W., and Kornhuber, B (1996). Antitumor activity of sodium valproate in
cultures of human neuroblastoma cells. Anticancer Drugs 7: 766-773.
Coleman, E.S. and Wooten, M.W. (1994). Nerve growth factor-induced differentiation
of PC 12 cells employs the PMA-insensitive protein kinase C-zeta isoform. J. Mol.
Neurosci. 5(1): 39-57.

127
Daston, G.P., Barnes, D., Elmore, E., Fitzgerald, M.P., and Sharma, S. (1995).
Evaluation of chick embryo neural retina cell culture as a screen for developmental
toxicants. Fund, and Applied. Tox. 26: 203-210.
Debus, E., Weber, K., and Osborn, M. (1983). Monoclonal antibodies specific for glial
fibrillary acidic (GFA) protein and for each of the neurofilament triplet polypeptides.
Differentiation 25: 193-203.
Dekker, L.V., deGraan, P.N.E., Oestreicher, A.B., Versteeg, D.H.G., and Gispen, W H.
(1989). Inhibition of noradrenaline release by antibodies to B-50 (GAP-43). Nature
342: 74-76.
Dixon, J.F. and Hokin, L.E. (1997). The mood-stabilizing drug valproate mimics lithium
in stimulating glutamate release and inositol 1,4,5-trisphosphate accumulation in brain
cortex slices but not accumulation of inositol monophosphates and bisphosphates
Proc. Natl. Acad. Sci. USA 94: 4757-4760.
Eadie, M.J (1991). Formation of active metabolites of anticonvulsant drugs: A review
of their pharmacokinetic and therapeutic significance. Clin. Pharmacokinet 21 (1):
27-41.
Ehlers, K., Elmazar, M.M.A., and Nau, H. (1996). Methionine reduces the valproic acid-
induced spina bifida rate in mice without altering valproic acid kinetics. J. Nutrition
126: 67-75.
Ekinci, F.J. and Shea, T.B (1997). Selective activation by bryostatin-1 demonstrates
unique roles for PKC epsilon in neurite extension and tau phosphorylation. Int. J.
Dev. Neurosci. 15 (7): 867-874.
Fagerstrom, S., Pahlman, S., Gestblom, C., and Nanberg, E. (1996). Protein kinase C-
epsilon is implicated in neurite outgrowth in differentiating human neuroblastoma cells.
Cell Growth Differ. 7 (6): 775-785.
Fawcett, J. (1989). Valproate use in acute mania and bipolar disorder: an international
perspective. J Clin. Psychiatry 50 |3, Suppl]: 10-12.
Friedman, E., Wang, H-Y., Levinson, D., Connell, T.A., and Singh, H. (1993). Altered
platelet protein kinase C activity in bipolar affective disorder, manic episode. Biol.
Psychiatry 33 (7): 520-525.
Fujise, A., Mizumo, K, Ueda, Y., Osada, S., Hirai, S., Takayanagi, A., Shimizu, N.,
Owada, M.K., Nakajima, H., and Ohno, S. (1994). Specificity of the high affinity
interaction of protein kinase C with a physiological substrate, myristoylated alanine-
rich protein kinase C substrate. J. Biol. Chem. 269: 31642-31648.
Gemer, R H. and Stanton, A (1992). Algorithm for patient management of acute manic
states: Lithium, valproate, or carbamazepine? J. Clin. Psychopharmacol. 12: 57S-63S.

128
Godfrey, P.P. (1989). Potentiation by lithium of CMP-phosphatidate formation in
carbachol-stimulated rat cerebral-cortical slices and its reversal by /wyo-inositol.
Biochem. J. 258: 621-624.
Gofflot, F., van Maele-Fabry, G, and Picard, J.J. (1996). Cranial nerves and ganglia are
altered after in vitro treatment of mouse embryos with valproic acid (VPA) and 4-en-
VPA. Dev. Brain Res. 93: 62-69.
Granneman, G.R , Schneck, D.W., Cavanaugh, J.H., and Witt, G.F. (1996).
Pharmacokinetic interactions and side effects resulting from concomitant
administration of lithium and divalproex sodium. J. Clin. Psychiatry 57 (5): 204-206.
Gressens, P., Marret, S., Martin, J.L., Laquerriere, A., Lombet, A., and Evrard, P.
(1998). Regulation of neuroprotective action of vasoactive intestinal peptide in the
murine developing brain by protein kinase C and mitogen-activated protein kinase
cascades: in vivo and in vitro studies. J. Neurochem. 70 (6): 2574-84.
Hayes, S .G. (1989). Long-term use of valproate in primary psychiatric disorders. J. Clin.
Psychiatry 50 [3, Suppl]: 35-39.
Heikkila, J.E., Akerlind, G, and Akerman, K.E. (1989). Protein kinase C activation and
down-regulation in relation to phorbol ester-induced differentiation of SH-SY5Y
human neuroblastoma cells. J. Cell Physiol. 140 (3): 593-600.
Hofmann, J. (1997). The potential for isozyme-selective modulation of protein kinase C.
FASEB J. 11: 649-669.
Holford, N.H.G. and Benet, L.Z. (1995). Pharmacokinetics and pharmacodynamics:
Rational dose selection and the time course of drug action. In Basic and Clinical
Pharmacology, 6th ed., ed. by B.G. Katzung, p. 35-36, Appleton and Lange, Norwalk,
CT.
Hollister, L.E. (1995). Valproic acid and clonazepam. In Basic and Clinical
Pharmacology, 6th ed., ed. by B.G. Katzung, p. 446, Appleton and Lange, Norwalk,
CT.
Hundle, B , McMahon, T., Dadgar, J., Chen, C.H., Mochly-Rosen, D , and Messing, R.O
(1997). An inhibitory fragment derived from protein kinase C-epsilon prevents
enhancement of nerve growth factor responses by ethanol and phorbol esters. J Biol.
Chem. 272 (23): 15028-15035.
Hundle, B , McMahon, T., Dadgar, J., and Messing, R.O. (1995). Overexpression of s-
protein kinase C enhances nerve growth factor-induced phosphorylation of mitogen-
activated protein kinases and neurite outgrowth. J. Biol. Chem. 270: 30134-30140.

129
Ishii, H., Jirousek, M R., Koya, D., Takagi, C., Xia, P., Clermont, A., Bursell, S-E., Kern,
T.S., Bailas, L.M., Heath, W.F., Stramm, L.E., Feener, E.P., and King, G.L. (1996).
Amelioration of vascular dysfunctions in diabetic rats by an oral PKC p inhibitor.
Science 272: 728-731.
Jackson, S. (1992). Regulating transcription factor activity by phosphorylation. Trends
Cell Biol. 2: 104-108.
Jalava, A., Akerman, K., and Heikkila, J. (1993). Protein kinase inhibitor, staurosporine,
induces a mature neuronal phenotype in SH-SY5Y human neuroblastoma cells through
an alpha-, beta-, and zeta-protein kinase C-independent pathway. J. Cell Physiol. 155
(2): 301-312.
Jallon, P. (1997). The problem of intractability: The continuing need for new medical
therapies in epilepsy. Epilepsia 38 (Suppl. 9): S37-S42.
Janicak, P.G., Sharma, R.P., Pandey, G, and Davis, J.M. (1998). Verapamil for the
treatment of acute mania: a double-blind, placebo-controlled trial. Am J. Psychiatry
155 (7): 972-973.
Jap Tjoen San, E R A., Schmidt-Michels, M., Oestreicher, A.B., Gispen, W.H., and
Schotman, P. (1992). Inhibition of nerve growth factor-induced B-50/GAP-43
expression by antisense oligomers interferes with neurite outgrowth of PC 12 cells.
Biochem. Biophys. Res. Comm. 187: 839-846.
Jirousek, M R., Gillig, JR, Gonzalez, C M., Heath, W.F., McDonald, J.H., III, Neel,
DA., Rito, C.J., Singh, U., Stramm, L.E., Melikian-Badalian, A., Baevsky, M , Balias,
L.M., Hall, S.E., Winneroski, L.L., and Faul, M.M. (1996). (5)-13-
[(Dimethylamino)methyl]-10,11,14,15-tetrahydro-4,9:16,21 -dimetheno-1//, 13 H-
dibenzo[e,£]pyrrolo[3,4-/?] [1,4,13 ]oxadiazacyclohexadecene-1,3(2//)-dione
(LY333531) and related analogues: Isozyme selective inhibitors of protein kinase Cp.
J. Med. Chem. 39: 2664-2671.
Joffe, R.T. (1993). Valproate in bipolar disorder: the Canadian perspective. Can. J
Psychiatry 38 (Suppl. 2): S46-S50.
Karin, M. and Smeal, T. (1992). Control of transcription factors by signal transduction
pathways: The beginning of the end. TIBS 17: 418-422.
Katz, F., Ellis, L., and Pfenninger, K.H. (1985). Nerve growth cones isolated from fetal
rat brain: III. Calcium-dependent protein phosphorylation. J. Neurosci. 5: 1402-1411.
Keck, P.E. and McElroy, S.L. (1996). Outcome in the pharmacologic treatment of
bipolar disorder. J. Clin. Psychopharm. 16 (2, Suppl. 1): 15S-23S.
Kelly, P G. and Regan, C M. (1992). Studies on valproate-induced perturbations of
neurulation in the explanted chick embryo. Toxicology 71: 137-144.

130
Kuntz-Simon, G. and Obert, G. (1995). Sodium valproate, an anti-convulsant drug,
stimulates human cytomegalovirus replication. J. Gen. Virol. 76: 1409-1415.
Kuo, J.F., Andersson, R.G.G., Wise, B.C., Mackerlova, L., Salomonsson, I., Brackett,
N.L., Katoh, N., Shoji, M, and Wrenn, R.W. (1980). Calcium-dependent protein
kinase: Widespread occurrence in various tissues and phyla of the animal kingdom and
comparison of effects of phospholipid, calmodulin, and trifluoperazine. Proc. Natl.
Acad. Sci. USA 77 (12): 7039-7043.
Lee, H.J., Hammond, D.N., Large, T.H., Roback, J.D., Sim, J A., Brown, DA., Otten,
U.H. and Wainer, B.H. (1990). Neuronal properties and trophic activities of
immortalized hippocampal cells from embryonic and young adult mice. J. Neurosci.
10: 1779-1787.
Lenox, R.H. (1987). Role of receptor coupling to phosphoinositide metabolism in the
therapeutic action of lithium. In Molecular Mechanisms of Neuronal Responsiveness.
Adv. Exp. Biol Med., Vol. 221. ed. by Y.H. Ehrlich, R.H. Lenox, E. Komecki, and
W.O. Berry, pp. 515-530, Plenum Press, New York.
Lenox, R.H., McNamara, R.K., Watterson, J.M., and Watson, D.G. (1996).
Myristoylated Alanine-Rich C Kinase Substrate (MARCKS): A molecular target for
the therapeutic action of mood stabilizers in the brain? J. Clin. Psychiatry 57 (suppl
13): 23-31.
Lenox, R.H. and Watson, D.G. (1994). Lithium and the brain: A psychopharmacological
strategy to a molecular basis for manic-depressive illness. Clin. Chem. 40: 309-314.
Lenox, R.H, Watson, D.G., Patel, J , and Ellis, J. (1992). Chronic lithium administration
alters a prominent PKC substrate in rat hippocampus. Brain Res. 570: 333-340.
Lobach, D.F., Rochelle, J.M., Watson, M L., Seldin, M.F., and Blackshear, P.J. (1993).
Nucleotide sequence, expression, and chromosomal mapping of Mrp and mapping of
five related sequences. Genomics 17: 194-204.
Loscher, W. (1993). Effects of the antiepileptic drug valproate on metabolism and
function of inhibitory and excitatory amino acids in the brain. Neurochem. Res. 18:
485-502.
Maiese, K. and Boccone, L. (1995). Neuroprotection by peptide growth factors against
anoxia and nitric oxide toxicity requires modulation of protein kinase C. J. Cereb.
Blood Flow Metab. 15 (3): 440-449.
Manji, H.K., Bebchuk, J.M., Moore, G.J., Glitz, D., Hasanat, K.A., and Chen, G. (1999).
Modulation of CNS signal transduction pathways and gene expression by mood-
stabilizing agents: Therapeutic implications. J. Clin. Psychiatry 60 (suppl 2): 27-39.

131
Manji, H.K., Bersudsky, Y., Chen, G, Belmaker, R.H., and Potter, W.Z. (1996a).
Modulation of protein kinase C isozymes and substrates by lithium: The role of myo¬
inositol Neuropsychopharmacology 15: 370-381.
Manji, H.K., Chen, G, Hsiao, J.K., Risby, ED, Masana, M.I., and Potter, W.Z. (1996b).
Regulation of signal transduction pathways by mood-stabilizing agents: Implications
for the delayed onset of therapeutic efficacy. J. Clin. Psychiatry 57 [suppl 13]: 34-46.
Manji, H.K., Etcheberrigaray, R , Chen, G, and Olds, J.L. (1993). Lithium decreases
membrane-associated protein kinase C in hippocampus: Selectivity for the a isozyme
J. Neurochem. 61: 2303-2310.
Manji, H.K. and Lenox, R.H.: (1994). Long-term action of lithium: A role for
transcriptional and posttranscriptional factors regulated by protein kinase C. Synapse
16: 11-28.
Manji, H.K., Moore, G.J., and Chen, G. (1999). Lithium at 50: Have the
neuroprotective effects of this unique cation been overlooked? Biol. Psychiatry 46
(7): 929-940.
Manji, H.K., Potter, W.Z., and Lenox, R.H. (1995). Signal transduction pathways:
Molecular targets for lithium's action. Arch. Gen. Psychiatry 52: 531-543.
Martin, M L., Breen, K.C., and Regan, C M. (1988). Perturbations of cellular functions
integral to neural tube formation by the putative teratogen sodium valproate. Tox. In
Vitro 2 (1): 43-48.
Martin, M L. and Regan, C M. (1988). The anticonvulsant sodium valproate specifically
induces the expression of a rat glial heat shock protein which is identified as the
collagen type IV receptor. Brain Res. 459: 131-137.
Martin, M L. and Regan, C M. (1991). The anticonvulsant valproate teratogen restricts
the glial cell cycle at a defined point in the mid-Gl phase. Brain Res 554: 223-228.
Masliah, E., Raber, J., Alford, M., Mallory, M., Mattson, M.P , Yang, D., Wong, D , and
Mucke, L. (1998). Amyloid protein precursor stimulates excitatory amino acid
transport. Implications for roles in neuroprotection and pathogenesis. J. Biol. Chem.
273(20): 12548-12554.
Mathis, P., Schmitt, L., and Moron, P (1988). [Efficacy of verapamil in mania crises],
Encephale 14 (3): 127-132.
McElroy, S.L. and Keck, P.E. (1993). Treatment guidelines for valproate in bipolar and
schizoaffective disorders. Can. J. Psychiatry 38 (Suppl. 2): S62-S66.

132
McElroy, S.L., Keck, PE., Pope, H.G., and Hudson, J.I. (1989). Valproate in
psychiatric disorders: Literature review and clinical guidelines. J. Clin. Psychiatry 50
[3, Suppl]: 23-29
McElroy, S.L., Keck, P.E., Pope, H.G., and Hudson, J.I. (1992). Valproate in the
treatment of bipolar disorder: Literature review and clinical guidelines. J. Clin.
Psychophamacol. 12: 42S-52S.
McElroy, S.L., Keck, PE., Stanton, S.P., Tugrul, K.C., Bennett, J.A., and Strakowski,
S .M. (1996). A randomized comparison of divalproex oral loading versus haloperidol
in the initial treatment of acute psychotic mania. J. Clin. Psychiatry 57: 142-146.
McNamara, R.K and Lenox, R.H. (1997). Comparative distribution of myristoylated
alanine-rich C kinase substrate (MARCKS) and FI/GAP-43 gene expression in the
adult rat brain. J. Comp. Neurology 379: 48-71.
Meiri, K.F., Bickerstaff, L.E., and Schwob, J.E. (1991). Monoclonal antibodies show
that kinase C phosphorylation of GAP-43 during axonogenesis is both spatially and
temporally restricted in vivo. J. Cell Biol. 112 (5): 991-1005.
Meiri, K.F., Saffell, J.L., Walsh, F.S., and Doherty, P (1998). Neurite outgrowth
stimulated by neural cell adhesion molecules requires growth-associated protein-43
(GAP-43) function and is associated with GAP-43 phosphorylation in growth cones.
J. Neuroscience 18 (24): 10429-10437.
Menegola, E., Broccia, M L., Nau, H., Prati, M., Ricolfi, R, and Giavini, E. (1996).
Teratogenic effects of sodium valproate in mice and rats at midgestation and at term.
Teratogen., Carcinogen., and Mutagen. 16: 97-108.
Meunier, H., Carraz, G, Meunier, Y., Eymard, P., and Aimard, M. (1963). Proprietes
pharmacodynamic de l’acide n-depropylacetique. ler Memoire: Proprietes
Antiepileptiques. Therapie 18: 435-438.
Minakuchi, R , Takai, Y., Yu, B., and Nishizuka, Y (1981). Widespread occurrence of
calcium-activated, phospholipid-dependent protein kinase in mammalian tissues. J.
Biochem. 89: 1651-1654.
Motohashi, N. (1990). GABA receptor alterations after chronic lithium - In comparison
with carbamazepine and sodium valproate. Clin. Neuropharmacol. 13: 207-208.
Mummery, C.L., Van den Brink, C.E., Van der Saag, P.T., and De Laat, S.W. (1984). A
short-term screening test for teratogens using differentiating neuroblastoma cells in
vitro. Teratology 29: 271-279.
Naumovski, L. and Cleary, M L. (1994). Bcl2 inhibits apoptosis associated with terminal
differentiation of HL-60 myeloid leukemia cells. Blood 83 (8): 2261-2267.

133
Newton, A C. (1995). Protein kinase C: Structure, function and regulation. J. Biol.
Chem. 270: 28495-28498.
Nishizuka, Y. (1995). Protein kinase C and lipid signalling for sustained cellular
responses. FASEB J. 9: 484-496.
Nonaka, S., Hough, CJ., and Chuang, D-M. (1998). Chronic lithium treatment robustly
protects neurons in the central nervous system against excitotoxicity by inhibiting N-
methyl-D-aspartate receptor-mediated calcium influx. Proc. Natl. Acad. Sci. USA 95:
2642-2647.
Nosel, P.G. and Klein, N.W. (1992). Methionine decreases the embryotoxicity of sodium
valproate in the rat: In vivo and in vitro observations. Teratology 46: 499-507.
O’Brien, E. and Regan, C M. (1998). Protein kinase C inhibitors arrest the C6 glioma
cell cycle at a mid-Gl phase restriction point: Implications for the antiproliferative
action of valproate. Tox. in Vitro 12 (1): 9-14.
Oehrlein, S.A., Maelicke, A., and Herget, T. (1998). Expression of protein kinase C gene
family members is temporally and spatially regulated during neural development in
vitro. Eur. J. Cell Biol. 77 (4): 323-337.
Ouimet, C.C., Wang, J.K.T., Walaas, S.I., Albert, K.A., and Greengard, P (1990).
Localization of the MARCKS (87 kDa) protein, a major specific substrate for protein
kinase C, in rat brain. J. Neuroscience 10: 1683-1698.
Parodi, M.T., Varesio, L., and Tonini, G.P. (1990). The specific inhibitor of protein
kinase C, l-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H7), induces morphological
change and cell differentiation of human neural crest-derived cell lineages. FEBS Lett.
269(1): 4-6.
Parrow, V., Fagerstrom, S., Meyerson, G., Nanberg, E., and Pahlman, S. (1995). Protein
kinase C-alpha and -epsilon are enriched in growth cones of differentiating SH-SY5Y
human neuroblastoma cells. J. Neurosci. Res. 41 (6): 782-791.
Parrow, V., Nanberg, E., Heikkila, J., Hammerling, U., and Pahlman, S. (1992). Protein
kinase C remains functionally active during TPA induced neuronal differentiation of
SH-SY5Y human neuroblastoma cells. J. Cell Physiol. 152 (3): 536-544.
Patel, J. and Kligman, D (1987). Purification and characterization of an Mr 87,000
protein kinase C substrate from rat brain J. Biol. Chem. 262: 16686-16691.
Penry, J.K. and Dean, J.C. (1989). The scope and use of valproate in epilepsy. J. Clin.
Psychiatry 50 [3, Suppl] 17-22.

134
Perez-Pinzon, M.A. and Born, J.G. (1999). Rapid preconditioning neuroprotection
following anoxia in hippocampal slices: role of the K ATP channel and protein kinase
C. Neuroscience. 89 (2): 453-459.
Petty, F. (1995). GABA and mood disorders: A brief review and hypothesis. J. Affect.
Disord. 34: 275-281.
Pizzi, M., Boroni, F., Benarese, M., Moraitis, C., Memo, M., and Spano, P (1999).
Neuroprotective effect of thyrotropin-releasing hormone against excitatory amino
acid-induced cell death in hippocampal slices. Eur. J. Pharmacol. 370 (2): 133-137.
Pizzi, M., Galli, P., Consolandi, O., Arrighi, V., Memo, M., ad Spano, P.F. (1996).
Metabotropic and ionotropic transducers of glutamate signal inversely control
cytoplasmic Ca2 concentration and excitotoxicity in cultured cerebellar granule cells:
pivotal role of protein kinase C. Mol. Pharmacol. 49 (4): 586-594.
Poltorak, M., Frye, M.A., Wright, R., Hemperly, J.J., George, M S., Pazzaglia, P.J.,
Jerrels, S.A., Post, R.M., and Freed, W.J. (1996). Increased neural cell adhesion
molecule in the CSF of patients with mood disorder. J. Neurochem. 66 (4): 1532-
1538.
Ponzoni, M., Lucarelli, E., Corrías, M.V., and Cornaglia-Ferraris, P (1993). Protein
kinase C isoenzymes in human neuroblasts. Involvement of PKC epsilon in cell
differentiation. FEBS Lett. 322 (2): 120-124.
Porter, R.J. and Meldrum, B.S. (1995). Antiepileptic drugs. In Basic and Clinical
Pharmacology, 6th ed., ed. by B.G. Katzung, pp 361-380, Appleton and Lange,
Norwalk, CT.
Post, R.M. (1989). Introduction: Emerging perspectives on valproate in affective
disorders. J. Clin. Psychiatry 50 (3 Suppl.): 3-9.
Post, R.M., Weiss, S.R.B ., and Chuang, D-M. (1992). Mechanisms of action of
anticonvulsants in affective disorders: Comparisons with lithium. J. Clin.
Psychopharmacol. 12: 23S-35S.
Ramsay, R E. (1984). Controlled and comparative trials with valproate: United States.
Epilepsia 25 (Suppl. 1): S40-S43.
Regan, C M. (1985). Therapeutic levels of sodium valproate inhibit mitotic indices in
cells of neural origin. Brain Res. 347: 394-398.
Rhoads, G.G. and Mills, J.L. (1986). Can vitamin supplements prevent neural tube
defects? Current evidence and ongoing investigations. Clin. Ob. Gyn. 29: 569-579.
Rimmer, E and Richens, A. (1985). An update on sodium valproate. Pharmacotherapy
5(3): 171-184.

135
Rosen, A., Keenan, K.F., Thelen, M, Nairn, AC, and Aderem, A. (1990). Activation of
protein kinase C results in the displacement of its myristoylated, alanine-rich substrate
from punctate structures in macrophage filopodia. J. Exp. Med. 172: 1211-1215.
Schlingensiepen, R and Schlingensiepen, K-H. (1997). Antisense oligodeoxynucleotides:
Highly specific tools for basic research and pharmacotherapy. In Antisense: From
Technology to Therapy, ed by R. Schlingensiepen, W Brysch, and K-H
Schlingensiepen, pp. 3-103, Blackwell Science, Malden, MA.
Shamloo, M. and Wieloch, T. (1999). Rapid decline in protein kinase C gamma levels in
the synaptosomal fraction of rat hippocampus after ischemic preconditioning.
Neuroreport 10 (5): 931-935.
Shaw, G. (1991). Neurofilament Proteins. In The Neuronal Cytoskeleton. pp. 185-214,
Wiley-Liss, New York.
Shea, T.B., Perrone-Bizzozero, N.I., Beermann, M L., and Benowitz, L.I. (1991).
Phospholipid-mediated delivery of anti-GAP-43 antibodies into neuroblastoma cells
prevents neuritogenesis. J. Neurosci. 11: 1685-1690.
Sheu, F.-S., Huang, F.L., and Huang, K.-P. (1995). Differential responses of protein
kinase C substrates (MARCKS, neuromodulin, and neurogranin) phosphorylation to
calmodulin and S100. Arch. Biochem. and Biophys. 316: 335-342.
Simon, G., Moog, C., and Obert, G. (1994). Valproic acid reduces the intracellular level
of glutathione and stimulates human immunodeficiency virus. Chem. Biol. Interact.
91: 111-121.
Smithells, R.W., Sheppard, S., Schorah, C.J., Sellar, M.J., Nevin, N.C., Harris, R., Read,
A.P., and Fielding, D.W. (1981). Apparent prevention of neural tube defects by
periconceptional vitamin supplementation. Arch. Dis. Child 56: 911-918.
Son, H, Davis, P.J., and Carpenter, D O. (1997). Time course and involvement of
protein kinase C-mediated phosphorylation of FI/GAP-43 in area CA3 after mossy
fiber stimulation. Cell. Molec. Neurobiol. 17 (2): 171-194.
Stabel, S. and Parker, P.J. (1991). Protein kinase C. Pharmac. Ther 51: 71-95.
Strittmatter, S.M., Fankhauser, C., Huang, P.L., Mashimo, H , and Fishman, M.C.
(1995). Neuronal pathfinding is abnormal in mice lacking the neuronal growth protein
GAP-43. Cell 80: 445-452.
Strittmatter, S.M., Valenzuela, D , and Fishman, M.C. (1994). An amino terminal
domain of the growth-associated protein GAP-43 mediates its effects on filopodial
formation and spreading. J. Cell Sci. 107: 195-204.

136
Stumpo, D.J., Bock, C B., Tuttle, J.S., and Blackshear, P.J. (1995). MARCKS
deficiency in mice leads to abnormal brain development and perinatal death. Proc.
Natl. Acad. Sci. USA 92: 944-948.
Swierczynski, S.L., Siddhanti, S.R., Tuttle, J.S., and Blackshear, P.J. (1996).
Nonmyristoylated MARCKS complements some but not all of the developmental
defects associated with MARCKS deficiency in mice. Dev Biol. 179: 135-147.
Tauskela, J.S., Chakravarthy, B.R., Murray, C.L., Wang, Y., Comas, T., Hogan, M.,
Hakim, A., and Morley, P. (1999). Evidence from cultured rat cortical neurons of
differences in the mechanism of ischemic preconditioning of brain and heart. Brain
Res. 827: 143-151.
Tonini, G.P., Parodi, M.T., DiMartino, D., and Varesio, L. (1991). Expression of protein
kinase C-alpha (PKC-alpha) and MYCN mRNAs in human neuroblastoma cells and
modulation during morphological differentiation induced by retinoic acid. FEBS Lett.
280 (2): 221-224.
Tremblay, R, Hewitt, K., Lesiuk, H., Mealing, G., Morley, P., Durkin, J.P (1999).
Evidence that brain-derived neurotrophic factor neuroprotection is linked to its ability
to reverse the NMDA-induced inactivation of protein kinase C in cortical neurons. J.
Neurochem. 72 (1): 102-111.
Trotz, M., Wegner, C., and Nau, H. (1987). Valproic acid-induced neural tube defects:
Reduction by folinic acid in the mouse. Life Sciences 41: 103-110.
Tsuneishi, S. (1992). Regulation of neurite outgrowth through protein kinase C and
protease nexin-1 in neuroblastoma cell. Kobe. J. Med. Sci. 38 (3): 147-159.
Uberall, F., Giselbrecht, S., Hellbert, K., Fresser, F, Bauer, B., Gschwendt, M., Grunicke,
H.H., and Baier, G. (1997). Conventional PKC-a, novel PKC-e and PKC-0, but not
atypical PKC-A. are MARCKS kinases in intact NIH 3T3 fibroblasts. J. Biol. Chem.
272 (7): 4072-4078.
Vadnal, R E. and Parthasarathy, R. (1995). A^yo-inositol monophosphatase: Diverse
effects of lithium, carbamazepine, and valproate. Neuropyschyopharm. 12: 277-285.
Vaux, D.L. (1993). Toward an understanding of the molecular mechanisms of
physiological cell death. Proc. Natl. Acad. Sci. USA 90: 786-789.
Waldmeier, P C.: (1987). Is there a common denominator for the antimanic effect of
lithium and antidepressants? Pharmacopsychiatry 20: 37-47.
Walton, S.A., Berk, M., and Brook, S. (1996). Superiority of lithium over verapamil in
mania: a randomized, controlled, single-blind trial. J. Clin. Psychiatry 57 (11): 543-
546.

137
Wang, H.Y. and Friedman, E. (1996). Enhanced protein kinase C activity and
translocation in bipolar affective disorder brains. Biol. Psychiatry 40 (7): 568-575.
Watson, D.G. and Lenox, R.H. (1996). Chronic lithium-induced down-regulation of
MARCKS in immortalized hippocampal cells: Potentiation by muscarinic receptor
activation. J. Neurochem. 67: 767-777.
Watson, D.G. and Lenox, R.H. (1997). Chronic lithium exposure leads to the down-
regulation of both MARCKS mRNA and protein in immortalized hippocampal cells
(#435.13). Soc. Neurosci. Abstr. 23: 1111.
Watson, D.G., Wainer, B.H., and Lenox, R.H. (1994). Phorbol ester- and retinoic acid-
induced regulation of the PKC substrate MARCKS in immortalized hippocampal cells.
J. Neurochem 63: 1666-1674.
Ways, D.K., Posekany, K., deVente, J., Garris, T, Chen, J., Hooker, J., Qin, W, Cook,
P, Fletcher, D , and Parker, P (1994). Overexpression of protein kinase C-Q
stimulates leukemic cell differentiation. Cell Growth Differ 5: 1195-1203.
Weiner, B. (ed) (1996). PDR Generics, 2nd ed., Medical Economics, Montvale, NJ.
Wiger, R., Stottum, A., and Brunborg, G. (1988). Estimating chemical developmental
hazard in a chicken embryo limb bud micromass system. Pharmacol. Toxicol. 62: 32-
37.
Wilk, A.L., Greenberg, J.H., Horigan, E.A., Pratt, R.M., and Martin, G.R (1980).
Detection of teratogenic compounds using differentiating embryonic cells in culture.
In Vitro 16 (4): 269-276.
Wooten, M.W. (1992). Differential expression of PKC isoforms and PC 12 cell
differentiation. Exp. Cell Res. 199 (1): 111-119.
Xie, X. and Hagan, R.M. (1998). Cellular and molecular actions of lamotrigine: Possible
mechanisms of efficacy in bipolar disorder. Neuropsychobiology 38 (3): 119-130.
Yang, E. and Korsmeyer, S.J. (1996). Molecular thanatopsis: A discourse on the BCL2
family and cell death. Blood 88 (2): 386-401.
Yankner, B.A., Benowitz, L.I., Villa-Komaroff, L., and Neve, R.L. (1990). Transfection
of PC 12 cells with the human GAP-43 gene: effects on neurite outgrowth and
regeneration. Mol. Brain Res. 7: 39-44.
Zeidman, R., Lofgren, B , Pahlman, S., and Larsson, C. (1999). PKC-epsilon, via its
regulatory domain and independently of its catalytic domain, induces neurite-like
processes in neuroblastoma cells. J Cell Biol. 145 (4): 713-726.

138
Zhang, K-Z , Westberg, J.A., Holtta, E., and Andersson, L.C (1996). BCL2 regulates
neural differentiation. Proc. Natl. Acad. Sci. USA 93: 4504-4508.
Zuber, M X., Goodman, D.W., Kams, L.R., and Fishman, M.C. (1989). The neuronal
growth associated protein GAP-43 induces filopodial formation in non-neuronal cells
Science 244: 1193-1195.

BIOGRAPHICAL SKETCH
The author was born in 1974 in Miami Shores, Florida, to two wonderful parents,
John and Celeste Watterson. She was raised on the east coast of Florida, the oldest of
three children. She earned her bachelor's of science degree in pre-medical biology at
Miami's Barry University in 1995, graduating in three years with honors. She then began
graduate studies in Pharmacology at the University of Florida, initially conducting research
in the laboratory of Robert H. Lenox, M .D , and finishing under the guidance of Edwin M.
Meyer, Ph D. She is currently single and the proud "mommy" of a beautiful chocolate
Labrador retriever-mix named Hershey.
139

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Edwin M. Meyer, Cha'ir
Associate Professor of Pharmacology and Therapeutics
I certify that I have read this study and that in
acceptable standards of scholarly^presentation and
as a dissertation for the degree/(5f Doctor qQPhüdsoj;
pinion it conforms to
'adequate, in scope and quality.
Robert LJenox, Cocháir
Professor of Pharmacology and Therapeutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy
Stephen P Baker
Prof/ssor of Pharmacology and Therapeutics
I certify that 1 have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy
K. Harrison
te Professor of Pharmacology and Therapeutics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy
c
Colin Sumners
Professor of Physiology

This dissertation was submitted to the Graduate Faculty of the College of
Medicine and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy
December 1999
—-— ca
Dean, College of Medicine
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08555 2924




PAGE 1

,1 9,752 678',(6 21 $ 387$7,9( 02/(&8/$5 0(&+$1,60 2) $&7,21 2) 9$/352,& $&,' ,1 75($70(17 2) %,32/$5 ',625'(5 %< -($11(77( 0 :$77(5621 $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

$&.12:/('*0(176 ZLVK WR WKDQN 'U (GZLQ 0H\HU ILUVW DQG IRUHPRVW IRU KH JUDFLRXVO\ DFFHSWHG PH fDV LVf LQWR KLV ODE WKHUHE\ VDYLQJ PH IURP D GHFLVLRQ GLG QRW ZDQW WR PDNH DP JUDWHIXO WKDW KH DOZD\V VDZ DQG KHOSHG PH WR VHH WKH SRVLWLYH DQG RSHQHG PH XS WR QHZ LGHDV WKDW QHYHU WKRXJKW WR H[DPLQH DP YHU\ DSSUHFLDWLYH RI 'U 5REHUW /HQR[fV DELOLW\ WR VWLPXODWH PH WR DOZD\V ORRN IRU PRUH DV ZHOO DV WR JLYH PRUH DQG WR QHYHU EH VDWLVILHG ZLWK OHVV WKDQ b DOVR FUHGLW KLP ZLWK UHGHILQLQJ P\ FRQFHSWLRQ RI fLQWHQVLW\f DQG fGULYHf DP LQGHEWHG WR 'U 'DYLG :DWVRQ IRU WUDLQLQJ PH ZLWK SDWLHQFH IURP VWDUW WR ILQLVK DQG IRU JLYLQJ PH GLUHFWLRQ DQG FRQILGHQFH DW WLPHV ZKHQ KDG OLWWOH DP JUDWHIXO WR P\ FRPPLWWHH PHPEHUV 'UV 6WHSKHQ %DNHU -HIIUH\ +DUULVRQ DQG &ROLQ 6XPQHUV IRU WKHLU JXLGDQFH DQG DSSURYDO ZKLFK QHHGHG LQ RUGHU WR PDLQWDLQ IRFXV DQG PRWLYDWLRQ DP DOVR WKDQNIXO IRU WKH VWHDGIDVW VXSSRUW DQG IULHQGO\ DWPRVSKHUH SURYLGHG E\ WKH GHSDUWPHQW VHFUHWDULHV WKH KHOSIXO WHFKQLFLDQV DQG WKH PDQ\ SHHUV KDYH ZRUNHG ZLWK RYHU WKH SDVW IRXU \HDUV ZLWKRXW ZKRVH KHOS FRXOG QHYHU KDYH UHDFKHG WKLV SRLQW )LQDOO\ ZRXOG OLNH WR WKDQN P\ SDUHQWV ZKR LQVWLOOHG LQ PH WKH DPELWLRQ WR DLP KLJK DQG WKH GULYH WR GR P\ EHVW OL

PAGE 3

7$%/( 2) &217(176 SDJH $&.12:/('*0(176 LL /,67 2) 7$%/(6 Y /,67 2) ),*85(6 YL $%%5(9,$7,216 $1' '(),1,7,216 L[ $%675$&7 [ &+$37(56 ,1752'8&7,21 93$ %LSRODU 'LVRUGHU 0HFKDQLVP RI $FWLRQ RI /LWKLXP LQ 7UHDWPHQW RI %LSRODU 'LVRUGHU 3.& DQG ,WV 6XEVWUDWHV 0$5&.6 DQG *$3 'HYHORSPHQWDO (IIHFWV RI 93$ 6XPPDU\ 6SHFLILF $LPV 0$7(5,$/6 $1' 0(7+2'6 &KHPLFDOV +1 &HOO &XOWXUH 3& &HOO &XOWXUH 1HXULWH 4XDQWLWDWLRQ 3URWHLQ 4XDQWLWDWLRQ E\ :HVWHUQ %ORWWLQJ 5LERQXFOHDVH 5QDVHf 3URWHFWLRQ $VVD\ 53$f 3.& ,QKLELWRUV 3.& $FWLYLW\ 6WDWLVWLFDO $QDO\VHV LLL

PAGE 4

())(&7 2) 93$ 21 0$5&.6 $1' *$3 ,QWURGXFWLRQ 5HVXOWV 'LVFXVVLRQ 52/( 2) 3.& ,1 0(&+$1,60 2) $&7,21 2) 93$ ,QWURGXFWLRQ 5HVXOWV 'LVFXVVLRQ ())(&7 2) 93$ 21 &(// 9,$%,/,7< *52:7+ $1' 0253+2/2*< ,QWURGXFWLRQ 5HVXOWV 'LVFXVVLRQ &21&/86,216 $1' )8785( ',5(&7,216 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ ,9

PAGE 5

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

PAGE 6

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
PAGE 7

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e H[SUHVVLRQ LQ +1 FHOOV D $FXWH 93$ ([SRVXUH 5HSUHVHQWDWLYH :HVWHUQ EORW E &KURQLF 93$ ([SRVXUH 5HSUHVHQWDWLYH :HVWHUQ EORW F 4XDQWLWDWLRQ 3.&FF DQWLVHQVH ROLJRQXFOHRWLGH GHVLJQ (IIHFW RI 3.&D DQWLVHQVH ROLJRQXFOHRWLGHV RQ 3.&D H[SUHVVLRQ LQ +1 FHOOV 5HSUHVHQWDWLYH :HVWHUQ EORW YLL

PAGE 8

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

PAGE 9

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f FHOO OLQH 3, SKRVSKRLQRVLWLGH 3.& SURWHLQ NLQDVH & 3'%X SKRUERO GLEXW\UDWH 53$ 51DVH SURWHFWLRQ DVVD\ 6'63$*( VRGLXP GRGHF\O VXOIDWH SRO\DFU\ODPLGH JHO HOHFWURSKRUHVLV 93$ YDOSURLF DFLG RU YDOSURDWH ,;

PAGE 10

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ ,1 9,752 678',(6 21 $ 387$7,9( 02/(&88$5 0(&+$1,60 2) $&7,21 2) 9$/352,& $&,' ,1 75($70(17 2) %,32/$5 ',625'(5 %\ -HDQQHWWH 0 :DWWHUVRQ 'HFHPEHU &KDLU (GZLQ 0 0H\HU 0DMRU 'HSDUWPHQW 3KDUPDFRORJ\ DQG 7KHUDSHXWLFV 9DOSURLF DFLG 93$ WUDGH QDPH 'HSDNRWHf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f FRQFRPLWDQW ZLWK LQFUHDVHG *$3 H[SUHVVLRQ 7KHVH HIIHFWV DUH DVVRFLDWHG ZLWK DOWHUHG DFWLYLW\ DQG H[SUHVVLRQ RI SURWHLQ NLQDVH & 3.&f DQ LPSRUWDQW [

PAGE 11

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f DQG PD\ SURYLGH IXUWKHU LQVLJKW LQWR WKH SDWKRSK\VLRORJ\ XQGHUO\LQJ ELSRODU GLVRUGHU [L

PAGE 12

&+$37(5 ,1752'8&7,21 93$ LV D EURDGVSHFWUXP DQWLFRQYXOVDQW VKRZQ LQ WKH ODVW WZR GHFDGHV WR EH HIILFDFLRXV LQ WKH WUHDWPHQW RI ELSRODU GLVRUGHU PDQLFGHSUHVVLYH LOOQHVVf 7KHVH WZR DFWLRQV FDQ EH GLVWLQJXLVKHG IURP HDFK RWKHU E\ WKHLU WLPHFRXUVHV ZLWK DQWLFRQYXOVDQW DFWLYLW\ EHLQJ REVHUYHG PXFK PRUH TXLFNO\ RQ D VFDOH RI KRXUV WR GD\V LQ FRQWUDVW WR WKH GD\V WR ZHHNV WUHDWPHQW UHTXLUHG IRU PRRGVWDELOL]LQJ DFWLYLW\ 7KH SUHFLVH PHFKDQLVPV E\ ZKLFK 93$ H[HUWV HLWKHU LWV DQWLHSLOHSWLF RU DQWLPDQLF HIIHFWV DUH VWLOO XQFOHDU 93$ KDV EHHQ IRXQG WR DIIHFW 1D IOX[ LQWUDFHOOXODU &DA DQG EUDLQ JDPPD K\GUR[\EXW\ULF DFLG *$%$f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f FKDUDFWHUL]H WKH WKHUDSHXWLFDOO\ UHOHYDQW WLPH DQG FRQFHQWUDWLRQ GHSHQGHQW HIIHFWV RI 93$ RQ 3.& DFWLYLW\ DQG LVR]\PH H[SUHVVLRQf DQG LWV VXEVWUDWHV 0$5&.6 DQG *$3f XVLQJ WKH KLSSRFDPSDOO\ GHULYHG +1 FHOOV DV D PRGHO V\VWHP IRU QHXURQDO HIIHFWV RI 93$ DQG f FKDUDFWHUL]H WKH WKHUDSHXWLFDOO\ UHOHYDQW HIIHFWV RI 93$ RQ WKH YLDELOLW\ JURZWK DQG PRUSKRORJ\ RI ERWK +1 FHOOV DQG 3&, FHOOV %DVHG RQ SUHOLPLQDU\ HYLGHQFH DQG SUHYLRXV UHSRUWV LW LV K\SRWKHVL]HG WKDW 93$ ZLOO KDYH D VLJQLILFDQW HIIHFW RQ WKH H[SUHVVLRQ RI 3.& VXEVWUDWHV 0$5&.6 DQG *$3 WKDW 3.&

PAGE 13

ZLOO EH LQWHJUDOO\ LQYROYHG LQ DQ\ VXFK HIIHFWV DQG WKDW FHOOV ZLOO XQGHUJR DQ DOWHUDWLRQ LQ FHOO JURZWK DQG PRUSKRORJ\ LQGLFDWLYH RI D WUDQVLWLRQ WR D PRUH GLIIHUHQWLDWHG VWDWH 93$ 93$ LV D VKRUWFKDLQ FDUER[\OLF DFLG WKDW ZDV ILUVW GLVFRYHUHG WR KDYH DQWLHSLOHSWLF HIIHFWV LQ DQG KDV EHHQ XWLOL]HG DV DQ DQWLFRQYXOVDQW LQ WKH 8QLWHG 6WDWHV IRU RYHU \HDUV 0HXQLHU HO DO 5DPVD\ f ,W ILUVW EHFDPH DYDLODEOH LQ WKH 8 6 LQ DV 'HSDNHQH WKH DFLG IRUP RI WKH GUXJ )LYH \HDUV ODWHU WKH HQWHULFFRDWHG IRUP 'HSDNRWH ZDV LQWURGXFHG 3HQU\ DQG 'HDQ f 7KLV SDUWLFXODU IRUPXODWLRQ WKH PRVW SRSXODU IRUP LQ XVH WRGD\ LV D VWDEOH FRRUGLQDWLRQ FRPSRXQG FRQWDLQLQJ HTXDO SURSRUWLRQV RI YDOSURLF DFLG DQG WKH VRGLXP YDOSURDWH VDOW 0F(OUR\ HW D f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f $ QXPEHU RI VWXGLHV KDYH LQYHVWLJDWHG WKH SXWDWLYH PHFKDQLVPVf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

