9:337

_UBRARY l
Michigan State
University

 

This is to certify that the
thesis entitled

MUSCLE ATROPHY AND ALTERATION OF METABOLIC
GENE EXPRESSION DURING EARLY STREPTOZOTOCIN
INDUCED DIABETES

presented by

Erika Templeton

has been accepted towards fulfillment
of the requirements for the

MS. degree in Physiology

 

 

Wow:

Major Professor’s Signature

% (2f 1001

Date

MSU is an affinnative-action. equal-opportunity employer

0 ’

'

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p'lClRC/DaIeDue.indd-p.1

MUSCLE ATROPHY AND ALTERATION OF METABOLIC GENE
EXPRESSION DURING EARLY STREPTOZOTOCIN INDUCED DIABETES
By

Erika Templeton

A THESIS

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

MASTER OF SCIENCE
Department of Physiology

2007

ABSTRACT

MUSCLE ATROPHY AND ALTERATION OF METABOLIC GENE EXPRESSION
DURING EARLY STREPTOZOTOCIN INDUCED DIABETES

By

Erika Templeton

Diabetes mellitus can lead to detrimental consequences on organ systems,
including that of skeletal muscle degeneration. A model emulating the effects of type I
diabetic insulin deficiency can be created in a murine model using the pharmacological
administration of streptozotocin (STZ). When treated with acute STZ, mice undergo
rapid weight loss and heightened blood glucose. Components of the insulin/IGF-l
signaling cascade were examined to determine whether protein degradation and protein
synthesis mechanisms were altered. PGC-la, a key regulator of metabolism as well as a
putative preventor of muscle atrophy was also investigated. To define the role of STZ
diabetes on muscle atrophy and altered expression of metabolic genes, several hindlimb
muscles of the diabetic mice were examined in parallel. In STZ mice, weight loss was
apparent accompanied by decreased muscle weight in the tibialis anterior. By contrast,
the muscle weights of the soleus and plantaris were unchanged. As verification of
muscle atrophy, molecular markers of atrophy were examined. MuRFl was unchanged
while atrogin-l mRN A expression increased in the plantaris and tibialis anterior. Akt-2
phosphorylation in the tibialis anterior and plantaris were also decreased, corresponding
to the increase in atrogin-l. These results indicate that in early acute STZ diabetes,
insulin signaling cascade encourages muscle atrophy in the predominantly glycolytic
muscles. In contrast, the soleus displayed decreased expression of PGC-l at this time
period while the plantaris and tibialis anterior remained unaltered. Thus, different
muscles respond differently in onset of muscle atrophy as well as metabolic regulation

during the early stages of acute STZ diabetes.

Dedicated to my parents, Dave and Lynn Templeton, and brother, Michael Templeton,

who continually provide their support in all of my endeavors.

iii

ACKNOWLEDGEMENTS

I would like to thank my committee members, Dr. Robert Wiseman, Dr. Laura
McCabe, and Dr. Ron Meyer for their support in this project. I would specifically like to
thank Dr. Wiseman for his guidance, patience, and support in the completion of this
study. Much thanks goes out to the McCabe laboratory, for providing the animals used in
this project and for their continued assistance. I would like to thank Dr. David Ankrapp
for his technical support in this project as well as Dr. Arthur Weber and Dr. Thomas

Adams for their guidance in completion of this BS/MS degree.

iv

 

 

 

 

 

 

 

 

 

 

 

TABLE OF CONTENTS

LIST OF TABLES vi
LIST OF FIGURES vii
KEY TO ABBREVIATIONS viii
SPECIFIC AIMS 1
INTRODUCTION 3
BACKGROUND AND SIGNIFICANCE 5
Diabetes Mellitus ............................................................................................................ 5
Insulin Signaling ............................................................................................................. 7
Mechanisms of Atrophy .................................................................................................. 8
Protein Degradation ................................................................................................... 9
Protein Synthesis ....................................................................................................... 11
Muscle Composition ..................................................................................................... 11
Metabolic Control: PGC-la .......................................................................................... 13
Type I Diabetes Model: Streptozotocin ........................................................................ 15
Current Project .............................................................................................................. 16
SPECIFIC AIM I i 19
Introduction 19
Materials and Methods .................................................................................................. 19
Preliminary Data ........................................................................................................... 26
Preliminary Conclusions ............................................................................................... 35
Continued Research ...................................................................................................... 37
SPECIFIC AIM H 39
Introduction ................................................................................................................... 39
Materials and Methods .................................................................................................. 40
Preliminary Data ........................................................................................................... 43
Preliminary Conclusions ............................................................................................... 48
Continued Research ...................................................................................................... 54
SPECIFIC AIM III - -- 57
Introduction ................................................................................................................... 57
Materials and Methods .................................................................................................. 58
Preliminary Data ........................................................................................................... 59
Preliminary Conclusions ............................................................................................... 61
Continued Research ...................................................................................................... 64
DISCUSSION ’ 67
BIBLIOGRAPHY 72

 

Table 1. Primer Sequences.

LIST OF TABLES

vi

25

LIST OF FIGURES

Figure 1. Insulin Receptor Signaing Pathway.
Figure 2. Mechanisms of Atrophy.

Figure 3. Blood Glucose levels.

Figure 4. Total Body Weight.

Figure 5. Average Muscle Weights.

Figure 6. Comparison of cDNA Preparations.
Figure 7. Comparison of Housekeeping Genes.
Figure 8. Atrogin-l mRNA Expression.
Figure 9. MuRFl mRNA Expression.

Figure 10. Western Blot Analysis: Akt-2
Figure 11. Western Blot Analysis: IRS-l.
Figure 12. Western Blot Analysis: p70 S6 kinase

Figure 13. PGC-la mRNA Expression.

vii

17

18

28

29

30

31

32

33

34

45

46

47

6O

STZ

SDS

RT-PCR

DTT

NTP

RNA

DNA

cDNA

KEY TO ABBREVIATIONS

Streptozotocin

Sodium Dodecylsulfate

Reverse Trancription-Polymerase Chain Reaction
Dithiothreitol

Nucleotide triphosphate

Ribonucleic acid

Deoxyribonucleic acid

Complementary Deoxyribonucleic acid

viii

SPECIFIC AIMS

PROJECT

Insulin is necessary for the majority of glucose uptake from the bloodstream and
is critical for energy homeostasis; however, this hormone is compromised in diabetes
mellitus. This disease causes muscle degeneration with a consequent loss in total body
weight as well as changes in metabolic regulation. This change in metabolic homeostasis
is likely induced by alterations in the insulin Signaling pathway influencing protein
expression as well as gene regulation of energetic processes. We propose to test the
hypothesis that streptozotocin (STZ) diabetes induces muscle atrophy and altered
regulation of metabolic genes. The current study investigates the role of the known
insulin signaling pathway in skeletal muscle atrophy using a mouse model of
streptozotocin induced diabetes.
SPECIFIC AIM I
Determine whether early STZ diabetes influences muscle weight and genetic markers of
muscle atrophy.
Physiologic measurements and quantitative real time PCR will be used.
SPECIFIC AIM II
Determine whether signaling enzymes that influence protein degradation and synthesis
are altered in STZ diabetes.
Western blot analysis and animal model measurements will be used.

SPECIFIC AIM III

Test whether streptozotocin diabetes alters expression of PGC-la, a transcriptional
regulator of glucose metabolism and overall cellular metabolic balance.

Quantitative real time PCR will be the used.

INTRODUCTION

Diabetes mellitus poses health threats by producing an abundance of
complications in many organ systems that include the cardiovascular, neurological,
gastrointestinal and skeletal muscle systems. The purpose of this study is to address the
role of streptozotocin (STZ) induced muscle atrophy and metabolic gene regulation
during diabetes.