PAGE 14

UHOHDVH RU E\ LQFUHDVLQJ SRVWV\QDSWLF UHVSRQVLYHQHVV WR *$%$ WKURXJK HIIHFWV RQ WKH *$%$UHFHSWRU FRPSOH[ RU DOWHUHG FRQGXFWDQFH HIIHFWLYHO\ GDPSHQLQJ QHXURQDO H[FLWDELOLW\ &KDSPDQ HO D/ 5LPPHU DQG 5LFKDUGV 0F(OUR\ HO D 3HQU\ DQG 'HDQ -RIIH f %HFDXVH DQWLFRQYXOVDQW HIIHFWV DUH RIWHQ REVHUYHG EHIRUH WRWDO EUDLQ *$%$ OHYHOV EHFRPH HOHYDWHG WKH 93$LQGXFHG LQFUHDVH LQ *$%$ OHYHOV DORQH PD\ QRW DFFRXQW IRU WKH GUXJfV DQWLFRQYXOVDQW HIIHFW 93 $ OLNH RWKHU DQWLFRQYXOVDQWV LV NQRZQ WR EORFN VXVWDLQHG KLJK IUHTXHQF\ UHSHWLWLYH ILULQJ RI FXOWXUHG QHXURQV GXH WR LWV HIIHFW RI VORZLQJ UHFRYHU\ RI WKH VRGLXP FKDQQHO IROORZLQJ LQDFWLYDWLRQ $OEXV DQG :LOOLDPVRQ f $W KLJK FRQFHQWUDWLRQV 93$ KDV EHHQ VKRZQ WR DOWHU PHPEUDQH SRWDVVLXP FRQGXFWDQFH DQG UHODWLYHO\ ORZ FRQFHQWUDWLRQV RI WKH GUXJ K\SHUSRODUL]H QHXURQV ERWK HIIHFWV VXJJHVWLYH RI GLUHFW HIIHFWV RQ QHXURQDO PHPEUDQHV 3RUWHU DQG 0HOGUXP f 2WKHU VWXGLHV KDYH SURYLGHG HYLGHQFH IRU UHGXFHG H[FLWDWRU\ QHXURWUDQVPLVVLRQ E\ WKH DPLQR DFLG DVSDUWDWH &KDSPDQ HO DO f 6WLOO RWKHU UHSRUWV KDYH VXJJHVWHG WKDW 93$ ZRUNV WKURXJK LQKLELWLRQ RI NLQGOLQJ D PRGHO RI HSLOHSV\ ZKLFK KROGV WKDW HSLOHSWLF DFWLYLW\ SUHGLVSRVHV WKH V\VWHP WR LQFUHDVHG IUHTXHQF\ RI VHL]XUHV WULJJHUHG E\ ORZHU WKUHVKROGV RYHU WLPH 3RUWHU DQG 0HOGUXP 0F(OUR\ HO DO f ,Q VXPPDU\ WKH YDULRXV PHFKDQLVPV SURSRVHG HQFRPSDVV WKUHH PDMRU QHW HIIHFWV RI 93$ f LQFUHDVHG *$%$HUJLF V\QDSWLF LQKLELWLRQ SRVWV\QDSWLFDOO\ f UHGXFWLRQ LQ WKH H[FLWDWRU\ V\QDSWLF WUDQVPLVVLRQ XOWLPDWHO\ UHVSRQVLEOH IRU HSLOHSWLF NLQGOLQJ DQG f UHGXFWLRQ LQ UHSHWLWLYH ILULQJ YLD D GLUHFW HIIHFW RQ YROWDJHVHQVLWLYH LRQ FKDQQHOV $OEXV DQG :LOOLDPVRQ f 7KH EURDG VSHFWUXP RI DQWLFRQYXOVDQW DFWLYLW\ RI 93$ LV SUREDEO\ UHODWHG WR PRUH WKDQ RQH RI WKH PROHFXODU PHFKDQLVPV GHPRQVWUDWHG WKXV IDU +RZHYHU LW LV XQFOHDU ZKHWKHU VXFK DQWLFRQYXOVLYH HIIHFWV PD\ DOVR DFFRXQW IRU WKH DQWLPDQLF HIILFDF\ RI 93$ DQG EHFDXVH OLWWOH LV NQRZQ RI WKH PRRGVWDELOL]LQJ PHFKDQLVP RI 93$ PXFK UHPDLQV WR EH H[SODLQHG ZLWK UHJDUG WR ZKDW NLQGV RI LQWUDFHOOXODU FKDQJHV PLJKW XQGHUOLH WKLV GUXJfV PRRGVWDELOL]LQJ HIIHFWV

PAGE 15

%LSRODU 'LVRUGHU ,Q UHFHQW \HDUV 93$ KDV HPHUJHG DV DQ HIIHFWLYH DOWHUQDWLYH WR OLWKLXP LQ WKH WUHDWPHQW RI DFXWH PDQLD LQ SDWLHQWV VXIIHULQJ IURP ELSRODU GLVRUGHU *HUQHU t 6WDQWRQ 0F(OUR\ HWDO %RZGHQ HW DO f DQG LV DOVR EHLQJ XVHG IRU ORQJWHUP SURSK\ODFWLF PDQDJHPHQW RI WKLV GLVRUGHU %LSRODU GLVRUGHU DOVR NQRZQ DV PDQLF GHSUHVVLYH LOOQHVV LV D GHELOLWDWLQJ SV\FKLDWULF FRQGLWLRQ IRU ZKLFK WKHUH LV QR NQRZQ FDXVH DQG RQO\ D OLPLWHG UDQJH RI WUHDWPHQW RSWLRQV DYDLODEOH ,W LV HVWLPDWHG WR DIIHFW DERXW b RI WKH SRSXODWLRQ LQ WKLV FRXQWU\ %RZGHQ HO DO f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f 7KH PDQLF HSLVRGHV HQWDLO VXFK GLVUXSWLYH V\PSWRPV DV K\SHUDFWLYLW\ H[SORVLYH WHPSHU LPSDLUHG MXGJHPHQW LQVRPQLD GLVRUJDQL]HG EHKDYLRU K\SHUVH[XDOLW\ DQG JUDQGLRVLW\ DQ\ FRPELQDWLRQ RI ZKLFK PD\ OHDG WR DOLHQDWLRQ IURP IDPLO\ DQG IULHQGV LQGHEWHGQHVV MRE ORVV DQG RWKHU PDMRU OLIH SUREOHPV %RZGHQ HW DO f ,I XQWUHDWHG WKLV GLVRUGHU KDV D WHQGHQF\ WR ZRUVHQ ZLWK LQFUHDVLQJ VHYHULW\ GXUDWLRQ DQG IUHTXHQF\ RI HSLVRGHV RYHU WLPH 3RVW f 7KHUH LV D KLJK UDWH RI VXLFLGH LQ SDWLHQWV VXIIHULQJ IURP WKLV GLVRUGHU HVWLPDWHG DW DERXW b LQ XQWUHDWHG FDVHV %RZGHQ HW DO f /LWKLXP KDV IRU RYHU \HDUV EHHQ WKH ILUVWOLQH GUXJ WKHUDS\ IRU ERWK DFXWH DQG SURSK\ODFWLF PDQDJHPHQW RI WKLV GLVRUGHU EXW GXH WR LWV QDUURZ WKHUDSHXWLF LQGH[ DQG PDQ\ XQGHVLUDEOH VLGH HIIHFWV RIWHQ UHVXOWLQJ LQ FRPSOLDQFH SUREOHPVf WKH QHHG IRU RWKHU YLDEOH WUHDWPHQW RSWLRQV KDV EHFRPH D PDMRU FRQFHUQ 3RVW %RZGHQ f $OWKRXJK QXPHURXV WUHDWPHQWV KDYH EHHQ XVHG H[SHULPHQWDOO\ LQ WKH SDVW LQFOXGLQJ

PAGE 16

DQWLSV\FKRWLF GUXJV FDOFLXP FKDQQHO EORFNHUV VXFK DV YHUDSDPLOf KLJK GRVHV RI WK\UR[LQH 7f VHURWRQLQ SUHFXUVRUV VXFK DV /WU\SWRSKDQf DQG HYHQ HOHFWURFRQYXOVLYH WKHUDS\ -RIIH f UHVXOWV KDYH QRW EHHQ FRQVLVWHQWO\ SRVLWLYH DQG QRQH LV XVHG H[WHQVLYHO\ DV D fPDLQVWUHDPf WUHDWPHQW DW WKH SUHVHQW WLPH )RU H[DPSOH WKH XVH RI DQWLGHSUHVVDQWV LV JHQHUDOO\ QRW FRQVLGHUHG EHFDXVH SDWLHQWV PD\ EH EURXJKW RXW RI WKHLU GHSUHVVLYH HSLVRGHV RQO\ WR EH LPPHGLDWHO\ VXEPHUJHG LQWR WKH PDQLF SKDVH IXUWKHU FHUWDLQ GUXJV PD\ LQFUHDVH WKH IUHTXHQF\ RI F\FOLQJ RI PRRG VZLQJV 3RVW f 7KH DQWLFRQYXOVDQWV FDUEDPD]HSLQH ODPRWULJLQH DQG JDEDSHQWLQ DV ZHOO DV RWKHU SKDUPDFRORJLFDO DJHQWV FOR]DSLQH FORQD]HSDP YHUDSDPLOf KDYH VKRZQ SURPLVH LQ WUHDWLQJ ELSRODU GLVRUGHU EXW YHU\ IHZ VWXGLHV KDYH EHHQ FRQGXFWHG DQG )'$ DSSURYDO KDV \HW WR EH REWDLQHG IRU WKLV FOLQLFDO LQGLFDWLRQ %RZGHQ -RIIH +ROOLVWHU f %HFDXVH OLWKLXP LV WKH DSSURYHG ILUVWOLQH WUHDWPHQW DQG KDV EHHQ XVHG DQG VWXGLHG H[WHQVLYHO\ LQ WKH SDVW WKLV SURMHFW ZLOO EH IRFXVHG LQ SDUW WRZDUG XQGHUVWDQGLQJ WKH PHFKDQLVP RI DFWLRQ RI ELSRODU GUXJV EDVHG RQ ZKDW LV NQRZQ DQG KDV EHHQ UHSRUWHG H[SHULPHQWDOO\ IRU OLWKLXP 93$ LV HPHUJLQJ DV D YLDEOH DOWHUQDWLYH RU FRPELQDWLRQDO GUXJ WUHDWPHQW IRU ELSRODU GLVRUGHU DQG LQ IDFW 93$ PD\ HYHQ KDYH D ZLGHU VSHFWUXP RI HIILFDF\ WKDQ OLWKLXP $OWKRXJK 93$ VHHPV WR EH OHVV HIIHFWLYH LQ WUHDWLQJ GHSUHVVLRQ WKDQ LQ WUHDWLQJ WKH PDQLF SKDVH RI WKH GLVRUGHU WKHUH LV DFFXPXODWLQJ HYLGHQFH RI LWV XVH LQ DW\SLFDO G\VSKRULFPL[HGf PDQLD UDSLG F\FOLQJ DQG VHFRQGDU\ PDQLDV ZKLFK KDYH SRRUHU SURJQRVHV DQG DUH JHQHUDOO\ OHVV UHVSRQVLYH WR OLWKLXP WKDQ WKH XVXDO IRUPV RI WKH GLVRUGHU &DODEUHVH t 'HOXFFKL &DODEUHVH HO DOf %RZGHQ HO DO -RIIH %URZQ f 93$ KDV DOVR EHHQ XVHG WR WUHDW YDULRXV FRQGLWLRQV ZKLFK DUH FRPRUELG ZLWK ELSRODU GLVRUGHU VXFK DV PLJUDLQHV SDQLF GLVRUGHU EXOLPLD DWWHQWLRQ GHILFLW K\SHUDFWLYLW\ GLVRUGHU $'+'f DQG RWKHU EUDLQUHODWHG LOOQHVVHV DQG 93$ DSSHDUV WR EH TXLWH HIIHFWLYH LQ WUHDWLQJ OLWKLXP QRQUHVSRQGHUV %RZGHQ -RIIH %URZQ f 7KH WKHUDSHXWLF LQGH[ RI 93$ LV FRQVLGHUDEO\ ZLGHU WKDQ WKDW RI OLWKLXP WKHUHE\ DOORZLQJ IRU JUHDWHU IUHHGRP LQ GRVDJH DGMXVWPHQW ZLWK OHVV ULVN RI WR[LFLW\

PAGE 17

)XUWKHUPRUH WKH VLGH HIIHFW SURILOH RI 93 $ LV OHVV LQWLPLGDWLQJ WKDQ WKDW RI OLWKLXP :KHUHDV WKH VLGH HIIHFWV RI OLWKLXP PD\ EH TXLWH VHYHUH DQG GLVWXUELQJ WR SDWLHQWV WK\URLG UHQDO DQG &16 HIIHFWV GHUPDWRORJLF DQG FDUGLRYDVFXODU FRPSOLFDWLRQVf 93 $ W\SLFDOO\ H[KLELWV D GLIIHUHQW PRUH EHQLJQ SURILOH RI VLGH HIIHFWV ZKLFK DUH JHQHUDOO\ PRUH DFFHSWDEOH WR SDWLHQWV JDVWURLQWHVWLQDO XSVHW ZHLJKW JDLQ KDLU WKLQQLQJ LQFUHDVHG KHSDWLF HQ]\PH OHYHOVf DQG PDQ\ RI LWV ERWKHUVRPH VLGH HIIHFWV DUH DOOHYLDWHG RYHU WLPH RU ZLWK D GHFUHDVH LQ GUXJ GRVDJH %RZGHQ 3HQU\ DQG 'HDQ f 7KH H[FHSWLRQ WR WKLV LV WKH LGLRV\QFUDWLF XQSUHGLFWDEOH DQG LUUHYHUVLEOH KHSDWLF WR[LFLW\ LQGXFHG E\ 93$ LQ SUHGLVSRVHG SDWLHQWV RQ UDUH RFFDVLRQ :KLOH WKLV FRQGLWLRQ PD\ EH IDWDO LI QRW GHWHFWHG LQ WLPH LW DIIHFWV RQO\ D YHU\ PLQLPDO SRUWLRQ RI WKH SRSXODWLRQ WKH KLJKHVW LQFLGHQFH UHSRUWHG ZDV D UDWH RI b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f WKHVH HIIHFWV JHQHUDOO\ RFFXU DW UDWKHU KLJK FRQFHQWUDWLRQV DQG PD\ UHODWH PRUH WR WKH DQWLFRQYXOVDQW SURSHUWLHV RI 93$ WKDQ WR LWV PRRG VWDELOL]LQJ HIIHFWV VHH )LJXUH :DOGPHLHU 0RWRKDVKL f )XUWKHU DV LV WKH FDVH IRU WKH WLPH FRXUVH RI OLWKLXP WKHUH LV D GHOD\ RI VHYHUDO GD\V LQ WKH RQVHW RI FOLQLFDO DQWLPDQLF DFWLRQ RI 93$ DOWKRXJK ORDGLQJ VWUDWHJLHV KDYH GHPRQVWUDWHG DQWLPDQLF HIIHFWV DW WLPHV DV HDUO\ DV G 0F(OUR\ HO DO f 7KXV DQ\ PHFKDQLVP SRVWXODWHG IRU WKH WKHUDSHXWLF DFWLRQ RI WKHVH GUXJV PXVW WDNH WKLV GHOD\HG RQVHW LQWR DFFRXQW 0DQML HO DO f %HFDXVH 93$ DQG OLWKLXP DUH FXUUHQWO\ WKH RQO\ WZR GUXJV DSSURYHG E\ WKH )'$ IRU WUHDWPHQW RI

PAGE 18

DFXWH PDQLD FRPSDULVRQV RI WKH DFWLRQV RI WKHVH DJHQWV DQG HIIRUWV WR HOXFLGDWH WKH PHFKDQLVPVf XQGHUO\LQJ WKH PRRGVWDELOL]LQJ DFWLYLW\ RI ERWK GUXJV LV LPSRUWDQW WR XQGHUVWDQGLQJ WKH SDWKRSK\VLRORJ\ RI WKLV LOOQHVV )XUWKHU WKLV DSSURDFK PD\ OHDG WR WKH LGHQWLILFDWLRQ RI QHZ RU H[LVWLQJ GUXJV ZKLFK PD\ EH XVHIXO LQ WKLV DSSOLFDWLRQ 0HFKDQLVP RI $FWLRQ RI /LWKLXP LQ 7UHDWPHQW RI %LSRODU 'LVRUGHU $FFXPXODWLQJ HYLGHQFH IURP RXU ODERUDWRU\ DQG RWKHUV VWURQJO\ LPSOLFDWHV UHFHSWRUPHGLDWHG 3, VLJQDOLQJ LQ WKH PHFKDQLVP RI DFWLRQ RI OLWKLXP LQ WKH EUDLQ %HUULGJH HW DOf *RGIUH\ /HQR[ DQG :DWVRQ f 1XPHURXV VWXGLHV KDYH HVWDEOLVKHG OLWKLXP DV DQ XQFRPSHWLWLYH LQKLELWRU RI WKH HQ]\PH LQRVLWRO PRQRSKRVSKDWDVH ,03DVHf DQG WKH UHVXOWLQJ DFFXPXODWLRQ RI LQWHUPHGLDWHV LQ WKH SDWKZD\ ZKLFK DUH UHVSRQVLEOH IRU DGGLWLRQDO GRZQVWUHDP HIIHFWV LQFOXGLQJ LQWUDFHOOXODU &DA UHOHDVH DQG 3.& DFWLYDWLRQ VHH )LJXUH 3RVW HW DO f ,W ZDV ILUVW VXJJHVWHG E\ %HUULGJH DQG FROOHDJXHV WKDW GHSOHWLRQ RI IUHH LQRVLWRO UHVXOWLQJ IURP WKH DFFXPXODWLRQ RI LQWHUPHGLDWHV LQ WKH 3, VLJQDOLQJ SDWKZD\ LV D PDMRU FRQVHTXHQFH RI ORQJWHUP OLWKLXP H[SRVXUH %HUULGJH HW DO %HUULGJH f 7KLV SKHQRPHQRQ ZRXOG KDYH JUHDWHVW LPSOLFDWLRQ LQ FHUWDLQ DUHDV RI WKH EUDLQ ZKLFK DUH UHODWLYHO\ PRUH LQRVLWROOLPLWHG E\ YLUWXH RI WKH EORRGEUDLQ EDUULHU DV ZHOO DV OHVV FDSDEOH RI SURGXFLQJ LQRVLWRO GH QRYR )XUWKHU EHFDXVH WKH 3, VLJQDOLQJ FDVFDGH LV OLQNHG WR SURWHLQFRXSOHG PXVFDULQLF UHFHSWRU PL QYc PVf DFWLYDWLRQ DUHDV RI WKH EUDLQ XQGHUJRLQJ WKH KLJKHVW UDWH RI UHFHSWRU DFWLYDWLRQ PD\ SUHGLFWDEO\ EH SUHIHUHQWLDOO\ DIIHFWHG E\ OLWKLXP ,QGHHG VWXGLHV LQ RXU ODERUDWRU\ KDYH SURYLGHG HYLGHQFH IRU DQ LQRVLWROGULYHQ DWWHQXDWLRQ RI WKH HIIHFWV RI OLWKLXP LQ +1 FHOOV DV ZHOO DV SRWHQWLDWLRQ E\ D PXVFDULQLF UHFHSWRU DJRQLVW FDUEDFKROf DQG SUHYHQWLRQUHYHUVDO E\ D PXVFDULQLF UHFHSWRU DQWDJRQLVW DWURSLQHf :DWVRQ DQG /HQR[ f )ROORZLQJ H[SRVXUH RI +1 FHOOV WR WKHUDSHXWLF FRQFHQWUDWLRQV RI OLWKLXP P0f 0$5&.6 SURWHLQ H[SUHVVLRQ LV VLJQLILFDQWO\ GRZQUHJXODWHG /HQR[ HW DO

PAGE 19

:DWVRQ DQG /HQR[ f 7KLV 0$5&.6 UHGXFWLRQ RQO\ RFFXUV LQ WKH UHODWLYH DEVHQFH RI LQRVLWRO IRU VXSSOHPHQWDWLRQ RI LQRVLWRO HYHQ DW FRQFHQWUDWLRQV DV ORZ DV L0 SUHYHQWV WKH OLWKLXPLQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 :DWVRQ DQG /HQR[ :DWVRQ HO DO VHH DOVR 7DEOH f 6LPLODUO\ WKH FRQFRPLWDQW H[SRVXUH RI +1 FHOOV WR ERWK OLWKLXP DQG WKH PXVFDULQLF DJRQLVW FDUEDFKRO UHVXOWV LQ DQ LQFUHDVH LQ WKH OHYHO RI 0$5&.6 SURWHLQ GRZQUHJXODWLRQ REVHUYHG DQG H[SRVXUH RI FHOOV WR WKH PXVFDULQLF DQWDJRQLVW DWURSLQH UHYHUVHV RU SUHYHQWV WKH OLWKLXPLQGXFHG UHGXFWLRQ LQ 0$5&.6 H[SUHVVLRQ :DWVRQ DQG /HQR[ :DWVRQ HO DO VHH DOVR 7DEOH f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f VKRZHG WKDW 93$ KDG QR HIIHFW HLWKHU VWLPXODWRU\ RU LQKLELWRU\ RQ ,03DVH SDUWLDOO\ SXULILHG IURP ERYLQH EUDLQ )XUWKHU OLWKLXP DQG FDUEDPD]HSLQH D OHVV FRPPRQO\ XVHG DQWLPDQLF DJHQWf H[KLELWHG RSSRVLWH HIIHFWV RQ ,03DVH VXJJHVWLQJ WKDW WKH HQ]\PH LV QRW D FRPPRQ VLWH RI DFWLRQ IRU WKHVH WZR PRRG VWDELOL]LQJ PHGLFDWLRQV 'L[RQ DQG +RNLQ f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

PAGE 20

RI VHOHFW SURWHLQ VXEVWUDWHV /HQR[ 0DQML DQG /HQR[ :DWVRQ DQG /HQR[ f )XUWKHU WKH FOLQLFDO HIIHFWV RI OLWKLXP PD\ EH PHGLDWHG DW OHDVW LQ SDUW WKURXJK WKHVH DFWLRQV $OWKRXJK 93$ DSSDUHQWO\ GRHV QRW VKDUH WKH SURSHUW\ RI 3, VLJQDOLQJ LQKLELWLRQ 9DGQDO DQG 3DUWKDVDUDWK\ 'L[RQ DQG +RNLQ f WKH TXHVWLRQ UHPDLQV DV WR ZKHWKHU LW DIIHFWV 3.& DV SUHYLRXV VWXGLHV KDYH VXJJHVWHG &KHQ HW DO 2f%ULHQ DQG 5HJDQ f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f 1XPHURXV VWXGLHV KDYH VXJJHVWHG D UROH IRU 3.& LQ WKH ORQJWHUP DFWLRQ RI OLWKLXP /HQR[ 0DQML DQG /HQR[ f DQG UHFHQW GDWD KDYH SURYLGHG HYLGHQFH WKDW FKURQLF 93$ H[SRVXUH DOWHUV 3.& DFWLYLW\ DV ZHOO DV WKH H[SUHVVLRQ RI D DQG LVR]\PHV LQ QRQQHXURQDO FHOOV &KHQ HW DO 0DQML HW DO Ef 3UHYLRXV VWXGLHV KDYH GHPRQVWUDWHG WKDW FKURQLF EXW QRW DFXWHf OLWKLXP WUHDWPHQW RI UDWV UHVXOWLQJ LQ FOLQLFDOO\ UHOHYDQW EUDLQ FRQFHQWUDWLRQV SURGXFHV D VLJQLILFDQW UHGXFWLRQ LQ WKH 3.& VXEVWUDWH 0$5&.6 LQ WKH KLSSRFDPSXV ZKLFK SHUVLVWV EH\RQG WUHDWPHQW GLVFRQWLQXDWLRQ /HQR[ HW DO 0DQML HW DO Df 7KLV OLWKLXPLQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 KDV DOVR EHHQ GHPRQVWUDWHG LQ DQ LPPRUWDOL]HG KLSSRFDPSDO FHOO OLQH :DWVRQ DQG /HQR[ f LQ ZKLFK LW ZDV SUHYLRXVO\ GHPRQVWUDWHG WKDW SKRUERO HVWHUV ZKLFK GLUHFWO\ DFWLYDWH 3.& GRZQUHJXODWH 0$5&.6 SURWHLQ H[SUHVVLRQ LQ D 3.&GHSHQGHQW PDQQHU :DWVRQ HW DO f

PAGE 21

0$5&.6 LV D SURPLQHQW DQG SUHIHUHQWLDO VXEVWUDWH LQ WKH EUDLQ IRU 3.& ZKLFK E\ YLUWXH RI SKRVSKRU\ODWLRQ UHJXODWHV WKH FHOOXODU ORFDOL]DWLRQ DQG DFWLYLW\ RI WKLV SURWHLQ 0$5&.6 ELQGV FDOPRGXOLQ LQ D FDOFLXPGHSHQGHQW IDVKLRQ DQG FURVVOLQNV ILODPHQWRXV DFWLQ DQG KDV EHHQ LPSOLFDWHG LQ FHOOXODU SURFHVVHV DVVRFLDWHG ZLWK F\WRVNHOHWDO UHVWUXFWXULQJ DQG QHXURSODVWLFLW\ H J WUDQVPHPEUDQH VLJQDOLQJ DQG QHXURWUDQVPLWWHU UHOHDVH $GHUHP %ODFNVKHDU f ,Q IDFW SUHYLRXV VWXGLHV LQ PXWDQW PLFH ODFNLQJ WKH 0$5&.6 JHQH KDYH VKRZQ WKDW 0$5&.6 H[SUHVVLRQ LV HVVHQWLDO WR QRUPDO &16 GHYHORSPHQW LQ DQLPDOV 6WXPSR HO DO f ,QVRIDU DV 0$5&.6 PD\ DOVR UHSUHVHQW D PROHFXODU WDUJHW IRU PRRG VWDELOL]HUV LQ WKH EUDLQ /HQR[ DQG :DWVRQ :DWVRQ DQG /HQR[ f LW VKRXOG EH RI EHQHILW WR H[DPLQH LWV SRWHQWLDO UHJXODWLRQ E\ 93$ DV ZHOO DV E\ RWKHU SV\FKRWURSLF DJHQWV 7KH JURZWKDVVRFLDWHG SURWHLQ *$3 LV D VHFRQG SURWHLQ WKRXJKW WR EH LQWHJUDO WR QRUPDO QHXURQDO GHYHORSPHQW DQG GLIIHUHQWLDWLRQ DV LW LV H[SUHVVHG DW KLJK OHYHOV GXULQJ QHUYRXV V\VWHP GHYHORSPHQW DQG LV DVVRFLDWHG ZLWK QHXULWH H[WHQVLRQ -DS 7MRHQ 6DQ HO DO $LJQHU DQG &DURQL 6WULWWPDWWHU HO DO f *$3 VKDUHV D QXPEHU RI SURSHUWLHV ZLWK 0$5&.6 LQFOXGLQJ DELOLW\ WR ELQG DFWLQ DQG FDOPRGXOLQ DOEHLW LQ D &DALQGHSHQGHQW IDVKLRQf DV ZHOO DV UHJXODWLRQ E\ 3.& UHYLHZHG LQ %HQRZLW] DQG 5RXWWHQEHUJ f ,QDVPXFK DV SUHYLRXV VWXGLHV KDYH SURYLGHG HYLGHQFH IRU DQ LQYHUVH UHJXODWLRQ RI 0$5&.6 DQG *$3 LQ FHUWDLQ UHJLRQV RI UDW EUDLQ 0F1DPDUD DQG /HQR[ f LW PD\ EH LQVWUXFWLYH WR DVVHVV WKH UHODWLYH OHYHOV RI *$3 REVHUYHG FRQFRPLWDQW ZLWK 93$LQGXFHG DOWHUDWLRQV LQ 0$5&.6 H[SUHVVLRQ ,QYHVWLJDWLRQ RI WKHVH WZR UHODWHG EXW GLIIHUHQWLDOO\ UHJXODWHG SURWHLQV ZLOO OLNHO\ SURYLGH LQVLJKW LQWR WKH PROHFXODU PHFKDQLVP RI WKLV GUXJ DV ZLOO H[DPLQDWLRQ RI WKLV GUXJfV HIIHFWV RQ 3.& WKH SULPDU\ HQ]\PH UHVSRQVLEOH IRU WKH GLUHFW UHJXODWLRQ RI ERWK RI WKHVH SURWHLQV

PAGE 22

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f LQ RIIVSULQJ RI 93 $WUHDWHG DQG QRQWUHDWHG PRWKHUV 6PLWKHOOV HWD 5KRDGV DQG 0LOOV 7URW] HO D 1RVHO DQG .OHLQ (KOHUV HO D f 1XPHURXV VWXGLHV KDYH LQYHVWLJDWHG WKLV SKHQRPHQRQ \HW LW UHPDLQV XQNQRZQ KRZ 93$ FDXVHV WKHVH GHYHORSPHQWDO GHIHFWV RU KRZ IRODWH VXSSOHPHQWV DUH DEOH WR SUHYHQW VXFK HIIHFWV 3UHYLRXV ZRUN KDV GHPRQVWUDWHG WKDW 93$ LQKLELWV FHOO SUROLIHUDWLRQ WKURXJK DUUHVW RI WKH FHOO F\FOH DW PLG *O SKDVH 0DUWLQ DQG 5HJDQ f DV ZHOO DV LQGXFHV GLIIHUHQWLDWLRQ RI QHXUREODVWRPD DQG JOLRPD FHOOV 5HJDQ f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

PAGE 23

WKRVH FRQGXFWHG SUHYLRXVO\ RU LQ SDUDOOHO IRU OLWKLXP VR WKDW FRPSDULVRQV PD\ EH PDGH ZKLFK PLJKW VKHG OLJKW RQ WKH EDVLV IRU WKH FRPPRQ PRRGVWDELOL]LQJ WKHUDSHXWLF DFWLRQ RI WKHVH WZR GUXJV ,Q VXPPDU\ WKXV IDU LW KDV EHHQ VKRZQ WKDW LQ WKH +1 FHOO PRGHO OLWKLXP HOLFLWV D GRZQUHJXODWLRQ LQ 0$5&.6 SULPDULO\ WKURXJK LWV HIIHFW RQ 3, VLJQDOLQJ DV HYLGHQFHG E\ WKH FDUEDFKROGULYHQ SRWHQWLDWLRQ DQG HLWKHU LQRVLWRO RU DWURSLQHGULYHQ DWWHQXDWLRQ RI HIIHFW :DWVRQ DQG /HQR[ :DWVRQ HW D f 7KLV OLWKLXP LQGXFHG 0$5&.6 UHGXFWLRQ RFFXUV RYHU D FKURQLF WLPH FRXUVH UHTXLULQJ GD\V WR ZHHNV DQG WKH HIIHFW LV JUHDWHU LQ WKH VROXEOH UDWKHU WKDQ PHPEUDQH IUDFWLRQ RI WKH FHOOV /HQR[ HW D :DWVRQ DQG /HQR[ f $ VLPLODU WLPH FRXUVH RI 0$5&.6 P51$ UHGXFWLRQ LV DOVR REVHUYHG :DWVRQ DQG /HQR[ f )XUWKHU FKURQLF OLWKLXP DGPLQLVWUDWLRQ HOLFLWV D VLJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ LQ YLYR LQ D UDW PRGHO /HQR[ HW D f $V IRU WKH HIIHFWV RI OLWKLXP RQ FHOO YLDELOLW\ DQG UHODWHG SDUDPHWHUV QR DOWHUDWLRQ LQ FHOO JURZWK UDWH RU PRUSKRORJ\ KDV EHHQ REVHUYHG IROORZLQJ DFXWH RU FKURQLF OLWKLXP H[SRVXUH :DWVRQ DQG /HQR[ f DQG QR DOWHUDWLRQ LQ H[SUHVVLRQ RI DQRWKHU SURPLQHQW 3.& VXEVWUDWH *$3 ZDV REVHUYHG :DWVRQ DQG /HQR[ f /LWKLXP H[SRVXUH UHVXOWV LQ SUHGLFWDEOH FKDQJHV LQ 3.& DFWLYLW\ ZLWK LQLWLDO DFWLYDWLRQ IROORZHG E\ FKURQLF GRZQUHJXODWLRQf DV ZHOO DV DOWHUDWLRQV LQ H[SUHVVLRQ RI WKH WZR 3.& LVRIRUPV D DQG V :DWVRQ DQG /HQR[ f VLPLODU WR ILQGLQJV SUHYLRXVO\ UHSRUWHG E\ 0DQML HW D DEf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

PAGE 24

6SHFLILF $LPV &KDUDFWHUL]H WKH 7KHUDSHXWLFDOO\ 5HOHYDQW (IIHFWV RI 93$ RQ 3.& DQG ,WV 6XEVWUDWHV 0$5&.6 DQG *$3 DUH NQRZQ WR EH LPSRUWDQW LQ PRGXODWLQJ FHQWUDO QHUYRXV V\VWHP &16f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f HVWDEOLVKLQJ WKH FRQFHQWUDWLRQUHVSRQVH UHODWLRQVKLS IRU WKH HIIHFWV RI 93$ RQ 0$5&.6 SURWHLQ ZLOO f HVWDEOLVK WKH WLPH FRXUVH RI DFWLRQ RI 93$ DW WKHUDSHXWLF FRQFHQWUDWLRQ DV ZHOO DV f WKH WLPH FRXUVH RI V\VWHP UHFRYHU\ DIWHU 93$ ZLWKGUDZDO ,Q DGGLWLRQ ZLOO DVVHVV WKH LPSRUWDQFH RI f LQRVLWRO FRQFHQWUDWLRQ DQG f PXVFDULQLF UHFHSWRU*T DFWLYDWLRQ LQ 93$LQGXFHG 0$5&.6 HIIHFWV )XUWKHUPRUH ZLOO f LQYHVWLJDWH WKH HIIHFWV RI FRPELQHG 93$ DQG OLWKLXP WUHDWPHQW (IIHFWV ZLOO EH H[DPLQHG LQ ERWK F\WRVROLF DQG PHPEUDQH IUDFWLRQV RI FHOOV ,W LV SUHGLFWHG WKDW 93$ ZLOO KDYH HIIHFWV VLPLODU WR WKRVH RI OLWKLXP LH VLJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ H[SUHVVLRQ IROORZLQJ FKURQLF EXW QRW DFXWHf H[SRVXUH WR FRQFHQWUDWLRQV RI WKH GUXJ ZLWKLQ D WKHUDSHXWLFDOO\ UHOHYDQW UDQJH DQG WKDW 0$5&.6 ZLOO UHWXUQ WR EDVDO OHYHOV DIWHU GUXJ UHPRYDO ,Q FRQWUDVW WR OLWKLXP WKHUH LV QR HYLGHQFH IRU 93$LQGXFHG PRGXODWLRQ RI WKH

PAGE 25

3, VLJQDOLQJ SDWKZD\ VR QR HIIHFWV E\ LQRVLWRO RU FDUEDFKRO PXVFDULQLF UHFHSWRU DJRQLVWf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f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f RYHU D EURDG WLPH FRXUVH IURP DFXWH WR FKURQLF WLPH SRLQWV 93$

PAGE 26

LQGXFHG DOWHUDWLRQV LQ 3.& DFWLYLW\ DUH H[SHFWHG IRU WKH VDPH UHDVRQV DV GHVFULEHG DERYH LQ 5ROH RI 3.& LQ PHFKDQLVP RI DFWLRQ RI 93$ 7KURXJK WKH XVH RI 3.& LQKLELWRUV DQG DQWLVHQVH ZKLFK EORFN 3.& DFWLYLW\ SODQ WR IXUWKHU FKDUDFWHUL]H $( DERYH VSHFLILFDOO\ 0$5&.6 DQG *$3 DOWHUDWLRQV DV ZHOO DV PRUSKRORJLFDO HIIHFWV VHH EHORZf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fV WKHUDSHXWLF SURILOH DQWLFRQYXOVDQW DQG PRRGVWDELOL]LQJ \HW WHUDWRJHQLFf 2EVHUYHG DOWHUDWLRQV LQ 0$5&.6

PAGE 27

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

PAGE 28

)LJXUH 0ROHFXODU (IIHFWV RI 93$ LQ %UDLQ 'HSLFWHG DUH f DW ULJKW WKH HIIHFWV RI 93$ DV SUHYLRXVO\ HVWDEOLVKHG E\ RWKHU JURXSV DQG WKRXJKW WR EH UHVSRQVLEOH IRU WKH DQWLFRQYXOVDQW HIILFDF\ RI WKH GUXJ DQG f DW OHIW WKH ORQJWHUP HIIHFWV K\SRWKHVL]HG WR FRQWULEXWH WR WKH PRRGVWDELOL]LQJ HIILFDF\ RI 93$ 2I WKH VKRUWWHUP HIIHFWV DW ULJKW LQFUHDVHG *$%$ QHXURWUDQVPLVVLRQ DQG HIIHFWV RQ LQWUDFHOOXODU FDOFLXP DUH JHQHUDOO\ WKRXJKW WR FRQWULEXWH OHVV WKDQ WKH HIIHFWV RI 93$ RQ YROWDJHVHQVLWLYH LRQ FKDQQHOV DQG WKH DVVRFLDWHG GHFUHDVH LQ H[FLWDWRU\ QHXURWUDQVPLVVLRQ DQG QHXURQDO ILULQJ 7KH K\SRWKHVL]HG HIIHFWV DW OHIW ZLOO EH DVVHVVHG LQ GHWDLO LQ WKH VWXGLHV ZKLFK IROORZ

PAGE 29

)LJXUH 0RGHO IRU /LWKLXP 5HJXODWLRQ RI 0$5&.6 DGDSWHG IURP /HQR[ HW DO f )ROORZLQJ DJRQLVW RFFXSDQF\ RI WKH UHFHSWRU ELQGLQJ VLWH *SURWHLQ FRXSOLQJ PHGLDWHV DFWLYDWLRQ RI WKH HQ]\PH SKRVSKROLSDVH & 3/&f ZKLFK FDWDO\]HV WKH K\GURO\VLV RI SKRVSKDWLG\OLQRVLWRO ELVSKRVSKDWH 3,3f LQWR VHFRQG PHVVHQJHUV GLDF\OJO\FHURO '$*f DQG P\RLQRVLWRO WULSKRVSKDWH ,QV 3f '$* LV DQ HQGRJHQRXV DFWLYDWRU RI 3.& DQG ,QV 3 LQGXFHV UHOHDVH RI FDOFLXP IURP LQWUDFHOOXODU VWRUHV VXFK DV WKH HQGRSODVPLF UHWLFXOXP /LWKLXP LQKLELWV XQFRPSHWLWLYHO\ WKH HQ]\PH LQRVLWRO PRQRSKRVSKDWDVH ,03DVHf WKHUHE\ LQKLELWLQJ WKH UHF\FOLQJ RI LQRVLWRO 7KLV UHVXOWV LQ DFFXPXODWLRQ RI PHWDEROLWHV LQ WKH '$* SDWKZD\ GXULQJ VWLPXODWLRQ RI UHFHSWRU VLJQDOLQJ $FWLYDWLRQ RI 3.&r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

PAGE 30

'1$

PAGE 31

&+$37(5 0$7(5,$/6 $1' 0(7+2'6 &KHPLFDOV 93$ SURS\OSHQWDQRLF DFLGf DQG RWKHU FKHPLFDOV ZHUH SXUFKDVHG IURP 6LJPD 6W /RXLV 02f XQOHVV RWKHUZLVH QRWHG )RU VRPH GUXJ FRPSDULVRQ VWXGLHV DGGLWLRQDO 93$ DQG 93$ PHWDEROLWHV ZHUH REWDLQHG IURP $EERWW /DERUDWRULHV $EERWW 3DUN ,/f DQG IOXR[HWLQH DQG /< D 3.& LQKLELWRUf ZHUH SURYLGHG E\ (OL /LOO\ DQG &RPSDQ\ ,QGLDQDSROLV ,1f 7LVVXH FXOWXUH UHDJHQWV ZHUH SXUFKDVHG WKURXJK *LEFR *DLWKHUVEXUJ 0' KRUVH VHUXP 530, DQG '0( PHGLDf +\FORQH /RJDQ 87 IHWDO ERYLQH VHUXPf DQG 6LJPD WU\SVLQ DQG JHQWDPLFLQf (OHFWURSKRUHVLV UHDJHQWV ZHUH SXUFKDVHG IURP %LR5DG +HUFXOHV &$f DQG 53$ UHDJHQWV ZHUH SXUFKDVHG IURP $PELRQ $XVWLQ 7;f +1 &HOO &XOWXUH 7KH LPPRUWDOL]HG KLSSRFDPSDO FHOO OLQH +1 GZf ZDV XVHG WR DVVHVV HIIHFWV RI 93$ RQ DOO SDUDPHWHUV LQFOXGLQJ 0$5&.6 SURWHLQ DQG P51$ *$3 3.& DFWLYLW\ DQG H[SUHVVLRQ HWF +1 FHOOV ZHUH NLQGO\ SURYLGHG E\ 'U %UXFH :DLQHU (PRU\ 8QLYHUVLW\ $WODQWD *$f 7KHVH DUH GHULYHG IURP WKH IXVLRQ RI SULPDU\ QHXURQV IURP WKH KLSSRFDPSXV RI SRVWQDWDO G PLFH ZLWK WKH 17* KXPDQ QHXUREODVWRPD FHOO OLQH DQG H[KLELW PRUSKRORJLFDO LPPXQRORJLFDO DQG HOHFWURSK\VLRORJLFDO SDUDPHWHUV FKDUDFWHULVWLF RI KLSSRFDPSDO QHXURQV LQ FXOWXUH /HH HW D f 7KHVH SURSHUWLHV DUH GHVFULEHG LQ 7DEOH +1 FHOOV KDYH EHHQ XVHG LQ RXU ODERUDWRU\ DV D PRGHO V\VWHP WR DGGUHVV WKH

PAGE 32

PHFKDQLVP RI DFWLRQ RI OLWKLXP DQG ZH SUHYLRXVO\ HVWDEOLVKHG WKH SUHVHQFH RI D IXQFWLRQDO VHFRQG PHVVHQJHU 3,f VLJQDOLQJ SDWKZD\ LQ WKHVH FHOOV :DWVRQ DQG /HQR[ f )XUWKHUPRUH +1 FHOOV KDYH UHFHQWO\ EHHQ DVVHVVHG DQG IRXQG WR H[SUHVV DW OHDVW WZR RI WKH QHXURILODPHQW WULSOHW SURWHLQV :DWWHUVRQ XQSXEOLVKHG GDWDf ZKLFK VXSSRUWV WKHLU FRQVLGHUDWLRQ DV SKHQRW\SLFDOO\ QHXURQDO 'HEXV HW D 6KDZ f &HOOV ZHUH JURZQ DW r& DQG b &2 LQ 'XOEHFFRnV PRGLILHG (DJOHnV PHGLXP '0(0f FRQWDLQLQJ b GHILQHG IHWDO ERYLQH VHUXP )%6f DQG PJPO JHQWDPLFLQ ZLWK RU ZLWKRXW WKH DGGLWLRQ RI P0 93$ 7KH GXUDWLRQ RI H[SRVXUHV UDQJHG IURP WR G 93$ ZDV DGGHG WR FXOWXUHV DW WLPH RI SODWLQJ DQG FXOWXUH PHGLXP ZDV FKDQJHG RU FHOOV ZHUH SDVVDJHG HYHU\ G ZLWK QR UHSOHQLVKPHQW RI 93$ EHWZHHQ SODWLQJV $VVD\V ZHUH SHUIRUPHG ZKHQ FHOOV UHDFKHG b FRQIOXHQFH 7DEOH OLVWV WKH SURSHUWLHV SUHVHQW f RU DEVHQW f LQ +1 FHOOV DQG WKH WZR SDUHQW FHOO OLQHV IURP ZKLFK WKH +1 ZHUH GHULYHG SULPDU\ PRXVH KLSSRFDPSDO QHXURQV Gf DQG 17* QHXUREODVWRPD FHOOV 7KLV WDEOH ZDV DGDSWHG IURP WKDW UHSRUWHG LQ /HH HW DO f LQ ZKLFK WKH +1 FHOO GHULYDWLRQ DQG FKDUDFWHULVWLFV ZHUH ILUVW UHSRUWHG 7DEOH 3URSHUWLHV RI ‘1 +\EULG DQG 3DUHQWDO &HOOV 7LVVXH RU +\EULG 1HXULOLF ([FLWDEOH 1*) 1*) &HOO /LQH *3, ,VR]\PH 3URFHVVHV 0HPEUDQH 1)0 3URWHLQ P51$ +LSSRFDPSXV 17* +1 3&, &HOO &XOWXUH 3& FHOOV REWDLQHG IURP $PHULFDQ 7\SH &XOWXUH &ROOHFWLRQ 5RFNYLOOH 0'f ZHUH JURZQ LQ 530, FRQWDLQLQJ b KHDWLQDFWLYDWHG KRUVH VHUXP b )%6 DQG P0 JOXWDPLQH &XOWXUHV ZHUH PDLQWDLQHG DW r& DQG b &2 DQG SDVVDJHG HYHU\ G 7R

PAGE 33

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f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b 7ULWRQ ; DQG VROXELOL]HG IRU PLQ 6ROXELOL]HG IUDFWLRQV ZHUH WKHQ FHQWULIXJHG DW [ J DQG WKH VXSHUQDWDQW FRQWDLQLQJ WKH VROXELOL]HG PHPEUDQH SURWHLQ FROOHFWHG 6DPSOHV ZHUH

PAGE 34

DGMXVWHG E\ DGGLWLRQ RU GLOXWLRQ WR b 7ULWRQ ; (TXDO DPRXQWV RI VROXEOH DQG PHPEUDQH FHOO SURWHLQ JJf DV GHWHUPLQHG E\ WKH %UDGIRUG PHWKRG ZHUH VHSDUDWHG E\ 6'63$*( DQG WUDQVIHUUHG WR SRO\YLQ\OLGHQH IOXRULGH PHPEUDQH 0LOOLSRUH &RUS %HGIRUG 0$f LQ D %LR5DG 7UDQV%ORW HOHFWURSKRUHVLV DSSDUDWXV DW 9 IRU K XVLQJ 7RZELQnV EXIIHU P0 7ULV S+ P0 JO\FLQH DQG b YYf PHWKDQROf $OO DQWLERGLHV ZHUH GLOXWHG LQ 76 EXIIHU P0 7ULV S+ DQG 0 1D&Of VHH 7DEOH IRU GLOXWLRQV DQG LQFXEDWLRQ WLPHV 'HWHFWLRQ RI WKH LPPXQH FRPSOH[ ZDV SHUIRUPHG XVLQJ +53FRQMXJDWHG DQWLUDEELW RU DQWLPRXVH ,J* DQG WKH 3LHUFH 5RFNIRUG ,/f HQKDQFHG FKHPLOXPLQHVFHQFH V\VWHP :HVWHUQ EORWV ZHUH TXDQWLILHG XVLQJ 1,+ ,PDJH VRIWZDUH IRU GHQVLWRPHWULF DQDO\VLV 5HVXOWV DUH H[SUHVVHG DV WKH SHUFHQWDJH RI SURWHLQ SUHVHQW LQ GUXJIUHH FRQWUROV JURZQ LQ SDUDOOHO DQG DVVD\HG RQ WKH VDPH :HVWHUQ EORW 5LERQXFOHDVH 51DVHf 3URWHFWLRQ $VVD\ 53$f 7KH 53$ ZDV SHUIRUPHG ZLWK WKH 53$,, NLW XVLQJ ELRWLQ\ODWHG DQWLVHQVH ULERSUREHV %,27,1VFULSW DQG %ULJKW6WDU V\VWHP $PELRQf %LRWLQ\ODWHG DQWLVHQVH ULERSUREHV IURP WKH 0DFV PRXVH JHQH FRGLQJ IRU 0$5&.6f DQG WKH F\FORSKLOLQ JHQHV ZHUH JHQHUDWHG DQG JHOSXULILHG $ IUDJPHQW RI 0DFV ZDV FORQHG E\ 573&5 LQWR D 7$ FORQLQJ YHFWRU ,Q9LWURJHQ &DUOVEDG &$f DQG UHVWULFWHG DW D XQLTXH ;PDO VLWH ,Q YLWUR WUDQVFULSWLRQ XVLQJ 63 51$ SRO\PHUDVH DQG WKLV FRQVWUXFW DV WKH WHPSODWH \LHOGV D QW ULERSUREH DQG SURWHFWV D ES IUDJPHQW RI 0$5&.6 P51$ 7KH F\FORSKLOLQ SUREH $PELRQf SURWHFWV D ES IUDJPHQW )RU HDFK VDPSOH QJ RI 51$ ZDV LQFXEDWHG ZLWK QJ RI HDFK SUREH DW r& RYHUQLJKW 1RQK\EULGL]HG 51$ ZDV GLJHVWHG ZLWK D PL[WXUH RI 51DVHV 3URWHFWHG IUDJPHQWV ZHUH VHSDUDWHG RQ D GHQDWXULQJ b 3$*( DQG WKHQ WUDQVIHUUHG RQWR D Q\ORQ PHPEUDQH WR DOORZ GHWHFWLRQ 5HVXOWLQJ DXWRUDGLRJUDSKV ZHUH WKHQ TXDQWLILHG DQG WKH DPRXQW RI 0DFV H[SUHVVLRQ ZDV QRUPDOL]HG WR WKH F\FORSKLOLQ

PAGE 35

LQWHUQDO FRQWURO 6KDUO\QQ 6ZHHQH\ LV DFNQRZOHGJHG IRU KHU H[SHUWLVH LQ GHYHORSPHQW RI WKH SUREHV DV ZHOO DV KHU WHFKQLFDO VNLOO LQ FDUU\LQJ RXW WKH 53$V DQG TXDQWLWDWLQJ UHVXOWV 3.& ,QKLELWRUV +1 FHOOV ZHUH H[SRVHG WR WKH 3.& LQKLELWRU /< DW 0 (OL /LOO\ DQG &RPSDQ\f FRQFRPLWDQWO\ ZLWK P0 93 $ IRU K $OWHUDWLRQV LQ 93$LQGXFHG 0$5&.6 GRZQUHJXODWLRQ DQG *$3 XSUHJXODWLRQ DV ZHOO DV FHOO GHQVLW\ DQG PRUSKRORJ\ ZHUH PHDVXUHG DQG FRPSDUHG EHWZHHQ FRQWURO JURXSV DQG /< WUHDWHG JURXSV HDFK LQ WKH SUHVHQFH RU DEVHQFH RI 93$ 3& FHOOV ZHUH H[SRVHG WR WKH 3.& LQKLELWRU ELVLQGRO\OPDOHLPLGH %,0 *LEFR *DLWKHUVEXUJ 0'f DW Q0 IRU K DQG DOWHUDWLRQV LQ FHOO GHQVLW\ DQG PRUSKRORJ\ ZHUH TXDQWLWDWHG DQG FRPSDUHG EHWZHHQ WUHDWPHQW JURXSV 3.& $FWLYLW\ 3.& DFWLYLW\ ZDV PHDVXUHG E\ WKH SKRVSKRU\ODWLRQ RI P\HOLQ EDVLF SURWHLQ 0%3f SHSWLGH 8SVWDWH %LRWHFKQRORJ\ /DNH 3ODFLG 1\3@$73 ZLWK RU ZLWKRXW _LJPO SKRVSKDWLG\OVHULQH DQG JJPO 30$ 7KH PL[WXUH ZDV WKHQ LQFXEDWHG DW r & IRU PLQ 7KH UHDFWLRQ ZDV VWRSSHG E\ WKH DGGLWLRQ RI QO RI LFHFROG P0 SKRVSKRULF DFLG )ROORZLQJ D PLQ LQFXEDWLRQ RQ LFH _G RI HDFK VDPSOH ZDV VSRWWHG RQWR :KDWPDQ 3 ILOWHU SDSHU )LOWHUV ZHUH ZDVKHG [f ZLWK D PL[WXUH RI P0 +32 DQG P0 VRGLXP S\URSKRVSKDWH ULQVHG LQ HWKDQRO DQG WKHQ DFHWRQH WKHQ GULHG DQG UDGLRDFWLYLW\ PHDVXUHG 6SHFLILF DFWLYLW\ IRU

PAGE 36

3.& ZDV PHDVXUHG DV QDQRPROHV RIf3PLQPJ SURWHLQ DQG H[SUHVVHG DV WKH GLIIHUHQFH EHWZHHQ DFWLYLW\ LQ WKH SUHVHQFH DQG DEVHQFH RI 30$SKRVSKDWLG\OVHULQH 'U 'DYLG :DWVRQ LV DFNQRZOHGJHG IRU KLV FRQWULEXWLRQ WR WKH GHYHORSPHQW RI WKH 3.& DFWLYLW\ DVVD\ DQG KLV DVVLVWDQFH LQ FDUU\LQJ RXW WKH H[SHULPHQWV 6WDWLVWLFDO $QDO\VHV )ROORZLQJ FRQVXOWDWLRQ ZLWK D ELRVWDWLVWLFLDQ VWDWLVWLFDO GLIIHUHQFHV EHWZHHQ WUHDWPHQWV ZHUH DQDO\]HG E\ RQHZD\ $129$ DQG SRVW KRF FRPSDULVRQV ZHUH FRQGXFWHG XVLQJ )LVKHUfV 3/6' WHVW ,Q DOO FDVHV PLQLPDO DFFHSWDEOH OHYHO IRU VLJQLILFDQFH ZDV S $OO UHVXOWV DUH UHSRUWHG DV PHDQ SHUFHQWDJH RI DQ XQWUHDWHG FRQWURO JURXS DVVD\HG LQ SDUDOOHO DQG HUURU EDUV UHSUHVHQW 6(0 'UV 5RQDOG 0DUNV 5REHUW 0F1DPDUD DQG 0DUN /HZLV DUH DFNQRZOHGJHG IRU WKHLU DVVLVWDQFH LQ LGHQWLI\LQJ YDOLG VWDWLVWLFDO PHWKRGV IRU DQDO\]LQJ WKHVH GDWD

PAGE 37

7DEOH $QWLERGLHV IRU :HVWHUQ %ORWV 7KLV WDEOH OLVWV WKH SULPDU\ DQWLERGLHV XWLOL]HG LQ ZHVWHUQ LPPXQREORWV WR GHWHFW HDFK RI WKH SURWHLQV RI LQWHUHVW )RU HDFK DQWLERG\ WKH IROORZLQJ LQIRUPDWLRQ LV UHSRUWHG VRXUFH IURP ZKLFK DQWLERG\ ZDV REWDLQHG GLOXWLRQ LQ WULVVDOLQH RU 76 EXIIHUf DW ZKLFK PHPEUDQH ZDV H[SRVHG WR DQWLERG\ OHQJWK RI H[SRVXUH RI PHPEUDQH WR DQWLERG\ DQG VHFRQGDU\ +53FRQMXJDWHG DQWLERG\ XVHG IRU DPSOLILFDWLRQ DQG GHWHFWLRQ RI SULPDU\ DQWLERG\ LWV VRXUFH GLOXWLRQ LQ 76 EXIIHU b 7ZHHQf DQG H[SRVXUH WLPH 7KH DQWL0$5&.6 SRO\FORQDO SULPDU\ DQWLERG\ ZDV UDLVHG DJDLQVW D JOXWDWKLRQH6WUDQVIHUDVH 0$5&.6 IXVLRQ SURWHLQ GHVLJQHG LQ RXU ODERUDWRU\ DV GHWDLOHG LQ :DWVRQ DQG /HQR[ f r $E 6RXUFH 'LOXWLRQ ,QFXEDWLRQ r $E 6RXUFH 'LOXWLRQ ,QFXEDWLRQ 0$5&.6 FXVWRP PDGHf RYHUQLJKW DQWLUDEELW %LR5DG K *$3 %RHKULQJHU 0DQQKHLP RYHUQLJKW DQWLPRXVH %LR5DG K 3.&D 7UDQVGXFWLRQ /DEV K DQWLPRXVH %LR5DG K 3.& 6DQWD &UX] K DQWLUDEELW %LR5DG K 3.&H 6DQWD &UX] K DQWLUDEELW %LR5DG K 3.& 7UDQVGXFWLRQ /DEV K DQWLPRXVH %LR5DG K

PAGE 38

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f XQGHUO\LQJ WKH PRRGVWDELOL]LQJ DFWLYLW\ RI ERWK GUXJV LV DQ LPSRUWDQW DUHD RI VWXG\ 0$5&.6 6WXGLHV FRQGXFWHG LQ RXU ODERUDWRU\ XVLQJ ERWK DQLPDO DQG FHOO PRGHOV KDYH SURYLGHG HYLGHQFH IRU D VLJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 IROORZLQJ FKURQLF EXW QRW DFXWH H[SRVXUH WR WKHUDSHXWLFDOO\ UHOHYDQW FRQFHQWUDWLRQV RI OLWKLXP /HQR[ HO D :DWVRQ DQG /HQR[ f ,Q LPPRUWDOL]HG KLSSRFDPSDO FHOOV :DWVRQ DQG /HQR[ f ILUVW GHPRQVWUDWHG WKDW ORQJWHUP Gf H[SRVXUH WR P0 /L&O UHVXOWHG LQ D UHYHUVLEOH WLPH DQG FRQFHQWUDWLRQGHSHQGHQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ

PAGE 39

6LPLODUO\ LQ YLYR VWXGLHV GHPRQVWUDWHG WKDW H[SRVXUH RI UDWV WR FKURQLF OLWKLXP ZHHNVf UHVXOWLQJ LQ FOLQLFDOO\ UHOHYDQW VHUXP FRQFHQWUDWLRQV RI P(TNJ LQ EUDLQ DOVR UHGXFHG 0$5&.6 LQ WKH KLSSRFDPSXV /HQR[ HO DO f $GGLWLRQDOO\ UHFHQW GDWD IURP RXU ODERUDWRU\ VXJJHVW D WLPH FRXUVH RI 0$5&.6 P51$ UHJXODWLRQ E\ OLWKLXP LQ +1 FHOOV VLPLODU WR WKH WLPH FRXUVH RI 0$5&.6 SURWHLQ GRZQUHJXODWLRQ ZLWK UHGXFWLRQV HYLGHQW E\ G DQG SHUVLVWLQJ EH\RQG Gf WKHUHE\ LQGLFDWLQJ DOWHUDWLRQV LQ HLWKHU WUDQVFULSWLRQ RU SRVWWUDQVFULSWLRQDO PHVVDJH VWDELOLW\ :DWVRQ DQG /HQR[ f 0$5&.6 H[SUHVVLRQ LV ZLGHO\ GLVWULEXWHG ZLWK KLJKHVW OHYHOV QRUPDOO\ H[SUHVVHG LQ EUDLQ VSOHHQ WHVWLV DQG OXQJ DQG ORZHU H[SUHVVLRQ LQ KHDUW NLGQH\ OLYHU DQG PXVFOH WLVVXHV RI PRXVH /REDFK HO DO 6ZLHUF]\QVNL HO DO f ,Q WKH EUDLQ WKH JUHDWHVW H[SUHVVLRQ LV LQ WKH PRVW SODVWLF UHJLRQV LQFOXGLQJ WKH DP\JGDOD VWULDWXP ROIDFWRU\ EXOE DQG FRUWH[ DQG VHSWXP 2XLPHW HO DO 0F1DPDUD DQG /HQR[ f DQG 0$5&.6 LV ORFDOL]HG WR VPDOO GHQGULWLF EUDQFKHV DQG D[RQ WHUPLQDOV RI QHXURQV DQG ERWK F\WRSODVPLF DQG PHPEUDQRXV FRPSRQHQWV RI JOLD .DW] HO DO 3DWHO DQG .OLJPDQ 2XLPHW HO DO 5RVHQ HO DO f 0$5&.6 FURVVOLQNV DFWLQ DQG ELQGV FDOPRGXOLQ LQ D &DAGHSHQGHQW PDQQHU WKHVH HYHQWV DUH PXWXDOO\ H[FOXVLYH 6KHX HO DO $OOHQ DQG $GHUHP f DQG DV D UHVXOW KDV EHHQ LPSOLFDWHG LQ FHOO PRWLOLW\ VHFUHWLRQ DQG WUDIILFNLQJ DQG FHOO F\FOH UHJXODWLRQ IRU UHYLHZV VHH $GHUHP DQG %ODFNVKHDU f $V LV LQGLFDWHG E\ LWV QDPH WKH DFWLYLW\ DQG H[SUHVVLRQ RI 0$5&.6 DUH UHJXODWHG LQ SDUW WKURXJK SKRVSKRU\ODWLRQ E\ 3.& WKRXJK 3.&LQGHSHQGHQW SDWKZD\V PD\ DOVR SOD\ D UROH LQ LWV SRVWWUDQVFULSWLRQDO UHJXODWLRQ %URRNV HW DO f 7KH LQKLELWLRQ RI 3.& RU PXWDWLRQ RI WKH 3.& SKRVSKRU\ODWLRQ VLWHV RI 0$5&.6 UHVXOWV LQ GHIHFWV LQ FHOOXODU PRWLOLW\ $OOHQ DQG $GHUHP f 0$5&.6 LV D SURWHLQ IRU ZKLFK WKH SUHFLVH IXQFWLRQ LV XQFOHDU EXW ZKLFK LV EHOLHYHG WR EH HVVHQWLDO WR QRUPDO &16 GHYHORSPHQW VLQFH PLFH ODFNLQJ GHWHFWDEOH P51$ DQG SURWHLQ GXH WR JHQH 0DFVf NQRFNRXW GLHG SHULQDWDOO\ HLWKHU EHIRUH RU ZLWKLQ MXVW KRXUV DIWHU ELUWK 6WXPSR HW D f 'HYHORSPHQWDO GHIHFWV LQFOXGHG GHIHFWV LQ PLGOLQH

PAGE 40

HYHQWV VXFK DV QHXUXODWLRQ IXVLRQ RI WKH FHUHEUDO KHPLVSKHUHV IRUPDWLRQ RI WKH JUHDW IRUHEUDLQ FRPPLVVXUHV DQG UHWLQDO DQG FRUWLFDO ODPLQDWLRQ 6WXPSR HO DO f 8OWUDVWUXFWXUDO H[DPLQDWLRQ UHYHDOHG ZLGHVSUHDG QHXURQDO HFWRSLD RU DEQRUPDO QHXURQDO SDWKILQGLQJ GXULQJ FHUHEURFRUWLFDO GHYHORSPHQW %ODFNVKHDU HO DO f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f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

PAGE 41

SURILOH RI OLWKLXP 6LPLODU UHVXOWV DUH K\SRWKHVL]HG IROORZLQJ 93$ H[SRVXUH VLQFH WKLV GUXJ H[KLELWV VLPLODU FOLQLFDO SURSHUWLHV RI GHOD\HG RQVHW RI DQWLPDQLF GUXJ DFWLRQ DQG GHOD\HG UHWXUQ WR EDVHOLQH IROORZLQJ GUXJ UHPRYDO RYHU D WKHUDSHXWLF FRQFHQWUDWLRQ UDQJH QJPOf 7KH ORQJWHUP HIIHFWV RI 93$ ZLOO EH WKH IRFXV RI WKHVH VWXGLHV IRU WKH\ PD\ EH LPSRUWDQW WR WKH PRRGVWDELOL]LQJ HIILFDF\ RI WKLV GUXJ ZKHUHDV HIIHFWV REVHUYHG PRUH DFXWHO\ DUH JHQHUDOO\ FUHGLWHG WR WKH DQWLFRQYXOVDQW SURSHUWLHV RI 93$ $V PHQWLRQHG LQ &KDSWHU DFFXPXODWLQJ HYLGHQFH LPSOLFDWHV UHFHSWRUPHGLDWHG 3, VLJQDOLQJ LQ WKH PHFKDQLVP RI DFWLRQ RI OLWKLXP LQ WKH EUDLQ %HUULGJH HO D *RGIUH\ /HQR[ DQG :DWVRQ f /LWKLXP LV DQ XQFRPSHWLWLYH LQKLELWRU RI WKH HQ]\PH ,03DVH DQG DV VXFK LW OHDGV WR DFFXPXODWLRQ RI LQWHUPHGLDWHV LQ WKH SDWKZD\ UHVSRQVLEOH IRU DGGLWLRQDO GRZQVWUHDP HIIHFWV LQFOXGLQJ LQWUDFHOOXODU &DA UHOHDVH DQG '$*PHGLDWHG 3.& DFWLYDWLRQ UHIHU WR )LJXUH 3RVW HO DO f ,W LV K\SRWKHVL]HG WKDW GHSOHWLRQ RI IUHH LQRVLWRO UHVXOWLQJ IURP WKH DFFXPXODWLRQ RI LQWHUPHGLDWHV GXH WR LQFRPSOHWH F\FOLQJ RI WKH 3, VLJQDOLQJ SDWKZD\ LV D PDMRU FRQVHTXHQFH RI ORQJWHUP OLWKLXP H[SRVXUH %HUULGJH HO D %HUULGJH f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f )LUVWO\ LQ VWXGLHV RI ),1 FHOOV H[SRVHG WR OLWKLXP LQRVLWRO VXSSOHPHQWDWLRQ DW FRQFHQWUDWLRQV DV ORZ DV Q0f ZDV DEOH WR SUHYHQW WKH GRZQUHJXODWLRQ RI 0$5&.6 LQGXFHG E\ P0 OLWKLXP WKHUHE\ VXSSRUWLQJ WKH LQRVLWROGHSOHWLRQ K\SRWKHVLV 6HFRQGO\ DGGLWLRQ RI WKH PXVFDULQLF UHFHSWRU DJRQLVW FDUEDFKRO P0f LQFUHDVHG WKH 0$5&.6 GRZQUHJXODWLRQ REVHUYHG :DWVRQ DQG /HQR[ f ZKLFK VXSSRUWV WKH QRWLRQ WKDW WKH DFWLRQ RI OLWKLXP LV DXJPHQWHG LQ FHOOV RU WLVVXH UHJLRQV XQGHUJRLQJ WKH KLJKHVW UDWH RI UHFHSWRU DFWLYDWLRQ

PAGE 42

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f ,Q DQRWKHU VWXG\ P0 YDOSURDWH IDLOHG WR LQFUHDVH DFFXPXODWLRQ RI LQRVLWRO SKRVSKDWHV LQFOXGLQJ WKH LQRVLWRO PRQRSKRVSKDWHV LQRVLWRO ELVSKRVSKDWH LQRVLWRO ELVSKRVSKDWH DQG LQRVLWRO WULVSKRVSKDWH ZKHUHDV OLWKLXP H[SRVXUH UHVXOWHG LQ DFFXPXODWLRQ RI DOO 'L[RQ DQG +RNLQ f ,Q D VWXG\ FRQGXFWHG LQ RXU ODERUDWRU\ 93$ IDLOHG WR UHVXOW LQ DFFXPXODWLRQ RI &033$ DOVR NQRZQ DV &'3'$* VHH )LJXUH f D PHWDEROLWH RI '$* DQG D VHQVLWLYH PHDVXUH RI 3, VLJQDOLQJ ZKHUHDV OLWKLXP UHVXOWHG LQ VLJQLILFDQW DFFXPXODWLRQ RI &033$ LQ &+2. DQG +1 FHOOV :DWVRQ HW D f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

PAGE 43

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f WLVVXHVSHFLILF H[SUHVVLRQ f GHYHORSPHQWDOO\ UHJXODWHG H[SUHVVLRQ f GLIIHUHQWLDOO\ UHJXODWHG WUDQVFULSWLRQ UDWHV DQG f DOWHUDWLRQV LQ P51$ OHYHOV LQ UHVSRQVH WR 3.& %ODFNVKHDU /REDFK HODO f 93$LQGXFHG HIIHFWV RQ 3.& LWVHOI ZLOO EH GLVFXVVHG LQ GHWDLO LQ &KDSWHU $FFXPXODWLQJ HYLGHQFH VKRZV WKDW OLWKLXP DQG 93$ KDYH UHJXODWRU\ HIIHFWV RQ YDULRXV WUDQVFULSWLRQ IDFWRUV WKHUHE\ VXJJHVWLQJ D UROH IRU WKHVH PRRG VWDELOL]HUV LQ DOWHULQJ JHQH H[SUHVVLRQ 6WXGLHV LQ WUDQVIHFWHG FXOWXUHG FHOOV KDYH LQGLFDWHG WKDW 93$ HQKDQFHV H[SUHVVLRQ RI JHQHV DQG UHSRUWHUV ZKLFK FRQWDLQ $3 HOHPHQWV LQ WKHLU UHJXODWRU\ GRPDLQV 6LPRQ HO DO .XQW]6LPRQ HO DO f 0DQML DQG FROOHDJXHV VKRZHG IXUWKHU WKDW 93$ H[SRVXUH UHVXOWHG LQ D WLPH DQG FRQFHQWUDWLRQGHSHQGHQW LQFUHDVH LQ '1$ ELQGLQJ DFWLYLW\ RI WKH WUDQVFULSWLRQ IDFWRU $3 LQ ERWK UDW & JOLRPD DQG KXPDQ QHXUREODVWRPD 6+6<
PAGE 44

WUDQVFULSWLRQDO PHFKDQLVPV VR WKDW FKDQJHV LQ P51$ VWDELOLW\ SOD\ D FUXFLDO UROH LQ WKH FRQWURO RI *$3 H[SUHVVLRQ IRU H[DPSOH 6LPLODU SRVWWUDQVFULSWLRQDO PHFKDQLVPV PLJKW DOVR EH DW ZRUN LQ WKH FDVH RI 0$5&.6 DQG LW LV WKH SXUSRVH RI WKHVH H[SHULPHQWV WR DVVHVV IXUWKHU WKLV SRVVLELOLW\ 3UHOLPLQDU\ VWXGLHV LQ RXU ODERUDWRU\ KDYH LQGLFDWHG WKDW IROORZLQJ OLWKLXP H[SRVXUH 0$5&.6 P51$ OHYHOV GR GHFOLQH DORQJ D WLPH FRXUVH VLPLODU WR WKDW RI WKH SURWHLQ GRZQUHJXODWLRQ :DWVRQ DQG /HQR[ f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f DQG PD\ SOD\ D UROH LQ H[RF\WRVLV 'HNNHU HW DW f *$3 OLNH 0$5&.6 DSSHDUV WR EH LQWHJUDO WR QRUPDO QHUYRXV V\VWHP GHYHORSPHQW EHFDXVH VWXGLHV RI PLFH ODFNLQJ WKH SURWHLQ DV D UHVXOW RI DQWLVHQVHPHGLDWHG JHQH NQRFNRXW VKRZ D KLJK IUHTXHQF\ RI HDUO\ SRVWQDWDO GHDWK ZLWK IHZ DQLPDOV VXUYLYLQJ SDVW WKUHH ZHHNV RI DJH DQG HYHQ IHZHU UHDFKLQJ PDWXULW\ 6WULWWPDWWHU HO DO f 7KRVH PLFH WKDW GR VXUYLYH H[KLELW DEQRUPDO QHXURQDO SDWKILQGLQJ *$3 LV D UDSLGO\ WUDQVSRUWHG SUHV\QDSWLF SURWHLQ PRVW KLJKO\ H[SUHVVHG GXULQJ GHYHORSPHQW DQG SRVWLQMXU\ UHJHQHUDWLRQ DQG LPPXQRKLVWRFKHPLVWU\ DQG LQ VLWX K\EULGL]DWLRQ VWXGLHV KDYH VKRZQ *$3 H[SUHVVLRQ WR EH DOPRVW H[FOXVLYHO\ QHXURQDO %HQRZLW] DQG 5RXWWHQEHUJ f 1HXULWH RXWJURZWK LV JHQHUDOO\ DFFRPSDQLHG E\ D ODUJH LQFUHDVH LQ *$3 H[SUHVVLRQ DQG IROORZLQJ GLIIHUHQWLDWLRQ DQG V\QDSWRJHQHVLV *$3 OHYHOV GHFOLQH VKDUSO\ DQG UHPDLQ ORZ H[FHSW LQ WKH PRVW SODVWLF UHJLRQV RI WKH

PAGE 45

EUDLQ VXFK DV DVVRFLDWLYH DQG OLPELF DUHDV ZKHUH LQWHQVH LPPXQRVWDLQLQJ SHUVLVWV LQ DGXOWV WKURXJKRXW OLIH 7KLV VHOHFWLYH H[SUHVVLRQ LV WKRXJKW WR FRQIHU WKH XQLTXH SRWHQWLDO IRU RQJRLQJ VWUXFWXUDO UHPRGHOLQJ %HQRZLW] DQG 3HUURQH%L]]R]HUR f *$3 LV DVVRFLDWHG ZLWK WKH F\WRVNHOHWRQ DQG LV FRQFHQWUDWHG LQ WKH JURZWK FRQHV RI HORQJDWLQJ D[RQV DQG QHUYH WHUPLQDOV WKRXJKW WR KHOS VWDELOL]H JURZWK FRQH PRUSKRORJ\ DQG SURPRWH JURZWK FRQH DGKHVLRQ $LJQHU DQG &DURQL f 6XSSRUW IRU WKLV QRWLRQ KDV EHHQ SURYLGHG E\ D QXPEHU RI VWXGLHV $QWLVHQVHGLUHFWHG GHSOHWLRQ RI *$3 LQ JURZWK FRQHV UHVXOWV LQ WKH DEVHQFH RI VSUHDGLQJ EUDQFKLQJ DQG DGKHVLRQ RI JURZWK FRQHV DQG WKHUHIRUH QHXULWH RXWJURZWKf LQ SULPDU\ VHQVRU\ QHXURQV $LJQHU DQG &DURQL f :KHUHDV VXSSUHVVLRQ RI *$3 H[SUHVVLRQ KDV EHHQ IRXQG WR \LHOG DGYHUVH HIIHFWV RQ D[RQ RXWJURZWK ERWK LQ YLYR DQG LQ YLWUR 0HLUL HW D %HQRZLW] DQG 5RXWWHQEHUJ 6WULWWPDWWHU HW DO -DS 7MRHQ 6DQ HW D 6KHD HW DO f RYHUH[SUHVVLRQ RI WKH SURWHLQ LQGXFHG QHUYH VSURXWLQJ LQ WKH DGXOW QHUYRXV V\VWHP RI WUDQVJHQLF PLFH $LJQHU HW D f (YHQ LQ QRQQHXURQDO FHOOV H[SUHVVLRQ RI D *$3 WUDQVJHQH KDV EHHQ IRXQG WR LQGXFH H[WHQVLYH SURFHVV RXWJURZWK DFFRPSDQLHG E\ UHRUJDQL]DWLRQ RI WKH PHPEUDQH F\WRVNHOHWRQ 6WULWWPDWWHU HW DO
PAGE 46

ZHOO DV WR UHJHQHUDWH D[RQV DIWHU LQMXU\ +RZHYHU WKH SUHVHQFH RI *$3 LV SUREDEO\ QRW QHFHVVDU\ WR QHXULWH RXWJURZWK SHU VH EXW UDWKHU SOD\V DQ LPSRUWDQW UROH LQ UHJXODWLQJ SUHFLVHO\ WKH YDULRXV SDUDPHWHUV LQYROYHG LQ WKH RYHUDOO HYHQW VXFK DV QDYLJDWLRQ RU SDWKILQGLQJ $LJQHU DQG &DURQL f $FFRUGLQJ WR WKLV f*$3 K\SRWKHVLVf QHUYH JURZWK LV FRQWUROOHG LQ SDUW E\ WKH H[SUHVVLRQ RI FHUWDLQ JURZWK DVVRFLDWHG SURWHLQV VXFK DV *$3 ZKLFK SURYLGH D PHFKDQLVP IRU UHJXODWLQJ QHXULWH RXWJURZWK DQG VWUXFWXUDO SODVWLFLW\ LQ WKH QHUYRXV V\VWHP 6NHQH DQG :LOODUG f %HFDXVH RI WKH YLWDO UROH RI *$3 LQ HIIHFWLQJ SODVWLF FKDQJHV LQ EUDLQ H[SHULPHQWV KDYH EHHQ GHVLJQHG WR DGGUHVV WKH HILIHFWVf RI 93 $ H[SRVXUH RQ *$3 H[SUHVVLRQ LQ +1 FHOOV ,W PD\ EH WKDW SDUW RI WKH HIILFDF\ RI 93 $ LQ HLWKHU ELSRODU GLVRUGHU RU HSLOHSV\ RU ERWK OLHV LQ LWV DELOLW\ WR HOLFLW VWUXFWXUDO PRGLILFDWLRQV RI SODVWLF UHJLRQV H[KLELWLQJ WKH SDWKRORJ\ ZKLFK OHDGV WR WKH PDQLIHVWDWLRQ RI WKH SDUWLFXODU GLVRUGHU ,QGHHG %HQRZLW] DQG 5RXWWHQEHUJ f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f 6LQFH QHXULWH RXWJURZWK DQG QHXURQDO GLIIHUHQWLDWLRQ DUH NQRZQ WR EH DVVRFLDWHG ZLWK XSUHJXODWLRQ RI *$3 SURWHLQ H[SUHVVLRQ LW ZLOO EH RI LQWHUHVW WR LQYHVWLJDWH WKH HIIHFW RI 93$ RQ WKLV SURWHLQ 7KH UHVXOWV RI WKHVH VWXGLHV ZLOO EH LQVWUXFWLYH IRU LQ DGGLWLRQ WR WKH GHYHORSPHQW RI D PRGHO ZKLFK PD\ KHOS H[SODLQ WKH UROH RI 0$5&.6 DQG RWKHU SODVWLFLW\ SURWHLQV VXFK DV *$3f LQ WKH PHFKDQLVP RI DFWLRQ RI ERWK 93$ DQG OLWKLXP ILQGLQJV ZLOO FRQWULEXWH WR D EHWWHU XQGHUVWDQGLQJ RI WKH SDWKRSK\VLRORJ\ LQYROYHG LQ ELSRODU GLVRUGHU

PAGE 47

7KH QRWLRQ RI V\QDSWLF UHPRGHOLQJ DV WKH XQGHUO\LQJ PHDQV IRU WKH GUXJ RU WKHUDS\ LQGXFHG VWDELOL]DWLRQ RI PRRG GLVRUGHUV KDV EHHQ VXSSRUWHG E\ VRPH UHSRUWV 2QH VWXG\ IRXQG LQFUHDVHG OHYHOV RI QHXUDO FHOO DGKHVLRQ PROHFXOH 1&$0fLPPXQRUHDFWLYH SURWHLQV LQ WKH FHUHEURVSLQDO IOXLG RI D SRSXODWLRQ RI ELSRODU SDWLHQWV 3ROWRUDN HW DK f $EQRUPDO IXQFWLRQ RU WXUQRYHU RI 1&$0 ZKLFK SOD\V D UROH LQ V\QDSWRJHQHVLV PRUSKRJHQHVLV DQG SODVWLFLW\ RI WKH QHUYRXV V\VWHP PD\ EH VXJJHVWLYH RI DEQRUPDO V\QDSWLF SODVWLFLW\ LQ DIIHFWHG LQGLYLGXDOV :LWK UHJDUG WR GUXJ WKHUDS\ IRU WKLV GLVRUGHU QRW RQO\ GR OLWKLXP DQG 93$ DIIHFW H[SUHVVLRQ RI SURWHLQV LQYROYHG LQ QHXUDO SODVWLFLW\ DQG FHOOXODU UHVSRQVH DV SUHYLRXVO\ GLVFXVVHGf EXW ERWK GUXJV KDYH DOVR EHHQ VKRZQ WR KDYH HIIHFWV RQ WUDQVFULSWLRQDO IDFWRUV ZKLFK UHJXODWH JHQH H[SUHVVLRQ DQG LW LV NQRZQ WKDW ORQJWHUP FKDQJHV LQ QHXURQDO V\QDSWLF IXQFWLRQ DUH GHSHQGHQW XSRQ WKH VHOHFWLYH UHJXODWLRQ RI JHQH H[SUHVVLRQ 0DQML DQG /HQR[ f /DPRWULJLQH DQRWKHU GUXJ XVHG LQ WUHDWPHQW RI ELSRODU GLVRUGHU ZDV UHFHQWO\ VKRZQ WR DOWHU QHXURQDO K\SHUH[FLWDELOLW\ YLD LQKLELWLRQ RI ERWK 1D DQG &D FKDQQHOV ILQGLQJV ZKLFK ZHUH LQWHUSUHWHG E\ WKH DXWKRUV WR VLJQLI\ PRGLILFDWLRQ RI V\QDSWLF SODVWLFLW\ ;LH DQG +DJDQ f 6XFK HYLGHQFH LV LQGLUHFW EXW RYHUDOO VXJJHVWLYH WKDW FKDQJHV LQ V\QDSWLF SODVWLFLW\ PD\ EH D PHFKDQLVP E\ ZKLFK DQWLELSRODU DQWLPDQLF DQG PRRGVWDELOL]LQJf WKHUDS\ IDFLOLWDWHV ORQJWHUP FKDQJHV LQ DIIHFW ƒ

PAGE 48

5HVXOWV $ OL )LJXUH 0$5&.6 SURWHLQ 6KRZQ DUH :HVWHUQ LPPXQREORWV FRPSDULQJ 0$5&.6 SURWHLQ H[SUHVVLRQ LQ f PRXVH EUDLQ DQG f +1 FHOOV $ VKRZV F\WRVROLF 0$5&.6 H[SUHVVLRQ DQG % VKRZV PHPEUDQH DVVRFLDWHG 0$5&.6 &RPSDUDWLYH 0$5&.6 H[SUHVVLRQ LQ +1 FHOOV XVHG LQ WKLV VWXG\f YV PRXVH EUDLQ LV VKRZQ LQ )LJXUH IROORZLQJ HQKDQFHG FKHPLOXPLQHVFHQW (&/f GHWHFWLRQ 'LIIHUHQFHV LQ DSSDUHQW PROHFXODU ZHLJKW PD\ UHVXOW IURP GLIIHUHQWLDO SKRVSKRU\ODWLRQ RU P\ULVWR\ODWLRQ RI WKH SURWHLQ 1RWH WKH PRUH SURQRXQFHG 0$5&.6 SURWHLQ H[SUHVVLRQ LQ PRXVH EUDLQ HYHQ ZLWK FRPSDUDEOH DPRXQWV RI WRWDO ORDGHG SURWHLQ (IIHFW RI 93$ ([SRVXUH RQ 0$5&.6 3URWHLQ LQ +1 &HOOV 93$ ZDV DGPLQLVWHUHG DW FRQFHQWUDWLRQV UDQJLQJ IURP WR P0 IRU RU G )LJXUHV DQG UHVSHFWLYHO\f 5HSUHVHQWDWLYH :HVWHUQ EORWV DUH SUHVHQWHG DORQJ ZLWK TXDQWLWDWLRQ RI PHDQ GDWD 6LJQLILFDQW FRQFHQWUDWLRQ DQG WLPHGHSHQGHQW HIIHFWV ZHUH REVHUYHG &KURQLF Gf 93$ H[SRVXUH UHVXOWV LQ D FRQFHQWUDWLRQGHSHQGHQW UHGXFWLRQ LQ 0$5&.6 OHYHOV ZLWK D PRUH SURQRXQFHG UHGXFWLRQ LQ PHPEUDQH WKDQ

PAGE 49

VROXEOH IUDFWLRQ RI WKHVH FHOOV 6LJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 LV REVHUYHG LQ WKH PHPEUDQH IUDFWLRQ DW FRQFHQWUDWLRQV DV ORZ DV P0 Sf DQG LQ WKH VROXEOH IUDFWLRQ DW P0 DQG KLJKHU FRQFHQWUDWLRQV Sf 7KHVH FRQFHQWUDWLRQV FRUUHODWH ZHOO ZLWK WKH WDUJHW WKHUDSHXWLF VHUXP FRQFHQWUDWLRQ UDQJH XVHG FOLQLFDOO\ IRU WUHDWPHQW RI ELSRODU GLVRUGHU D _93$_ P0 )LJXUH &RQFHQWUDWLRQGHSHQGHQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ LQ +1 FHOOV IROORZLQJ G 93$ H[SRVXUH Df 5HSUHVHQWDWLYH :HVWHUQ EORW RI PHPEUDQHDVVRFLDWHG 0$5&.6 H[SUHVVLRQ IROORZLQJ G 93$ H[SRVXUH PHDQ GDWD DUH GHSLFWHG JUDSKLFDOO\ LQ E /DQHV ZHUH WUHDWHG IRU G DV IROORZV XQWUHDWHG FRQWURO P0 P0 P0 P0 P0 DQG P0 93$ UHVSHFWLYHO\ Ef 7KLV ILJXUH VKRZV TXDQWLWDWLYHO\ WKH 0$5&.6 SURWHLQ OHYHOV LQ +1 FHOOV IROORZLQJ G H[SRVXUH WR YDU\LQJ FRQFHQWUDWLRQV RI 93$ 'DWD H[SUHVVHG DV SHUFHQW RI 0$5&.6 FRPSDUHG WR XQWUHDWHG FRQWURO FHOOV JURZQ LQ SDUDOOHO ZHUH GHULYHG IURP :HVWHUQ LPPXQREORWV HJ )LJXUH Df DQG DUH WKH PHDQ RI DW OHDVW IRXU GHWHUPLQDWLRQV (UURU EDUV UHSUHVHQW 6(0

PAGE 50

E >93$f LX0 )LJXUH &RQFHQWUDWLRQGHSHQGHQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ LQ +1 FHOOV IROORZLQJ G 93$ H[SRVXUH Df 5HSUHVHQWDWLYH '$% :HVWHUQ EORW RI 0$5&.6 H[SUHVVLRQ IROORZLQJ G 93$ H[SRVXUH /DQHV VKRZ VROXEOH 0$5&.6 IROORZLQJ G WUHDWPHQW ZLWK DQG P0 93$ UHVSHFWLYHO\ DQG ODQHV VKRZ PHPEUDQH 0$5&.6 IROORZLQJ H[SRVXUH WR WKH VDPH GRVHV Ef 7KLV ILJXUH VKRZV TXDQWLWDWLYH GDWD IRU PHDQ 0$5&.6 OHYHOV IROORZLQJ G 93$ H[SRVXUH 'DWD H[SUHVVHG DV SHUFHQW RI 0$5&.6 FRPSDUHG WR XQWUHDWHG FRQWURO FHOOV JURZQ LQ SDUDOOHO ZHUH GHULYHG IURP :HVWHUQ LPPXQREORWV DQG DUH WKH PHDQ RI DW OHDVW IRXU GHWHUPLQDWLRQV (UURU EDUV UHSUHVHQW 6(0 rS

PAGE 51

7LPH &RXUVH RI 93$,QGXFHG 0$5&.6 'RZQ5HJXODWLRQ LQ +1 &HOOV +1 FHOOV ZHUH H[SRVHG FRQWLQXRXVO\ WR P0 93$ IRU SHULRGV UDQJLQJ IURP WR G 7KLV FRQFHQWUDWLRQ LV HTXLYDOHQW WR QJPO ZKLFK LV ZLWKLQ WKH UHSRUWHG WKHUDSHXWLF UDQJH IRU WKH PRRG VWDELOL]LQJ DFWLRQ RI 93$ QJPOf &HOOV ZHUH REVHUYHG WR XQGHUJR D VLJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ H[SUHVVLRQ LQ ERWK IUDFWLRQV DV HDUO\ DV G ZLWK QR HIIHFW RQ 0$5&.6 DW G VHH )LJXUH f 7KH 0$5&.6 UHGXFWLRQ LQ WKH PHPEUDQH IUDFWLRQ LV PRUH SURQRXQFHG WKDQ WKDW LQ WKH F\WRVRO 7KUHH G G DQG G YDOXHV DUH VLJQLILFDQWO\ GLIIHUHQW Sf IURP G 0$5&.6 YDOXHV LQ ERWK VROXEOH DQG PHPEUDQH IUDFWLRQV RI WKHVH FHOOV 7KHVH GDWD DUH FRQVLVWHQW ZLWK WKH FOLQLFDO SURILOH RI WKLV GUXJ ZKLFK H[KLELWV D GHOD\ LQ RQVHW RI DQWLPDQLF DFWLRQ DW WDUJHW WKHUDSHXWLF FRQFHQWUDWLRQV 7LPH 'D\Vf )LJXUH 7LPH FRXUVH RI 93$LQGXFHG 0$5&.6 GRZQUHJXODWLRQ LQ )71 FHOOV 7KLV ILJXUH LOOXVWUDWHV WKH WLPHGHSHQGHQF\ RI 93$LQGXFHG 0$5&.6 SURWHLQ GRZQUHJXODWLRQ +1 FHOOV ZHUH H[SRVHG WR P0 93$ D FOLQLFDOO\ UHOHYDQW WKHUDSHXWLF FRQFHQWUDWLRQ IRU RU G DQG 0$5&.6 H[SUHVVLRQ ZDV DVVHVVHG E\ TXDQWLWDWLYH LPPXQREORW DQDO\VLV 5HVXOWV H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWUROV UHSUHVHQW WKH PHDQ s 6(0 RI DW OHDVW IRXU GHWHUPLQDWLRQV rS

PAGE 52

E )LJXUH 5HFRYHU\ RI 0$5&.6 IROORZLQJ 93$LQGXFHG ORQJWHUP GRZQ UHJXODWLRQ Df 7KLV ILJXUH VKRZV WKH WLPH FRXUVH RI 93$LQGXFHG 0$5&.6 GRZQ UHJXODWLRQ LQ +1 FHOOV DV ZHOO DV WKH UHFRYHU\ RI 0$5&.6 SURWHLQ WR EDVDO OHYHOV IROORZLQJ GUXJ ZLWKGUDZDO +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU G DW ZKLFK WLPH WKH GUXJ ZDV HLWKHU UHPRYHG RU WUHDWPHQW FRQWLQXHG 0$5&.6 OHYHOV ZHUH DVVHVVHG WKURXJK :HVWHUQ LPPXQREORW )LJXUH E LV UHSUHVHQWDWLYHf DQG UHVXOWV DUH H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO FXOWXUHV JURZQ LQ SDUDOOHO 5HVXOWV UHSUHVHQW PHDQ s 6(0 RI DW OHDVW WKUHH GHWHUPLQDWLRQV rS Ef 6KRZQ LV D UHSUHVHQWDWLYH :HVWHUQ EORW RI 0$5&.6 H[SUHVVLRQ LQ WKH PHPEUDQH IUDFWLRQ RI +1 FHOOV IROORZLQJ FKURQLF 93$ H[SRVXUH DQG VXEVHTXHQW ZLWKGUDZDO /DQHV DQG VKRZ FRQWURO OHYHOV RI 0$5&.6 SURWHLQ /DQH VKRZV 0$5&.6 H[SUHVVLRQ IROORZLQJ G 93$ H[SRVXUH FRPSDUH WR /DQH f DQG /DQH VKRZV 0$5&.6 H[SUHVVLRQ IROORZLQJ G 93$ H[SRVXUH FRPSDUH WR /DQH f /DQH UHSUHVHQWV D 93$ ZLWKGUDZDOUHFRYHU\ JURXS DQG VKRZV 0$5&.6 LQ FHOOV H[SRVHG FRQWLQXRXVO\ IRU G WR P0 93$ DQG WKHQ IRU G WR PHGLD IUHH RI GUXJ 5HFRYHU\ RI 0$5&.6 IROORZLQJ 93$LQGXFHG /RQJ7HUP 'RZQ5HJXODWLRQ $V VKRZQ LQ )LJXUH D +1 FHOOV H[SRVHG WR P0 93$ IRU WR G H[KLELWHG D VLJQLILFDQW GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ H[SUHVVLRQ LQ ERWK IUDFWLRQV