In both human diabetic patients and animal models of diabetes, skeletal muscle
deterioration and remodeling ensues (Feczko et al. 1994; Oberbach et al. 2006). In
uncontrolled diabetes, muscle loss occurs immediately following nutrient imbalance.
This evidence suggests the onset of diabetes signals skeletal muscle remodeling very
early in the disease. In the absence of insulin, ubiquitin li gases capable of protein
degradation are actively transcribed while protein synthesis machinery is also affected
(Kandarian et al. 2006). Therefore, determining the mechanism of atrophy requires
inspection of both protein synthesis and degradation pathways. Skeletal muscle system is
comprised of many different types of proteins and isoforms that offer specialization
among each of the muscles. Further complicating the investigation of muscle is the
varying composition of muscle types.

Following exercise, peroxisome proliferator activated receptor-y coactivator-l
(PGC—l), a protein responsible for upregulation of genes involved in oxidative
metabolism is highly expressed. Although the increase in PGC-la expression in
hypertropht has been well-characterized, the role it plays in atrophy, if at all, is unclear.

When PGC-la is overexpressed, muscles are more resistant to atrophy (Sandri et al.

2006). This group also suggests that PGC-la is capable of physically interacting with
FOXO, implicating PGC-la as an effector of the insulin signaling cascade. PGC-la
levels have been shown to decrease with type H diabetes in muscle (Patti et al. 2003).
The role of PGC-la STZ diabetes mediated muscle atrophy is not entirely understood.

The goals of the following proposed project are to: 1) demonstrate skeletal muscle
atrophy is induced by streptozotocin (STZ) diabetes; 2) elucidate activation levels of
insulin receptor signaling components in early type I diabetes; 3) determine activation
levels of the insulin dependent degradation pathways; 4) determine activation of the
insulin-regulated protein synthesis pathway; 5) ascertain whether PGC-la expression is
altered in the muscle during STZ diabetes. Preliminary experiments have been

completed for each of the Specific Aims that support proposed experiments.

BACKGROUND AND SIGNIFICANCE

Diabetes Mellitus

Diabetes mellitus is a disease affecting over six percent of the United States
population and growing in epidemic proportions according to the Center for Disease
Control (Center for Disease Control 2005). Diabetes is the sixth leading cause of death
but could be larger because many diabetes related deaths are underreported as a cause
(Center for Disease Control 2005). Currently, there is no known cure for diabetes, but
successful management can occur through diet, exercise, and pharrnacologic intervention.
However, even with treatment, the disease may generate debilitating consequences and
an overall poor prognosis. Long term consequences include retinopathy, neuropathy,
osteoporosis, kidney disease, stroke, heart disease, and myopathy (Center for Disease
Control 2005).

Three different forms of diabetes mellitus have been characterized. Though all
types are marked by hyperglycemia, they can be distinguished by their origination and
progressiveness. Gestational diabetes is acquired during pregnancy in women who do
not normally have diabetes (Phillips et al. 2006). Type H diabetes, formerly called adult-
onset, is chronic and progressive. Insulin resistance leads to overproduction of insulin
itself eventually compromising B-cell function in the pancreatic islets of Langerhaans.
Type I diabetes, once known as juvenile, actually occurs with approximately the same
frequency in adults and juveniles (National Institutes of Health 2006). Patients with type
I diabetes mellitus develop hypoinsulinemia and consequently hyperglycemia due to lack

of insulin. Currently, the mechanism of B-cell destruction is unknown, but potential

factors are thought to be genetic, autoimmune, or environmental. In all forms of diabetes
mellitus, glucose, protein and lipid balance is altered (Beck-Nielson et a1. 1992). The
metabolic imbalance produced by diabetes causes altered gene expression and protein
modification (Antonetti 1995; Crozier et al. 2003). Alterations at the protein level allow
the body to compensate for perturbations in homeostasis; however, these modifications
can be maladaptive.