PAGE 53

PHPEUDQH VROXEOHf RYHU WLPH DV SUHYLRXVO\ GHPRQVWUDWHG LQ )LJXUH $IWHU G RI FRQWLQXRXV 93$ H[SRVXUH DQG VXEVHTXHQW GUXJ ZDVKRXW DQG ZLWKGUDZDO 0$5&.6 SURWHLQ OHYHOV UHFRYHUHG ZLWK VLJQLILFDQW LQFUHDVH Sf WR FRQWURO OHYHOV ZLWKLQ G RI FXOWXUH LQ GUXJIUHH PHGLXP )LJXUH Df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nV HIIHFWV RQ 0$5&.6 FHOOV ZHUH H[SRVHG WR 93$ DORQH RU LQ FRPELQDWLRQ ZLWK HLWKHU FDUEDFKRO P0f RU Z\RLQRVLWRO XS WR P0f ,Q FRQWUDVW WR SUHYLRXV UHVXOWV REWDLQHG ZLWK /L&O :DWVRQ DQG /HQR[ f WKH 93$LQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 SURWHLQ H[SUHVVLRQ ZDV QRW DOWHUHG E\ DGGLWLRQ RI FDUEDFKRO 7DEOH DQG )LJXUH f WR WKH FXOWXUH PHGLXP 6LPLODUO\ VXSSOHPHQWDWLRQ RI Z\RLQRVLWRO 7DEOH DQG )LJXUH f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

PAGE 54

7DEOH $GGLWLRQ RI &DUEDFKRO )DLOV WR 3RWHQWLDWH WKH 93$,QGXFHG 0$5&.6 'RZQ 5HJXODWLRQ LQ +1 &HOOV +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU G RU P0 OLWKLXP IRU G LQ WKH SUHVHQFH RU DEVHQFH RI WKH PXVFDULQLF DJRQLVW FDUEDFKRO 5HVXOWV DUH SUHVHQWHG DV SHUFHQW RI XQWUHDWHG FRQWURO FXOWXUHV JURZQ LQ SDUDOOHO DQG UHSUHVHQW WKH PHDQ s 6(0 RI DW OHDVW WKUHH GHWHUPLQDWLRQV rS 0$5&.6 3URWHLQ b &RQWURO 0HDQ s 6(0f 93$([SRVHG /L&O([SRVHG >&DUEDFKRO@ 6ROXEOH 0HPEUDQH 6ROXEOH 0HPEUDQH P0 s s s s P0 s s srr s rr )LJXUH (IIHFW RI FDUEDFKRO RQ 93$LQGXFHG 0$5&.6 GRZQ UHJXODWLRQ 6KRZQ LV D UHSUHVHQWDWLYH :HVWHUQ EORW RI 0$5&.6 H[SUHVVLRQ IROORZLQJ 93$ H[SRVXUH ZLWK RU ZLWKRXW WKH FRQFRPLWDQW DGGLWLRQ RI FDUEDFKRO /DQH VKRZV PHPEUDQHDVVRFLDWHG 0$5&.6 OHYHOV LQ XQWUHDWHG FRQWURO +1 FHOOV /DQH VKRZV 0$5&.6 LQ G 93$WUHDWHG FHOOV /DQH VKRZV 0$5&.6 LQ FDUEDFKROWUHDWHG FHOOV DQG /DQH VKRZV 0$5&.6 LQ FHOOV WUHDWHG FRQFRPLWDQWO\ ZLWK 93$ DQG FDUEDFKRO 0HDQ GDWD DUH UHSRUWHG LQ 7DEOH

PAGE 55

7DEOH ,QRVLWRO 6XSSOHPHQWDWLRQ )DLOV WR $WWHQXDWH WKH 93$,QGXFHG 0$5&.6 'RZQ5HJXODWLRQ LQ +1 &HOOV +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU G RU P0 OLWKLXP P0 FDUEDFKROf IRU G LQ WKH SUHVHQFH RU DEVHQFHf RI YDU\LQJ FRQFHQWUDWLRQV RI P\RLQRVLWRO 5HVXOWV DUH SUHVHQWHG DV SHUFHQW RI XQWUHDWHG FRQWURO FXOWXUHV JURZQ LQ SDUDOOHO DQG UHSUHVHQW WKH PHDQ s 6(0 RI DW OHDVW WKUHH GHWHUPLQDWLRQV 1' QRW GHWHUPLQHGf rS 0$5&.6 3URWHLQ b &RQWURO 0HDQ s 6(0f 93$([SRVHG /L&OFDUEDFKRO([SRVHG >P\RLQRVLWRO@ 6ROXEOH 0HPEUDQH 6ROXEOH 0HPEUDQH s s s s S0 1 1' s s S0 1' 1' sr sr S0 s s sr sr P0 s s 1' 1' )LJXUH (IIHFW RI LQRVLWRO RQ 93$LQGXFHG 0$5&.6 GRZQ UHJXODWLRQ 6KRZQ LV D UHSUHVHQWDWLYH :HVWHUQ EORW RI 0$5&.6 H[SUHVVLRQ IROORZLQJ 93$ H[SRVXUH ZLWK RU ZLWKRXW FRQFRPLWDQW DGGLWLRQ RI Q\RLQRVLWRO /DQHV VKRZ PHPEUDQHDVVRFLDWHG 0$5&.6 LQ f FRQWURO FHOOV f G P0 93$WUHDWHG FHOOV f LQRVLWROWUHDWHG FHOOV DQG f FRPELQHG 93$ DQG LQRVLWRO WUHDWHG FHOOV 0HDQ GDWD DUH UHSRUWHG LQ 7DEOH

PAGE 56

93$ /L&O&DUEDFKRO 93$ /L&O&DUEDFKRO 7UHDWPHQW E $ % ‘SP )LJXUH $GGLWLYH HIIHFW RI FRPELQHG 93$ DQG OLWKLXP H[SRVXUH RQ 0$5&.6 SURWHLQ H[SUHVVLRQ Df +1 FHOOV ZHUH H[SRVHG WR P0 93$ DQGRU P0 OLWKLXP P0 FDUEDFKROf LQ LQRVLWROIUHH '0(0 IRU G 0$5&.6 SURWHLQ OHYHOV H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV VHH )LJXUH Ef DQG YDOXHV UHSUHVHQW PHDQ s 6 ( 0 RI DW OHDVW WKUHH GHWHUPLQDWLRQV rS Ef 5HSUHVHQWDWLYH :HVWHUQ EORW RI 0$5&.6 SURWHLQ IROORZLQJ G H[SRVXUH RI FHOOV WR 93$ DQGRU OLWKLXP $ VKRZV F\WRVROLF 0$5&.6 LQ +1 FHOOV DIWHU WKH IROORZLQJ WUHDWPHQWV QR GUXJ f P0 93$ f P0 93$ f P0 OLWKLXP f P0 OLWKLXP P0 FDUEDFKRO f P0 OLWKLXP P0 FDUEDFKRO P0 93$ f DQG P0 OLWKLXP P0 FDUEDFKRO P0 93$ f % VKRZV 0$5&.6 LQ WKH PHPEUDQH IUDFWLRQ RI +1 FHOOV IROORZLQJ WKH VDPH WUHDWPHQWV IRU G $GGLWLYH (IIHFW RI &RPELQHG 93$ DQG /L&, ([SRVXUH RQ 0$5&.6 3URWHLQ ([SUHVVLRQ LQ +1 &HOOV +1 FHOOV ZHUH H[SRVHG WR P0 93$ DQGRU P0 /L&O P0 FDUEDFKRO IRU D SHULRG RI G LQ RUGHU WR DVVHVV WKH HIIHFWV RI WKH 93 $OLWKLXP FRPELQDWLRQ RQ 0$5&.6 SURWHLQ H[SUHVVLRQ DQG GLVWULEXWLRQ 5HVXOWV SUHVHQWHG LQ )LJXUH D VKRZ WKDW D JUHDWHU UHGXFWLRQ RI 0$5&.6 SURWHLQ ZDV REVHUYHG LQ ERWK VROXEOH DQG PHPEUDQH

PAGE 57

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f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f DQG QSURS\OJOXWDULF DFLG 3*$f IRU RU G 7DEOH )LJXUH DEf 7KHVH 93$ PHWDEROLWHV WKH VWUXFWXUHV RI ZKLFK DUH GHSLFWHG LQ )LJXUH F ZHUH SURYLGHG E\ $EERWW /DERUDWRULHV $EERWW 3DUN ,/f DQG KDYH EHHQ IRXQG WR

PAGE 58

KDYH OLWWOH RU QR DQWLFRQYXOVDQW HIILFDF\ LQ YLYR &KDSPDQ HW DO 6XOOLYDQ SHUVRQDO FRPPXQLFDWLRQf &HOOV ZHUH H[SRVHG WR FRQFHQWUDWLRQV RI DQG P0 8QOLNH 93 $ ZKLFK SURGXFHG QRWLFHDEOH FKDQJHV LQ FHOO PRUSKRORJ\ DW P0 VHH &KDSWHU f WKHVH DQDORJV SURGXFHG QR VXFK FKDQJH DW WKLV RU KLJKHU FRQFHQWUDWLRQV )ROORZLQJ +93$ H[SRVXUH QR VLJQLILFDQW UHGXFWLRQ LQ 0$5&.6 SURWHLQ ZDV HYLGHQW LQ WKH VROXEOH IUDFWLRQ HYHQ DW FRQFHQWUDWLRQV WLPHV KLJKHU WKDQ WKH PLQLPDO HIIHFWLYH FRQFHQWUDWLRQ RI 93$ +RZHYHU LQ WKH PHPEUDQH IUDFWLRQ D VLJQLILFDQW UHGXFWLRQ LQ 0$5&.6 SURWHLQ ZDV REVHUYHG RQO\ LQ FHOOV H[SRVHG WR P0 +93$ S DW G S DW Gf $W WKLV KLJKHVW FRQFHQWUDWLRQ WHVWHG P0 ; WKH WKHUDSHXWLF OHYHO RI 93$f D b UHGXFWLRQ LQ 0$5&.6 SURWHLQ ZDV REVHUYHG IROORZLQJ G H[SRVXUH )ROORZLQJ 3*$ H[SRVXUH QR VLJQLILFDQW UHGXFWLRQ LQ 0$5&.6 SURWHLQ ZDV DSSDUHQW LQ WKH VROXEOH IUDFWLRQ HYHQ DW FRQFHQWUDWLRQV WLPHV KLJKHU WKDQ WKH PLQLPDO HIIHFWLYH FRQFHQWUDWLRQ RI 93$ $ VPDOO EXW LQVLJQLILFDQW UHGXFWLRQ LQ 0$5&.6 SURWHLQ LQ WKH PHPEUDQH IUDFWLRQ ZDV REVHUYHG IROORZLQJ P0 H[SRVXUH WR 3*$ IRU G $ UHSUHVHQWDWLYH :HVWHUQ EORW LV SUHVHQWHG LQ )LJXUH E D R k X e W f r R ; : X ;

PAGE 59

$ % & )LJXUH (IIHFWV RI RWKHU SV\FKRWURSLF DJHQWV RQ 0$5&.6 SURWHLQ H[SUHVVLRQ Df 7KLV ILJXUH VKRZV WKH UHODWLYH ODFN RI HIIHFW RI YDULRXV RWKHU SV\FKRWURSLF DJHQWV RQ 0$5&.6 SURWHLQ H[SUHVVLRQ LQ +1 FHOOV &HOOV ZHUH H[SRVHG WR GUXJV RWKHU WKDQ 93$ RU OLWKLXP VHH 7DEOH IRU GHWDLOVf DV IROORZV P0 K\GUR[\YDOSURLF DFLG G P0 SURS\OJOXWDULF DFLG G S0 FDUEDPD]HSLQH G SJPO IOXR[HWLQH G SJPO LPLSUDPLQH G SJPO SKHQ\WRLQ G SJPO GLD]HSDP G QJPO KDORSHULGRO G SJPO PRUSKLQH G QJPO YHUDSDPLO G 0$5&.6 OHYHOV ZHUH DVVHVVHG DV SUHYLRXVO\ GHVFULEHG GDWD UHSUHVHQW PHDQ s 6(0 RI DW OHDVW WKUHH GHWHUPLQDWLRQV rS Ef 0$5&.6 H[SUHVVLRQ IROORZLQJ H[SRVXUH WR 93$ DQG WZR PHWDEROLWHV 6KRZQ LV D UHSUHVHQWDWLYH :HVWHUQ EORW RI 0$5&.6 SURWHLQ LQ +1 FHOOV DIWHU G H[SRVXUH WR WKH IROORZLQJ QR GUXJ f P0 93$ SXUFKDVHG IURP 6LJPD &KHPLFDO &RPSDQ\ f P0 93$ SURYLGHG E\ $EERWW /DERUDWRULHV IRU FRPSDULVRQ SXUSRVHV f P0 f P0 f DQG P0 f SURS\OJOXWDULF DFLG P0 f P0 f DQG P0 f K\GUR[\YDOSURLF DFLG $ VKRZV F\WRVROLF 0$5&.6 H[SUHVVLRQ DQG % VKRZV PHPEUDQHDVVRFLDWHG 0$5&.6 Ff 6WUXFWXUDO FRPSDULVRQV RI 93$ DQG PHWDEROLWHV 6KRZQ DUH WKH FKHPLFDO VWUXFWXUHV RI 93$ DQG WZR RI LWV PHWDEROLWHV K\GUR[\YDOSURLF DFLG RU QSURS\OK\GUR[\SHQWDQRLF DFLG DQG QSURS\OJOXWDULF DFLG

PAGE 60

7DEOH 'UXJV 6RXUFHV DQG ([SRVXUH &RQGLWLRQV IRU +1 &HOOV 7KLV WDEOH OLVWV WKH GUXJV WR ZKLFK +1 FHOOV ZHUH H[SRVHG WKHLU VRXUFHV WDUJHW FLUFXODWLQJ SODVPDf FRQFHQWUDWLRQV LQ KXPDQV FRQFHQWUDWLRQV WHVWHG LQ +1 FHOOV DQG YHKLFOH LQ ZKLFK GUXJ ZDV VXVSHQGHG 'UXJ FRQFHQWUDWLRQV WHVWHG HQFRPSDVVHG D EURDG UDQJH LQFOXGLQJ VXSUDWKHUDSHXWLF FRQFHQWUDWLRQV LQ RUGHU WR DFFRXQW IRU GLVFUHSDQFLHV VXFK DV WKRVH SUHVHQWHG E\ YDULDWLRQ LQ FLUFXODWLQJ VHUXP YV EUDLQ FRQFHQWUDWLRQV RI GUXJ 'UXJ 6RXUFH 7DUJHW 3ODVPD &RQH LQ +XPDQV &RQFHQWUDWLRQV 7HVWHG 9HKLFOH 6RGLXP 9DOSURDWH 6LJPD 3 $EERWW $ QJPO P0 VHL]XUH FRQWUROff P0 G+ 3URS\OJOXWDULF $FLG $EERWW $ P0 G+ +\GUR[\YDOSURDWH VRGLXP VDOW $EERWW $ P0 G+ &DUEDPD]HSLQH 6LJPD & QJPO Q0 VHL]XUH FRQWUROfEfF >0 (72+ )OXR[HWLQH+&O /LOO\ /) YDULHV QJPO G+ ,PLSUDPLQH),&O 6LJPD QJPO DQWLGHSUHVVDQWf QJPO G+ 3KHQ\WRLQ VRGLXP VDOW 6LJPD QJPO DQWLFRQYXOVDQWff QJPO (72+ 'LD]HSDP 6LJPD QJPO DQ[LRO\WLFff QJPO VHL]XUH FRQWUROf QJPO (72+ +DORSHULGRO 6LJPD + QJPO DQWLSV\FKRWLFfE QJPO G+ 0RUSKLQH 6XOIDWH 6LJPD 0 QJPO QJPO DQDOJHVLF LQ QRQWROHUDQW SDWLHQWVfF G+ 9HUDSDPLO+&O 6LJPD 9 YDULHV QJPO (72+ D+ROIRUG DQG %HQHW E%HQHW DQG :LOOLDPV F:HLQHU G3RUWHU DQG 0HOGUXP

PAGE 61

(IIHFW RI 93$ RQ 0$5&.6 P51$ +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU WR G DQG 0$5&.6 P51$ OHYHOV ZHUH PHDVXUHG E\ 53$ DV GHVFULEHG LQ 0DWHULDOV DQG 0HWKRGV $V VKRZQ LQ )LJXUH D 0$5&.6 P51$ OHYHOV GHFUHDVH RYHU WLPH ZLWK FRQWLQXDO 93$ H[SRVXUH Sf $ UHSUHVHQWDWLYH 53$ LV VKRZQ LQ )LJXUH E D 7LPH Gf E P P 0$5&.6 F\FORSKLOLQ $ % & ( ) + )LJXUH 7LPH FRXUVH RI 93 $LQGXFHG 0$5&.6 P51$ DOWHUDWLRQV Df 7KLV ILJXUH VKRZV WKH WLPH FRXUVH RI 93 $LQGXFHG DOWHUDWLRQV LQ 0$5&.6 P51$ +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU G DQG 0$5&.6 P51$ OHYHOV ZHUH DVVHVVHG E\ 53$ ZLWK WKH DLG RI D F\FORSKLOLQ LQWHUQDO FRQWURO DV GHVFULEHG LQ 0DWHULDOV DQG 0HWKRGV 5HVXOWV IRU HDFK WUHDWPHQW DUH FDOFXODWHG DV D UDWLR RI 0$5&.6 WR F\FORSKLOLQ DQG WKHQ HDFK YDOXH LV H[SUHVVHG DV D SHUFHQWDJH s 6 (0 f RI XQWUHDWHG FRQWURO VDPSOHV SUHSDUHG LQ SDUDOOHO rS 6KRZQ LQ Ef LV D UHSUHVHQWDWLYH 53$ 8SSHU EDQG UHIOHFWV 0$5&.6 P51$ DQG ORZHU EDQG UHIOHFWV F\FORSKLOLQ P51$ 7UHDWPHQWV DUH DV IROORZV $ FRQWURO % G 93$ & G 93$ FRQWURO ( G 93$ ) G 93$ FRQWURO + G 93$ FRQWURO G 93$

PAGE 62

(IIHFW RI 93$ RQ *$3 ([SUHVVLRQ LQ +1 &HOOV *$3 H[SUHVVLRQ LQ +1 FHOOV LV FRPSDUHG WR LWV H[SUHVVLRQ LQ UDW EUDLQ LQ )LJXUH )ROORZLQJ G P0 93$ H[SRVXUH *$3 OHYHOV LQ +1 FHOOV LQFUHDVH DV VKRZQ LQ )LJXUH 7KLV LQFUHDVH LV FRQFHQWUDWLRQGHSHQGHQW DV VKRZQ LQ )LJXUH Sf +1 FHOOV ZHUH H[SRVHG IRU G WR FRQFHQWUDWLRQV RI 93$ UDQJLQJ IURP P0 WR P0 DQG PHPEUDQHDVVRFLDWHG *$3 H[SUHVVLRQ LV VKRZQ )LJXUH f $ % )LJXUH *$3 6KRZQ LV D :HVWHUQ LPPXQREORW FRPSDULQJ *$3 H[SUHVVLRQ LQ f VROXEOH DQG f PHPEUDQH IUDFWLRQV RI $f +1 FHOOV DQG %f UDW EUDLQ )LJXUH (IIHFW RI FKURQLF 93$ H[SRVXUH RQ *$3 H[SUHVVLRQ LQ +1 FHOOV 6KRZQ LV D UHSUHVHQWDWLYH :HVWHUQ EORW RI *$3 H[SUHVVLRQ LQ WKH PHPEUDQH IUDFWLRQ RI +1 FHOOV IROORZLQJ 93$ H[SRVXUH /DQH VKRZV *$3 H[SUHVVLRQ LQ XQWUHDWHG FRQWURO FHOOV DQG /DQH VKRZV *$3 LQ FHOOV WUHDWHG IRU G ZLWK P0 93$

PAGE 63

>93$@ P0 )LJXUH &RQFHQWUDWLRQGHSHQGHQW LQFUHDVH LQ *$3 H[SUHVVLRQ LQ +1 FHOOV IROORZLQJ 93$ H[SRVXUH +1 FHOOV ZHUH H[SRVHG IRU G WR YDU\LQJ FRQFHQWUDWLRQV RI 93$ DQG PHPEUDQHDVVRFLDWHG *$3 H[SUHVVLRQ ZDV DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV UHSUHVHQW WKH PHDQ RI WZR RU WKUHH GHWHUPLQDWLRQV DQG DUH H[SUHVVHG DV WKH SHUFHQW RI *$3 SURWHLQ H[SUHVVHG LQ XQWUHDWHG FRQWURO FXOWXUHV JURZQ LQ SDUDOOHO rS 7LPH Gf )LJXUH 7LPH FRXUVH RI 93$LQGXFHG *$3 LQFUHDVH LQ +1 FHOOV +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU WLPHV UDQJLQJ IURP G WR G DQG *$3 OHYHOV ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV UHIOHFW WKH PHDQ RI DW OHDVW WKUHH GHWHUPLQDWLRQV DQG DUH H[SUHVVHG DV SHUFHQW RI *$3 LQ DQ XQWUHDWHG FRQWURO rS

PAGE 64

7LPH &RXUVH RI 93$,QGXFHG *$3 ,QFUHDVH LQ +1 &HOOV +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU G UHVXOWV DUH VKRZQ LQ )LJXUH 8SUHJXODWLRQ RI *$3 ZDV REVHUYHG LQ PHPEUDQH DQG VROXEOH IUDFWLRQV RI FHOOV H[SRVHG WR 93$ IRU G DQG G UHVSHFWLYHO\ Sf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

PAGE 65

EHIRUH RQVHW RI DFWLRQ DOWKRXJK ORDGLQJ VWUDWHJLHV KDYH SURYHQ XVHIXO LQ DFKLHYLQJ IDVWHU UHVSRQVH ZLWKLQ Gf %RZGHQ HW DO 0F(OUR\ HW DO f &RQVLVWHQW ZLWK VXFK FOLQLFDO REVHUYDWLRQV D VLJQLILFDQW UHGXFWLRQ RI 0$5&.6 SURWHLQ LQ ERWK VROXEOH DQG PHPEUDQH IUDFWLRQV RI +1 FHOOV ZDV QRW REVHUYHG IROORZLQJ DFXWH Gf 93 $ DGPLQLVWUDWLRQ H[FHSW DW OHYHOV DERYH WKH PD[LPDO WDUJHW WKHUDSHXWLF VHUXP FRQFHQWUDWLRQ P0f $W P0 93 $ WKH GRZQUHJXODWLRQ RI 0$5&.6 ZDV JUHDWHVW EHWZHHQ G DQG G DQG OLWWOH IXUWKHU UHGXFWLRQ ZDV REVHUYHG XSRQ FRQWLQXHG H[SRVXUH Gf :LWK UHJDUG WR WKHUDSHXWLF FRQFHQWUDWLRQV WKHUH KDYH EHHQ UHSRUWV RI FOLQLFDO UHVSRQVH DW SODVPD FRQFHQWUDWLRQV RI 93$ DV ORZ DV fLJPO %RZGHQ HW DO f ZKLFK LV FRPSDUDEOH WR WKH REVHUYHG HIIHFWV RI 93$ RQ 0$5&.6 GRZQUHJXODWLRQ DW FRQFHQWUDWLRQV DV ORZ DV P0 LQ WKLV VWXG\ +RZHYHU 93$ LV JHQHUDOO\ WKRXJKW WR EH PRVW HIIHFWLYH DW WKHUDSHXWLF FRQFHQWUDWLRQV RI LJPO P0f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b LQ WKH VROXEOH IUDFWLRQ DQG b LQ WKH PHPEUDQH DQG OLWKLXPFDUEDFKRO HOLFLWHG PHDQ UHGXFWLRQV RI b DQG b LQ VROXEOH DQG PHPEUDQH UHVSHFWLYHO\ WKH FRPELQDWLRQ RI 93$ SOXV OLWKLXPFDUEDFKRO UHVXOWHG LQ D b GURS LQ VROXEOH 0$5&.6 DQG D b GURS LQ PHPEUDQHDVVRFLDWHG 0$5&.6 7KHVH GDWD DUH LQ DJUHHPHQW ZLWK FOLQLFDO VWXGLHV ZKLFK VKRZ 93 $OLWKLXP FRPELQDWLRQ WKHUDS\ WR KDYH JUHDWHU WKHUDSHXWLF HIILFDF\ IRU

PAGE 66

WUHDWPHQW RI ELSRODU GLVRUGHU WKDQ HLWKHU DJHQW DGPLQLVWHUHG LQGLYLGXDOO\ &DODEUHVH DQG 'HOXFFKL +D\HV f ,Q FRQWUDVW WR OLWKLXP ZKLFK H[HUWV D SUHIHUHQWLDO GRZQ UHJXODWLRQ RI 0$5&.6 LQ WKH VROXEOH IUDFWLRQ RI +1 FHOOV :DWVRQ DQG /HQR[ f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f WR WKH FXOWXUH PHGLXP KDG QR HIIHFW RQ WKH 93$LQGXFHG UHGXFWLRQ LQ 0$5&.6 DQG WKHVH ILQGLQJV DUH FRQVLVWHQW ZLWK WKH ODFN RI HIIHFW RI 93$ RQ UHFHSWRU FRXSOHG DFFXPXODWLRQ RI &033$ LQ WKH &+2. FHOOV :DWVRQ HW DO f &033$ DOVR NQRZQ DV &'3'$* LV D E\SURGXFW RI '$* PHWDEROLVP DQG SUHFXUVRU IRU 3, UHIRUPDWLRQ DQG VHUYHV DV D VHQVLWLYH PHDVXUH RI F\FOLQJ WKURXJK WKH 3, VLJQDOLQJ FDVFDGH UHIHU WR )LJXUH f 7KXV LQ VSLWH RI WKH VKDUHG SURSHUW\ RI OLWKLXP DQG 93$ LQ GRZQ UHJXODWLQJ WKH H[SUHVVLRQ RI 0$5&.6 WKHVH GDWD VXJJHVW WKDW WKH PHFKDQLVP WKURXJK ZKLFK WKLV RFFXUV PD\ LQYROYH GLIIHUHQW SDWKZD\V 1RQHWKHOHVV WKHUH DSSHDUV WR EH HYLGHQFH IRU D UROH RI 3.& UHJXODWLRQ LQ WKH DFWLRQ RI ERWK OLWKLXP DQG 93$ IRU SUHYLRXV VWXGLHV KDYH GHPRQVWUDWHG WKDW SKRUERO HVWHUV GRZQUHJXODWH 0$5&.6 SURWHLQ LQ QHXURQDOO\ GHULYHG FHOO SRSXODWLRQV LQ D 3.&GHSHQGHQW PDQQHU :DWVRQ HW DO f 7KLV ZLOO EH IXUWKHU GLVFXVVHG LQ ODWHU FKDSWHUV

PAGE 67

7ZR VWUXFWXUDO DQDORJV RI 93$ VHH )LJXUH Ff ZKLFK DUH VLJQLILFDQWO\ OHVV SRWHQW DV DQWLFRQYXOVDQWV LQ FRPSDULVRQ WR 93$ &KDSPDQ HW D f ZHUH VWXGLHG %RWK DQDORJV DUH NQRZQ PHWDEROLWHV RI 93$ DOWKRXJK WKH\ DUH SURGXFHG WKURXJK PLQRU PHWDEROLF SDWKZD\V DQG UHSUHVHQW D PLQLPDO SHUFHQWDJH RI WRWDO EUDLQ SODVPD FRQFHQWUDWLRQ &KDSPDQ HW DO (DGLH f ,Q SUHYLRXV VWXGLHV QHLWKHU K\GUR[\YDOSURLF DFLG QRU SURS\OJOXWDULF DFLG ZDV VKRZQ WR EH HIIHFWLYH LQ SUHYHQWLQJ SHQW\OHQHWHWUD]ROLQGXFHG VHL]XUHV LQ PLFH &KDSPDQ HW DO 6XOOLYDQ SHUVRQDO FRPPXQLFDWLRQf DQG QHLWKHU DJHQW KDV EHHQ WHVWHG FOLQLFDOO\ IRU HIILFDF\ LQ WKH WUHDWPHQW RI ELSRODU GLVRUGHU ,Q RXU VWXGLHV K\GUR[\YDOSURLF DFLG SURGXFHG D VWDWLVWLFDOO\ VLJQLILFDQW HIIHFW LQ GRZQUHJXODWLQJ PHPEUDQHDVVRFLDWHG 0$5&.6 EXW DW D FRQFHQWUDWLRQ ZHOO DERYH WKDW REVHUYHG IRU 93$ P0 K\GUR[\YDOSURLF DFLG YV P0 93$f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f GLG QRW KDYH DQ\ HIIHFW RQ 0$5&.6 H[SUHVVLRQ &DUEDPD]HSLQH KDV EHHQ VKRZQ WR KDYH HIIHFWV ZLWKLQ WKH DGHQ\O\O F\FODVH FDVFDGH D SURSHUW\ VKDUHG E\ OLWKLXP DW KLJKHU FRQFHQWUDWLRQV 0DQML HW DO 0RUN HW DO f +RZHYHU FDUEDPD]HSLQH KDV JHQHUDOO\ EHHQ VKRZQ WR EH OHVV HIIHFWLYH RYHUDOO WKDQ

PAGE 68

OLWKLXP RU 93 $ DV D PRRG VWDELOL]HU .HFN HOm %RZGHQ f DQG YHUDSDPLO D FDOFLXP FKDQQHO EORFNHU KDV \HW WR EH VWXGLHG H[WHQVLYHO\ IRU WKLV DSSOLFDWLRQ )ROORZLQJ H[SRVXUH WR D WKHUDSHXWLF VHUXP FRQFHQWUDWLRQ P0f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
PAGE 69

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n PLWRWLF QHXURQV DQG LV KLJKO\ H[SUHVVHG GXULQJ D[RQDO JURZWK DQG V\QDSVH IRUPDWLRQ $LJQHU HW D f )RU H[DPSOH LQFUHDVHG *$3 H[SUHVVLRQ PLJKW VLJQLI\ V\QDSWLF UHJURZWK DQGRU UHPRGHOLQJ F\WRSURWHFWLRQ RU HYHQ GHYHORSPHQW RI DEHUUDQW SDWKZD\V 6\QDSWLF UHPRGHOLQJ ZDV GLVFXVVHG SUHYLRXVO\ DV D SRWHQWLDO PHFKDQLVP IRU ORQJWHUP PRRG VWDELOL]DWLRQ $ QRYHO ILQGLQJ REVHUYHG LQ FDUU\LQJ RXW WKHVH VWXGLHV ZDV WKDW IROORZLQJ FKURQLF Gf 93$ H[SRVXUH +1 FHOOV DSSHDUHG WR EHFRPH PRUH DGKHUHQW WR WKH WLVVXH FXOWXUH IODVNV LQ ZKLFK WKH\ ZHUH JURZQ 7KLV LV RI LQWHUHVW LQ OLJKW RI WKH ILQGLQJ WKDW FKURQLF 93$ H[SRVXUH DOVR HOLFLWHG DQ LQFUHDVH LQ H[SUHVVLRQ RI *$3 ZKLFK LV NQRZQ WR EH LPSRUWDQW IRU SURPRWLQJ DQG PDLQWDLQLQJ JURZWK FRQH DGKHVLRQ GXULQJ QHXULWH RXWJURZWK $LJQHU DQG &DURQL f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

PAGE 70

UHODWH HLWKHU WR WKHLU GLIIHULQJ HIILFDFLHV LQ WUHDWLQJ YDULRXV W\SHV RI ELSRODU GLVRUGHUV LH 93$ LV JHQHUDOO\ PRUH HIILFDFLRXV LQ WUHDWLQJ UDSLGF\FOLQJ ELSRODU GLVRUGHUf RU DOWHUQDWLYHO\ WR WKHLU YHU\ GLIIHUHQW SKDUPDFRORJLFDO SURILOHV DVLGH IURP WKHLU PRRG VWDELOL]LQJ HIILFDF\ LH 93$ LV DOVR DQ DQWLHSLOHSWLF ZKHUHDV OLWKLXP LV QRWf &RQFOXVLRQV :LWK UHJDUG WR DOWHUDWLRQV LQ SURWHLQ H[SUHVVLRQ DQG WKHLU QHFHVVLW\ DQGRU VXIILFLHQF\ LQ EULQJLQJ DERXW WKH GHVLUHG HQG UHVXOW WKDW RI DWWDLQLQJ VXFFHVVIXO PRRG VWDELOL]LQJ WUHDWPHQWf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

PAGE 71

FDUEDPD]HSLQH DQRWKHU ZLGHO\ XVHG PRRGVWDELOL]LQJ GUXJ WKHUDS\ DOWKRXJK QRW )'$ DSSURYHG IRU WKLV DSSOLFDWLRQf DV ZHOO DV VRPH OHVV FRPPRQO\ XVHG PRRGVWDELOL]LQJ DJHQWV LQFOXGLQJ ODPRWULJLQH GDWD QRW VKRZQf DQG YHUDSDPLO GLG QRW DOWHU 0$5&.6 H[SUHVVLRQ SRLQW WR WKH SRVVLELOLW\ WKDW HLWKHU 0$5&.6 LV QRW LQWHJUDO WR PRRGVWDELOL]LQJ HIILFDF\ RU WKDW WKH LQ YLWUR H[SRVXUH FRQGLWLRQV DQG PRGHO ZH HPSOR\HG ZHUH QRW RSWLPDO WR DOORZ WKH GRZQUHJXODWLRQ ZH PD\ KDYH REVHUYHG IROORZLQJ FKURQLF H[SRVXUH LQ YLYR $QRWKHU LPSRUWDQW TXHVWLRQ FRQFHUQLQJ WKH HIIHFWV REVHUYHG LV WKDW RI VXIILFLHQF\ 7KH LGHQWLILFDWLRQ RI 0$5&.6 DV D SRWHQWLDO WDUJHW IRU WKH DFWLRQ RI GUXJV XVHG LQ WKH WUHDWPHQW RI ELSRODU GLVRUGHU VXJJHVWV WKDW WKH GRZQUHJXODWLRQ RI 0$5&.6 DORQH E\ DQ\ PHWKRG ZRXOG EH HQRXJK WR \LHOG VXFFHVVIXO DPHOLRUDWLRQ RI ELSRODU V\PSWRPV +RZHYHU WKLV GLPLQLVKHV WKH SRWHQWLDO UROH RI RWKHU PHFKDQLVPV VXFK DV WKH DOWHUDWLRQV LQ 3, VLJQDOLQJ LQGXFHG E\ OLWKLXP RU WKH HIIHFWV RI 93$ RQ *$3 H[SUHVVLRQ ,W LV FRQFHLYDEOH WKDW WKH DOWHUDWLRQV LQ 0$5&.6 H[SUHVVLRQ REVHUYHG LQ YLWUR ZRXOG OHDG LQ YLYR WR IXUWKHU GRZQVWUHDP HYHQWV LPSRUWDQW IRU DOWHULQJ SODVWLFLW\ RU RWKHU SDUDPHWHUV LQYROYHG LQ FKDQJLQJ WKH FRXUVH RI WKH GLVRUGHU ,Q RWKHU ZRUGV 0$5&.6 PD\ EH MXVW RQH WDUJHW RI PDQ\ ZKLFK PD\ XOWLPDWHO\ UHVXOW LQ WKH HIILFDF\ RI WKLV JURXS RI GUXJV WKRVH ZKLFK H[KLELW DQWLPDQLF DQGRU PRRGVWDELOL]LQJ DFWLRQ 7KLV LV PRUH FRPSDWLEOH ZLWK WKH QRWLRQ WKDW RWKHU GUXJV PD\ EH DEOH WR SRVLWLYHO\ DIIHFW WKH FRXUVH RI ELSRODU WKHUDS\ :KHUHDV 93$ DQG OLWKLXP PD\ RSHUDWH DW OHDVW LQ SDUW WKURXJK WKHLU HIIHFWV RQ 0$5&.6 WKH LQYROYHPHQW RI 0$5&.6 LQ ELSRODU WKHUDS\ LV LQGLUHFW DQG RWKHU GUXJV FRXOG WKHRUHWLFDOO\ HOLFLW WKH VDPH HQG UHVXOW \HW WKURXJK D VHSDUDWH SDWKZD\ SHUKDSV E\SDVVLQJ 0$5&.6 DOWRJHWKHU ,W LV LQWHUHVWLQJ WKDW 93$ OHG WR RSSRVLWH HIIHFWV RQ WZR &16 SURWHLQV ZKLFK DUH ERWK LQYROYHG LQ FHOOXODU SODVWLFLW\ DQG LQWUDFHOOXODU VLJQDOLQJ DQG ERWK UHJXODWHG E\ 3.& 7KLV LV GHVSLWH WKH IDFW WKDW WKH 93$LQGXFHG DOWHUDWLRQV RI WKHVH SURWHLQV RFFXUUHG RYHU D YHU\ VLPLODU WLPH FRXUVH DQG GRVH UDQJH 7KHVH GDWD VXSSRUW UHFHQW LQ YLYR ILQGLQJV LQ RXU ODERUDWRU\ ZKLFK LPSO\ DQ DSSDUHQW LQYHUVH UHJXODWLRQ RI 0$5&.6 DQG *$3 LQ

PAGE 72

FHUWDLQ KLJKO\ SODVWLF EUDLQ UHJLRQV VXFK DV WKH KLSSRFDPSXV LQ WKDW OHYHOV RI *$3 DUH KLJK LQ VRPH UHJLRQV ZKLOH 0$5&.6 OHYHOV DUH ORZ DQG YLFH YHUVD 0F1DPDUD DQG /HQR[ f 7KLV ZRXOG VXJJHVW WKDW DW FHUWDLQ GHYHORSPHQWDO VWDJHV RU LQ SDUWLFXODU EUDLQ UHJLRQV 0$5&.6 DQG *$3 SOD\ DOWHUQDWLYH RU RSSRVLQJ UROHV HYHQ WKRXJK WKH\ FDQ EH UHJXODWHG VLPLODUO\ 6XFK FRPSDUDEOH SKDUPDFRG\QDPLF SURSHUWLHV VXJJHVW WKDW ERWK SURWHLQV SOD\ LPSRUWDQW UROHV LQ WKH ORQJWHUP WKHUDSHXWLF DFWLRQ RI 93$

PAGE 73

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f 3.& LV DFWXDOO\ FRPSULVHG RI D ODUJH IDPLO\ RI DW OHDVW PDPPDOLDQ LVRIRUPV FODVVLILHG LQWR WKUHH JURXSV EDVHG RQ WKHLU DFWLYDWLRQ SURILOHV 7KH FRQYHQWLRQDO F3.&f JURXS LQFOXGHV D SLSLO DQG \ ZKLFK DUH UHJXODWHG E\ &D GLDF\OJO\FHUROV '$*f QHJDWLYHO\ FKDUJHG SKRVSKROLSLGV DQG SKRUERO HVWHUV SRWHQW WXPRU SURPRWHUV ZKLFK PLPLF '$*f 7KH QRYHO Q3.&f JURXS FRQWDLQV H U_ /f DQG DQG LV FKDUDFWHUL]HG E\ DFWLYDWLRQ UHTXLUHPHQWV VLPLODU WR WKRVH RI WKH F3.&fV EXW ODFN RI &D DFWLYDWLRQ 7KH DW\SLFDO D3.&f PHPEHUV DUH e W !f DQG _L'f DQG DUH DOO LQGHSHQGHQW RI &D DQG '$* RU SKRUEROV UHTXLULQJ RQO\ SKRVSKROLSLGV IRU DFWLYDWLRQ )XMLVH HO D +RIPDQQ f 7KH 3.& LVR]\PHV GLIIHU LQ WKHLU SULPDU\ VWUXFWXUH DQG JHQRPLF RULJLQ VR WKDW WKH\ DUH QRW GHULYHG IURP D VLQJOH P51$ EXW UDWKHU IURP D VHULHV RI VWUXFWXUDOO\ UHODWHG P51$V FRGLQJ IRU D IDPLO\ RI HQ]\PHV ZLWK YHU\ VLPLODU SURSHUWLHV ,Q KXPDQV WKH JHQHV IRU WKH YDULRXV LVR]\PHV DUH ORFDWHG RQ GLIIHUHQW

PAGE 74

FKURPRVRPHV ZLWK WKH H[FHSWLRQ RI WKH SL DQG S+ LVRIRUPV ZKLFK DUH WKH UHVXOW RI GLIIHUHQWLDO VSOLFLQJ 6WDEHO DQG 3DUNHU f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f DQG SKRUERO HVWHUV ELQG WR WUDQVLHQWO\ DFWLYDWH WKH HQ]\PH HYHQWXDOO\ UHVXOWLQJ LQ DQ LQFUHDVHG UDWH RI GHJUDGDWLRQ GRZQUHJXODWLRQf $OO PHPEHUV RI WKH 3.& VXSHUIDPLO\ FDUU\ LQ WKH 1WHUPLQDO UHJLRQ DW OHDVW RQH ]LQFILQJHU VWUXFWXUH PRUH FRPPRQO\ WZRf ZKLFK LV WKRXJKW WR DLG LQ SURWHLQSURWHLQ LQWHUDFWLRQV 7KH 1WHUPLQDO UHJLRQ DOVR FRQWDLQV D SVHXGRVXEVWUDWH VHTXHQFH FRQIHUULQJ WKH HQ]\PHfV DXWRLQKLELWRU\ DFWLYLW\ 7KLV UHJLRQ VHUYHV WR UHJXODWH DFWLYLW\ RI WKH HQ]\PH DQG UHPRYDO RI WKLV UHJLRQ UHVXOWV LQ FRQVWLWXWLYH DQG DFWLYDWRULQGHSHQGHQW NLQDVH DFWLYLW\ 7KH SVHXGRVXEVWUDWH GRPDLQ UHVHPEOHV WKH FRQVHQVXV SKRVSKRU\ODWLRQ VHTXHQFH EXW FRQWDLQV QR SKRVSKRU\ODWDEOH UHVLGXH LW FRQWUROV NLQDVH DFWLYLW\ E\ RFFXS\LQJ WKH FDWDO\WLF VLWH DQG WKHUHE\ EORFNLQJ DFFHVV RI WKH VXEVWUDWH 6WDEHO DQG 3DUNHU f F3.& DFWLYDWLRQ LV GHSHQGHQW XSRQ WZR PDMRU HYHQWV f IRUPDWLRQ RI WKH WHUQDU\ HQ]\PH&DSKRVSKROLSLG FRPSOH[ WKH SUHVHQFH RI WKH SKRVSKROLSLG IXUWKHU HQKDQFLQJ WKH DIILQLW\ RI WKH HQ]\PH IRU FDOFLXPf DQG f ELQGLQJ RI '$* RU SKRUEROf OHDGLQJ WR

PAGE 75

WKH FRQIRUPDWLRQDO FKDQJHV ZKLFK UHVXOW LQ DFWLYDWLRQ 'RZQUHJXODWLRQ RI WKH HQ]\PH IROORZV SURORQJHG DFWLYDWLRQ DQG LV DFFRPSOLVKHG WKURXJK ERWK LQFUHDVHG SURWHRO\VLV DQG VHTXHVWUDWLRQ DVVRFLDWLRQ RI WKH HQ]\PH ZLWK F\WRVNHOHWDO RU QXFOHDU FRPSRQHQWVf $XWRSKRVSKRU\ODWLRQ RI WKH HQ]\PH LV QRW D UHTXLUHPHQW EXW SUREDEO\ SOD\V D UROH 6WDEHO DQG 3DUNHU f 'DWD IURP RXU ODERUDWRU\ LQGLFDWH WKDW 3.& DFWLYLW\ LV GHFUHDVHG RYHU WLPH IROORZLQJ FKURQLF OLWKLXP H[SRVXUH DV LV H[SUHVVLRQ RI 3.& LVR]\PHV D DQG V :DWVRQ DQG /HQR[ f 7KHVH ILQGLQJV DUH LQ DJUHHPHQW ZLWK WKRVH UHSRUWHG E\ 0DQML DQG FROOHDJXHV ZKLFK UHIOHFWHG UHGXFWLRQV LQ 3.& DFWLYLW\ DQG H[SUHVVLRQ RI D DQG H LVRIRUPV IROORZLQJ HLWKHU OLWKLXP RU 93 $ H[SRVXUH LQ HLWKHU UDWV RU & JOLRPD FHOOV 0DQML HW DO &KHQ HW DO 0DQML HW DO f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f :KHUHDV 3.&D DQG V KDYH EHHQ LPSOLFDWHG LQ WXPRU JURZWK DQG VXSSUHVVLRQ RI DSRSWRVLV PD\ LQGXFH RSSRVLWH HIIHFWV +RIPDQQ f %RWK 3.&D DQG DUH WKRXJKW WR EH LPSRUWDQW LQ HDUO\ EUDLQ GHYHORSPHQW %ODFNVKHDU HW DO f ZKLOH 3.&& PD\ SOD\ D UROH LQ SUROLIHUDWLRQ DQG VHYHUDO LVRIRUPV KDYH UHSRUWHGO\ EHHQ LPSOLFDWHG LQ QHXULWH RXWJURZWK DQGRU FHOO GLIIHUHQWLDWLRQ :D\V HW DO +XQGOH HW DO %RUJDWWL HW DO f :H KDYH SUHYLRXVO\ HVWDEOLVKHG WKDW WKH LPPRUWDOL]HG KLSSRFDPSDO FHOO OLQH +1 H[SUHVVHV 3.& LVRIRUPV D V DQG DQG RXU ODERUDWRU\ KDV GHPRQVWUDWHG LQ YLWUR HIIHFWV RI OLWKLXP RQ 3.& DFWLYLW\ DV ZHOO DV RQ H[SUHVVLRQ OHYHOV RI YDULRXV 3.& LVR]\PHV :DWVRQ DQG /HQR[ f ,Q OLJKW RI SUHYLRXV UHSRUWV

PAGE 76

GHPRQVWUDWLQJ 93$LQGXFHG DOWHUDWLRQV LQ 3.& LQ D QRQQHXURQDO FHOO OLQH LW ZLOO EH RI LQWHUHVW WR LQYHVWLJDWH WKH HIIHFWV RI 93 $ H[SRVXUH RQ WKHVH SDUDPHWHUV LQ +1 FHOOV HVSHFLDOO\ JLYHQ SUHOLPLQDU\ HYLGHQFH IURP RXU ODERUDWRU\ IRU HIIHFWV RI 93 $ RQ 0$5&.6 DQG *$3 LQ WKHVH FHOOV ,W KDV EHHQ VXJJHVWHG WKDW WKH FRORFDOL]DWLRQ RI VRPH LVRIRUPV ZLWK VXEVWUDWHVf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

PAGE 77

DQWLVHQVH ROLJRQXFOHRWLGHV ZLOO EH XVHG WR QDUURZ GRZQ ZKLFK LVRIRUPV DUH LPSRUWDQW IRU HDFK SDUDPHWHU ,W LV K\SRWKHVL]HG WKDW 3.& DFWLYLW\ ZLOO EH QHFHVVDU\ WR DQ\ REVHUYHG HIIHFWV E\ 93$ RQ 0$5&.6 RU *$3 DQG WKDW LQKLELWLRQ RI WKH HQ]\PH ZLOO EORFN VXFK REVHUYHG HIIHFWV

PAGE 78

5HVXOWV (IIHFW RI 93$ ([SRVXUH RQ 3.& $FWLYLW\ LQ +1 &HOOV +1 FHOOV H[SRVHG WR P0 93$ IRU PLQ G ZHUH DVVD\HG IRU HIIHFWV RQ 3.& DFWLYLW\ DV PHDVXUHG E\ 0%3 SKRVSKRU\ODWLRQ )LJXUH f $FXWH 93$ H[SRVXUH PLQ Kf UHVXOWHG LQ D WHQGHQF\ WRZDUG LQFUHDVHG PHPEUDQHDVVRFLDWHG DFWLYLW\ DQG D FRUUHODWHG GHFUHDVH LQ 3.& DFWLYLW\ LQ WKH F\WRVROLF IUDFWLRQ 7KLV SDUDGLJP PRVW FORVHO\ IROORZV D WUDQVORFDWLRQ SKHQRPHQRQ ZKHUHE\ F\WRVROLF 3.& LV VKXWWOHG LQWR WKH PHPEUDQH IRU DFWLYDWLRQ %\ WKH G WLPH SRLQW DFWLYLW\ LQ WKH PHPEUDQH IUDFWLRQV LV GRZQUHJXODWHG Sf 6LPLODU UHVXOWV ZHUH REVHUYHG XVLQJ 0$5&.6 SHSWLGH DV WKH SKRVSKRU\ODWLRQ VXEVWUDWH GDWD QRW VKRZQf 7LPH KRXUVf )LJXUH (IIHFW RI 93$ H[SRVXUH RQ 3.& DFWLYLW\ LQ +1 FHOOV 93$ H[SRVXUH OHDGV WR DSSDUHQW 3.& DFWLYDWLRQ IROORZHG E\ GRZQUHJXODWLRQ +1 FHOOV H[SRVHG WR P0 93$ ZHUH SURFHVVHG DQG VROXEOH DQG PHPEUDQH IUDFWLRQV LVRODWHG DV GHWDLOHG LQ 0DWHULDOV DQG 0HWKRGV &KDSWHU f $VVD\ ZDV SHUIRUPHG XWLOL]LQJ WKH 3.& DFWLYDWRUV 30$ SKRUEROf &D&O DQG SKRVSKDWLG\OVHULQH 3.& DFWLYDWLRQ ZDV PHDVXUHG E\ WKH \ 3 SKRVSKRU\ODWLRQ LQ SPROPLQPJf RI WKH 3.& VXEVWUDWH 0%3 0HDQ DFWLYLW\ RI FRQWURO VDPSOHV ZDV s SPROPLQPJ 7KH ILJXUH VKRZV WKH RYHUDOO SDWWHUQ RI 3.& DFWLYLW\ IURP PLQ DFXWHf WR G FKURQLFf 93$ H[SRVXUH WKH LQVHW VKRZV 3.& DFWLYDWLRQ RQO\ DW DFXWH WLPH SRLQWV XS WR Kf 5HVXOWV DUH PHDQ s 6(0 RI DW OHDVW IRXU GHWHUPLQDWLRQV DQG DUH H[SUHVVHG DV SHUFHQW RI DFWLYLW\ LQ XQWUHDWHG FRQWURO VDPSOHV rS

PAGE 79

(IIHFW RI 93$ RQ 3.&D 3.&D LV REVHUYHG HQWLUHO\ LQ WKH VROXEOH IUDFWLRQ RI )71 FHOOV $V VKRZQ LQ )LJXUH H[SRVXUH WR P0 93$ UHVXOWV LQ DQ DFXWH PLQ Kf LQFUHDVH LQ 3.& D H[SUHVVLRQ IROORZHG E\ GRZQUHJXODWLRQ RYHU WLPH Sf + -:YX:X m1I80.2M. 7,0( KRXUVf )LJXUH (IIHFW RI 93$ RQ 3.&D H[SUHVVLRQ LQ VROXEOH IUDFWLRQ RI +1 FHOOV Df 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORW RI VROXEOH 3.&D H[SUHVVLRQ IROORZLQJ DFXWH K RU OHVVf 93$ H[SRVXUH # P0 3URWHLQ VDPSOHV ZHUH ORDGHG LQWR ODQHV DV IROORZV /DQH XQWUHDWHG FRQWURO /DQH PLQ 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ Ef 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORW RI F\WRVROLF 3.&D H[SUHVVLRQ IROORZLQJ FKURQLF Gf 93$ H[SRVXUH # P0 /DQH ZDV ORDGHG ZLWK SURWHLQ IURP DQ XQWUHDWHG FRQWURO DQG /DQH ZDV ORDGHG ZLWK SURWHLQ IURP D VDPSOH H[SRVHG WR 93$ IRU G Ff +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU PLQ G DQG 3.&D OHYHOV SUHVHQW RQO\ LQ WKH F\WRVROLF IUDFWLRQ RI WKHVH FHOOVf ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV DUH H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO DQG DUH WKH PHDQ s 6(0 RI IRXU GHWHUPLQDWLRQV rS (IIHFW RI 93$ RQ 3.&6 3.& XQGHUJRHV D VLJQLILFDQW LQFUHDVH LQ H[SUHVVLRQ OHYHOV LQ ERWK VROXEOH Sf DQG PHPEUDQH Sf IUDFWLRQV RI FHOOV GXULQJ FKURQLF 93$ H[SRVXUH t Gf DQG DV ODWH DV G WKHUH DSSHDUV WR EH QR GRZQUHJXODWLRQ )LJXUH f

PAGE 80

)LJXUH (IIHFW RI 93 $ RQ 3.& H[SUHVVLRQ LQ +1 FHOOV Df 5HSUHVHQWDWLYH :HVWHUQ EORW RI 3.& H[SUHVVLRQ IROORZLQJ DFXWH Kf 93$ H[SRVXUH # P0 7KH EORW DW OHIW /DQHV f VKRZV F\WRVROLF 3.& LQ +1 FHOOV ZKLOH WKH EORW DW ULJKW /DQHV f VKRZV PHPEUDQHDVVRFLDWHG 3.& 3URWHLQ VDPSOHV ZHUH ORDGHG LQWR ODQHV DV IROORZV /DQH XQWUHDWHG FRQWURO /DQH PLQ 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH XQWUHDWHG FRQWURO /DQH PLQ 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ Ef 5HSUHVHQWDWLYH :HVWHUQ RI 3.& H[SUHVVLRQ IROORZLQJ FKURQLF Gf 93$ H[SRVXUH # P0 7KH EORW DW OHIW VKRZV F\WRVROLF 3.& LQ +1 FHOOV ZKLOH WKH EORW DW ULJKW VKRZV PHPEUDQHDVVRFLDWHG 3.& /DQH ZDV ORDGHG ZLWK SURWHLQ IURP DQ XQWUHDWHG FRQWURO DQG /DQH ZDV ORDGHG ZLWK SURWHLQ IURP D VDPSOH H[SRVHG WR 93$ IRU G Ff +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU PLQ G DQG 3.& OHYHOV ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV DUH H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO DQG DUH WKH PHDQ s 6 (0 RI IRXU GHWHUPLQDWLRQV rS VROXEOH DQG rS PHPEUDQH

PAGE 81

D E )LJXUH (IIHFW RI 93 $ RQ 3.&H H[SUHVVLRQ LQ +1 FHOOV Df 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORW RI 3.&H H[SUHVVLRQ IROORZLQJ DFXWH K RU OHVVf 93$ H[SRVXUH # P0 7KH VL[ ODQHV DW OHIW VKRZ F\WRVROLF RU VROXEOHf 3.& LQ +1 FHOOV ZKLOH WKH VL[ ODQHV DW ULJKW VKRZ PHPEUDQHDVVRFLDWHG 3.& 3URWHLQ VDPSOHV ZHUH ORDGHG LQWR ODQHV DV IROORZV /DQH XQWUHDWHG FRQWURO /DQH PLQ 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ Ef 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORW RI 3.&H H[SUHVVLRQ IROORZLQJ FKURQLF Gf 93$ H[SRVXUH # P0 7KH EORW DW OHIW VKRZV F\WRVROLF RU VROXEOHf 3.& LQ +1 FHOOV ZKLOH WKH EORW DW ULJKW VKRZV PHPEUDQHDVVRFLDWHG 3.& /DQH ZDV ORDGHG ZLWK SURWHLQ IURP DQ XQWUHDWHG FRQWURO DQG /DQH ZDV ORDGHG ZLWK SURWHLQ IURP D VDPSOH H[SRVHG WR 93$ IRU G Ff &HOOV ZHUH H[SRVHG WR P0 93$ IRU PLQ G DQG 3.&H OHYHOV ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV DUH H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO DQG DUH WKH PHDQ s 6(0 RI IRXU GHWHUPLQDWLRQV rS

PAGE 82

(IIHFW RI 93$ RQ 3.&H $V VKRZQ LQ )LJXUH PHPEUDQHDVVRFLDWHG 3.&V LQFUHDVHV DFXWHO\ PLQ Kf DQG WKHQ UHWXUQV WR EDVDO OHYHOV E\ WKH G WLPH SRLQW &\WRVROLF 3.&V H[SUHVVLRQ UHPDLQV FORVH WR EDVHOLQH DFXWHO\ DQG WKHQ LQFUHDVHV VLJQLILFDQWO\ IROORZLQJ SURORQJHG Gf 93$ H[SRVXUH Sf 7KLV SDWWHUQ RI H[SUHVVLRQ UHIOHFWV DQ DSSDUHQW DFWLYDWLRQ DQG WUDQVORFDWLRQ RI 3.&V ZKHUHE\ 3.& LV DFWLYDWHG VKXWWOHG IURP F\WRVRO LQWR WKH PHPEUDQH ZKHUH WKH HQ]\PH LV DFWLYHf DQG WKHQ GRZQUHJXODWHG DQG UHWXUQHG IURP PHPEUDQH WR F\WRVRO IROORZLQJ SURORQJHG DFWLYDWLRQ (IIHFW RI 93$ RQ 3.& 3.&& VKRZV OLWWOH FKDQJH LQ H[SUHVVLRQ DW HDUO\ WLPH SRLQWV DQG WKHQ LV XS UHJXODWHG VRPHWLPH EHWZHHQ G DQG Sf ZLWK D VLPLODU WHQGHQF\ WRZDUG XS UHJXODWLRQ REVHUYHG LQ WKH VROXEOH IUDFWLRQ )LJXUH f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

PAGE 83

E )LJXUH (IIHFW RI 93 $ RQ 3.&e H[SUHVVLRQ LQ +1 FHOOV Df 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORWV RI 3.&e H[SUHVVLRQ IROORZLQJ DFXWH K RU OHVVf 93$ H[SRVXUH # P0 7KH EORW DW OHIW VKRZV F\WRVROLF RU VROXEOHf 3.& LQ +1 FHOOV ZKLOH WKH EORW DW ULJKW VKRZV PHPEUDQHDVVRFLDWHG 3.& 3URWHLQ VDPSOHV ZHUH ORDGHG LQWR ODQHV DV IROORZV /DQH XQWUHDWHG FRQWURO /DQH PLQ 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ /DQH K 93$ Ef 5HSUHVHQWDWLYH :HVWHUQ LPPXQREORW RI 3.&A H[SUHVVLRQ IROORZLQJ FKURQLF Gf 93$ H[SRVXUH # P0 7KH EORW DW OHIW VKRZV F\WRVROLF RU VROXEOHf 3.& LQ +1 FHOOV ZKLOH WKH EORW DW ULJKW VKRZV PHPEUDQHDVVRFLDWHG 3.& /DQH ZDV ORDGHG ZLWK SURWHLQ IURP DQ XQWUHDWHG FRQWURO DQG /DQH ZDV ORDGHG ZLWK SURWHLQ IURP D VDPSOH H[SRVHG WR 93$ IRU G Ff +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU PLQ G DQG 3.&e OHYHOV ZHUH DVVHVVHG E\ :HVWHUQ LPPXQREORW DQDO\VLV 5HVXOWV DUH H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWURO DQG DUH WKH PHDQ s 6(0 RI IRXU GHWHUPLQDWLRQV rS PHPEUDQH

PAGE 84

f W VWDUW PRXVH 3.& DOSKD JHQH )LJXUH 3.&D DQWLVHQVH ROLJRQXFOHRWLGH GHVLJQ 7KLV ILJXUH GHSLFWV WKH DQWLVHQVH ROLJRQXFOHRWLGHV GHVLJQHG WR ELQG WR 3.&D P51$ DQG EORFN LWV WUDQVODWLRQ 6KRZQ LV WKH UHSUHVHQWDWLYH 3.&D JHQH IURP PRXVH DV UHSRUWHG LQ WKH 1&%, GDWDEDVHf DQG WKH ILYH ROLJRQXFOHRWLGHV f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f WR fQXFOHRWLGH VHTXHQFHV RI HDFK 3.&D ROLJRQXFOHRWLGH 21f E\ QXPEHU DQG WKH VSHFLHV VSHFLILFLW\ RI HDFK 7DEOH $QWLVHQVH ROLJRQXFOHRWLGH VHTXHQFHV 21 6HTXHQFH RI 1XFOHRWLGHV 6SHFLHV 6SHFLILFLW\ $&&$7**&7*$&*77 PRXVH t KXPDQ $$&*7&$*&&$7**7 PRXVH t KXPDQ &$*&&$7**77&&&&&&$$& PRXVH $*$$**7$***&77&&*7$7 PRXVH t KXPDQ *77 $&& $$& $**$&&7 $*&7 PRXVH 7 *$*77 & $*& $7 *$& $77*7 PRXVH

PAGE 85

,QLWLDO H[SHULPHQWV DWWHPSWHG WR HVWDEOLVK WKH RSWLPDO GRVH UDQJH IRU ROLJRQXFOHRWLGH H[SRVXUH DQG SURWHLQ GRZQUHJXODWLRQ 2OLJRQXFOHRWLGH FRQFHQWUDWLRQV UDQJLQJ IURP WR S0 WKH VWDQGDUG FRQFHQWUDWLRQV UHSRUWHG LQ VLPLODU VWXGLHVf KDG QR HIIHFW RQ FHOO YLDELOLW\ RU 3.&D H[SUHVVLRQ IROORZLQJ K H[SRVXUH GDWD QRW VKRZQf ,QFUHDVLQJ ROLJRQXFOHRWLGH FRQFHQWUDWLRQ WR S0 DQG OHQJWKHQLQJ H[SRVXUH WLPH WR K KDG QR HIIHFW RQ 3.&D H[SUHVVLRQ GDWD QRW VKRZQf 1H[W PHWKRGV ZHUH HPSOR\HG WR LQFUHDVH ROLJRQXFOHRWLGH XSWDNH DQG FRPSHQVDWH IRU SRWHQWLDO ROLJRQXFOHRWLGH GHJUDGDWLRQ ,Q NHHSLQJ ZLWK SUHYLRXV UHSRUWV RI VXFFHVVIXO DQWLVHQVHPHGLDWHG SURWHLQ GRZQUHJXODWLRQ WKH FDWLRQLF FDUULHU OLSLG /LSRIHFW$0,1( *LEFR %5/ *DLWKHUVEXUJ 0'f ZDV XVHG DW SJPO DQG SJPO DQG FHOOV ZHUH VHUXP GHSULYHG GXULQJ ROLJRQXFOHRWLGH H[SRVXUH ,Q DGGLWLRQ PHGLXP DQG LWV FRQVWLWXHQWV ZHUH UHSODFHG IUHVK HYHU\ K GXULQJ H[SRVXUH 3.&D H[SUHVVLRQ UHPDLQHG DW EDVDO OHYHOV IROORZLQJ RU K H[SRVXUH WR ROLJRQXFOHRWLGH XQGHU WKHVH FRQGLWLRQV GDWD QRW VKRZQf 7KH ILQDO DSSURDFK XWLOL]HG SKRUERO WR GRZQUHJXODWH 3.& H[SUHVVLRQ QRQ VHOHFWLYHO\f WKHQ 3.&D DQWLVHQVH ROLJRQXFOHRWLGHV ZHUH DGGHG IROORZLQJ SKRUERO UHPRYDO LQ DQ HIIRUW WR VHOHFWLYHO\ EORFN 3.&D UHFRYHU\ $ UHSUHVHQWDWLYH :HVWHUQ LV VKRZQ LQ )LJXUH +1 FHOOV ZHUH H[SRVHG WR S0 SKRUERO GLEXW\UDWH 3'%Xf IRU K DIWHU ZKLFK FHOOV ZHUH ZDVKHG ZLWK PHGLD WR UHPRYH GUXJ 1H[W FHOOV ZHUH H[SRVHG WR S0 VHQVH RU DQWLVHQVH ROLJRQXFOHRWLGHV WRWDO RI ILYH DQWLVHQVH ROLJRQXFOHRWLGHVf GLUHFWHG WR 3.&D )LJXUH f &HOOV ZHUH H[SRVHG WR ROLJRQXFOHRWLGHV IRU D WRWDO RI K EXW DQWLVHQVHWUHDWHG FHOOV VKRZHG QR GLIIHUHQFH LQ 3.&D H[SUHVVLRQ WKDQ VHQVHWUHDWHG RU XQWUHDWHG FRQWURO FHOOV $V VKRZQ LQ )LJXUH WKHUH LV FOHDUO\ QR GLIIHUHQFH LQ 3.&D H[SUHVVLRQ EHWZHHQ WKRVH 3'%XSUHWUHDWHG FHOOV ZKLFK DUH DQWLVHQVHWUHDWHG RU XQWUHDWHG 3'%X H[SRVXUH UHVXOWHG LQ D UHYHUVLEOH GRZQUHJXODWLRQ RI 3.&D /DQH YV f DQG 3.&D DQWLVHQVH IDLOHG WR SUHYHQW WKH UHFRYHU\ RI 3.&D H[SUHVVLRQ DW UDWHV FRPSDUDEOH WR FRQWURO OHYHOV /DQHV t YV f :KHUHDV WR[LFLW\ FRPPRQO\ DULVHV DV D SUREOHP LQ DQWLVHQVH DSSOLFDWLRQV LQ FHOO DQG WLVVXH FXOWXUH WKH +1

PAGE 86

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f 0$5&.6 H[SUHVVLRQ LQ ERWK F\WRVRO DQG PHPEUDQH RI 93$WUHDWHG FHOOV ZDV VLJQLILFDQWO\ GLIIHUHQW IURP WKDW RI /
PAGE 87

U R V r V R 8 H f ‘0 k f§ &$ 8 6& \ 6ROXEOH 0HPEUDQH &RQWURO 93$ /< 93$ /< 7UHDWPHQW E )LJXUH (IIHFW RI 3.& LQKLELWRU /< RQ 93$LQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 Df +1 FHOOV ZHUH H[SRVHG IRU K WR P0 93$ ZLWK RU ZLWKRXW L0 /< 0$5&.6 H[SUHVVLRQ ZDV TXDQWLWDWHG E\ :HVWHUQ EORW DQDO\VLV DQG UHVXOWV H[SUHVVHG DV SHUFHQW RI XQWUHDWHG FRQWUROf UHIOHFW WKH PHDQ s 6(0 RI DW OHDVW ILYH GHWHUPLQDWLRQV rS VROXEOH rS PHPEUDQH Ef 5HSUHVHQWDWLYH :HVWHUQ EORW VKRZLQJ 0$5&.6 H[SUHVVLRQ LQ WKH SUHVHQFH RI 93$ s /< /DQHV UHSUHVHQW F\WRVROLF 0$5&.6 DQG ZHUH WUHDWHG IRU K DV IROORZV FRQWURO XQWUHDWHG f P0 93$ f S0 /< f P0 93$ S0 /< f /DQHV VKRZ PHPEUDQHDVVRFLDWHG 0$5&.6 DQG ZHUH WUHDWHG IRU K DV GHVFULEHG IRU UHVSHFWLYHO\f DERYH (IIHFW RI 3.& ,QKLELWRU /< RQ 93$LQGXFHG *$3 ,QGXFWLRQ /< DORQH RU LQ FRPELQDWLRQ ZLWK 93$ KDG QR HIIHFW RQ *$3 H[SUHVVLRQ LQ +1 FHOOV )LJXUH f *$3 OHYHOV LQ ERWK VROXEOH DQG PHPEUDQH IUDFWLRQV DUH VLJQLILFDQWO\ LQFUHDVHG Sf LQ ERWK 93$WUHDWHG JURXSV FRPSDUHG WR

PAGE 88

FRQWUROV XQWUHDWHG RU /
PAGE 89

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f DOWKRXJK ERWK GUXJV KDYH VLPLODU HIIHFWV RQ 0$5&.6 H[SUHVVLRQ 2XU ILQGLQJV LQGLFDWH WKDW OLNH OLWKLXP 93$ LQGXFHV DOWHUDWLRQV LQ H[SUHVVLRQ RI 3.& LVRIRUPV D DQG V :DWVRQ DQG /HQR[ f +RZHYHU WKH WLPH FRXUVH IRU 93$ LQGXFHG FKDQJHV LV HDUOLHU WKDQ WKDW IRU OLWKLXP DQG 3.&H H[SUHVVLRQ DFWXDOO\ LQFUHDVHV DFXWHO\ LQ PHPEUDQH FKURQLFDOO\ LQ F\WRVROf LQ FRQWUDVW WR WKH UHGXFWLRQ SUHYLRXVO\ UHSRUWHG IRU ERWK OLWKLXP DQG 93$ &KHQ HW DO 0DQML HW DO :DWVRQ DQG /HQR[ f &KHQ HW DO f UHSRUWHG WKDW FKURQLF 93$ H[SRVXUH RI & JOLRPD FHOOV UHGXFHG WKH H[SUHVVLRQ RI 3.& LVR]\PHV D DQG H LQ LQWDFW FHOOV D ILQGLQJ VLPLODU WR WKDW REVHUYHG IROORZLQJ FKURQLF OLWKLXP DGPLQLVWUDWLRQ LQ WKH VDPH FHOO PRGHO 0DQML HW DO f 6XEVHTXHQW VWXGLHV UHYHDOHG 3.&LQGXFHG DOWHUDWLRQV LQ PXOWLSOH FRPSRQHQWV RI WKH DGUHQHUJLF UHFHSWRUFRXSOHG F$03JHQHUDWLQJ V\VWHP LQFOXGLQJ 3L$5 *DV DQG DGHQ\O\O F\FODVH HDFK RI ZKLFK LV SKRVSKRU\ODWHG E\ 3.& &KHQ HW DO f 7KH VLJQLILFDQFH RI WKHVH FKDQJHV LV XQFOHDU EXW WRJHWKHU WKH\ VXJJHVW D PDMRU UROH IRU 3.& LQ WKH 93$ RU OLWKLXPLQGXFHG PROHFXODU HIIHFWV XQGHUO\LQJ WKHLU HIILFDF\ DQG PD\ EH LQVWUXFWLYH ZKHQ RWKHU 3.&UHODWHG HIIHFWV DUH FRPSDUHG EHWZHHQ WKH WZR GUXJV ,Q WKLV VWXG\ 93$ H[SRVXUH UHVXOWHG LQ DQ DSSDUHQW WUDQVLHQW DFWLYDWLRQ RI 3.& ZLWKLQ Kf IROORZHG E\ D JUDGXDO UHGXFWLRQ LQ DFWLYLW\ ZLWK FKURQLF H[SRVXUH K DQG EH\RQGf 3.& DFWLYLW\ ZDV VLJQLILFDQWO\ GRZQUHJXODWHG E\ WKH K WLPH SRLQW 6LPLODU

PAGE 90

UHVXOWV ZHUH UHSRUWHG IROORZLQJ H[SRVXUH RI +1 FHOOV WR WKH SKRUERO 30$ WKRXJK DFWLYDWLRQ RFFXUUHG ZLWKLQ PLQXWHV DQG GRZQUHJXODWLRQ RI 3.& DFWLYLW\ ZDV HYLGHQW E\ K :DWVRQ HW DO f 7KLV ILQGLQJ LV IXUWKHU LQ DJUHHPHQW ZLWK SUHYLRXV UHSRUWV ZKLFK VKRZHG WKDW FKURQLF 93$ RU OLWKLXP H[SRVXUH RI & JOLRPD FHOOV VLJQLILFDQWO\ UHGXFHG 3.& DFWLYLW\ LQ ERWK PHPEUDQH DQG VROXEOH IUDFWLRQV &KHQ HW DO 0DQML HW DO f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e LV XQDEOH WR SKRVSKRU\ODWH 0$5&.6 LQ YLWUR )XMLVH HWDO +HUJHW HW DO 8EHUDOO HW DO f 2QH FRXOG K\SRWKHVL]H D SURFHVV LQ ZKLFK 93$ GLUHFWO\ RU LQGLUHFWO\ DFWLYDWHV 3.& HVSHFLDOO\ LVRIRUPV D DQGRU H DW HDUO\ WLPH SRLQWV EHWZHHQ DQG Kf %\ G WKHUH LV D VLJQLILFDQW GRZQUHJXODWLRQ RI DFWLYLW\ LQ WKH PHPEUDQH DQG D FRPSDUDEOH GURS LQ WKH F\WRVRO &RQVLVWHQW ZLWK WKLV K\SRWKHVLV DUH GDWD ZKLFK VKRZ DQ LQFUHDVH LQ 3.& DFWLYLW\ ZLWKLQ VHFRQGV IROORZLQJ SKRUERO H[SRVXUH 6WDEHO DQG 3DUNHU f DQG D WLPH FRXUVH IRU 0$5&.6 GRZQ UHJXODWLRQ ZKLFK FORVHO\ IROORZV WKH WLPH FRXUVH RI 3.& DFWLYDWLRQ LQ +1 FHOOV H[SRVHG WR SKRUERO :DWVRQ HW DO f :KLOH 93$ DQG OLWKLXP ERWK DIIHFW 3.& LVR]\PH D DQG SHUKDSV Vf WKH WZR GUXJV KDYH YHU\ GLIIHUHQW HIIHFWV RQ 3.& 93$ LQGXFHV D FKURQLF XSUHJXODWLRQ LQ 3.& ZKHUHDV OLWKLXP KDV OLWWOH RU QR HIIHFW RQ 3.& H[SUHVVLRQ :DWVRQ DQG /HQR[ f ,Q FRPSDULQJ 93$ DQG OLWKLXP RQH PLJKW K\SRWKHVL]H WKDW VXFK GLIIHUHQFHV LQ SDWWHUQV RI 3.& DFWLYDWLRQ XQGHUOLH WKH GLIIHUHQFHV REVHUYHG LQ GRZQVWUHDP HIIHFWV VXFK DV SURWHLQ H[SUHVVLRQ RU FHOO JURZWK DQG PRUSKRORJ\ )RU H[DPSOH WKH XSUHJXODWLRQ RI 3.& UHVXOWLQJ XQLTXHO\ IURP 93$ H[SRVXUH LV SUREDEO\ QRW LPSRUWDQW IRU WKH 0$5&.6 GRZQ

PAGE 91

UHJXODWLRQ REVHUYHG EXW PLJKW EH LQYROYHG LQ WKH *$3 LQGXFWLRQ REVHUYHG IROORZLQJ 93$ EXW QRW OLWKLXP H[SRVXUH LQ VXSSRUW RI WKLV K\SRWKHVLV LV HYLGHQFH LQGLFDWLQJ WKDW 3.& DQG H SKRVSKRU\ODWH *$3 2HKUOHLQ HW DO f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f /RQJWHUP GRZQUHJXODWLRQ RI 3.& DFWLYLW\ ZLWK SKRUEROV VXFK DV 30$ KDV SURYLGHG VLPLODU UHVXOWV 7VXQHLVKL &DUOVRQ HW DO -DODYD HW D (NLQFL DQG 6KHD f ZLWK RQH VWXG\ VKRZLQJ HQKDQFHG QHXULWH RXWJURZWK IROORZLQJ FRPELQHG SKRUERO DQG 3.& LQKLELWRU WUHDWPHQW +HLNNLOD HW DO f +RZHYHU VRPH RI WKHVH VDPH VWXGLHV KDYH IXUWKHU VKRZQ WKDW GRZQ UHJXODWLRQ RI 3.& DFWLYLW\ DORQH LV LQVXIILFLHQW WR HOLFLW D GLIIHUHQWLDWHG SKHQRW\SH (NLQFL DQG 6KHD &DUOVRQ HW DO :RRWHQ f 2Q WKH RWKHU KDQG QXPHURXV UHSRUWV KDYH SUHVHQWHG HTXDOO\ FRQYLQFLQJ HYLGHQFH IRU D SRVLWLYH UROH IRU 3.& LQ UHJXODWLQJ QHXURQDO GLIIHUHQWLDWLRQ DQG D SRVLWLYH UROH LQ PDLQWDLQLQJ WKH IXQFWLRQDOO\ DFWLYH VWDWH &ROHPDQ DQG :RRWHQ &DEHOO DQG $XGHVLUN 3DUURZ HW DO $EUDKDP HW DO 7RQLQL HW DO f WKRXJK WKLV UROH LV SUREDEO\ QRW XQLTXH WR 3.& 3UHYLRXV GDWD KDYH LPSOLFDWHG D QXPEHU RI GLIIHUHQW 3.& LVRIRUPV LQ FHOOXODU GLIIHUHQWLDWLRQ LQFOXGLQJ DOO IRXU RI WKH LVRW\SHV SUHVHQW LQ +1 FHOOV D H DQG & :DGD HW DO /HOL HW DO :D\V HW DO +XQGOH HW DO %RUJDWWL HW DO f ,QGHHG WKH VXEFHOOXODU ORFDOL]DWLRQ DQG

PAGE 92

DFWLYDWLRQ RI 3.& LVR]\PHV DUH UHJXODWHG LQ DQ LVRIRUPVSHFLILF PDQQHU GXULQJ QHXURJHQHVLV VXJJHVWLQJ WKH\ DUH LQYROYHG LQ FRQWURO RI QHXUDO GHYHORSPHQW DQG LQ QHXURQDO GLIIHUHQWLDWLRQ 2HKUOHLQ HW DO f 7KH UROH RI 3.& LQ SUROLIHUDWLRQ DQG GLIIHUHQWLDWLRQ UHPDLQV XQFOHDU IRU UHSRUWV GLIIHU DV WR ZKLFK 3.& LVRIRUPVf LVDUH UHVSRQVLEOH IRU YDULRXV DVSHFWV RI FHOOXODU GLIIHUHQWLDWLRQ 6RPH VWXGLHV KDYH DWWULEXWHG WR 3.&D D UROH DV WKH LQWHUPHGLDWH LQ FHOOXODU GLIIHUHQWLDWLRQ DV D UHVXOW RI DOWHUDWLRQV LQ 3.&D P51$ DQGRU SURWHLQ H[SUHVVLRQ REVHUYHG FRQFRPLWDQW ZLWK GLIIHUHQWLDWLRQ RI 3& DQG RWKHU FHOO OLQHV %RUJDWWL HW DO 3DUURZ HW DO 0XUUD\ HW DO /HOL HW DO 7RQLQL HW DO :DGD HW DO f +RZHYHU WKHVH UHSRUWV GLIIHU LQ WKHLU PHFKDQLVWLF FRQFOXVLRQV IRU LQ WKUHH VWXGLHV 3.&D H[SUHVVLRQ LQFUHDVHG ZLWK GLIIHUHQWLDWLRQ %RUJDWWL HW DO 3DUURZ HW DO 0XUUD\ HW DO f ZKLOH LQ WKUHH RWKHU VWXGLHV 3.&D DQG H OHYHOV KDG GHFUHDVHG IROORZLQJ FRPSDUDEOH WUHDWPHQW DQG QHXULWRJHQHVLV /HOL HW DO 7RQLQL HW DO :DGD HW DO f /HOLnV JURXS IXUWKHU VKRZHG WKDW WKH LQWUDFHOOXODU GHOLYHU\ RI 3.&D DQG H LVRIRUPVSHFLILF DQWLERGLHV UHVXOWHG LQ D PRUSKRORJLFDOO\ GLIIHUHQWLDWHG SKHQRW\SH LQ 6+6<< FHOOV /HOL HW DO f :KLOH VRPH JURXSV REVHUYHG QR GLIIHUHQWLDWLRQDVVRFLDWHG DOWHUDWLRQV LQ LVRIRUPV RWKHU WKDQ 3.&D D QXPEHU RI UHSRUWV KDYH LQGLFDWHG WKDW LQ YDULRXV FHOO OLQHV RWKHU PHPEHUV RI WKH 3.& VXSHUIDPLO\ DUH LQYROYHG %RUJDWWLnV JURXS ZKR VKRZHG WKDW 3.&D ZDV XSUHJXODWHG GXULQJ 3& FHOO GLIIHUHQWLDWLRQ IRXQG LQ WKH VDPH VWXG\ WKDW 3.& V DQG & ZHUH GRZQUHJXODWHG %RUJDWWL HW DO f $ QXPEHU RI VWXGLHV KDYH SURYLGHG HYLGHQFH IRU D UROH RI 3.&H LQ UHJXODWLQJ QHXURQDO GLIIHUHQWLDWLRQ VLQFH QHXULWH RXWJURZWK KDV EHHQ DVVRFLDWHG ZLWK DOWHUHG 3.&H OHYHOV 3RQ]RQL HW DO 3DUURZ HW DO )DJHUVWURP HW DO (NLQFL DQG 6KHD +XQGOH HW DO =HLGPDQ HW DO f ,Q VXSSRUW RI WKH QRWLRQ WKDW 3.&H PD\ EH LQYROYHG LQ WKH SURFHVV DV VXJJHVWHG E\ /HOLnV DQG :DGDnV ZRUN /HOL HW DO :DGD HW DO f D VHSDUDWH

PAGE 93

VWXG\ VKRZHG WKDW RYHUH[SUHVVLRQ RI 3.&H EXW QRW 3.& OHG WR DQ LQFUHDVH LQ 1*) LQGXFHG QHXULWH RXWJURZWK LQ VWDEO\ WUDQVIHFWHG 3& FHOO OLQHV +XQGOH HW DO f ,Q VWDUN FRQWUDVW WR WKHVH ILQGLQJV \HW DQRWKHU JURXS UHSRUWHG MXVW WKH RSSRVLWH LQ 1,+ 7 FHOOV LQ WKDW 3.& RYHUH[SUHVVLRQ LQGXFHG D PRUSKRORJLFDO FKDQJH DQG UHGXFHG JURZWK UDWH FRQVLVWHQW ZLWK GLIIHUHQWLDWLRQ WKDW ZDV QRW REVHUYHG LQ FHOOV RYHUH[SUHVVLQJ 3.&V 0LVFKDN HW DO f )XUWKHU 3.&e KDV EHHQ VLPLODUO\ H[DPLQHG LQ WUDQVIHFWHG OHXNHPLF FHOOV DQG UHVXOWV LQGLFDWH WKDW 3.&& RYHUH[SUHVVLRQ LV DEOH WR LQGXFH FHOOXODU GLIIHUHQWLDWLRQ :D\V HW DO f $QRWKHU VWXG\ WKDW XWLOL]HG DQWLVHQVH WHFKQLTXHV DOVR HPSKDVL]HG WKH LPSRUWDQFH RI 3.&& LQ 1*)LQGXFHG QHXURQDO GLIIHUHQWLDWLRQ &ROHPDQ DQG :RRWHQ f $OO RI WKHVH VWXGLHV SRLQW WR GLIIHUHQW 3.& LVR]\PHV DV PHGLDWRUV RI VLPLODU HIIHFWV 7KHVH DSSDUHQWO\ FRQWUDVWLQJ UHVXOWV PD\ PHUHO\ EH D IXQFWLRQ RI WKH YDU\LQJ H[SHULPHQWDO PHWKRGV DQG FRQGLWLRQV XVHG IURP RQH VWXG\ WR WKH QH[W VXFK DV GLIIHUHQW FHOO OLQHV H[SRVXUH FRQGLWLRQV DQG RWKHU SDUDPHWHUV 7KHVH ILQGLQJV PD\ DOWHUQDWLYHO\ LQGLFDWH WKDW f HDFK LVR]\PH SOD\V LWV RZQ VSHFLDOL]HG UROH LQ WKH FRPSOH[ SURFHVV RI FHOOXODU GLIIHUHQWLDWLRQ DQG f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

PAGE 94

3DUNHU f VXJJHVW WKDW WKH GLVWULEXWLRQ RI 0$5&.6 FORVHO\ IROORZV WKH GLVWULEXWLRQ RI 3.& EDVHG RQ HDUO\ VWXGLHV RI 3.& DFWLYLW\ VKRZLQJ ZLGH GLVSDULW\ DFURVV UHJLRQV .XR HW DO 0LQDNXFKL HWDO f +DYLQJ HVWDEOLVKHG DQ DSSDUHQW FDXVDO UHODWLRQVKLS EHWZHHQ 93$ DQG 3.& WKH SXUSRVH RI WKH QH[W H[SHULPHQWV ZDV WR HVWDEOLVK PRUH SUHFLVHO\ WKH QHFHVVLW\ RI 3.& LQ HOLFLWLQJ WKH REHVUYHG 93$LQGXFHG HIIHFWV RQ 0$5&.6 DQG *$3 ,Q RUGHU WR EHWWHU HVWDEOLVK WKH UROH RI YDULRXV 3.& LVRIRUPV LQ WKH 0$5&.6 GRZQUHJXODWLRQ DQG *$3 LQFUHDVH REVHUYHG H[SHULPHQWV ZHUH GHVLJQHG HPSOR\LQJ DQWLVHQVH WHFKQRORJ\ 7KURXJK WKLV PHWKRG DQWLVHQVH ROLJRQXFOHRWLGHV DUH GHVLJQHG DQG GLUHFWHG WR VSHFLILFDOO\ ELQG DQG fNQRFN GRZQf WKH QXFOHLF DFLG SUHFXUVRUV RI HDFK LQGLYLGXDO 3.& LVR]\PH UHVXOWLQJ LQ WUDQVLHQW HOLPLQDWLRQ RI WKH SURWHLQ LQ TXHVWLRQ 7KHUH DUH IRXU PDLQ PHFKDQLVPV E\ ZKLFK DQWLVHQVH ROLJRQXFOHRWLGHV DUH K\SRWKHVL]HG WR ZRUN f E\ LQKLELWLQJ WKH QHFHVVDU\ SURFHVVLQJ DQG WUDQVSRUW RI 51$ f E\ LQWHUIHULQJ ZLWK ELQGLQJ RI WKH ULERVRPH WR WKH QXFOHRWLGH VHTXHQFH VR WKDW WUDQVODWLRQ LV QRW LQLWLDWHG f E\ LQKLELWLQJ WKH FRPSOHWLRQ RI WUDQVODWLRQ DIWHU WKH SURFHVV KDV EHJXQ DQG f E\ ELQGLQJ WR WKH VHTXHQFH WKHUHE\ FKDUDFWHUL]LQJ LW DV DEQRUPDO DQG UHVXOWLQJ LQ GHJUDGDWLRQ E\ 51DVH + 6FKOLQJHQVLHSHQ HW DO f )RU WKLV VWXG\ D QXPEHU RI WULDOV ZHUH SHUIRUPHG LQ DQ HIIRUW WR GRZQUHJXODWH 3.&D LQ WKH +1 FHOOV 2QH VHQVH DQG ILYH DQWLVHQVH ROLJRQXFOHRWLGHV ZHUH SUHSDUHG E\ *HPLQL %LRWHFK /WG 7KH :RRGODQGV 7;f WKURXJK WKH '1$ 6\QWKHVLV &RUH RI WKH 8) ,QWHUGLVFLSOLQDU\ &HQWHU IRU %LRWHFKQRORJ\ 5HVHDUFK ,&%5f (DFK ROLJRQXFOHRWLGH ZDV GHVLJQHG WR EHDU XQLTXH LGHQWLW\ WR WKH 3.&D JHQRPH RI HLWKHU PRXVH RU KXPDQ DV UHSRUWHG LQ WKH 1DWLRQDO &HQWHU IRU %LRWHFKQRORJ\ ,QIRUPDWLRQ GDWDEDVH DQG VFUHHQHG WKURXJK *HQ%DQN 9DULRXV UHJLRQV RI WKH F'1$ VHTXHQFH ZHUH WDUJHWHG LQFOXGLQJ WKH VWDUW FRGRQ VHH )LJXUH f ,Q DQ HIIRUW WR PD[LPL]H VWDELOLW\ DQG ELQGLQJ RI ROLJRQXFOHRWLGHV DOO ZHUH FRPSRVHG RI RU EDVHV HDFK DQG KDG SKRVSKRURWKLRDWH OLQNDJHV LQ SODFH RI WKH HQGRJHQRXV SKRVSKRGLHVWHU EDFNERQH RI '1$ :KHUHDV WR[LFLW\

PAGE 95

FRPPRQO\ DULVHV DV D SUREOHP LQ DQWLVHQVH DSSOLFDWLRQV LQ FHOO DQG WLVVXH FXOWXUH WKH +1 FHOOV VKRZHG QR HYLGHQFH RI F\WRWR[LFLW\ IROORZLQJ WR G H[SRVXUH WR DQG _0 FRQFHQWUDWLRQV RI 3.&D ROLJRQXFOHRWLGHV ,Q LQLWLDO H[SHULPHQWV FHOOV ZHUH H[SRVHG WR ROLJRQXFOHRWLGHVf IRU K LQ DQ HIIRUW WR HVWDEOLVK D SURWRFRO IRU WKH FRQVLVWHQW UHGXFWLRQ RI 3.&D OHYHOV SULRU WR 93 $ H[SRVXUH :KHQ QXPHURXV HIIRUWV IDLOHG WR HOLFLW DQ\ UHGXFWLRQ LQ 3.&D H[SUHVVLRQ 3.& ZDV ILUVW GRZQUHJXODWHG WKURXJK K AL0 3'%X SKRUEROf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f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

PAGE 96

VLJQLILFDQW WLPH DQG H[SHQVH PD\ RIWHQ EH LQYHVWHG EHIRUH GRZQUHJXODWLRQ RI WKH SURWHLQ LQ TXHVWLRQ LV RSWLPL]HG :KHUHDV H[SUHVVLRQ RI HDFK 3.& LVRIRUP FRXOG QRW EH GRZQUHJXODWHG LQGLYLGXDOO\ RYHUDOO 3.& DFWLYLW\ LQKLELWLRQ ZDV DFKLHYHG WKURXJK WKH XVH RI WKH 3.& LQKLELWRU /< (OL /LOO\ DQG &RPSDQ\ ,QGLDQDSROLV ,1f D GLUHFW FRPSHWLWRU IRU $73 ELQGLQJ -LURXVHN HW DO f /< LV D 3.&3VHOHFWLYH LQKLELWRU DW ORZ QP FRQFHQWUDWLRQVf ZKLFK DOVR LQKLELWV 3.& LVR]\PHV D DQG H DW FRQFHQWUDWLRQV LQ WKH ORZ Q0 UDQJH ,VKLL HW DO -LURXVHN HW DO f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
PAGE 97