With disease progression, further health problems arise as a consequence of
diabetes mellitus. One complication attributable to hyperglycemia caused by diabetes is
retinopathy, eventually leading to blindness. Eventually, cardiopathy may become a
problem coinciding with obesity, hyperlipidemia and hypertension. Diabetes associated
neuropathy and osteoporosis also develop (Bouillon et al. 1995). Myopathy is a
prolonged consequence of diabetes, but immediate alterations are also visible. In the
long term, muscle may begin to be altered in growth patterns (Feczko et al. 1994). In the
study by Feczko et al. (1994) mice were induced with type I diabetes by pharmacologic
means. After 42 days of diabetes, the extensor digitorum longus muscle of the mouse
hindlimb was inspected histochemically. The authors found a shift in the distribution of
fibers and destruction of the normal “checkerboard” pattern of fiber distribution (Feczko
et a]. 1994). This data displays the impact that STZ diabetes has on muscle over an
extended period of time.

Skeletal muscle comprises 40-50% of total body mass (Song et al. 1999). It
provides structural and mechanical support for the body, but also serves as a potential
reservoir of amino acids that can be utilized when necessary. Furthermore, skeletal

muscle is responsible for the majority (70%) of total glucose uptake from the

bloodstream (Mas et al. 2006). In order for skeletal muscle to accomplish this feat,
insulin is a necessary component (Zierath et al. 2000).

When insulin binds to receptors on the outer membrane of the muscle cell, it
elicits a cascade that ends with the translocation of more glucose transporters to the
surface of the muscle in order to relieve the bloodstream of hyperglycemia. As insulin
signaling is lost in type I diabetes, the ability to signal the muscle to take up more glucose
is also inoperative. Without insulin in appropriate proportions, the muscle perceives that
there is not enough available energy in the body. At this point, the muscle begins to
breakdown its own supply of proteins in order to accommodate the negative energy
balance. This is maladaptive in the diabetic condition since there is an abundance of
glucose available for use in the bloodstream, but it is unable to enter the skeletal muscle.
When the muscle begins to breakdown more than it is able to build, it is referred to as
muscle atrophy or wasting. The signaling components and mechanisms that lead to
muscle atrophy have not completely been resolved, though during diabetes lack of insulin
may play a role. In this study we investigate the role of insulin signaling using a model
that mimics the insulin loss as seen in type I diabetes mellitus.

Insulin Signaling

The insulin signaling cascade is depicted in Figure 1, initiating with binding of the
ligand to the receptor. As a tyrosine kinase receptor it has the ability to
autophosphorylate on the B-subunit at multiple residues in order to activate further
Signaling proteins (Kahn 1994). Insulin receptor phosphorylation consequently initiates
phosphorylation of the IRS (insulin receptor substrate) adaptor proteins thus causing the

signal to be passed (Virkamaki A et a1. 1999). IRS is able to activate

phosphatidylinositol (PI) 3-kinase which then causes activation of Akt, also known as
protein kinase B (PKB) (Oku ,et al. 2001). Akt is activated by phosphorylation at amino
acid positions Thr 308 and Ser 473 by phosphatidylinositol 3, 4, 5-triphosphate
dependent protein kinase (PDK) (Morino et al. 2006). PI3K and Akt have the ability to
transduce Si gnals across multiple pathways imposing regulation on enzymatic processes.

This intricate signaling cascade also employs proteins that exist in multiple
isoforms. Several isoforms of IRS adaptor proteins have been identified although IRS-l
and IRS-2 predominate in skeletal muscle (Bouzakri et al. 2006). Bouzakri et al. showed
in myotubes that IRS-2 is dispensable in insulin Signaling while IRS-1 is necessary
(2006). Three isoforms of Akt have been shown to exist, but this group determined that
Akt—2 is essential to insulin signaling. They concluded that IRS-1 and Akt-2 are crucial
to myoblast differentiation as well as glucose metabolism.

When the insulin Signaling cascade reaches PI3K and Akt multiple proteins can
be activated. The putative scheme of atrophy mechanisms under the control of P13K are
displayed in Figure 2. The major pathways associated with insulin Si gnaling include the
caspase-3 apoptotic pathway, FOXO transcription factors, and mTOR (mammalian target
of raparnycin) pathway. Activation of FOXO and mTOR are mediated through Akt
Signaling. These pathways are discussed in the following section.