UHGXFWLRQ LQ 0$5&.6 WKRXJK 3.& DFWLYDWLRQ PD\ QRW EH WKH RQO\ URXWH WKURXJK ZKLFK WKH 0$5&.6 DOWHUDWLRQV PD\ RFFXU ,QGHHG %URRNV HW DO f UHSRUWHG HYLGHQFH RI UHJXODWLRQ RI 0$5&.6 WKURXJK ERWK 3.&GHSHQGHQW DQG LQGHSHQGHQW SDWKZD\V ZLWK SODWHOHWGHULYHG JURZWK IDFWRU 3'*)fLQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 WKURXJK D SRVWWUDQVFULSWLRQDO PHFKDQLVP $V IRU *$3 3.& DFWLYDWLRQ LV SUREDEO\ QRW QHFHVVDU\ WR HOLFLW WKH DOWHUDWLRQV LQ *$3 REVHUYHG 7KLV LV HYLGHQFHG E\ WKH IDFW WKDW 3.& LQKLELWLRQ IDLOHG WR EORFN WKH 93$LQGXFHG LQFUHDVH LQ *$3 LQ WKH +1 FHOOV (YHQ ZLWK WKH VWURQJ *$3 H[SUHVVLRQ REVHUYHG LQ WKH PHPEUDQH RI WKHVH FHOOV DQG WKH VLJQLILFDQW XSUHJXODWLRQ RI WKLV SURWHLQ WR VHYHUDO WLPHV EDVDO OHYHOV /< FDXVHG QR DWWHQXDWLRQ RI WKH 93$ LQGXFHG *$3 LQFUHDVH REVHUYHG $VVXPLQJ FRQGLWLRQV ZHUH RSWLPDO IRU VLJQLILFDQW 3.& LQKLELWLRQ ZH FDQ FRQFOXGH WKDW 3.& DFWLYDWLRQ LV QRW HQWLUHO\ UHVSRQVLEOH IRU WKH 93$LQGXFHG *$3 LQFUHDVH REVHUYHG QRU LV 3.& DFWLYDWLRQ DORQH OLNHO\ WR EH VXIILFLHQW IRU WKH HIIHFW VLQFH OLWKLXPLQGXFHG DOWHUDWLRQV LQ 3.& DFWLYLW\ DQG LVRIRUP H[SUHVVLRQ GLG QRW OHDG WR DQ\ FKDQJH LQ *$3 H[SUHVVLRQ 7KH ILQGLQJV GR OLWWOH WR DGGUHVV WKH LVVXH RI WKH LQYROYHPHQW RI 3.& LQ ELSRODU GLVRUGHU DQG VXFFHVVIXO PRRGVWDELOL]LQJ WKHUDS\ +RZHYHU RQH FDQ K\SRWKHVL]H WKDW LI 0$5&.6 G\VUHJXODWLRQ LV LQWHJUDO WR ELSRODU GLVRUGHU DQG 3.& LV WKH SULPDU\ PRGXODWRU RI 0$5&.6 WKHQ SHUKDSV WKH HQ]\PH LV SHUWXUEHG LQ VRPH SDWLHQWV H[KLELWLQJ V\PSWRPV RI WKH GLVRUGHU 7KLV ZRXOG EH HVSHFLDOO\ DSSOLFDEOH WR FDVHV LQ ZKLFK WKH SDWLHQW GHPRQVWUDWHV QRUPDO OHYHOV RI 0$5&.6 IRU 0$5&.6 EHLQJ GRZQVWUHDP IURP 3.& LV UHJXODWHG E\ 3.& DQG LI 3.& SDUDPHWHUV DUH DEQRUPDO WKHQ 3.&LQGXFHG 0$5&.6 UHJXODWLRQ ZLOO OLNHO\ DOVR EH DEQRUPDO (YHQ LI 0$5&.6 LWVHOI LV QRW LQYROYHG LQ WKH PDQLIHVWDWLRQ RI WKH GLVRUGHU 3.& PD\ EH D PDMRU SOD\HU ,QGHHG SUHYLRXV VWXGLHV KDYH SURYLGHG HYLGHQFH IRU DOWHUDWLRQV LQ 3.&UHODWHG VLJQDOLQJ LQ SDWLHQWV ZLWK ELSRODU GLVRUGHU 5HSRUWV KDYH VKRZQ WKDW LQ ERWK SODWHOHWV DQG SRVWPRUWHP EUDLQ VDPSOHV RI LQGLYLGXDOV GLDJQRVHG ZLWK WKH GLVHDVH 3.& WUDQVORFDWLRQ DQG DFWLYLW\ OHYHOV ZHUH

PAGE 98

VLJQLILFDQWO\ LQFUHDVHG DV FRPSDUHG WR FRQWUROV DV PHDVXUHG E\ SKRUERO RU VHURWRQLQ VWLPXODWHG KLVWRQH SKRVSKRU\ODWLRQ )ULHGPDQ HW DO :DQJ DQG )ULHGPDQ f ,Q DGGLWLRQ 3.& LVRIRUP H[SUHVVLRQ ZDV DOWHUHG LQ FRPSDULVRQ WR FRQWUROV ZLWK KLJKHU OHYHOV RI 3.&D DQG & DQG ORZHU 3.&H H[SUHVVLRQ LQ ELSRODU SDWLHQWV :DQJ DQG )ULHGPDQ f )XUWKHU WZRZHHN OLWKLXP WUHDWPHQW RI VXEMHFWV UHVXOWHG LQ D UHGXFWLRQ LQ ERWK 3.& WUDQVORFDWLRQ DQG DFWLYLW\ LQ SODWHOHWV )ULHGPDQ HW DO f ,Q D PRUH UHFHQW FOLQLFDO VWXG\ WKH 3.&VHOHFWLYH LQKLELWRU WDPR[LIHQ DQ HVWURJHQ UHFHSWRU DQWDJRQLVW ZLGHO\ XVHG LQ WKH WUHDWPHQW RI EUHDVW FDQFHUf KDV EHHQ VKRZQ WR EH HIIHFWLYH LQ WUHDWLQJ DFXWH PDQLD LQ D VPDOO VDPSOH RI SDWLHQWV 0DQML HW DO f ,QDVPXFK DV 3.& LV EHOLHYHG WR SOD\ D SLYRWDO UROH LQ UHJXODWLQJ QHXURQDO VLJQDO WUDQVGXFWLRQ DQG LQ PRGXODWLQJ LQWUDFHOOXODU FURVVWDON EHWZHHQ QHXURWUDQVPLWWHU V\VWHPV 1LVKL]XND f WKHVH VWXGLHV DUH VXJJHVWLYH RI WKH SRWHQWLDO IRU DIIHFWLYH PRRGf G\VIXQFWLRQ DV D UHVXOW RI DOWHUDWLRQV RU G\VIXQFWLRQ LQ 3.& DQG DVVRFLDWHG QHXURQDO VLJQDOLQJ SDWKZD\V $ GHILQLWLYH DVVRFLDWLRQ EHWZHHQ 3.& DOWHUDWLRQ DQG PRRG VWDELOL]DWLRQ PD\ IDFLOLWDWH WKH GHYHORSPHQW RI D ZKROH QHZ OLQH RI DJHQWV XVHIXO LQ WKH WUHDWPHQW RI ELSRODU GLVRUGHU

PAGE 99

&+$37(5 ())(&7 2) 93$ 21 &(// 9,$%,/,7< *52:7+ $1' 0253+2/2*< ,QWURGXFWLRQ :KLOH 93$ KDV SURYHQ YHU\ XVHIXO LQ LWV FOLQLFDO DSSOLFDWLRQV WR VHL]XUHV DQG ELSRODU GLVRUGHUV WKH GHWULPHQWDO HIIHFWV RI WKLV GUXJ RQ WKH GHYHORSLQJ HPEU\R KDYH EHHQ ZHOO GRFXPHQWHG :LJHU HW DO *RIIORW HO DO 0HQHJROD HO DO f +LJK IUHTXHQFLHV RI QHXUDO WXEH GHIHFWV 17'Vf KDYH EHHQ OLQNHG WR SUHQDWDO 93$ H[SRVXUH %MHUNHGDO HW D 5REHUW DQG *XLEDXG /LQGKRXW DQG 6FKPLGW :LJHU HW DO *RIIORW HW DO 0HQHJROD HW D f 6XFK GHIHFWV PD\ LQFOXGH RSHQ QHXUDO WXEH VSLQD ELILGD DQHQFHSKDO\ PHQLQJRP\HORFHOH DQG HQFHSKDORFHOH DQG DUH WKRXJKW WR EH FDXVHG E\ D QXPEHU RI GLIIHUHQW IDFWRUV VXFK DV HQYLURQPHQWDO RU SKDUPDFRORJLFDO WHUDWRJHQV JHQHWLF DQRPDOLHV DQG YDULRXV PDWHUQDO SUHGLVSRVLQJ IDFWRUV 5KRDGV DQG 0LOOV 1RUPDQ HW DO f ,Q PRUH UHFHQW \HDUV IROLF DFLG DQGRU PHWKLRQLQH VXSSOHPHQWDWLRQ KDYH EHHQ VKRZQ WR UHGXFH WKH IUHTXHQF\ RI 17'V ERWK LQ WKH JHQHUDO SRSXODWLRQ DQG LQ RIIVSULQJ RI PRWKHUV XQGHUJRLQJ DQWLFRQYXOVDQW WKHUDS\ /DXUHQFH HW DO 6PLWKHOOV HW DO 5KRDGV DQG 0LOOV 7URW] HW DO 0XOLQDUH HW DO 1RVHO DQG .OHLQ (KOHUV HW DO f 7KHUHIRUH ZRPHQ DUH DGYLVHG WR GLVFRQWLQXH 93$ WKHUDS\ RU WDNH IROLF DFLG VXSSOHPHQWV GXULQJ DQG MXVW EHIRUH SUHJQDQF\ 7KH PHFKDQLVP RI WKLV SURWHFWLYH HIIHFW LV QRW XQGHUVWRRG )RODWH LV NQRZQ WR EH D SUHFXUVRU RI PHWKLRQLQH DQ LPSRUWDQW DPLQR DFLG IRU FHOO SUROLIHUDWLRQ DQG YLWDPLQ % LV DQ HVVHQWLDO FRIDFWRU IRU PHWKLRQLQH V\QWKHVLV ,Q PRWKHUV RI 17'DIIHFWHG RIIVSULQJ FLUFXODWLQJ VHUXP IRODWH OHYHOV DUH ORZ EXW QRW FOLQLFDOO\ GHILFLHQW WKH VDPH LV

PAGE 100

WUXH IRU % OHYHOV /DXUHQFH HW DK 5KRDGV DQG 0LOOV .LUNH HW DK f ,Q DGGLWLRQ KRPRF\VWHLQH WKH GLUHFW SUHFXUVRU IRU PHWKLRQLQH LV IRXQG DW KLJK OHYHOV LQ WKH VHUXP RI VRPH PRWKHUV LQ WKLV SRSXODWLRQ 6WHHJHUV7KHXQLVVHQ HW DK 0LOOV HW DO f 7KLV HYLGHQFH VXJJHVWV WKDW LQ DW OHDVW D IUDFWLRQ RI 17' FDVHV WKH PRWKHUV RI 17'DIIOLFWHG RIIVSULQJ H[KLELW D PHWDEROLF LQVXIILFLHQF\ SHUKDSV DW WKH OHYHO RI WKH KRPRF\VWHLQHWRIRODWH FRQYHUVLRQ VWHS 7KLV PHWDEROLF GHIHFW K\SRWKHVLV LV LQ FRQWUDVW WR WKH VLPSOHU QRWLRQ RI D QXWULWLRQDO GHILFLHQF\ RI HLWKHU IRODWH RU PHWKLRQLQHf DQG LV FRQVLVWHQW ZLWK WKH SRVVLELOLW\ WKDW 93$ WKHUDS\ LV RQH IDFWRU ZKLFK PD\ LQGXFH RU HQKDQFH WKLV PHWDEROLF GHIHFW ,W LV RI QRWH WKDW LQ H[SHULPHQWDO VWXGLHV LQ YLYR VXSSOHPHQWDWLRQ RI HLWKHU QXWULHQW WR SUHJQDQW DQLPDOV UHVXOWHG LQ GHFUHDVHG IUHTXHQFLHV RI 17'V LQ RIIVSULQJ ZKLOH LQ YLWUR VWXGLHV GLG QRW UHSURGXFH WKH SURWHFWLYH HIIHFW 7URW] HW DK +DQVHQ DQG *UDIWRQ 1RVHO DQG .OHLQ +DQVHQ HW DK (KOHUV HW DK f 3UHOLPLQDU\ VWXGLHV LQ RXU ODERUDWRU\ IXUWKHU VXSSRUW WKH QRWLRQ WKDW WKH HIIHFW LV RI D FRPSOLFDWHG PHWDEROLF QDWXUH IRU IRODWH RU PHWKLRQLQH VXSSOHPHQWDWLRQ WR FHOO JURZWK PHGLXP GLG QRW SUHYHQW WKH 93$LQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 REVHUYHG LQ +1 FHOOV :DWWHUVRQ XQSXEOLVKHG GDWDf ,Q OLJKW RI WKH GHOHWHULRXV GHYHORSPHQWDO HIIHFWV RI WKLV GUXJ WKH HIIHFWV RI 93$ RQ FHOO JURZWK DQG GLIIHUHQWLDWLRQ PHULW LQGHSWK LQYHVWLJDWLRQ 3UHYLRXV VWXGLHV ZKLOH IHZ LQ QXPEHU KDYH SURYLGHG HYLGHQFH IRU D GRVHGHSHQGHQW 93$LQGXFHG LQKLELWLRQ RI FHOO SUROLIHUDWLRQ YLD DUUHVW DW PLG*O RI WKH FHOO F\FOH 0DUWLQ DQG 5HJDQ f DV ZHOO DV 93$LQGXFHG GLIIHUHQWLDWLRQ RI QHXUREODVWRPD DQG JOLRPD FHOOV 5HJDQ &LQDWO HW DK f 7KH LQKLELWLRQ RI FHOO SUROLIHUDWLRQ DOWHUDWLRQV LQ FHOOXODU LQWHUDFWLRQV DQG DOWHUHG GLIIHUHQWLDWLRQ KDYH DOO EHHQ FRUUHODWHG ZLWK SUHQDWDO GUXJ WHUDWRJHQLFLW\ HVSHFLDOO\ RI WKH GHYHORSLQJ QHXUDO WXEH :LON HW DK 0XPPHU\ HW DK 0DUWLQ HW DK f ,W LV SODXVLEOH WKDW DUUHVW RI FHOO SUROLIHUDWLRQ DW D VSHFLILF SKDVH LQ WKH FHOO F\FOH DFFRPSDQLHG E\ LUUHYHUVLEOH GLIIHUHQWLDWLRQ PD\ UHVXOW LQ WKH HIIHFWV REVHUYHG WR EH FRQVLVWHQW ZLWK SUHQDWDO WHUDWRJHQLFLW\

PAGE 101

7KH H[SHULPHQWV RI WKLV FKDSWHU KDYH EHHQ GHVLJQHG WR DGGUHVV IXUWKHU WKH LVVXHV RI FHOO JURZWK DQG PRUSKRORJ\ DQG WKH HIIHFWV 93$ KDV RQ WKHVH SDUDPHWHUV 7KH SXUSRVH RI WKH IROORZLQJ H[SHULPHQWV LV WR DVFHUWDLQ ZKHWKHU 93$ LQGXFHV DOWHUDWLRQV DVVRFLDWHG ZLWK GLIIHUHQWLDWLRQ LQ WKH VDPH FHOO PRGHO LQ ZKLFK DOWHUHG 0$5&.6 DQG *$3 H[SUHVVLRQ ZHUH REVHUYHG %DVHG RQ WKH VLJQLILFDQWO\ LQFUHDVHG *$3 OHYHOV GLVFXVVHG SUHYLRXVO\ LQ &KDSWHU f LW LV K\SRWKHVL]HG WKDW 93$ ZLOO LQFUHDVH QHXULWH RXWJURZWK VLQFH ERWK SURFHVVHV LQFUHDVHG *$3 DQG RXWJURZWK RI QHXULWHVf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f ,I 93$ ZHUH LQGHHG IRXQG WR HOLFLW DOWHUDWLRQV LQ FHOO JURZWK DQG PRUSKRORJ\ WKH QH[W VWHS ZRXOG EH WR DGGUHVV WKH PHFKDQLVP RI WKHVH HIIHFWV E\ H[DPLQLQJ WKH UROH RI 3.& $V GLVFXVVHG LQ &KDSWHU 3.&VSHFLILF LQKLELWRUV /< DQG ELVLQGRO\OPDOHLPLGH RU %,0 &DOELRFKHP /D -ROOD &$f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

PAGE 102

7KHVH H[SHULPHQWV ZLOO IXUWKHU DGGUHVV WKH TXHVWLRQ RI WKH HIIHFWV RI 93 $ RQ FHOO SODVWLFLW\ 7KH DELOLW\ RI 93 $ WR DOWHU FHOO SODVWLFLW\ LV UHOHYDQW WR LWV XQLTXH SKDUPDFRORJLFDO SURILOH ZKLFK LQFOXGHV ERWK DQWLFRQYXOVDQW DQG ORQJWHUP PRRG VWDELOL]LQJ HIIHFWV LQ WKH DGXOW EUDLQ DV ZHOO DV WHUDWRJHQLF HIIHFWV RQ WKH GHYHORSLQJ QHUYRXV V\VWHP $OO RI WKHVH PDMRU HIIHFWV ZRXOG VHHP WR VXJJHVW DOWHUHG SODVWLFLW\ HLWKHU WKHUDSHXWLF RU DEQRUPDOf DV D FRQWULEXWLQJ IDFWRU 5HVXOWV (IIHFW RI 93$ RQ +1 &HOO 'LYLVLRQ +1 FHOOV ZHUH H[SRVHG WR P0 93$ IRU XS WR G $W WKLV WKHUDSHXWLFDOO\ UHOHYDQW FRQFHQWUDWLRQ DQ LQKLELWLRQ RI FHOO JURZWK ZDV REVHUYHG HYLGHQW LQ WKH LQFUHDVHG FHOO GRXEOLQJ WLPH S 7DEOH f 'HVSLWH WKLV HIIHFW RQ FHOO QXPEHU QR HIIHFW RQ FHOO YLDELOLW\ ZDV REVHUYHG ZKHQ FHOOV ZHUH SODWHG DW KLJK FRQIOXHQFH HYHQ DW D KLJKHU FRQFHQWUDWLRQ RI 93$ P0f DV VKRZQ LQ 7DEOH QR VLJQLILFDQFHf 7KLV VXSSRUWV WKH QRWLRQ WKDW WKH UHGXFWLRQ LQ FHOO QXPEHU LV D UHVXOW RI GHFUHDVHG JURZWK UDWH DQG QRW GXH WR FHOO GHDWK VLQFH LW ZDV RQO\ REVHUYHG XQGHU FRQGLWLRQV LQ ZKLFK FHOO GLYLVLRQ ZDV SRVVLEOH 7DEOH VKRZV WKH PHDQ s 6(0f WLPH UHTXLUHG IRU WKH +1 FHOO SRSXODWLRQ WR GRXEOH LQ QXPEHU ZKHQ H[SRVHG WR HLWKHU QR GUXJ FRQWUROf RU FKURQLF P0 93$ rS WWHVWf 7DEOH &HOO 'RXEOLQJ 7LPH &RQWURO s K P0 93$ s K r 7DEOH VKRZV WKH PHDQ s 6(0f QXPEHU RI FHOOV SHU YLHZLQJ ILHOG LQ FXOWXUHV SODWHG WRR GHQVHO\ WR DOORZ IRU QHXULWH RXWJURZWK 1R VLJQLILFDQW HIIHFW ZDV REVHUYHG 7DEOH &HOOV)LHOG &RQWURO s P0 93$ s

PAGE 103

)LJXUH 93$LQGXFHG QHXULWH RXWJURZWK LQ +1 FHOOV 7KHVH SKRWRPLFURJUDSKV VKRZ Df XQWUHDWHG +1 FHOOV Ef +1 FHOOV H[SRVHG WR P0 93$ IRU G DQG Ff +1 FHOOV H[SRVHG WR P0 93$ IRU G 7KH ILJXUH Gf VXPPDUL]HV TXDQWLWDWLYHO\ WKH HIIHFW RI 93$ RQ +0 FHOO PRUSKRORJ\ &HOOV ZHUH H[SRVHG WR P0 93$ IRU YDULHG OHQJWKV RI WLPH DQG WKHQ FHOOV ZHUH SKRWRJUDSKHG DQG DQDO\]HG 'DWD DUH SUHVHQWHG DV PHDQ SHUFHQW s 6(0f RI FHOOV ZLWK RQH QHXULWH RU PRUH DQG PHDQ SHUFHQW s 6(0f RI FHOOV ZLWK WZR QHXULWHV RU PRUH D QHXULWH ZDV GHILQHG DV KDYLQJ OHQJWK HTXDOLQJ DW OHDVW WKH GLDPHWHU RI WKH UHVSHFWLYH FHOO ERG\ rS

PAGE 104

(IIHFW RI 93$ RQ +1 &HOO 0RUSKRORJ\ ,Q FRPSDULVRQ WR FRQWURO FHOOV JURZQ LQ SDUDOOHO LQ WKH DEVHQFH RI GUXJ DQ LQFUHDVH LQ QHXULWH RXWJURZWK ZDV REVHUYHG LQ 93$H[SRVHG FHOOV DV HDUO\ DV G LQ FXOWXUH DQG SHUVLVWHG IRU WKH GXUDWLRQ RI WKH H[SRVXUH )LJXUH f 7KH SHUFHQWDJH RI FHOOV ZLWK RQH QHXULWH RU PRUH LQFUHDVHG VLJQLILFDQWO\ Sf EHWZHHQ DQG G DQG UHPDLQHG DW DERXW b WKURXJKRXW WKH GXUDWLRQ RI 93$ H[SRVXUH IURP WR G 7KH SHUFHQWDJH RI FHOOV ZLWK WZR QHXULWHV RU PRUH DOVR LQFUHDVHG Sf EHWZHHQ DQG G IURP b LQ FRQWUROV WR b LQ 93$WUHDWHG FHOOVf DQG LQFUHDVHG DJDLQ Sf EHWZHHQ DQG G WR bf ,W ZDV DOVR QRWHG GXULQJ FHOO FROOHFWLRQ DQG KDUYHVWLQJ WKDW +1 FHOOV H[SRVHG WR 93$ FRQFHQWUDWLRQV KLJKHU WKDQ P0 IRU DW OHDVW G H[KLELWHG DQ LQFUHDVHG DGKHUHQFH WR FXOWXUH IODVNV DV HYLGHQFHG E\ WKH JUHDWHU HIIRUW UHTXLUHG WR UHPRYH FHOOV E\ VORXJKLQJ ZLWK YHUVHQH GDWD QRW VKRZQf 7KLV JUHDWHU DGKHUHQFH WR VXEVWUDWH LV FRQVLVWHQW ZLWK WKH LQFUHDVHG RXWJURZWK RI SURFHVVHV DQG LQFUHDVHG H[SUHVVLRQ RI *$3 DQG WRJHWKHU DUH VXJJHVWLYH RI GLIIHUHQWLDWLRQ WR D PRUH PDWXUH VWDWH ,W ZDV DOVR REVHUYHG WKDW FHOO GHQVLW\ VLJQLILFDQWO\ LQIOXHQFHG QHXULWH RXWJURZWK RI +1 FHOOV (YHQ LQ WKH DEVHQFH RI GUXJ WUHDWPHQW WKHUH ZDV DQ LQYHUVH FRUUHODWLRQ EHWZHHQ QXPEHU RI FHOOV LQ WKH YLHZLQJ ILHOG DQG SHUFHQW RI WKHVH FHOOV H[KLELWLQJ QHXULWHV S GDWD QRW VKRZQf (IIHFWV RI WKH 3.& ,QKLELWRU /< RQ 93$,QGXFHG $OWHUDWLRQV LQ +1 &HOO 1XPEHU DQG 1HXULWH 2XWJURZWK +1 FHOOV ZHUH H[SRVHG WR P0 93$ DQGRU _L0 /< IRU K $V GHSLFWHG LQ )LJXUH D FHOO QXPEHU ZDV GHFUHDVHG VLJQLILFDQWO\ RQO\ LQ FXOWXUHV WUHDWHG ZLWK WKH FRPELQDWLRQ RI 93$ DQG /< Sf WKRXJK FXOWXUHV WUHDWHG ZLWK HLWKHU GUXJ DORQH H[KLELWHG D WHQGHQF\ WRZDUG GHFUHDVHG JURZWK )LJXUH E VKRZV WKH VLJQLILFDQW HIIHFW RI HLWKHU 93$ RU /< RQ QHXULWH RXWJURZWK LQ +1 FHOOV (LWKHU DJHQW LQFUHDVHG QHXULWH RXWJURZWK DV FRPSDUHG WR FRQWUROV Sf DQG WKH WZR DJHQWV

PAGE 105

FRPELQHG \LHOGHG D JUHDWHU HIIHFW WKDQ HLWKHU DJHQW DORQH Sf 7KHVH UHVXOWV VXJJHVW WKDW 3.& H[HUWV D QHJDWLYH LQIOXHQFH RQ FHOO GLIIHUHQWLDWLRQ VR WKDW DGGLWLRQ RI D 3.& LQKLELWRU DOORZV IRU LQFUHDVHG QHXULWH RXWJURZWK )XUWKHU WKH PHFKDQLVP RI 93$LQGXFHG LQFUHDVH LQ QHXULWHV LV SHUKDSV QRW DFWLYDWLRQ RI 3.& EXW HLWKHU ORQJWHUP GRZQ UHJXODWLRQ RI 3.& RU VRPH DOWHUQDWH PHFKDQVLP DOWRJHWKHU 7KH REVHUYDWLRQV WKDW 93 $ DQG /< H[KLELWHG DGGLWLYH HIIHFWV RQ ERWK QHXULWH RXWJURZWK DQG FHOO QXPEHU DUH VXJJHVWLYH RI GLIIHUHQW PHFKDQLVPV D 7UHDWPHQW E X = } 8 &RQWURO 93$ /< 93$ /< 7UHDWPHQW )LJXUH (IIHFWV RI 93$ DQG /< RQ +1 FHOO GHQVLW\ DQG QHXULWH RXWJURZWK +1 FHOOV ZHUH H[SRVHG WR HLWKHU P0 93$ RU P0 /< RU WKH FRPELQDWLRQ IRU K &HOOV ZHUH WKHQ SKRWRJUDSKHG DQG DQDO\]HG WR GHWHUPLQH Df FHOO QXPEHU DQG Ef SHUFHQW RI FHOO SRSXODWLRQ ZLWK QHXULWHV 'DWD DUH SUHVHQWHG DV Df PHDQ QXPEHU RI FHOOV LQ WKH YLHZLQJ ILHOG s 6(0 rSf DQG Ef PHDQ SHUFHQW RI FHOO SRSXODWLRQ ZLWK QHXULWHV s 6(0 rS rrSf

PAGE 106

(IIHFW RI 93 $ RQ 1HXULWH 2XWJURZWK LQ 3& &HOOV 6LPLODU WR WKH HIIHFWV RI 93 $ REVHUYHG LQ +1 FHOOV D FRQFHQWUDWLRQGHSHQGHQW LQFUHDVH LQ QHXULWH RXWJURZWK LQ 3& FHOOV RFFXUUHG IROORZLQJ G H[SRVXUH WR WKHUDSHXWLF FRQFHQWUDWLRQV RI 93 $ Sf DV VKRZQ LQ )LJXUH ,Q FHOOV H[SRVHG WR D KLJKHU FRQFHQWUDWLRQ RI 93 $ P0f QHXULWH RXWJURZWK DSSHDUHG WR EH QR GLIIHUHQW IURP WKDW RI FRQWUROV ,W LV RI QRWH WKDW VLPLODU WR HIIHFWV REVHUYHG LQ +1 FHOOV WKH 3& FHOOV H[KLELWHG DQ LQYHUVH FRUUHODWLRQ EHWZHHQ FHOO GHQVLW\ DQG QHXULWH RXWJURZWK S GDWD QRW VKRZQf )LJXUH 93$LQGXFHG QHXULWH RXWJURZWK LQ 3& FHOOV &HOOV ZHUH H[SRVHG WR 93$ IRU G DQG WKHQ SKRWRJUDSKV ZHUH WDNHQ IRU FRPSDULVRQ 'DWD DUH SUHVHQWHG DV PHDQ SHUFHQW s 6(0f RI FHOOV GLVSOD\LQJ RQH QHXULWH RU PRUH rS (IIHFWV RI 93$ DQG WKH 3.& ,QKLELWRU %,0 RQ 3& &HOO 1XPEHU DQG 1HXULWH 2XWJURZWK 1RQGLIIHUHQWLDWHG 3& FHOOV H[SRVHG WR HLWKHU P0 RU P0 93$ H[KLELWHG QR DOWHUDWLRQ LQ JURZWK UDWH FRPSDUHG WR FRQWUROV DV HYLGHQFHG E\ ODFN RI GLIIHUHQFH LQ FHOO GHQVLW\ )LJXUH D DQG Ef 6LPLODUO\ Q0 %,0 HLWKHU DORQH RU LQ FRPELQDWLRQ ZLWK 93$ HOLFLWHG QR DOWHUDWLRQ LQ RYHUDOO FHOO GHQVLW\ )ROORZLQJ H[SRVXUH WR P0

PAGE 107

93 $ 3& FHOOV H[KLELWHG LQFUHDVHG QHXULWH RXWJURZWK ZKLFK ZDV XQDIIHFWHG E\ WKH DGGLWLRQ RI Q0 %,0 S )LJXUH Ff 1HLWKHU %,0 QRU P0 93 $ DORQH RU LQ FRPELQDWLRQ KDG DQ\ HIIHFW RQ QHXULWH RXWJURZWK LQ 3& FHOOV )LJXUH Gf 7KHVH UHVXOWV VXJJHVW WKDW DW FRQFHQWUDWLRQV RI 93 $ WKHUDSHXWLF IRU PRRGVWDELOL]DWLRQ HJ P0f 3& FHOOV H[KLELW LQFUHDVHG QHXULWH RXWJURZWK DQG WKDW WKLV HIIHFW LV LQGHSHQGHQW RI 3.& D E &RQWURO P0 93$ Q0 %,0 93$ %,0 7UHDWPHQW &RQWURO P0 93$ Q0 %,0 93$ %,0 7UHDWPHQW )LJXUH (IIHFWV RI 93$ DQG %,0 D 3.& ,QKLELWRU RQ 3& &HOO 'HQVLW\ DQG 1HXULWH 2XWJURZWK 1RQGLIIHUHQWLDWHG 3& FHOOV ZHUH H[SRVHG WR P0 93$ P0 93$ Q0 %,0 RU VRPH FRPELQDWLRQ WKHUHRI IRU K 6KRZQ LQ Df DQG Ef DUH PHDQ FHOO GHQVLW\ s 6(0f IROORZLQJ H[SRVXUH WR P0 93$ DQG P0 93$ UHVSHFWLYHO\ 6KRZQ LQ Ff DQG Gf DUH SHUFHQW RI WRWDO SRSXODWLRQ H[KLELWLQJ QHXULWHV s 6(0f IROORZLQJ H[SRVXUH WR P0 93$ P0 93$ Q0 %,0 RU D FRPELQDWLRQ rS

PAGE 108

3RVW1*) 7 UHDWPHQW )LJXUH (IIHFW RI 93 $ RQ 1*)'LIIHUHQWLDWHG 3& &HOOV 'LIIHUHQWLDWLRQ RI 3& FHOOV ZDV LQGXFHG E\ G H[SRVXUH WR QJPO 1*) DQG WKHQ 1*) ZDV HLWKHU UHPRYHG RU UHSODFHG DQG 93$ ZDV DGGHG DW YDULRXV FRQFHQWUDWLRQV P0f 1*)SRVLWLYH FRQWURO LV UHSUHVHQWHG DV EDVHOLQH RQ \D[LVf Df &HOO 'HQVLW\ 6KRZQ LV PHDQ s 6(0f QXPEHU RI FHOOV H[SUHVVHG DV D SHUFHQW RI 1*)SRVLWLYH FRQWURO YDOXHV 1R SURWHFWLYH HIIHFW ZDV REVHUYHG Ef 1HXULWH 2XWJURZWK 6KRZQ LV PHDQ SHUFHQW s 6(0f RI FHOO SRSXODWLRQ H[KLELWLQJ QHXULWHV LQ WUHDWPHQW JURXSV FRPSDUHG WR 1*)SRVLWLYH FRQWUROV 3URWHFWLRQ DJDLQVW ORVV RI QHXULWHV ZDV REVHUYHG rS

PAGE 109

1HXURSURWHFWLYH (IIHFW RI 93 $ RQ 1*)'LIIHUHQWLDWHG 3& &HOOV )ROORZLQJ 1*)LQGXFHG GLIIHUHQWLDWLRQ RI 3& FHOOV YDU\LQJ FRQFHQWUDWLRQV RI 93$ P0f ZHUH DGGHG DORQH RU LQ FRPELQDWLRQ ZLWK 1*) QJPOf 1*)GHSULYHG FXOWXUHV H[SHULHQFHG D ORVV RI QHXULWHV DQG GHFUHDVHG FHOO QXPEHU DWWULEXWDEOH WR FHOO GHDWKf DV VKRZQ LQ )LJXUH 93$WUHDWHG FXOWXUHV DORQH RU LQ FRPELQDWLRQ ZLWK 1*)f H[KLELWHG D GHFUHDVHG UDWH RI QHXULWH ORVV WKRXJK WKLV HIIHFW ZDV QR JUHDWHU WKDQ WKDW HOLFLWHG E\ 1*) DORQH S )LJXUH Ef ([SRVXUH WR WKH FRPELQDWLRQ RI 93$ DQG 1*) UHVXOWHG LQ D WHQGHQF\ DOEHLW LQVLJQLILFDQW WRZDUG GHFUHDVHG FHOO GHDWK ZKLOH 93$ DORQH KDG QR DSSDUHQW HIIHFW )LJXUH Df 7KHVH UHVXOWV VXJJHVW WKDW 93$ LV SURWHFWLYH DJDLQVW QHXULWH ORVV LQ 1*)GLIIHUHQWLDWHG 3& FHOOV EXW QRW SURWHFWLYH DJDLQVW FHOO ORVV DVVRFLDWHG ZLWK WURSKLF IDFWRU GHSULYDWLRQ 'LVFXVVLRQ +1 &HOOV ([SRVXUH RI +1 FHOOV WR WKHUDSHXWLF FRQFHQWUDWLRQV XS WR P0f RI 93$ IRU XS WR G SURGXFHG QR DSSDUHQW F\WRWR[LFLW\ DQG FHOOV DSSHDUHG KHDOWK\ DQG YLDEOH ZKHQ DVVHVVHG E\ 7U\SDQ EOXH G\H H[FOXVLRQ DVVD\ GDWD QRW VKRZQf +RZHYHU D FRQFHQWUDWLRQ GHSHQGHQW LQFUHDVH LQ FHOO GRXEOLQJ WLPH ZDV QRWHG )ROORZLQJ ORQJWHUP H[SRVXUH FHOOXODU JURZWK UDWHV ZHUH QRWDEO\ VORZHU IRU +1 FHOOV H[SRVHG WR P0 93$ 7KLV HIIHFW ZDV UHYHUVLEOH XSRQ UHPRYDO RI WKH GUXJ DQG FRXOG EH DYRLGHG DOWRJHWKHU E\ SODWLQJ FHOOV PRUH GHQVHO\ 7DEOH f 7KHVH ILQGLQJV DUH VLPLODU WR UHSRUWHG HIIHFWV RI 93$ RQ JOLRPD &f DQG QHXUREODVWRPD QHXUR $ 8.)1% 8.)1%f FHOOV LQ ZKLFK 93$ H[KLELWHG D UHYHUVLEOH DQWLSUROLIHUDWLYH HIIHFW ZKLFK ZDV QRW DWWULEXWDEOH WR F\WRWR[LFLW\

PAGE 110

0DUWLQ DQG 5HJDQ &LQDWO HW DO f ,Q RQH SDUWLFXODU VWXG\ 93 $ ZDV IRXQG WR DUUHVW FHOOV DW WKH SRVWPLWRWLF PLG*O SKDVH RI WKH FHOO F\FOH 0DUWLQ DQG 5HJDQ f 7KH QH[W SDUDPHWHU DVVHVVHG LQ WKH +1 FHOOV ZDV WKDW RI GLIIHUHQWLDWLRQ )ROORZLQJ FKURQLF Gf H[SRVXUH WR WKHUDSHXWLF FRQFHQWUDWLRQV RI 93 $ FHOOV H[KLELWHG D VLJQLILFDQW LQFUHDVH LQ QHXULWH RXWJURZWK FRQVLVWHQW ZLWK SUHYLRXV UHSRUWV 0DUWLQ DQG 5HJDQ &LQDWO HW DO f 7KH LQFUHDVH LQ QHXULWH RXWJURZWK REVHUYHG LQ WKLV VWXG\ RFFXUUHG RYHU D WLPH FRXUVH VLPLODU WR WKDW RI WKH 0$5&.6 GRZQUHJXODWLRQ REVHUYHG VHH &KDSWHU f :KLOH QR VLJQLILFDQW DOWHUDWLRQV LQ HLWKHU SDUDPHWHU ZHUH REVHUYHG DW G IROORZLQJ P0 93 $ H[SRVXUH ERWK FHOO PRUSKRORJ\ QHXULWH RXWJURZWKf DQG 0$5&.6 H[SUHVVLRQ KDG XQGHUJRQH VLJQLILFDQW FKDQJHV E\ G )ROORZLQJ FRQWLQXHG 93$ H[SRVXUH OLWWOH RU QR FKDQJH ZDV HYLGHQW LQ HLWKHU 0$5&.6 RU FHOO PRUSKRORJ\ EHWZHHQ G DQG RU G EXW D VLJQLILFDQW LQFUHDVH LQ QHXULWH RXWJURZWK ZDV REVHUYHG E\ WKH G WLPH SRLQW ZKLFK FRUUHVSRQGV URXJKO\ WR WKH IXUWKHU GHFUHDVH LQ 0$5&.6 SURWHLQ REVHUYHG EHWZHHQ DQG G VHH )LJXUH f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f )XUWKHU H[SRVXUH RI +1 FHOOV WR HLWKHU RI WKH 93$ PHWDEROLWHV K\GUR[\YDOSURLF DFLG RU SURS\OJOXWDULF DFLG ZKRVH VWUXFWXUHV DUH DQDODJRXV WR WKDW RI 93$ SURGXFHG QR FKDQJH LQ FHOO PRUSKRORJ\ DQG RQO\ RQH PHWDEROLWH DW KLJK FRQFHQWUDWLRQV KDG D QRWDEOH HIIHFW RQ 0$5&.6 2Q WKH RWKHU KDQG UHWLQRLF DFLG D NQRZQ GLIIHUHQWLDWLQJ DJHQW KDV EHHQ VKRZQ WR HOLFLW ERWK 0$5&.6 GRZQUHJXODWLRQ DQG QHXULWH RXWJURZWK LQ WKHVH FHOOV :DWVRQ HW DO f )XUWKHU WKH HIIHFW RI OLWKLXP RQ 0$5&.6 GLIIHUV IURP WKDW RI 93$ LQ WKH +1

PAGE 111

FHOOV :KHUHDV OLWKLXP DSSHDUV WR H[HUW LWV JUHDWHVW HIIHFW RQ 0$5&.6 LQ WKH VROXEOH IUDFWLRQ RI WKHVH FHOOV DQG UHTXLUHV D ORQJHU H[SRVXUH 93$ H[SRVXUH UHVXOWV LQ D JUHDWHU UHGXFWLRQ LQ 0$5&.6 H[SUHVVLRQ DW D PXFK HDUOLHU WLPH SRLQW G IRU 93$ YV G IRU OLWKLXPf DQG WKH HIIHFW LV PRVW SURQRXQFHG LQ WKH PHPEUDQH IUDFWLRQ RI WKHVH FHOOV $V KDV EHHQ VXJJHVWHG SUHYLRXVO\ 93$ DQG OLWKLXP OLNHO\ RSHUDWH WKURXJK WZR GLIIHUHQW RU RQO\ SDUWLDOO\ RYHUODSSLQJ PHFKDQLVPV :KLOH OLWKLXP ZRUNV LQ SDUW WKURXJK LWV HIIHFW RQ WKH 3, VLJQDOLQJ V\VWHP 93$ GRHV QRW 6LPLODUO\ 93$ H[SRVXUH OHDGV WR HIIHFWV RQ FHOO JURZWK DQG PRUSKRORJ\ ZKLFK PD\ RU PD\ QRW EH UHODWHG WR LWV WKHUDSHXWLF HIILFDF\ ,QGHHG WKH REVHUYHG 93$LQGXFHG HIIHFWV PRUH WKDQ OLNHO\ FRQWULEXWH WR WKLV GUXJf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f FRQWLQXHG WR LQFUHDVH WKURXJKRXW WKH OHQJWK RI 93$ H[SRVXUH H[DPLQHG 7KLV DORQJ ZLWK WKH IDFW WKDW OLWKLXP KDG QR HIIHFW RQ HLWKHU *$3 H[SUHVVLRQ RU QHXULWH RXWJURZWK GDWD QRW VKRZQf VXJJHVWV WKDW WKH HIIHFWV DUH VRPHKRZ UHODWHG $OWKRXJK QHFHVVLW\ DQGRU VXIILFLHQF\ KDYH QRW EHHQ DGGUHVVHG VXIILFLHQWO\ E\ WKH GDWD SUHVHQWHG WKHUH LV D ZHDOWK RI HYLGHQFH ZKLFK VXSSRUWV WKH QRWLRQ WKDW LQFUHDVHG *$3 OHYHOV JUHDWO\ FRQWULEXWH WR WKH RYHUDOO SURFHVV RI QHXULWH RXWJURZWK DQG FHOO GLIIHUHQWLDWLRQ $V

PAGE 112

GLVFXVVHG LQ &KDSWHU QHXULWH RXWJURZWK LV DFFRPSDQLHG E\ D ODUJH LQFUHDVH LQ *$3 H[SUHVVLRQ DQG LV WKRXJKW WR KHOS VWDELOL]H JURZWK FRQH PRUSKRORJ\ DQG SURPRWH JURZWK FRQH DGKHVLRQ $LJQHU DQG &DURQL f $QWLVHQVHGLUHFWHG GHSOHWLRQ RI *$3 LQ JURZWK FRQHV UHVXOWV LQ WKH DEVHQFH RI VSUHDGLQJ EUDQFKLQJ DQG DGKHVLRQ RI JURZWK FRQHV DQG WKHUHIRUH QHXULWH RXWJURZWKf LQ SULPDU\ VHQVRU\ QHXURQV $LJQHU DQG &DURQL f 6XSSUHVVLRQ RI *$3 H[SUHVVLRQ KDV EHHQ IRXQG WR \LHOG DGYHUVH HIIHFWV RQ D[RQ RXWJURZWK ERWK LQ YLYR DQG LQ YLWUR 0HLUL HW DO %HQRZLW] DQG 5RXWWHQEHUJ 6WULWWPDWWHU HW DO -DS 7MRHQ 6DQ HW DO 6KHD HW DO f DQG RYHUH[SUHVVLRQ RI *$3 LQGXFHG QHUYH VSURXWLQJ LQ WKH DGXOW QHUYRXV V\VWHP RI WUDQVJHQLF PLFH $LJQHU HW DO f (YHQ LQ QRQQHXURQDO FHOOV H[SUHVVLRQ RI D *$3 WUDQVJHQH KDV EHHQ IRXQG WR LQGXFH H[WHQVLYH SURFHVV RXWJURZWK DFFRPSDQLHG E\ UHRUJDQL]DWLRQ RI WKH PHPEUDQH F\WRVNHOHWRQ 6WULWWPDWWHU HW DO
PAGE 113

IROORZHG E\ GRZQUHJXODWLRQ RFFXUUHG ZLWKLQ KRXUV IROORZLQJ DGGLWLRQ RI 93 $ WR WKH FXOWXUH PHGLXP ZKHUHDV QHXULWH RXWJURZWK ZDV REVHUYHG RQO\ IROORZLQJ FKURQLF 93$ H[SRVXUH 6LPLODUO\ DOWHUDWLRQV LQ 3.& LVRIRUP H[SUHVVLRQ ZHUH REVHUYHG DFXWHO\ ZLWKLQ K IROORZLQJ 93$ DGGLWLRQ DV ZHOO DV IROORZLQJ ORQJHUWHUP 93$ H[SRVXUH Gf $OWKRXJK UHSRUWV DUH LQFRQFOXVLYH DQG HYHQ FRQWUDGLFWRU\ DV WR WKH HLWKHU SRVLWLYH $EUDKDP HW DO 3DUURZ HW DO &DEHO DQG $XGHVLUN &ROHPDQ DQG :RRWHQ f RU QHJDWLYH +HLNNLODH DO 3DURGLH DO 7VXQHLVKL &DUOVRQ HW DO -DODYD HW DO &KDNUDYDUWK\ HW DO (NLQFL DQG 6KHD f UHJXODWRU\ UROH RI 3.& LQ QHXURQDO GLIIHUHQWLDWLRQ HDFK RI WKH IRXU 3.& LVR]\PHV SUHVHQW LQ +1 FHOOV KDV LQ RQH UHSRUW RU DQRWKHU EHHQ LPSOLFDWHG LQ QHXULWH RXWJURZWK DQGRU FHOOXODU GLIIHUHQWLDWLRQ DV GLVFXVVHG DW OHQJWK LQ &KDSWHU 7KH SUHVHQW VWXG\ VXSSRUWV WKH DVVRFLDWLRQ RI QHXULWH RXWJURZWK ZLWK HLWKHU RQH RU DQ\ FRPELQDWLRQ RI WKH IROORZLQJ Df DFXWH LQGXFWLRQ IROORZHG E\ GRZQUHJXODWLRQ RI 3.&D H[SUHVVLRQ Ef FKURQLF XSUHJXODWLRQ RI 3.& Ff DSSDUHQW WUDQVORFDWLRQ FRQVLVWHQW ZLWK DFWLYDWLRQf RI 3.&H IURP F\WRVRO WR PHPEUDQH IROORZHG E\ GRZQUHJXODWLRQ DQGRU Gf FKURQLF XS UHJXODWLRQ RI 3.&7KH ILQGLQJV GLVFXVVHG KHUHLQ DUH FRQVLVWHQW ZLWK SUHYLRXV VWXGLHV VKRZLQJ DOWHUDWLRQV LQ 3.&D P51$ DQGRU SURWHLQ H[SUHVVLRQ REVHUYHG FRQFRPLWDQW ZLWK GLIIHUHQWLDWLRQ RI 3& DQG RWKHU FHOO OLQHV %RUJDWWL HW DO 3DUURZ HW DO 0XUUD\ HW DO /HOL HWDO 7RQLQL HWDO :DGD HW DO f +RZHYHU WKHVH UHSRUWV GLIIHU LQ WKHLU PHFKDQLVWLF FRQFOXVLRQV IRU LQ WKUHH VWXGLHV 3.&D H[SUHVVLRQ LQFUHDVHG ZLWK GLIIHUHQWLDWLRQ %RUJDWWL HW DO 3DUURZ HW DO 0XUUD\ HW DO f ZKLOH LQ WKUHH RWKHU VWXGLHV 3.&D OHYHOV KDG GHFUHDVHG IROORZLQJ FRPSDUDEOH WUHDWPHQW DQG QHXULWRJHQHVLV /HOL HW DO 7RQLQL HW DO :DGD HW DO f 7KH ILQGLQJV SUHVHQWHG KHUHLQ DUH DOVR FRQVLVWHQW ZLWK UHSRUWV ZKLFK KDYH SURYLGHG HYLGHQFH IRU D UROH RI 3.&V LQ UHJXODWLQJ QHXURQDO GLIIHUHQWLDWLRQ VLQFH QHXULWH RXWJURZWK KDV EHHQ DVVRFLDWHG ZLWK DOWHUHG 3.&H OHYHOV 3RQ]RQL HW DO 3DUURZ HW DO )DJHUVWURP HW DO (NLQFL DQG 6KHD +XQGOH HW DO =HLGPDQ HW

PAGE 114

DO f /HOL DQG :DGD REVHUYHG LQFUHDVHG H[SUHVVLRQ RI 3.&H FRQFRPLWDQW ZLWK QHXULWH RXWJURZWK /HOL HW DO :DGD HW DO f DQG DQRWKHU UHSRUW LQGLFDWHG WKDW RYHUH[SUHVVLRQ RI 3.&H OHG WR DQ LQFUHDVH LQ 1*)LQGXFHG QHXULWH RXWJURZWK LQ VWDEO\ WUDQVIHFWHG 3& FHOO OLQHV +XQGOH HW DO f 6WLOO RWKHU UHSRUWV HPSKDVL]HG WKH LPSRUWDQFH RI 3.&e LQ QHXURQDO GLIIHUHQWLDWLRQ VKRZLQJ WKDW 3.&e RYHUH[SUHVVLRQ LV DEOH WR LQGXFH FHOOXODU GLIIHUHQWLDWLRQ LQ WUDQVIHFWHG OHXNHPLF FHOOV :D\V HW DO f DQG DQWLVHQVHPHGLDWHG GRZQUHJXODWLRQ RI 3.&e LQKLELWV 1*)LQGXFHG QHXULWH RXWJURZWK LQ 3& FHOOV &ROHPDQ DQG :RRWHQ f ,Q WKLV VWXG\ LQKLELWLRQ RI 3.& E\ /< DORQH LQFUHDVHG QHXULWH RXWJURZWK LQ +1 FHOOV FRPSDUHG WR WKDW RI XQWUHDWHG FRQWUROV /< DOVR HQKDQFHG WKH 93$ LQGXFHG LQFUHDVH LQ QHXULWH RXWJURZWK DQG GHFUHDVHG +1 FHOO GHQVLW\ ZKHQ WKH WZR GUXJV ZHUH FRPELQHG 7KHVH GDWD VXJJHVW WKDW UDWKHU WKDQ VHUYLQJ DV WKH QHFHVVDU\ LQWHUPHGLDWH IRU WKH HIIHFWV RI 93 $ RQ +1 FHOO JURZWK DQG PRUSKRORJ\ DV LQLWLDOO\ K\SRWKHVL]HG RQH RU PRUH LVRIRUPV RI 3.& PD\ DFWXDOO\ EH LQKLELWRU\ WRZDUG FHOO GLIIHUHQWLDWLRQ 3.& LQKLELWLRQ QRW RQO\ HOLFLWHG HIIHFWV RQ LWV RZQ EXW HQKDQFHG WKH REVHUYHG HIIHFWV RI 93 $ RQ WKHVH SDUDPHWHUV 6XFK ILQGLQJV VXJJHVW WKDW 93$ PD\ RSHUDWH WKURXJK LQDFWLYDWLRQ RI 3.& WR HOLFLW REVHUYHG HIIHFWV RQ FHOO JURZWK DQG PRUSKRORJ\ RU SHUKDSV WKURXJK D VHSDUDWH PHFKDQLVP DOWRJHWKHU 7KH ILQGLQJ WKDW 3.& ZDV QRW D QHFHVVDU\ LQWHUPHGLDWH LQ WKH 93$LQGXFHG *$3 LQGXFWLRQ REVHUYHG &KDSWHU f LV FRQVLVWHQW ZLWK WKHVH GDWD VLQFH LQFUHDVHG *$3 H[SUHVVLRQ LV DVVRFLDWHG ZLWK FHOO GLIIHUHQWLDWLRQ DV GLVFXVVHG LQ &KDSWHU $ QRYHO ILQGLQJ REVHUYHG EXW QRW TXDQWLILHG LQ WKLV VWXG\ ZDV WKH LQFUHDVHG DGKHUHQFH RI 93$WUHDWHG FHOOV WR WKH WLVVXH FXOWXUH IODVNV LQ ZKLFK WKH\ ZHUH JURZQ $QRWKHU JURXS UHSRUWHG D VLPLODU HIIHFW LQ & JOLRPD DQG SULPDU\ DVWURF\WHV 0DUWLQ HW DO f )XUWKHU LQYHVWLJDWLRQ UHYHDOHG WKDW 93$ LQGXFHG D VLJQLILFDQW LQFUHDVH LQ WKH H[SUHVVLRQ RI WKH FROODJHQ W\SH ,9 UHFHSWRU D KHDWVHQVLWLYH SURWHLQ WKRXJKW WR EH UHVSRQVLEOH IRU WKH LQFUHDVHG DGKHUHQFH WR VXEVWUDWH 0DUWLQ DQG 5HJDQ f 7KH KHDW

PAGE 115

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f KDG QR HIIHFW RQ HLWKHU SDUDPHWHU $W D WKHUDSHXWLF FRQFHQWUDWLRQ RI 93$ P0f KRZHYHU QHXULWH RXWJURZWK LQFUHDVHG ZLWK RU ZLWKRXW DGGLWLRQ RI %,0 VXJJHVWLQJ WKDW f 93$ DW FRQFHQWUDWLRQV WKHUDSHXWLF IRU PRRG VWDELOL]DWLRQ PD\ VHUYH DV D GLIIHUHQWLDWLQJ DJHQW LQ 3& FHOOV DQG f 93$LQGXFHG GLIIHUHQWLDWLRQ RI WKHVH FHOOV GRHV QRW LQYROYH 3.& DV LQKLELWLRQ RI WKH HQ]\PH KDG QHLWKHU D SRVLWLYH QRU QHJDWLYH LPSDFW RQ WKH HIIHFW RI 93$ 7KHVH ILQGLQJV DUH FRQVLVWHQW ZLWK WKRVH REVHUYHG IROORZLQJ 93$ H[SRVXUH LQ WKH +1 FHOOV LQVRIDU DV WKHUDSHXWLF FRQFHQWUDWLRQV RI 93$ HOLFLWHG LQFUHDVHG QHXULWH RXWJURZWK D SURFHVV ZKLFK GLG QRW DSSHDU WR UHO\ RQ 3.& +RZHYHU WKH ILQGLQJV LQ 3& FHOOV GLIIHU IURP WKRVH RI WKH +1 FHOOV LQ WKDW 3.& LQKLELWLRQ GLG QRW HQKDQFH WKH

PAGE 116

93$LQGXFHG QHXULWH RXWJURZWK QRU GLG LW KDYH DQ\ HIIHFW ZKHQ DGGHG DORQH )XUWKHU 3& FHOO JURZWK ZDV XQDIIHFWHG E\ HLWKHU 93$ RU %,0 RU WKH FRPELQDWLRQ $ QXPEHU RI UHFHQW VWXGLHV KDYH LQYHVWLJDWHG WKH UROH RI 3.& LQ QHXURSURWHFWLRQ 5HVXOWV DUH FRQIOLFWLQJ EXW RIWHQ VXSSRUW VXFK D UROH IRU 3.& ,Q D PRGHO RI LVFKHPLF SUHFRQGLWLRQLQJLQGXFHG QHXURSURWHFWLRQ XWLOL]LQJ D WZRYHVVHO RFFOXVLRQ SURWRFROf DOWHUHG GLVWULEXWLRQ DQG OHYHOV RI 3.&\ ZHUH REVHUYHG LQ UDW KLSSRFDPSDO &$ DQG QHRFRUWH[ UHJLRQV VXJJHVWLQJ DQ DVVRFLDWLRQ EHWZHHQ 3.& DQG QHXURSURWHFWLRQ 6KDPORR DQG :LHORFK f
PAGE 117

2QH VHULHV RI H[SHULPHQWV ZDV GHVLJQHG WR WHVW WKH K\SRWKHVLV WKDW 93$ KDG QHXURSURWHFWLYH DFWLYLW\ LQ DGGLWLRQ WR LWV QHXULWHVWDELOL]LQJ UROH ,QDVPXFK DV YDULRXV DQLPDO PRGHOV KDYH SURYLGHG HYLGHQFH IRU SURJUHVVLYH VHL]XUHLQGXFHG QHXURQDO GDPDJH 93$ KDV EHHQ LQYHVWLJDWHG DV D SRWHQWLDO QHXURSURWHFWLYH DJHQW -DOORQ f )XUWKHUPRUH OLWKLXP ZDV VKRZQ LQ D UHFHQW UHSRUW WR EH QHXURSURWHFWLYH DJDLQVW 10'$ PHGLDWHG JOXWDPDWHLQGXFHG H[FLWRWR[LFLW\ DSRSWRVLVf LQ FRUWLFDO KLSSRFDPSDO DQG FHUHEHOODU JUDQXOH FHOOV 7KLV HIIHFW ZKLFK DSSDUHQWO\ LQYROYHG PRGXODWLRQ RI 10' $ PHGLDWHG &D LQIOX[ ZDV ERWK WLPH DQG FRQFHQWUDWLRQGHSHQGHQW DQG LQGHSHQGHQW RI LQRVLWRO FRQFHQWUDWLRQ WKHUHIRUH LQGHSHQGHQW RI ,03DVHf DQG QRW D UHVXOW RI DOWHUDWLRQV LQ HLWKHU 10'$ UHFHSWRU FRPSRVLWLRQ RU QXPEHU 1RQDND HW DO f 2WKHU VWXGLHV LQ FXOWXUHG FHUHEHOODU JUDQXOH FHOOV DV ZHOO DV UDW IURQWDO FRUWH[ KLSSRFDPSXV DQG VWULDWXP VKRZHG FKURQLF OLWKLXPLQGXFHG XSUHJXODWLRQ RI %&/ P51$ DQG SURWHLQ OHYHOV DV ZHOO DV SURWHFWLRQ DJDLQVW JOXWDPDWHLQGXFHG F\WRWR[LFLW\ &KHQ DQG &KXDQJ 0DQML HW DO f %FO LV D SURWRRQFRJHQH WKRXJKW WR EH SURWHFWLYH DJDLQVW FHOO GHDWK DULVLQJ IURP D QXPEHU RI GLIIHUHQW SDWKZD\V LQFOXGLQJ DSRSWRVLV DQG PD\ DOVR SOD\ D UROH LQ WKH UHJXODWLRQ RI QHXUDO FHOO GLIIHUHQWLDWLRQ 9DX[ 1DXPRYVNL DQG &OHDU\
PAGE 118

RQ GLIIHUHQW FHOO SRSXODWLRQV :LWK UHJDUG WR WKH QHXURSURWHFWLYH SRWHQWLDO RI 93 $ DJDLQVW DSRSWRVLV ILQGLQJV DUH LQFRQFOXVLYH 93$ H[KLELWHG SURWHFWLYH WHQGHQFLHV WRZDUG QHXULWHV EXW GLG QRW SURWHFW DJDLQVW FHOO ORVV IROORZLQJ 1*)LQGXFHG GLIIHUHQWLDWLRQ DQG VXEVHTXHQW ZLWKGUDZDO LQ 3& FHOOV 7KHVH ILQGLQJV DJUHH ZLWK UHFHQW SUHOLPLQDU\ REVHUYDWLRQV E\ :DWVRQ SHUVRQDO FRPPXQLFDWLRQf ZKLFK VKRZ QR HYLGHQFH IRU QHXURSURWHFWLRQ E\ DFXWH Kf 93$ H[SRVXUH DJDLQVW JOXWDPDWHLQGXFHG F\WRWR[LFLW\ LQ KLSSRFDPSDOO\GHULYHG +7 FHOOV 7KH DELOLW\ RI DQWLFRQYXOVDQWV VXFK DV 93$ WR GHFUHDVH WKH IUHTXHQF\ RU VHYHULW\ RI HSLOHSWLF HSLVRGHV GRHV QRW QHFHVVDULO\ FRQIHU RU UHIOHFW QHXURSURWHFWLYH SURSHUWLHV 6XFK GUXJV DUH SURSK\ODFWLF DJDLQVW IXUWKHU VHL]XUHLQGXFHG GDPDJH EXW WKHVH GUXJV DUH QRW JHQHUDOO\ FRQVLGHUHG WR SRVVHVV WKH SRZHU WR FRUUHFW WKH SDWKRORJ\ XQGHUO\LQJ HSLOHSV\ -DOORQ f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f 93$ KDV EHHQ WKH IRFXV RI PDQ\ VWXGLHV LQ WKLV UHJDUG DV LWV SUHQDWDO XVH KDV EHHQ DVVRFLDWHG ZLWK D KLJK LQFLGHQFH RI RIIVSULQJ ERP ZLWK YDULRXV G\VPRUSKLF ELUWK GHIHFWV FROOHFWLYHO\ UHIHUUHG WR DV WKH )HWDO 9DOSURDWH 6\QGURPH 'DVWRQ HW DO $UGLQJHU HW DO :LJHU HW DO :LQWHU HW DO &KHVVD DQG ,DQQHWWL

PAGE 119

7HLQ DQG 0DF*UHJRU 'L/LEHUWL HW DO &OD\ HW DO 1DX HW DO 'DOHQV HW DO f ,Q DGGLWLRQ WR D QXPEHU RI VRPDWLF GHIHFWV FRQVLVWHQWO\ REVHUYHG LQIDQWV H[SRVHG SUHQDWDOO\ WR 93$ DOVR H[KLELWHG D VWULNLQJO\ KLJKHU LQFLGHQFH RI QHXUDO WXEH GHIHFWV WKDQ WKH JHQHUDO SRSXODWLRQ %MHUNHGDO HW D 5REHUW DQG *XLEDXG /LQGKRXW DQG 6FKPLGW f $ QXPEHU RI DQLPDO VWXGLHV KDYH VKRZQ WKDW SUHQDWDO H[SRVXUH WR 93$ DW SK\VLRORJLFDOO\ UHOHYDQW SODVPD OHYHOV UHVXOWV LQ D WHQGHQF\ WRZDUG GHFUHDVHG VL]H DQG ZHLJKW RI RIIVSULQJ (KOHUV HW DO +DQVHQ DQG *UDIWRQ f DV ZHOO DV LQFUHDVHG LQFLGHQFH RI HPEU\ROHWKDOLW\ +DQVHQ HW DO E (KOHUV HW DO f 7KHUH LV DOVR RYHUZKHOPLQJ HYLGHQFH IRU WKH WHUDWRJHQLF HIIHFWV RI 93$ RQ WKH URGHQW QHXUDO WXEH GDWD LQGLFDWH WKDW HYHQ D VLQJOH WUHDWPHQW ZLWK 93$ DW FULWLFDO GHYHORSPHQWDO WLPHV FDQ UHVXOW LQ VXFK QHXUDO WXEH GHIHFWV DV H[HQFHSKDO\ VSLQD ELILGD DSHUWD DQG VSLQD ELILGD RFFXOWD LQ RIIVSULQJ +DQVHQ HW DO E 7URW] HW DO f +DQVHQnV JURXS VKRZHG WKDW RIIVSULQJ RI PLFH H[SRVHG LQ YLYR RQ JG WR RQH VXEFXWDQHRXV LQMHFWLRQ RI 93$ H[KLELWHG D VLJQLILFDQW LQFUHDVH LQ WKH UDWH RI H[HQFHSKDO\ DV FRPSDUHG WR FRQWUROV +DQVHQ HW DO Ef LQ D VHSDUDWH VWXG\ LQ YLWUR UDW HPEU\RV H[SRVHG WR 93$ IRU K EHJLQQLQJ RQ JG H[KLELWHG D KLJKHU LQFLGHQFH RI RSHQ QHXUDO WXEHV +DQVHQ DQG *UDIWRQ f 2WKHU VWXGLHV KDYH VKRZQ VLPLODU WHUDWRJHQLF DEQRUPDOLWLHV DV ZHOO DV GHIHFWV LQ FUDQLDO QHUYH GHYHORSPHQW RI LQ YLWUR FXOWXUHG UDW PRXVH RU FKLFN HPEU\RV H[SRVHG WR 93$ IRU MXVW K &KDWRW HW DO .HOO\ DQG 5HJDQ 1RVHO DQG .OHLQ *RIIORW HW DO f 5HSRUWV UHIOHFW FRPSDUDEOH LQFUHDVHV LQ H[HQFHSKDO\ DQG RWKHU QHXUDO WXEH GHIHFWV LQ PRXVH IHWXVHV EULHIO\ H[SRVHG WR 93$ LQ YLYR GXULQJ HDUO\ JHVWDWLRQ 7URW] HW DO (KOHUV HW DO f 7KHVH HIIHFWV DUH LQGLFDWLYH RI D GLVUXSWLRQ LQ IRUPDWLRQ RI WKH QHXUDO WXEH UDWKHU WKDQ VLPSO\ D QHXURGHYHORSPHQWDO GHOD\ .HOO\ DQG 5HJDQ f %RWK UHWURVSHFWLYH DQG SURVSHFWLYH VWXGLHV KDYH DWWHPSWHG WR H[DPLQH WKLV SKHQRPHQRQ EXW WKH SDWWHUQ RI PDOIRUPDWLRQV FDXVHG E\ 93$ DSSHDUV WR YDU\ JUHDWO\ DFURVV VSHFLHV (KOHUV HW DO f DQG ILQGLQJV GLIIHU IURP RQH UHSRUW WR WKH QH[W 2QH

PAGE 120

VWULNLQJ FRQVLVWHQF\ DPRQJ DOO VWXGLHV LV WKH VHOHFWLYLW\ RI 93 $ IRU GHIHFWV RI WKH PLGOLQH 3HUKDSV WKH VSHFLILFLW\ RI DFWLRQ RI 93 $ RQ SKDVH RI WKH FHOO F\FOH PD\ KHOS H[SODLQ WKH GUXJnV VSHFLILFLW\ LQ LQGXFLQJ QHXUDO WXEH GHIHFWV 0DUWLQ DQG 5HJDQ f 6RPH VWXGLHV VXJJHVW SHDN SODVPD OHYHOV RI 93 $ LWVHOI DQG QRW LWV PHWDEROLWHV DUH UHVSRQVLEOH IRU WKH GUXJnV WHUDWRJHQLFLW\ LW KDV DOVR EHHQ VXJJHVWHG WKDW WKH GUXJ DFFXPXODWHV WR D JUHDWHU GHJUHH LQ HPEU\RQLF WDUJHW WLVVXHV UHODWLYH WR PDWHUQDO SODVPD FRQFHQWUDWLRQV 0HQHJROD HW DO .HOO\ DQG 5HJDQ 0DUWLQ DQG 5HJDQ f 7KXV IDU QR FRQVHQVXV KDV EHHQ UHDFKHG DV WR WKH SXWDWLYH PHFKDQLVP RI WKLV GUXJnV WHUDWRJHQLF HIIHFWV DQG GDWD DUH HTXDOO\ LQFRQFOXVLYH ZLWK UHJDUG WR SUHYHQWLYH UHJLPHQV :KHUHDV IROLF DFLG VXSSOHPHQWDWLRQ KDV EHHQ UHSRUWHG WR SUHYHQW DQWLFRQYXOVDQWLQGXFHG QHXUDO WXEH GHIHFWV LQ UDWV DQG PLFH LQ YLYR (KOHUV HW DO 1RVHO DQG .OHLQ 7URW] HW DO f WKH LQ YLWUR HYLGHQFH LV QRW DV FRPSHOOLQJ ,Q YLWUR UHSRUWV LQ JHQHUDO KDYH VKRZQ D IDLOXUH RI IROLF DFLG RU LWV PHWDEROLWHV WR SUHYHQW 93$LQGXFHG QHXUDO WXEH GHIHFWV LQ PRXVH RU UDW HPEU\RV +DQVHQ DQG *UDIWRQ +DQVHQ HW DO f $GGLWLRQDOO\ SUHOLPLQDU\ ILQGLQJV IURP RXU ODERUDWRU\ LQ ZKLFK ZH XVHG WKH LPPRUWDOL]HG KLSSRFDPSDO FHOO PRGHO WR LQYHVWLJDWH WKH SUHYHQWLYH HIIHFW RI IROLF DFLG RU PHWKLRQLQH D IROLF DFLG PHWDEROLWHf RQ 93$LQGXFHG 0$5&.6 GRZQUHJXODWLRQ VKRZHG QR SUHYHQWLRQ RU UHYHUVDO RI WKH 0$5&.6 HIIHFW GDWD QRW VKRZQf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

PAGE 121

GXULQJ ZKLFK JURZWK VWRSV DQG VSHFLDOL]HG IXQFWLRQDO GLIIHUHQWLDWLRQ EHJLQV 7KH NH\ LV WKDW WKLV GLIIHUHQWLDWLRQ FRPPHQFHV DW D WLPH SULRU WR JURZWK FRPSOHWLRQ DQG WKHUHIRUH EHIRUH WKH FHOOV DUH RUJDQL]HG DW D WLVVXH OHYHO DV D UHVXOW WKH ZKROH RUJDQLVP H[KLELWV VXFK JURVV DEQRUPDOLWLHV DV RSHQ QHXUDO WXEH 2XU ILQGLQJV DQG WKRVH RI RWKHUV 0DUWLQ HW DO f LQ ZKLFK FKURQLF 93$ H[SRVXUH UHVXOWHG LQ GHFUHDVHG JURZWK UDWHV DQG LQFUHDVHG PRUSKRORJLFDO GLIIHUHQWLDWLRQ DUH FRQVLVWHQW ZLWK WKLV K\SRWKHVLV 5HJDQfV JURXS IXUWKHU VXJJHVWV WKDW DJHQWV VXFK DV 93$ ZKLFK VLJQLILFDQWO\ DOWHU PLWRVLV DQG FHOO VXEVWUDWXP DGKHVLYLW\ FDQ EH SUHGLFWHG WR LQGXFH QHXUDO WXEH GHIHFWV IRU FHOO SUROLIHUDWLRQ DQG UHODWHG IXQFWLRQV DUH NQRZQ WR EH LQWHJUDO WR QHXUDO WXEH IRUPDWLRQ 0DUWLQ DQG 5HJDQ f ,W LV DOVR RI QRWH WKDW ZKLOH SUHQDWDO 93$ H[SRVXUH RIWHQ OHDGV WR QHXUDO WXEH GHIHFWV ODFN RI 0$5&.6 SURWHLQ UHVXOWV LQ VLPLODU HIIHFWV ,Q D WUDQVJHQLF PRXVH PRGHO KRPR]\JRXV QXOO IRU WKH 0$5&.6 JHQH 0DFV DOO SXSV GLHG SHULQDWDOO\ DQG H[KLELWHG D QXPEHU RI QHXUDO WXEH DQG RWKHU PLGOLQH GHIHFWV 6WXPSR HW D %ODFNVKHDU HW D f 6ZLHUF]\QVNL HW DO f VXEVHTXHQWO\ VKRZHG WKDW WUDQVJHQLF H[SUHVVLRQ RI WKH KXPDQ 0$5&.6 JHQH 0$&6 LQ WKHVH NQRFNRXW PLFH HIIHFWLYHO\ UHVFXHG WKH 0$5&.6 GHILFLHQW SKHQRW\SH DQG DOORZHG IRU VXUYLYDO RI WKH WUDQVJHQLF PLFH 7KHVH VWXGLHV VXJJHVW WKDW 0$5&.6 LV HVVHQWLDO IRU QRUPDO &16 GHYHORSPHQW DQG WKDW ODFN RI 0$5&.6 UHVXOWV LQ DEQRUPDO GHYHORSPHQW RI YDULRXV PLGOLQH VWUXFWXUHV DQG HYHQ GHDWK 7KHVH ILQGLQJV DUH RI LQWHUHVW LQ OLJKW RI WKH SUHVHQW VWXG\ LQ ZKLFK ZH KDYH SUHVHQWHG HYLGHQFH IRU 93$LQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 DQG DVVRFLDWHG PRUSKRORJLFDO DOWHUDWLRQV 2XU ILQGLQJV VXJJHVW D SXWDWLYH PHFKDQLVP IRU WKH GHYHORSPHQWDO DEQRUPDOLWLHV DVVRFLDWHG ZLWK SUHQDWDO 93$ H[SRVXUH 7KDW WKHUH LV H[WHQVLYH HYLGHQFH RI 93$LQGXFHG QHXUDO WXEH GHIHFWV DQG WKDW 93$ HOLFLWV D GRZQUHJXODWLRQ LQ 0$5&.6 D SURWHLQ ZLWKRXW ZKLFK QRUPDO QHXUDO WXEH GHYHORSPHQW FDQQRW SURFHHG LV LQWULJXLQJ :KLOH RXU ILQGLQJV DUH PHUHO\ VXJJHVWLYH DQG PRUH H[WHQVLYH VWXGLHV ZLOO QHHG WR EH GRQH WKLV LV SHUKDSV WKH

PAGE 122

,OO ILUVW DWWHPSW WR DWWULEXWH 0$5&.6 RU ODFN WKHUHRIf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f 7KH QRWLRQ RI V\QDSWLF UHPRGHOLQJ DV WKH XQGHUO\LQJ PHDQV IRU WKH GUXJ RU WKHUDS\LQGXFHG VWDELOL]DWLRQ RI PRRG GLVRUGHUV KDV DOVR EHHQ VXSSRUWHG E\ VRPH UHSRUWV DV SUHYLRXVO\ GLVFXVVHG 2QH VWXG\ IRXQG LQFUHDVHG OHYHOV RI 1&$0LPPXQRUHDFWLYH SURWHLQV LQ WKH FHUHEURVSLQDO IOXLG RI D SRSXODWLRQ RI ELSRODU SDWLHQWV 3ROWRUDN HO D f ZKLFK PD\ EH VXJJHVWLYH RI DEQRUPDO V\QDSWLF SODVWLFLW\ LQ DIIHFWHG LQGLYLGXDOV /LWKLXP DQG 93$ DIIHFW WUDQVFULSWLRQDO IDFWRUV ZKLFK UHJXODWH JHQH H[SUHVVLRQ DQG LW LV NQRZQ WKDW ORQJn WHUP FKDQJHV LQ QHXURQDO V\QDSWLF IXQFWLRQ DUH GHSHQGHQW XSRQ WKH VHOHFWLYH UHJXODWLRQ RI JHQH H[SUHVVLRQ 0DQML DQG /HQR[ f /DPRWULJLQH ZDV UHFHQWO\ VKRZQ WR DOWHU QHXURQDO K\SHUH[FLWDELOLW\ YLD LQKLELWLRQ RI ERWK 1D DQG &D FKDQQHOV SHUKDSV VLJQLI\LQJ PRGLILFDWLRQ RI V\QDSWLF SODVWLFLW\ DV WKH DXWKRUV VXJJHVW ;LH DQG +DJDQ f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

PAGE 123

ZH KDYH VKRZQ WKHVH PROHFXODU HYHQWV RFFXU IROORZLQJ FKURQLF EXW QRW DFXWH 93 $ H[SRVXUH DQG WKLV VSHDNV WR WKH WKHUDSHXWLF UHOHYDQFH RI RXU ILQGLQJV WKH IXQFWLRQDO FKDQJHV QHFHVVDU\ IRU ORQJWHUP PRRG VWDELOL]DWLRQ PXVW EH VXFK WKDW WKH\ ODJ LQ RQVHW DUH QRW HDVLO\ UHYHUVHG

PAGE 124

&+$37(5 &21&/86,216 $1' )8785( ',5(&7,216 93$ KDV SURYHQ WR EH YHU\ XVHIXO LQ LWV DSSOLFDWLRQ WR ERWK VHL]XUH GLVRUGHUV DQG ELSRODU GLVRUGHUV %HFDXVH RI WKH UHODWLYH GHDUWK RI PRRGVWDELOL]LQJ WKHUDSLHV DYDLODEOH WRGD\ LW LV WKH ODWWHU DSSOLFDWLRQ RQ ZKLFK WKHVH UHVHDUFK HIIRUWV KDYH EHHQ IRFXVHG 2YHU WKH SDVW WZR GHFDGHV D QXPEHU RI UHYLHZV KDYH EHHQ SXEOLVKHG RQ WKH HIILFDF\ RI 93$ LQ WUHDWLQJ ELSRODU DIIHFWLYH GLVRUGHU .HFN DQG 0F(OUR\ 0F(OUR\ DQG .HFN 0F(OUR\ +D\HV %URZQ )DZFHWW f )URP YDULRXV VWXGLHV LW KDV EHFRPH FOHDU WKDW 93$ HOLFLWV D VLJQLILFDQW DQWLPDQLF UHVSRQVH UDWH HYHQ LQ OLWKLXP UHIUDFWRU\ SDWLHQWV DQG RWKHUV ZKR GLVSOD\ PRUH UHVLVWDQW IRUPV RI ELSRODU GLVRUGHU VXFK DV WKH UDSLG F\FOLQJ YDULDQW &DODEUHVH DQG 'HOXFFKL %URZQ f )XUWKHU 93$ KDV SURYHQ WR EH TXLWH XVHIXO DQG VDIH LQ FRPELQDWLRQ ZLWK OLWKLXP RU DV D ORQJWHUP SURSK\ODFWLF GUXJ WKHUDS\ *UDQQHPDQ HO D +D\HV f %HFDXVH RI LWV EURDG WKHUDSHXWLF UDQJH DFWXDOO\ WKRXJKW WR EH HYHQ EURDGHU WKDQ WKH IRUPHUO\ HVWDEOLVKHG QJPO UDQJHf 93$ FDQ EH SUHVFULEHG ZLWK OHVV FRQVLGHUDWLRQ RI ULVN DQG OHVV FORVH PRQLWRULQJ ,Q IDFW 93$ KDV UHFHLYHG PRUH DWWHQWLRQ DV D ILUVWOLQH PRRG VWDELOL]HU EHIRUH OLWKLXP RU FDUEDPD]HSLQHf EHFDXVH RI LWV HIILFDF\ UHODWLYHO\ UDSLG RQVHW RI DFWLRQ IDYRUDEOH VLGH HIIHFW SURILOH DQG UHODWLYH HDVH RI DGPLQLVWUDWLRQ 0F(OUR\ DQG .HFN f 7KH ILQGLQJV RI RXU FXUUHQW DQG SUHYLRXV VWXGLHV DUH VXPPDUL]HG LQ 7DEOH %RWK FKURQLF 93$ DQG OLWKLXP VKDUH WKH SURSHUW\ RI UHJXODWLQJ 0$5&.6 H[SUHVVLRQ DW FOLQLFDOO\ UHOHYDQW FRQFHQWUDWLRQV P0 93$ ZDV XVHG IRU DOO H[SHULPHQWV LQ WKLV VWXG\f WR WKH H[FOXVLRQ RI DOO RWKHU SV\FKRWURSLF DJHQWV H[DPLQHG WR GDWH &RPSDULVRQV EHWZHHQ 93$ DQG OLWKLXP DUH LQVWUXFWLYH IRU WKHVH WZR )'$DSSURYHG PRRGVWDELOL]LQJ

PAGE 125

DJHQWV VKDUH D VWULNLQJ QXPEHU RI GLIIHUHQFHV DQG IXUWKHU LQYHVWLJDWLRQ PD\ EHJLQ WR DQVZHU VRPH YHU\ LPSRUWDQW TXHVWLRQV UHJDUGLQJ SDWKRSK\VLRORJ\ DV ZHOO DV GUXJ WKHUDS\ ,QGHHG WKH PRRGVWDELOL]LQJ SURSHUWLHV RI WKHVH WZR GUXJV PD\ EH SDUWLDOO\ DWWULEXWDEOH WR WKH VLPLODULWLHV REVHUYHG RQ D FHOOXODU OHYHO VXFK DV WKH VLPLODU HIIHFWV RQ 3.&Df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f DV ZHOO DV VRPH OHVV FRPPRQO\ XVHG PRRGVWDELOL]LQJ DJHQWV LQFOXGLQJ ODPRWULJLQH GDWD QRW VKRZQf DQG YHUDSDPLO GLG QRW UHVXOW LQ DOWHUDWLRQV LQ 0$5&.6 H[SUHVVLRQ SRLQWV WR WKH SRVVLELOLW\ WKDW HLWKHU f 0$5&.6 GRZQUHJXODWLRQ LWVHOI LV QRW LQWHJUDO WR PRRGVWDELOL]LQJ HIILFDF\ RU f WKH LQ YLWUR H[SRVXUH FRQGLWLRQV DQG PRGHO HPSOR\HG ZHUH QRW RSWLPDO WR

PAGE 126

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

PAGE 127

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f NQRZQ WR EH LQWHJUDO WR QHXUDO WXEH IRUPDWLRQ WKH ZKROH RUJDQLVP H[KLELWV VXFK JURVV DEQRUPDOLWLHV DV RSHQ QHXUDO WXEH 2WKHU UHSRUWV DUH FRQVLVWHQW ZLWK WKLV K\SRWKHVLV DQG WKH ILQGLQJV GLVFXVVHG KHUHLQ :LON HW DO 0XPPHU\ HW DO 0DUWLQ DQG 5HJDQ 0DUWLQ HW DO f ,Q IDFW WKH 93$LQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 DOVR ILWV ZHOO DQG PD\ FRQWULEXWH IXUWKHU WR WKLV PRGHO $V PHQWLRQHG SUHYLRXVO\ PLFH ODFNLQJ 0$5&.6 GXH WR DQWLVHQVH PHGLDWHG JHQH NQRFNRXW GLH SHULQDWDOO\ H[KLELWLQJ D QXPEHU RI JURVV DEQRUPDOLWLHV LQ PLGOLQH VWUXFWXUH IRUPDWLRQ 6WXPSR HW D %ODFNVKHDU HW DO f 3HUKDSV WKH UDWLR RI 0$5&.6 WR *$3 LV DQ LPSRUWDQW VLJQDO ZKLFK ZKHQ RXW RI EDODQFH OHDGV WR D SDWKRORJLFDO HYHQ IDWDO SKHQRW\SH $FFRUGLQJ WR WKLV K\SRWKHVLV

PAGE 128

UHODWLYHO\ KLJKHU 0$5&.6 H[SUHVVLRQ FRQFRPLWDQW ZLWK ORZ *$3 H[SUHVVLRQ ZRXOG VLJQDO IRU JURZWK FHOO GLYLVLRQf DQG ORZ 0$5&.6 KLJK *$3 ZRXOG VLJQDO IRU JURZWK DUUHVW DQG LQGXFWLRQ RI GLIIHUHQWLDWLRQ 7KH GDWD DUH WKXV IDU FRQVLVWHQW ZLWK VXFK D PRGHO DQG IXWXUH VWXGLHV PLJKW IXUWKHU DGGUHVV WKH LVVXH E\ H[DPLQLQJ DQG FRPSDULQJ UHODWLYH 0$5&.6 DQG *$3 OHYHOV LQ ERWK FRQWURO XQDIIHFWHGf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f DQG LV LPSOLFDWHG LQ /73 D PRGHO IRU PHPRU\ DQG OHDUQLQJ 6WDEHO DQG 3DUNHU %HQRZLW] DQG 5RXWWHQEHUJ 6RQ HW DO f LQ SDUW GXH WR LWV DVVRFLDWLRQ ZLWK VWDWH RI JURZWK DQG VSURXWLQJ RU QHXULWH H[WHQVLRQ %HQRZLW] DQG 5RXWWHQEHUJ f 0HLUL HW DO f VKRZHG WKDW QHXULWH RXWJURZWK VWLPXODWHG E\ QHXUDO FHOO DGKHVLRQ PROHFXOHV UHTXLUHG *$3 SKRVSKRU\ODWLRQ
PAGE 129

LV LQWHJUDO WR QHXULWH RXWJURZWK +RZHYHU WKH UROH RI 3.& LQ QHXURQDO GLIIHUHQWLDWLRQ UHPDLQV LQ TXHVWLRQ :KLOH WKHUH LV VXEVWDQWLDO HYLGHQFH IRU D SRVLWLYH UROH RI 3.& LQ UHJXODWLQJ QHXURQDO GLIIHUHQWLDWLRQ DQG PDLQWDLQLQJ WKH IXQFWLRQDOO\ DFWLYH VWDWH &ROHPDQ DQG :RRWHQ &DEHOO DQG $XGHVLUN 3DUURZ HW DO $EUDKDP HW DO 7RQLQL HW DO f DQG SUHYLRXV GDWD KDYH LPSOLFDWHG D QXPEHU RI GLIIHUHQW 3.& LVRIRUPV LQ FHOOXODU GLIIHUHQWLDWLRQ :DGD HW DO /HOL HWDO :D\V HW DO +XQGOH HW DO %RUJDWWL HW DO f RWKHU VWXGLHV KDYH VXJJHVWHG WKDW UDWKHU WKDW EHLQJ D SRVLWLYH UHJXODWRU RI WKH SURFHVV 3.& PD\ DFWXDOO\ EH DQ LQKLELWRU\ LQIOXHQFH VR WKDW VXSSUHVVLRQ RI WKH HQ]\PH UHVXOWV LQ LQFUHDVHG SURFHVV RXWJURZWK +HLNNLOD HW DO 3DURGL HW DO 7VXQHLVKL :RRWHQ &DUOVRQ HW DO -DODYD HW DO (NLQFL DQG 6KHD f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

PAGE 130

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f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f

PAGE 131

$QRWKHU LPSRUWDQW DUHD RI VWXG\ UHODWHG WR GLIIHUHQWLDWLRQ LV WKH LQYHVWLJDWLRQ RI WKH HIIHFWV RI 93 $ RQ JHQH UHJXODWLRQ IRU LQ DGGLWLRQ WR WKH UROH RI 3.& LQ DOWHULQJ SURWHLQ H[SUHVVLRQ E\ YLUWXH RI SKRVSKRU\ODWLRQ 3.& DQG RWKHU SURWHLQ NLQDVHV VXFK DV SURWHLQ NLQDVH $ DQG W\URVLQH NLQDVHf DV ZHOO DV SURWHLQ SKRVSKDWDVHV DUH NQRZQ WR EH LQYROYHG LQ DFWLYDWLRQ DQG LQDFWLYDWLRQ RI WUDQVFULSWLRQ IDFWRUV $3 DQG &5(% -DFNVRQ .DULQ DQG 6PHDO f $3 DQG &5(% DUH NQRZQ WR PRGXODWH H[SUHVVLRQ RI NH\ SURWHLQV ZKLFK DUH LQYROYHG LQ QHXURQDO H[FLWDELOLW\ VXFK DV UHFHSWRUV QHXURSHSWLGHV DQG NH\ HQ]\PHV LQ QHXURWUDQVPLWWHU ELRV\QWKHVLV DQG WKHVH SDUDPHWHUV DUH FHUWDLQO\ RI LQWHUHVW ZLWK UHJDUG WR HLWKHU HSLOHSV\ RU ELSRODU GLVRUGHU 0DQMLfV JURXS VKRZHG WKDW LQ & JOLRPD DQG KXPDQ QHXUREODVWRPD 6+6<
PAGE 132

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f DW QRQWR[LF FRQFHQWUDWLRQV 93$ QRW RQO\ DUUHVWHG FHOO JURZWK DQG LQGXFHG SKHQRW\SLF GLIIHUHQWLDWLRQ EXW DOVR LQFUHDVHG WKH LPPXQRJHQLFLW\ RI FHOOV 7KLV RFFXUUHG WKURXJK LQFUHDVHG H[SUHVVLRQ RI VXUIDFH QHXUDO FHOO DGKHVLRQ PROHFXOHV 1&$0Vf DQG LQFUHDVHG VHQVLWLYLW\ WR QDWXUDO NLOOHUO\PSKRNLQH DFWLYDWHG NLOOHU 1./$.f F\WRO\VLV WKURXJK QRQWR[LF PHFKDQLVPV &LQDWO HW DO f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

PAGE 133

H[SHULHQFHG LQ DWWHPSWLQJ WR DWWDLQ WKHUDSHXWLF VHUXP OHYHOV RI 93 $ IRU WKH UDWf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

PAGE 134

7DEOH (IIHFWV RI WZR PRRGVWDELOL]LQJ GUXJV 93$ DQG OLWKLXP DV REVHUYHG LQ +1 FHOOV IROORZLQJ FKURQLF H[SRVXUH $ 6XPPDU\ 7KH SUHVHQFH RI WKH GHVFULEHG HIIHFW LV LQGLFDWHG E\ ff DQG LQGLFDWHV WKH DEVHQFH RI WKH HIIHFW (IIHFW 93$ /LWKLXP fL0$5&.6 3URWHLQ D 0$5&.6 P51$ E 3, 6LJQDOLQJ F 7*$3 B E 3.& $FWLYLW\ E 3.&D E 73.& B E L3.&H E 7SNF L&HOO *URZWK B G 71HXULWH 2XWJURZWK B G D :DWVRQ HW DL E :DWVRQ DQG /HQR[ F /HQR[ HW DL G :DWVRQ DQG /HQR[

PAGE 135

5()(5(1&(6 $EUDKDP 6DPSVRQ .( 3RZHUV ( $ 0D\R 5XII 9$ DQG /HDFK ./ f ,QFUHDVHG 3.$ DQG 3.& DFWLYLWLHV DFFRPSDQ\ QHXURQDO GLIIHUHQWLDWLRQ RI 17' FHOOV 1HXURVFL 5HV f $GHUHP $ f 7KH 0$5&.6 EURWKHUV $ IDPLO\ RI SURWHLQ NLQDVH & VXEVWUDWHV &HOO $LJQHU / $UEHU 6 .DSIKDPPHU -3 /DX[ 7 6FKQHLGHU & %RWWHUL ) %UHQQHU + 5 DQG &DURQL 3 f 2YHUH[SUHVVLRQ RI WKH QHXUDO JURZWKDVVRFLDWHG SURWHLQ *$3 LQGXFHV QHUYH VSURXWLQJ LQ WKH DGXOW QHUYRXV V\VWHP RI WUDQVJHQLF PLFH &HOO $LJQHU / DQG &DURQL 3 f 'HSOHWLRQ RI N' JURZWKDVVRFLDWHG SURWHLQ LQ SULPDU\ VHQVRU\ QHXURQV OHDGV WR GLPLQLVKHG IRUPDWLRQ DQG VSUHDGLQJ RI JURZWK FRQHV &HOO %LRO $LJQHU / DQG &DURQL 3 f $EVHQFH RI SHUVLVWHQW VSUHDGLQJ EUDQFKLQJ DQG DGKHVLRQ LQ *$3GHSOHWHG JURZWK FRQHV &HOO %LRO f $OEXV + DQG :LOOLDPVRQ 5 f (OHFWURSK\VLRORJLF DQDO\VLV RI WKH DFWLRQV RI YDOSURDWH RQ S\UDPLGDO QHXURQV LQ WKH UDW KLSSRFDPSDO VOLFH (SLOHSVLD f $OOHQ /$+ DQG $GHUHP $ f $ UROH IRU 0$5&.6 WKH D LVR]\PH RI SURWHLQ NLQDVH & DQG P\RVLQ LQ ]\PRVDQ SKDJRF\WRVLV E\ PDFURSKDJHV ([S 0HG %DHWJH (( DQG +DPPDQJ -3 f 1HXULWH RXWJURZWK LQ 3& FHOOV GHILFLHQW LQ *$3 1HXURQ %DUWRQ %0 DQG *LWOLQ 0f 9HUDSDPLO LQ WUHDWPHQWUHVLVWDQW PDQLD DQ RSHQ WULDO &OLQ 3V\FKRSKDUPDFRO f %HQHW /= DQG :LOOLDPV 5/ f 3KDUPDFRNLQHWLF GDWD *RRGPDQ DQG *LOPDQfV 7KH 3KDUPDFRORJLFDO %DVLV RI 7KHUDSHXWLFV WK HG HG E\ $* *LOPDQ 7 : 5DLO $ 6 1LHV DQG 3 7D\ORU SS 3HUJDPRQ 3UHVV (OPVIRUG 1< %HQRZLW] /, DQG 3HUURQH%L]]R]HUR 1, f 7KH H[SUHVVLRQ RI *$3 LQ UHODWLRQ WR QHXURQDO JURZWK DQG SODVWLFLW\ ZKHQ ZKHUH KRZ DQG ZK\ 3URJUHVV LQ %UDLQ 5HVHDUFK

PAGE 136

%HQRZLW] /, DQG 5RXWWHQEHUJ $ f *$3 $Q LQWULQVLF GHWHUPLQDQW RI QHXURQDO GHYHORSPHQW DQG SODVWLFLW\ 7,16 %HUULGJH 0f ,QRVLWRO WULVSKRVSKDWH FDOFLXP OLWKLXP DQG FHOO VLJQDOLQJ -$0$ %HUULGJH 0'RZQHV &3 DQG +DQOH\ 0 5 f /LWKLXP DPSOLILHV DJRQLVW GHSHQGHQW SKRVSKDWLG\OLQRVLWRO UHVSRQVHV LQ EUDLQ DQG VDOLYDU\ JODQGV %LRFKHP %ODFNVKHDU 3 f 7KH 0$5&.6 IDPLO\ RI FHOOXODU SURWHLQ NLQDVH & VXEVWUDWHV %LRO &KHP %RQLHFH ,5 DQG :DJQHU -$ f *URZWK IDFWRUV SURWHFW 3& FHOOV DJDLQVW LVFKHPLD E\ D PHFKDQLVP WKDW LV LQGHSHQGHQW RI 3.$ 3.& DQG SURWHLQ V\QWKHVLV 1HXURVFL f %RUJDWWL 3 0D]]RQL 0 &DULQL & 1HUL /0 0DUFKLVLR 0 %HUWRODVR / 3UHYLDWL 0 =DXOL DQG &DSLWDQL 6 f &KDQJHV RI QXFOHDU SURWHLQ NLQDVH & DFWLYLW\ DQG LVRW\SH FRPSRVLWLRQ LQ 3& FHOO SUROLIHUDWLRQ DQG GLIIHUHQWLDWLRQ ([SHULPHQWDO &HOO 5HVHDUFK %RZGHQ &/ f 3UHGLFWRUV RI UHVSRQVH WR GLYDOSURH[ DQG OLWKLXP &OLQ 3V\FKLDWU\ >6XSSO @ %RZGHQ &/ f 5ROH RI QHZHU PHGLFDWLRQV IRU ELSRODU GLVRUGHU &OLQ 3V\FKRSKDUPDFRORJ\ 6XSSO f 66 %RZGHQ &/ %UXJJHU $ 0 6ZDQQ $& &DODEUHVH -5 -DQLFDN 3* 3HWW\ ) 'LOVDYHU 6& 'DYLV -0 5XVK $6PDOO -* *DU]D7UHYLQR (6 5LVFK 6& *RRGQLFN 3DQG 0RUULV '' f (IILFDF\ RI GLYDOSURH[ YV OLWKLXP DQG SODFHER LQ WKH WUHDWPHQW RI PDQLD -$0$ %URRNV 6) +HUJHW 7 %URDG 6 DQG 5R]HQJXUW ( f 7KH H[SUHVVLRQ RI .0$5&.6 D PDMRU VXEVWUDWH RI SURWHLQ NLQDVH & 3.&f LV GRZQUHJXODWHG WKURXJK ERWK 3.&GHSHQGHQW DQG LQGHSHQGHQW SDWKZD\V -%& f %URZQ 5 f 8 6 H[SHULHQFH ZLWK YDOSURDWH LQ PDQLF GHSUHVVLYH LOOQHVV D PXOWLFHQWHU WULDO &OLQ 3V\FKLDWU\ > 6XSSO@ &DEHOO / DQG $XGHVLUN f (IIHFWV RI VHOHFWLYH LQKLELWLRQ RI SURWHLQ NLQDVH & F\FOLF $03GHSHQGHQW SURWHLQ NLQDVH DQG &DfFDOPRGXOLQGHSHQGHQW SURWHLQ NLQDVH RQ QHXULWH GHYHORSPHQW LQ FXOWXUHG UDW KLSSRFDPSDO QHXURQV ,QW 'HY 1HXURVFL f

PAGE 137

&DODEUHVH -5 DQG 'HOXFFKL *$ f 3KHQRPHQRORJ\ RI UDSLG F\FOLQJ PDQLF GHSUHVVLRQ DQG LWV WUHDWPHQW ZLWK YDOSURDWH &OLQ 3V\FKLDWU\ > 6XSSO@ &DODEUHVH -5 DQG 'HOXFFKL *$ f 6SHFWUXP RI HIILFDF\ RI YDOSURDWH LQ SDWLHQWV ZLWK UDSLGF\FOLQJ ELSRODU GLVRUGHU $P 3V\FKLDWU\ &DODEUHVH -5 0DUNRYLW] 3.LPPHO 6( DQG :DJQHU 6& f 6SHFWUXP RI HIILFDF\ RI YDOSURDWH LQ UDSLGF\FOLQJ ELSRODU SDWLHQWV &OLQ 3V\FKRSKDUPDFRO 66 &DODEUHVH -5 5DSSRUW '.LPPHO 6( 5HHFH % DQG :R\VKYLOOH 0f 5DSLGF\FOLQJ ELRODU GLVRUGHU DQG LWV WUHDWPHQW ZLWK YDOSURDWH &DQ 3V\FKLDWU\ VXSSO f 66 &DUOVRQ 52 0DUWLQ 0 )HOGPDQ (/ DQG $JUDQRII %: f 3.& DFWLYLW\ DQG 3.&DOSKD P51$ FRQWHQW DUH UHGXFHG LQ VHUXPGHULYHG KXPDQ QHXUREODVWRPD FHOOV ZLWKRXW FRQFRPLWDQW LQGXFWLRQ RI GLIIHUHQWLDWLRQ ([S &HOO 5HV f &KDNUDYDUWK\ %5 :RQJ DQG 'XUNLQ -3 f (YLGHQFH WKDW WKH PRGXODWLRQ RI PHPEUDQHDVVRFLDWHG SURWHLQ NLQDVH & DFWLYLW\ E\ DQ HQGRJHQRXV LQKLELWRU SOD\V D UROH LQ 1( PXULQH QHXUREODVWRPD FHOO GLIIHUHQWLDWLRQ 1HXURFKHP &KDSPDQ $ .HDQH 3( 0HOGUXP %6 6LPLDQG DQG 9HPLHUHV -& f 0HFKDQLVP RI DQWLFRQYXOVDQW DFWLRQ RI YDOSURDWH 3URJUHVV LQ 1HXURELRO &KHQ 0DQML +. +DZYHU '% :ULJKW &% DQG 3RWWHU := f &KURQLF VRGLXP YDOSURDWH VHOHFWLYHO\ GHFUHDVHV SURWHLQ NLQDVH & D DQG H LQ YLWUR 1HXURFKHP &KHQ
PAGE 138

'DVWRQ *3 %DUQHV (OPRUH ( )LW]JHUDOG 03 DQG 6KDUPD 6 f (YDOXDWLRQ RI FKLFN HPEU\R QHXUDO UHWLQD FHOO FXOWXUH DV D VFUHHQ IRU GHYHORSPHQWDO WR[LFDQWV )XQG DQG $SSOLHG 7R[ 'HEXV ( :HEHU DQG 2VERUQ 0 f 0RQRFORQDO DQWLERGLHV VSHFLILF IRU JOLDO ILEULOODU\ DFLGLF *)$f SURWHLQ DQG IRU HDFK RI WKH QHXURILODPHQW WULSOHW SRO\SHSWLGHV 'LIIHUHQWLDWLRQ 'HNNHU /9 GH*UDDQ 31( 2HVWUHLFKHU $% 9HUVWHHJ '+* DQG *LVSHQ : + f ,QKLELWLRQ RI QRUDGUHQDOLQH UHOHDVH E\ DQWLERGLHV WR % *$3f 1DWXUH 'L[RQ -) DQG +RNLQ /( f 7KH PRRGVWDELOL]LQJ GUXJ YDOSURDWH PLPLFV OLWKLXP LQ VWLPXODWLQJ JOXWDPDWH UHOHDVH DQG LQRVLWRO WULVSKRVSKDWH DFFXPXODWLRQ LQ EUDLQ FRUWH[ VOLFHV EXW QRW DFFXPXODWLRQ RI LQRVLWRO PRQRSKRVSKDWHV DQG ELVSKRVSKDWHV 3URF 1DWO $FDG 6FL 86$ (DGLH 0f )RUPDWLRQ RI DFWLYH PHWDEROLWHV RI DQWLFRQYXOVDQW GUXJV $ UHYLHZ RI WKHLU SKDUPDFRNLQHWLF DQG WKHUDSHXWLF VLJQLILFDQFH &OLQ 3KDUPDFRNLQHW f (KOHUV (OPD]DU 00$ DQG 1DX + f 0HWKLRQLQH UHGXFHV WKH YDOSURLF DFLG LQGXFHG VSLQD ELILGD UDWH LQ PLFH ZLWKRXW DOWHULQJ YDOSURLF DFLG NLQHWLFV 1XWULWLRQ (NLQFL )DQG 6KHD 7% f 6HOHFWLYH DFWLYDWLRQ E\ EU\RVWDWLQ GHPRQVWUDWHV XQLTXH UROHV IRU 3.& HSVLORQ LQ QHXULWH H[WHQVLRQ DQG WDX SKRVSKRU\ODWLRQ ,QW 'HY 1HXURVFL f )DJHUVWURP 6 3DKOPDQ 6 *HVWEORP & DQG 1DQEHUJ ( f 3URWHLQ NLQDVH & HSVLORQ LV LPSOLFDWHG LQ QHXULWH RXWJURZWK LQ GLIIHUHQWLDWLQJ KXPDQ QHXUREODVWRPD FHOOV &HOO *URZWK 'LIIHU f )DZFHWW f 9DOSURDWH XVH LQ DFXWH PDQLD DQG ELSRODU GLVRUGHU DQ LQWHUQDWLRQDO SHUVSHFWLYH &OLQ 3V\FKLDWU\ 6XSSO@ )ULHGPDQ ( :DQJ +< /HYLQVRQ &RQQHOO 7$ DQG 6LQJK + f $OWHUHG SODWHOHW SURWHLQ NLQDVH & DFWLYLW\ LQ ELSRODU DIIHFWLYH GLVRUGHU PDQLF HSLVRGH %LRO 3V\FKLDWU\ f )XMLVH $ 0L]XPR 8HGD < 2VDGD 6 +LUDL 6 7DND\DQDJL $ 6KLPL]X 1 2ZDGD 0. 1DNDMLPD + DQG 2KQR 6 f 6SHFLILFLW\ RI WKH KLJK DIILQLW\ LQWHUDFWLRQ RI SURWHLQ NLQDVH & ZLWK D SK\VLRORJLFDO VXEVWUDWH P\ULVWR\ODWHG DODQLQH ULFK SURWHLQ NLQDVH & VXEVWUDWH %LRO &KHP *HPHU 5 + DQG 6WDQWRQ $ f $OJRULWKP IRU SDWLHQW PDQDJHPHQW RI DFXWH PDQLF VWDWHV /LWKLXP YDOSURDWH RU FDUEDPD]HSLQH" &OLQ 3V\FKRSKDUPDFRO 66

PAGE 139

*RGIUH\ 33 f 3RWHQWLDWLRQ E\ OLWKLXP RI &03SKRVSKDWLGDWH IRUPDWLRQ LQ FDUEDFKROVWLPXODWHG UDW FHUHEUDOFRUWLFDO VOLFHV DQG LWV UHYHUVDO E\ Z\RLQRVLWRO %LRFKHP *RIIORW ) YDQ 0DHOH)DEU\ DQG 3LFDUG -f &UDQLDO QHUYHV DQG JDQJOLD DUH DOWHUHG DIWHU LQ YLWUR WUHDWPHQW RI PRXVH HPEU\RV ZLWK YDOSURLF DFLG 93$f DQG HQ 93$ 'HY %UDLQ 5HV *UDQQHPDQ *5 6FKQHFN ': &DYDQDXJK -+ DQG :LWW *) f 3KDUPDFRNLQHWLF LQWHUDFWLRQV DQG VLGH HIIHFWV UHVXOWLQJ IURP FRQFRPLWDQW DGPLQLVWUDWLRQ RI OLWKLXP DQG GLYDOSURH[ VRGLXP &OLQ 3V\FKLDWU\ f *UHVVHQV 3 0DUUHW 6 0DUWLQ -/ /DTXHUULHUH $ /RPEHW $ DQG (YUDUG 3 f 5HJXODWLRQ RI QHXURSURWHFWLYH DFWLRQ RI YDVRDFWLYH LQWHVWLQDO SHSWLGH LQ WKH PXULQH GHYHORSLQJ EUDLQ E\ SURWHLQ NLQDVH & DQG PLWRJHQDFWLYDWHG SURWHLQ NLQDVH FDVFDGHV LQ YLYR DQG LQ YLWUR VWXGLHV 1HXURFKHP f +D\HV 6 f /RQJWHUP XVH RI YDOSURDWH LQ SULPDU\ SV\FKLDWULF GLVRUGHUV &OLQ 3V\FKLDWU\ > 6XSSO@ +HLNNLOD -( $NHUOLQG DQG $NHUPDQ .( f 3URWHLQ NLQDVH & DFWLYDWLRQ DQG GRZQUHJXODWLRQ LQ UHODWLRQ WR SKRUERO HVWHULQGXFHG GLIIHUHQWLDWLRQ RI 6+6<< KXPDQ QHXUREODVWRPD FHOOV &HOO 3K\VLRO f +RIPDQQ f 7KH SRWHQWLDO IRU LVR]\PHVHOHFWLYH PRGXODWLRQ RI SURWHLQ NLQDVH & )$6(% +ROIRUG 1+* DQG %HQHW /= f 3KDUPDFRNLQHWLFV DQG SKDUPDFRG\QDPLFV 5DWLRQDO GRVH VHOHFWLRQ DQG WKH WLPH FRXUVH RI GUXJ DFWLRQ ,Q %DVLF DQG &OLQLFDO 3KDUPDFRORJ\ WK HG HG E\ %* .DW]XQJ S $SSOHWRQ DQG /DQJH 1RUZDON &7 +ROOLVWHU /( f 9DOSURLF DFLG DQG FORQD]HSDP ,Q %DVLF DQG &OLQLFDO 3KDUPDFRORJ\ WK HG HG E\ %* .DW]XQJ S $SSOHWRQ DQG /DQJH 1RUZDON &7 +XQGOH % 0F0DKRQ 7 'DGJDU &KHQ &+ 0RFKO\5RVHQ DQG 0HVVLQJ 52 f $Q LQKLELWRU\ IUDJPHQW GHULYHG IURP SURWHLQ NLQDVH &HSVLORQ SUHYHQWV HQKDQFHPHQW RI QHUYH JURZWK IDFWRU UHVSRQVHV E\ HWKDQRO DQG SKRUERO HVWHUV %LRO &KHP f +XQGOH % 0F0DKRQ 7 'DGJDU DQG 0HVVLQJ 52 f 2YHUH[SUHVVLRQ RI V SURWHLQ NLQDVH & HQKDQFHV QHUYH JURZWK IDFWRULQGXFHG SKRVSKRU\ODWLRQ RI PLWRJHQ DFWLYDWHG SURWHLQ NLQDVHV DQG QHXULWH RXWJURZWK %LRO &KHP

PAGE 140

,VKLL + -LURXVHN 0 5 .R\D 7DNDJL & ;LD 3 &OHUPRQW $ %XUVHOO 6( .HUQ 76 %DLODV /0 +HDWK :) 6WUDPP /( )HHQHU (3 DQG .LQJ */ f $PHOLRUDWLRQ RI YDVFXODU G\VIXQFWLRQV LQ GLDEHWLF UDWV E\ DQ RUDO 3.& 3 LQKLELWRU 6FLHQFH -DFNVRQ 6 f 5HJXODWLQJ WUDQVFULSWLRQ IDFWRU DFWLYLW\ E\ SKRVSKRU\ODWLRQ 7UHQGV &HOO %LRO -DODYD $ $NHUPDQ DQG +HLNNLOD f 3URWHLQ NLQDVH LQKLELWRU VWDXURVSRULQH LQGXFHV D PDWXUH QHXURQDO SKHQRW\SH LQ 6+6<< KXPDQ QHXUREODVWRPD FHOOV WKURXJK DQ DOSKD EHWD DQG ]HWDSURWHLQ NLQDVH &LQGHSHQGHQW SDWKZD\ &HOO 3K\VLRO f -DOORQ 3 f 7KH SUREOHP RI LQWUDFWDELOLW\ 7KH FRQWLQXLQJ QHHG IRU QHZ PHGLFDO WKHUDSLHV LQ HSLOHSV\ (SLOHSVLD 6XSSO f 66 -DQLFDN 3* 6KDUPD 53 3DQGH\ DQG 'DYLV -0 f 9HUDSDPLO IRU WKH WUHDWPHQW RI DFXWH PDQLD D GRXEOHEOLQG SODFHERFRQWUROOHG WULDO $P 3V\FKLDWU\ f -DS 7MRHQ 6DQ ( 5 $ 6FKPLGW0LFKHOV 0 2HVWUHLFKHU $% *LVSHQ :+ DQG 6FKRWPDQ 3 f ,QKLELWLRQ RI QHUYH JURZWK IDFWRULQGXFHG %*$3 H[SUHVVLRQ E\ DQWLVHQVH ROLJRPHUV LQWHUIHUHV ZLWK QHXULWH RXWJURZWK RI 3& FHOOV %LRFKHP %LRSK\V 5HV &RPP -LURXVHN 0 5 *LOOLJ 5 *RQ]DOH] & 0 +HDWK :) 0F'RQDOG -+ ,,, 1HHO '$ 5LWR &6LQJK 8 6WUDPP /( 0HOLNLDQ%DGDOLDQ $ %DHYVN\ 0 %DOLDV /0 +DOO 6( :LQQHURVNL // DQG )DXO 00 f f >'LPHWK\ODPLQRfPHWK\O@WHWUDK\GUR GLPHWKHQR + GLEHQ]R>He@S\UUROR>"@ > @R[DGLD]DF\FORKH[DGHFHQHfGLRQH /
PAGE 141

.XQW]6LPRQ DQG 2EHUW f 6RGLXP YDOSURDWH DQ DQWLFRQYXOVDQW GUXJ VWLPXODWHV KXPDQ F\WRPHJDORYLUXV UHSOLFDWLRQ *HQ 9LURO .XR -) $QGHUVVRQ 5** :LVH %& 0DFNHUORYD / 6DORPRQVVRQ %UDFNHWW 1/ .DWRK 1 6KRML 0 DQG :UHQQ 5: f &DOFLXPGHSHQGHQW SURWHLQ NLQDVH :LGHVSUHDG RFFXUUHQFH LQ YDULRXV WLVVXHV DQG SK\OD RI WKH DQLPDO NLQJGRP DQG FRPSDULVRQ RI HIIHFWV RI SKRVSKROLSLG FDOPRGXOLQ DQG WULIOXRSHUD]LQH 3URF 1DWO $FDG 6FL 86$ f /HH ++DPPRQG '1 /DUJH 7+ 5REDFN -' 6LP $ %URZQ '$ 2WWHQ 8+ DQG :DLQHU %+ f 1HXURQDO SURSHUWLHV DQG WURSKLF DFWLYLWLHV RI LPPRUWDOL]HG KLSSRFDPSDO FHOOV IURP HPEU\RQLF DQG \RXQJ DGXOW PLFH 1HXURVFL /HQR[ 5+ f 5ROH RI UHFHSWRU FRXSOLQJ WR SKRVSKRLQRVLWLGH PHWDEROLVP LQ WKH WKHUDSHXWLF DFWLRQ RI OLWKLXP ,Q 0ROHFXODU 0HFKDQLVPV RI 1HXURQDO 5HVSRQVLYHQHVV $GY ([S %LRO 0HG 9RO HG E\ <+ (KUOLFK 5+ /HQR[ ( .RPHFNL DQG :2 %HUU\ SS 3OHQXP 3UHVV 1HZ
PAGE 142

0DQML +. %HUVXGVN\ < &KHQ %HOPDNHU 5+ DQG 3RWWHU := Df 0RGXODWLRQ RI SURWHLQ NLQDVH & LVR]\PHV DQG VXEVWUDWHV E\ OLWKLXP 7KH UROH RI P\Rn LQRVLWRO 1HXURSV\FKRSKDUPDFRORJ\ 0DQML +. &KHQ +VLDR -. 5LVE\ (' 0DVDQD 0, DQG 3RWWHU := Ef 5HJXODWLRQ RI VLJQDO WUDQVGXFWLRQ SDWKZD\V E\ PRRGVWDELOL]LQJ DJHQWV ,PSOLFDWLRQV IRU WKH GHOD\HG RQVHW RI WKHUDSHXWLF HIILFDF\ &OLQ 3V\FKLDWU\ >VXSSO f 0DQML +. (WFKHEHUULJDUD\ 5 &KHQ DQG 2OGV -/ f /LWKLXP GHFUHDVHV PHPEUDQHDVVRFLDWHG SURWHLQ NLQDVH & LQ KLSSRFDPSXV 6HOHFWLYLW\ IRU WKH D LVR]\PH 1HXURFKHP 0DQML +. DQG /HQR[ 5+ f /RQJWHUP DFWLRQ RI OLWKLXP $ UROH IRU WUDQVFULSWLRQDO DQG SRVWWUDQVFULSWLRQDO IDFWRUV UHJXODWHG E\ SURWHLQ NLQDVH & 6\QDSVH 0DQML +. 0RRUH *DQG &KHQ f /LWKLXP DW +DYH WKH QHXURSURWHFWLYH HIIHFWV RI WKLV XQLTXH FDWLRQ EHHQ RYHUORRNHG" %LRO 3V\FKLDWU\ f 0DQML +. 3RWWHU := DQG /HQR[ 5+ f 6LJQDO WUDQVGXFWLRQ SDWKZD\V 0ROHFXODU WDUJHWV IRU OLWKLXPnV DFWLRQ $UFK *HQ 3V\FKLDWU\ 0DUWLQ 0 / %UHHQ .& DQG 5HJDQ & 0 f 3HUWXUEDWLRQV RI FHOOXODU IXQFWLRQV LQWHJUDO WR QHXUDO WXEH IRUPDWLRQ E\ WKH SXWDWLYH WHUDWRJHQ VRGLXP YDOSURDWH 7R[ ,Q 9LWUR f 0DUWLQ 0 / DQG 5HJDQ & 0 f 7KH DQWLFRQYXOVDQW VRGLXP YDOSURDWH VSHFLILFDOO\ LQGXFHV WKH H[SUHVVLRQ RI D UDW JOLDO KHDW VKRFN SURWHLQ ZKLFK LV LGHQWLILHG DV WKH FROODJHQ W\SH ,9 UHFHSWRU %UDLQ 5HV 0DUWLQ 0 / DQG 5HJDQ & 0 f 7KH DQWLFRQYXOVDQW YDOSURDWH WHUDWRJHQ UHVWULFWV WKH JOLDO FHOO F\FOH DW D GHILQHG SRLQW LQ WKH PLG*O SKDVH %UDLQ 5HV 0DVOLDK ( 5DEHU $OIRUG 0 0DOORU\ 0 0DWWVRQ 03 (IILFDF\ RI YHUDSDPLO LQ PDQLD FULVHV@ (QFHSKDOH f 0F(OUR\ 6/ DQG .HFN 3( f 7UHDWPHQW JXLGHOLQHV IRU YDOSURDWH LQ ELSRODU DQG VFKL]RDIIHFWLYH GLVRUGHUV &DQ 3V\FKLDWU\ 6XSSO f 66

PAGE 143

0F(OUR\ 6/ .HFN 3( 3RSH +* DQG +XGVRQ -, f 9DOSURDWH LQ SV\FKLDWULF GLVRUGHUV /LWHUDWXUH UHYLHZ DQG FOLQLFDO JXLGHOLQHV &OLQ 3V\FKLDWU\ > 6XSSO@ 0F(OUR\ 6/ .HFN 3( 3RSH +* DQG +XGVRQ -, f 9DOSURDWH LQ WKH WUHDWPHQW RI ELSRODU GLVRUGHU /LWHUDWXUH UHYLHZ DQG FOLQLFDO JXLGHOLQHV &OLQ 3V\FKRSKDPDFRO 66 0F(OUR\ 6/ .HFN 3( 6WDQWRQ 63 7XJUXO .& %HQQHWW -$ DQG 6WUDNRZVNL 6 0 f $ UDQGRPL]HG FRPSDULVRQ RI GLYDOSURH[ RUDO ORDGLQJ YHUVXV KDORSHULGRO LQ WKH LQLWLDO WUHDWPHQW RI DFXWH SV\FKRWLF PDQLD &OLQ 3V\FKLDWU\ 0F1DPDUD 5. DQG /HQR[ 5+ f &RPSDUDWLYH GLVWULEXWLRQ RI P\ULVWR\ODWHG DODQLQHULFK & NLQDVH VXEVWUDWH 0$5&.6f DQG ),*$3 JHQH H[SUHVVLRQ LQ WKH DGXOW UDW EUDLQ &RPS 1HXURORJ\ 0HLUL .) %LFNHUVWDII /( DQG 6FKZRE -( f 0RQRFORQDO DQWLERGLHV VKRZ WKDW NLQDVH & SKRVSKRU\ODWLRQ RI *$3 GXULQJ D[RQRJHQHVLV LV ERWK VSDWLDOO\ DQG WHPSRUDOO\ UHVWULFWHG LQ YLYR &HOO %LRO f 0HLUL .) 6DIIHOO -/ :DOVK )6 DQG 'RKHUW\ 3 f 1HXULWH RXWJURZWK VWLPXODWHG E\ QHXUDO FHOO DGKHVLRQ PROHFXOHV UHTXLUHV JURZWKDVVRFLDWHG SURWHLQ *$3f IXQFWLRQ DQG LV DVVRFLDWHG ZLWK *$3 SKRVSKRU\ODWLRQ LQ JURZWK FRQHV 1HXURVFLHQFH f 0HQHJROD ( %URFFLD 0 / 1DX + 3UDWL 0 5LFROIL 5 DQG *LDYLQL ( f 7HUDWRJHQLF HIIHFWV RI VRGLXP YDOSURDWH LQ PLFH DQG UDWV DW PLGJHVWDWLRQ DQG DW WHUP 7HUDWRJHQ &DUFLQRJHQ DQG 0XWDJHQ 0HXQLHU + &DUUD] 0HXQLHU < (\PDUG 3 DQG $LPDUG 0 f 3URSULHWHV SKDUPDFRG\QDPLF GH OfDFLGH QGHSURS\ODFHWLTXH OHU 0HPRLUH 3URSULHWHV $QWLHSLOHSWLTXHV 7KHUDSLH 0LQDNXFKL 5 7DNDL <
PAGE 144

1HZWRQ $ & f 3URWHLQ NLQDVH & 6WUXFWXUH IXQFWLRQ DQG UHJXODWLRQ %LRO &KHP 1LVKL]XND < f 3URWHLQ NLQDVH & DQG OLSLG VLJQDOOLQJ IRU VXVWDLQHG FHOOXODU UHVSRQVHV )$6(% 1RQDND 6 +RXJK &DQG &KXDQJ '0 f &KURQLF OLWKLXP WUHDWPHQW UREXVWO\ SURWHFWV QHXURQV LQ WKH FHQWUDO QHUYRXV V\VWHP DJDLQVW H[FLWRWR[LFLW\ E\ LQKLELWLQJ 1 PHWK\O'DVSDUWDWH UHFHSWRUPHGLDWHG FDOFLXP LQIOX[ 3URF 1DWO $FDG 6FL 86$ 1RVHO 3* DQG .OHLQ 1: f 0HWKLRQLQH GHFUHDVHV WKH HPEU\RWR[LFLW\ RI VRGLXP YDOSURDWH LQ WKH UDW ,Q YLYR DQG LQ YLWUR REVHUYDWLRQV 7HUDWRORJ\ 2f%ULHQ ( DQG 5HJDQ & 0 f 3URWHLQ NLQDVH & LQKLELWRUV DUUHVW WKH & JOLRPD FHOO F\FOH DW D PLG*O SKDVH UHVWULFWLRQ SRLQW ,PSOLFDWLRQV IRU WKH DQWLSUROLIHUDWLYH DFWLRQ RI YDOSURDWH 7R[ LQ 9LWUR f 2HKUOHLQ 6$ 0DHOLFNH $ DQG +HUJHW 7 f ([SUHVVLRQ RI SURWHLQ NLQDVH & JHQH IDPLO\ PHPEHUV LV WHPSRUDOO\ DQG VSDWLDOO\ UHJXODWHG GXULQJ QHXUDO GHYHORSPHQW LQ YLWUR (XU &HOO %LRO f 2XLPHW && :DQJ -.7 :DODDV 6, $OEHUW .$ DQG *UHHQJDUG 3 f /RFDOL]DWLRQ RI WKH 0$5&.6 N'Df SURWHLQ D PDMRU VSHFLILF VXEVWUDWH IRU SURWHLQ NLQDVH & LQ UDW EUDLQ 1HXURVFLHQFH 3DURGL 07 9DUHVLR / DQG 7RQLQL *3 f 7KH VSHFLILF LQKLELWRU RI SURWHLQ NLQDVH & OLVRTXLQROLQ\OVXOIRQ\OfPHWK\OSLSHUD]LQH +f LQGXFHV PRUSKRORJLFDO FKDQJH DQG FHOO GLIIHUHQWLDWLRQ RI KXPDQ QHXUDO FUHVWGHULYHG FHOO OLQHDJHV )(%6 /HWW f 3DUURZ 9 )DJHUVWURP 6 0H\HUVRQ 1DQEHUJ ( DQG 3DKOPDQ 6 f 3URWHLQ NLQDVH &DOSKD DQG HSVLORQ DUH HQULFKHG LQ JURZWK FRQHV RI GLIIHUHQWLDWLQJ 6+6<< KXPDQ QHXUREODVWRPD FHOOV 1HXURVFL 5HV f 3DUURZ 9 1DQEHUJ ( +HLNNLOD +DPPHUOLQJ 8 DQG 3DKOPDQ 6 f 3URWHLQ NLQDVH & UHPDLQV IXQFWLRQDOO\ DFWLYH GXULQJ 73$ LQGXFHG QHXURQDO GLIIHUHQWLDWLRQ RI 6+6<< KXPDQ QHXUREODVWRPD FHOOV &HOO 3K\VLRO f 3DWHO DQG .OLJPDQ f 3XULILFDWLRQ DQG FKDUDFWHUL]DWLRQ RI DQ 0U SURWHLQ NLQDVH & VXEVWUDWH IURP UDW EUDLQ %LRO &KHP 3HQU\ -. DQG 'HDQ -& f 7KH VFRSH DQG XVH RI YDOSURDWH LQ HSLOHSV\ &OLQ 3V\FKLDWU\ > 6XSSO@

PAGE 145

3HUH]3LQ]RQ 0$ DQG %RUQ -* f 5DSLG SUHFRQGLWLRQLQJ QHXURSURWHFWLRQ IROORZLQJ DQR[LD LQ KLSSRFDPSDO VOLFHV UROH RI WKH $73 FKDQQHO DQG SURWHLQ NLQDVH & 1HXURVFLHQFH f 3HWW\ ) f *$%$ DQG PRRG GLVRUGHUV $ EULHI UHYLHZ DQG K\SRWKHVLV $IIHFW 'LVRUG 3L]]L 0 %RURQL ) %HQDUHVH 0 0RUDLWLV & 0HPR 0 DQG 6SDQR 3 f 1HXURSURWHFWLYH HIIHFW RI WK\URWURSLQUHOHDVLQJ KRUPRQH DJDLQVW H[FLWDWRU\ DPLQR DFLGLQGXFHG FHOO GHDWK LQ KLSSRFDPSDO VOLFHV (XU 3KDUPDFRO f 3L]]L 0 *DOOL 3 &RQVRODQGL 2 $UULJKL 9 0HPR 0 DG 6SDQR 3) f 0HWDERWURSLF DQG LRQRWURSLF WUDQVGXFHUV RI JOXWDPDWH VLJQDO LQYHUVHO\ FRQWURO F\WRSODVPLF &D FRQFHQWUDWLRQ DQG H[FLWRWR[LFLW\ LQ FXOWXUHG FHUHEHOODU JUDQXOH FHOOV SLYRWDO UROH RI SURWHLQ NLQDVH & 0RO 3KDUPDFRO f 3ROWRUDN 0 )U\H 0$ :ULJKW 5 +HPSHUO\ -*HRUJH 0 6 3D]]DJOLD 3-HUUHOV 6$ 3RVW 50 DQG )UHHG :f ,QFUHDVHG QHXUDO FHOO DGKHVLRQ PROHFXOH LQ WKH &6) RI SDWLHQWV ZLWK PRRG GLVRUGHU 1HXURFKHP f 3RQ]RQL 0 /XFDUHOOL ( &RUUDV 09 DQG &RUQDJOLD)HUUDULV 3 f 3URWHLQ NLQDVH & LVRHQ]\PHV LQ KXPDQ QHXUREODVWV ,QYROYHPHQW RI 3.& HSVLORQ LQ FHOO GLIIHUHQWLDWLRQ )(%6 /HWW f 3RUWHU 5DQG 0HOGUXP %6 f $QWLHSLOHSWLF GUXJV ,Q %DVLF DQG &OLQLFDO 3KDUPDFRORJ\ WK HG HG E\ %* .DW]XQJ SS $SSOHWRQ DQG /DQJH 1RUZDON &7 3RVW 50 f ,QWURGXFWLRQ (PHUJLQJ SHUVSHFWLYHV RQ YDOSURDWH LQ DIIHFWLYH GLVRUGHUV &OLQ 3V\FKLDWU\ 6XSSOf 3RVW 50 :HLVV 65% DQG &KXDQJ '0 f 0HFKDQLVPV RI DFWLRQ RI DQWLFRQYXOVDQWV LQ DIIHFWLYH GLVRUGHUV &RPSDULVRQV ZLWK OLWKLXP &OLQ 3V\FKRSKDUPDFRO 66 5DPVD\ 5( f &RQWUROOHG DQG FRPSDUDWLYH WULDOV ZLWK YDOSURDWH 8QLWHG 6WDWHV (SLOHSVLD 6XSSO f 66 5HJDQ & 0 f 7KHUDSHXWLF OHYHOV RI VRGLXP YDOSURDWH LQKLELW PLWRWLF LQGLFHV LQ FHOOV RI QHXUDO RULJLQ %UDLQ 5HV 5KRDGV ** DQG 0LOOV -/ f &DQ YLWDPLQ VXSSOHPHQWV SUHYHQW QHXUDO WXEH GHIHFWV" &XUUHQW HYLGHQFH DQG RQJRLQJ LQYHVWLJDWLRQV &OLQ 2E *\Q 5LPPHU ( DQG 5LFKHQV $ f $Q XSGDWH RQ VRGLXP YDOSURDWH 3KDUPDFRWKHUDS\ f

PAGE 146

5RVHQ $ .HHQDQ .) 7KHOHQ 0 1DLUQ $& DQG $GHUHP $ f $FWLYDWLRQ RI SURWHLQ NLQDVH & UHVXOWV LQ WKH GLVSODFHPHQW RI LWV P\ULVWR\ODWHG DODQLQHULFK VXEVWUDWH IURP SXQFWDWH VWUXFWXUHV LQ PDFURSKDJH ILORSRGLD ([S 0HG 6FKOLQJHQVLHSHQ 5 DQG 6FKOLQJHQVLHSHQ .+ f $QWLVHQVH ROLJRGHR[\QXFOHRWLGHV +LJKO\ VSHFLILF WRROV IRU EDVLF UHVHDUFK DQG SKDUPDFRWKHUDS\ ,Q $QWLVHQVH )URP 7HFKQRORJ\ WR 7KHUDS\ HG E\ 5 6FKOLQJHQVLHSHQ : %U\VFK DQG .+ 6FKOLQJHQVLHSHQ SS %ODFNZHOO 6FLHQFH 0DOGHQ 0$ 6KDPORR 0 DQG :LHORFK 7 f 5DSLG GHFOLQH LQ SURWHLQ NLQDVH & JDPPD OHYHOV LQ WKH V\QDSWRVRPDO IUDFWLRQ RI UDW KLSSRFDPSXV DIWHU LVFKHPLF SUHFRQGLWLRQLQJ 1HXURUHSRUW f 6KDZ f 1HXURILODPHQW 3URWHLQV ,Q 7KH 1HXURQDO &\WRVNHOHWRQ SS :LOH\/LVV 1HZ
PAGE 147

6WXPSR '%RFN & % 7XWWOH -6 DQG %ODFNVKHDU 3f 0$5&.6 GHILFLHQF\ LQ PLFH OHDGV WR DEQRUPDO EUDLQ GHYHORSPHQW DQG SHULQDWDO GHDWK 3URF 1DWO $FDG 6FL 86$ 6ZLHUF]\QVNL 6/ 6LGGKDQWL 65 7XWWOH -6 DQG %ODFNVKHDU 3f 1RQP\ULVWR\ODWHG 0$5&.6 FRPSOHPHQWV VRPH EXW QRW DOO RI WKH GHYHORSPHQWDO GHIHFWV DVVRFLDWHG ZLWK 0$5&.6 GHILFLHQF\ LQ PLFH 'HY %LRO 7DXVNHOD -6 &KDNUDYDUWK\ %5 0XUUD\ &/ :DQJ < &RPDV 7 +RJDQ 0 +DNLP $ DQG 0RUOH\ 3 f (YLGHQFH IURP FXOWXUHG UDW FRUWLFDO QHXURQV RI GLIIHUHQFHV LQ WKH PHFKDQLVP RI LVFKHPLF SUHFRQGLWLRQLQJ RI EUDLQ DQG KHDUW %UDLQ 5HV 7RQLQL *3 3DURGL 07 'L0DUWLQR DQG 9DUHVLR / f ([SUHVVLRQ RI SURWHLQ NLQDVH &DOSKD 3.&DOSKDf DQG 0<&1 P51$V LQ KXPDQ QHXUREODVWRPD FHOOV DQG PRGXODWLRQ GXULQJ PRUSKRORJLFDO GLIIHUHQWLDWLRQ LQGXFHG E\ UHWLQRLF DFLG )(%6 /HWW f 7UHPEOD\ 5 +HZLWW /HVLXN + 0HDOLQJ 0RUOH\ 3 'XUNLQ -3 f (YLGHQFH WKDW EUDLQGHULYHG QHXURWURSKLF IDFWRU QHXURSURWHFWLRQ LV OLQNHG WR LWV DELOLW\ WR UHYHUVH WKH 10'$LQGXFHG LQDFWLYDWLRQ RI SURWHLQ NLQDVH & LQ FRUWLFDO QHXURQV 1HXURFKHP f 7URW] 0 :HJQHU & DQG 1DX + f 9DOSURLF DFLGLQGXFHG QHXUDO WXEH GHIHFWV 5HGXFWLRQ E\ IROLQLF DFLG LQ WKH PRXVH /LIH 6FLHQFHV 7VXQHLVKL 6 f 5HJXODWLRQ RI QHXULWH RXWJURZWK WKURXJK SURWHLQ NLQDVH & DQG SURWHDVH QH[LQ LQ QHXUREODVWRPD FHOO .REH 0HG 6FL f 8EHUDOO ) *LVHOEUHFKW 6 +HOOEHUW )UHVVHU ) %DXHU % *VFKZHQGW 0 *UXQLFNH ++ DQG %DLHU f &RQYHQWLRQDO 3.&D QRYHO 3.&H DQG 3.& EXW QRW DW\SLFDO 3.&$ DUH 0$5&.6 NLQDVHV LQ LQWDFW 1,+ 7 ILEUREODVWV %LRO &KHP f 9DGQDO 5 ( DQG 3DUWKDVDUDWK\ 5 f $A\RLQRVLWRO PRQRSKRVSKDWDVH 'LYHUVH HIIHFWV RI OLWKLXP FDUEDPD]HSLQH DQG YDOSURDWH 1HXURS\VFK\RSKDUP 9DX[ '/ f 7RZDUG DQ XQGHUVWDQGLQJ RI WKH PROHFXODU PHFKDQLVPV RI SK\VLRORJLFDO FHOO GHDWK 3URF 1DWO $FDG 6FL 86$ :DOGPHLHU 3 & f ,V WKHUH D FRPPRQ GHQRPLQDWRU IRU WKH DQWLPDQLF HIIHFW RI OLWKLXP DQG DQWLGHSUHVVDQWV" 3KDUPDFRSV\FKLDWU\ :DOWRQ 6$ %HUN 0 DQG %URRN 6 f 6XSHULRULW\ RI OLWKLXP RYHU YHUDSDPLO LQ PDQLD D UDQGRPL]HG FRQWUROOHG VLQJOHEOLQG WULDO &OLQ 3V\FKLDWU\ f

PAGE 148

:DQJ +< DQG )ULHGPDQ ( f (QKDQFHG SURWHLQ NLQDVH & DFWLYLW\ DQG WUDQVORFDWLRQ LQ ELSRODU DIIHFWLYH GLVRUGHU EUDLQV %LRO 3V\FKLDWU\ f :DWVRQ '* DQG /HQR[ 5+ f &KURQLF OLWKLXPLQGXFHG GRZQUHJXODWLRQ RI 0$5&.6 LQ LPPRUWDOL]HG KLSSRFDPSDO FHOOV 3RWHQWLDWLRQ E\ PXVFDULQLF UHFHSWRU DFWLYDWLRQ 1HXURFKHP :DWVRQ '* DQG /HQR[ 5+ f &KURQLF OLWKLXP H[SRVXUH OHDGV WR WKH GRZQ UHJXODWLRQ RI ERWK 0$5&.6 P51$ DQG SURWHLQ LQ LPPRUWDOL]HG KLSSRFDPSDO FHOOV f 6RF 1HXURVFL $EVWU :DWVRQ '* :DLQHU %+ DQG /HQR[ 5+ f 3KRUERO HVWHU DQG UHWLQRLF DFLG LQGXFHG UHJXODWLRQ RI WKH 3.& VXEVWUDWH 0$5&.6 LQ LPPRUWDOL]HG KLSSRFDPSDO FHOOV 1HXURFKHP :D\V '. 3RVHNDQ\ GH9HQWH *DUULV 7 &KHQ +RRNHU 4LQ : &RRN 3 )OHWFKHU DQG 3DUNHU 3 f 2YHUH[SUHVVLRQ RI SURWHLQ NLQDVH &4 VWLPXODWHV OHXNHPLF FHOO GLIIHUHQWLDWLRQ &HOO *URZWK 'LIIHU :HLQHU % HGf f 3'5 *HQHULFV QG HG 0HGLFDO (FRQRPLFV 0RQWYDOH 1:LJHU 5 6WRWWXP $ DQG %UXQERUJ f (VWLPDWLQJ FKHPLFDO GHYHORSPHQWDO KD]DUG LQ D FKLFNHQ HPEU\R OLPE EXG PLFURPDVV V\VWHP 3KDUPDFRO 7R[LFRO :LON $/ *UHHQEHUJ -+ +RULJDQ ($ 3UDWW 50 DQG 0DUWLQ *5 f 'HWHFWLRQ RI WHUDWRJHQLF FRPSRXQGV XVLQJ GLIIHUHQWLDWLQJ HPEU\RQLF FHOOV LQ FXOWXUH ,Q 9LWUR f :RRWHQ 0: f 'LIIHUHQWLDO H[SUHVVLRQ RI 3.& LVRIRUPV DQG 3& FHOO GLIIHUHQWLDWLRQ ([S &HOO 5HV f ;LH ; DQG +DJDQ 50 f &HOOXODU DQG PROHFXODU DFWLRQV RI ODPRWULJLQH 3RVVLEOH PHFKDQLVPV RI HIILFDF\ LQ ELSRODU GLVRUGHU 1HXURSV\FKRELRORJ\ f
PAGE 149

=KDQJ .= :HVWEHUJ -$ +ROWWD ( DQG $QGHUVVRQ /& f %&/ UHJXODWHV QHXUDO GLIIHUHQWLDWLRQ 3URF 1DWO $FDG 6FL 86$ =XEHU 0 ; *RRGPDQ ': .DPV /5 DQG )LVKPDQ 0& f 7KH QHXURQDO JURZWK DVVRFLDWHG SURWHLQ *$3 LQGXFHV ILORSRGLDO IRUPDWLRQ LQ QRQQHXURQDO FHOOV 6FLHQFH

PAGE 150

%,2*5$3+,&$/ 6.(7&+ 7KH DXWKRU ZDV ERUQ LQ LQ 0LDPL 6KRUHV )ORULGD WR WZR ZRQGHUIXO SDUHQWV -RKQ DQG &HOHVWH :DWWHUVRQ 6KH ZDV UDLVHG RQ WKH HDVW FRDVW RI )ORULGD WKH ROGHVW RI WKUHH FKLOGUHQ 6KH HDUQHG KHU EDFKHORUnV RI VFLHQFH GHJUHH LQ SUHPHGLFDO ELRORJ\ DW 0LDPLnV %DUU\ 8QLYHUVLW\ LQ JUDGXDWLQJ LQ WKUHH \HDUV ZLWK KRQRUV 6KH WKHQ EHJDQ JUDGXDWH VWXGLHV LQ 3KDUPDFRORJ\ DW WKH 8QLYHUVLW\ RI )ORULGD LQLWLDOO\ FRQGXFWLQJ UHVHDUFK LQ WKH ODERUDWRU\ RI 5REHUW + /HQR[ 0 DQG ILQLVKLQJ XQGHU WKH JXLGDQFH RI (GZLQ 0 0H\HU 3K 6KH LV FXUUHQWO\ VLQJOH DQG WKH SURXG PRPP\ RI D EHDXWLIXO FKRFRODWH /DEUDGRU UHWULHYHUPL[ QDPHG +HUVKH\

PAGE 151

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ (GZLQ 0 0H\HU &KDnLU $VVRFLDWH 3URIHVVRU RI 3KDUPDFRORJ\ DQG 7KHUDSHXWLFV FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ASUHVHQWDWLRQ DQG Mt7XOO DV D GLVVHUWDWLRQ IRU WKH GHJUHHI 'RFWRU RMD3KLOGVRL LLQLRQ LW FRQIRUPV WR nDGHTXDWH LQ VFRSH DQG TXDOLW\ 5REHUW /HQR[ &RFK£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n LWH 3URIHVVRU RI 3KDUPDFRORJ\ DQG 7KHUDSHXWLFV FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ F &ROLQ 6XPQHUV 3URIHVVRU RI 3K\VLRORJ\

PAGE 152

7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI 0HGLFLQH DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 'HFHPEHU f§f§ V! Fƒ &YYnn[ 'HDQ &ROOHJH RI 0HGLFLQH 'HDQ *UDGXDWH 6FKRRO

PAGE 153

81,9(56,7< 2) )/25,'$