Mechanisms of Atrophy

Skeletal muscle responds quickly to drastic declines in energy demands, rapidly
degenerating proteins into amino acids that can be used for energy. During diabetes,
muscle continually remodels. When the muscle is in a state in which protein

degeneration exceeds protein synthesis, it is referred to as atrophy. Atrophy has been

characterized by Lecker and colleagues (2004) as both a decrease in protein synthesis and
an increase in protein degradation.
Protein Degradation

The first Akt-mediated mechanism of muscle breakdown in need of investigation
involves members of the ubiquitin proteosome pathway. The drastic increase in muscle
proteolysis is due to an increase in ubiquitination (Lecker et al. 2004). Degradation via
the ubiquitin proteosome path is ATP-dependent and involves li gases that tag proteins
destined for degradation with a string of ubiquitin molecules (Hershko et al. 1998).
Ubiquitination allows the proteins to be identified and destroyed by the 26S proteosome
(Hershko et al. 1998). Many ubiquitin ligases are thought to exist, but two are
dramatically upregulated during Skeletal muscle atrophy. Atrogin-l/MAFbx (muscle
atrophy F box) was originally identified simultaneously by Gomes et al. (2001) and
Bodine et al. (2001). Discovery of MuRFl was also accomplished by Bodine and
colleagues (2001). Both atrogin-l and MuRF 1 have been shown to increase in thirteen
different forms of atrophy (Bodine et al. 2001). Collectively atrogin-l and MuRFl can
be referred to as atrogenes.

Found in skeletal and cardiac muscle, atrogin-l and MuRFl have been
characterized as E3 ubiquitin li gases (Cao et al. 2005). The Ub-protein ligases are
specific for a certain class of proteins to mark for breakdown (Hershko et al. 1998). The
E3 portion of the enzyme is possesses a recognition site to confer specificity on the
substrate (Passmore et al. 2004). Several groups have demonstrated that atrogin-l plays a
role in both initiation and overall maintenance of protein degradation (Gomes et al. 2001;

Lee et al. 2004; Sandri et al. 2004). Both atrogin-l and MuRFl null mice appear

phenotypically identical to litterrnates until they undergo hindlimb suspension, whereby
the null mice are resistant to muscle atrophy. The atrogin-l null mouse exhibits a greater
resistance to muscle wasting than the MuRFl null mouse model (Bodine et al. 2001).
Lecker et al. Showed that in many forms of skeletal muscle atrophy, atrogin~1 and
MuRFl were upregulated in mRNA expression (2004). As these genes are shown to
readily increase with atrophy, they can be used as molecular markers of muscle wasting.

Atrogin-l expression is controlled through Akt signaling. Transcription of this
ubiquitin ligase is controlled directly by the FOXO family of transcription factors (Lee et
al. 2004; Sandri et a1. 2004). FOXO transcription factors are effective activators of
atrogin-l as FOXO is able to bind directly to the promoter for atrogin-l (Sandri et al.
2004). Though the target of atrogin-l is not completely clear, MyoD and calcineurin
have been identified as potential substrates (Lee et al. 2004; Tintignac et al. 2005).

MuRF-l (muscle-specific RING finger protein 1) is a 40 kDa protein belonging to
the family of TRIM proteins or tripartite because it contains three important domains
(Bodine et al. 2001; Kamura et al. 1999; Centner et al. 2001). These domains include a
RING finger domain, allowing for ligase activity, a B-Box domain with an unknown
function, and a coiled-coil domain which may be necessary for the formation of
heterodimers (Kamura et al. 1999; Centner et al. 2001). Kedar et al. (2004) have shown
that MuRFl has the ability to bind cardiac troponin I, while Weinert et al. (2006)
determined that the structural protein titin contains a binding site for MuRF 1.

FOXO transcriptional activators regulate atrogin-l expression, but whether they
are able to activate MuRFl is controversial. When phosphorylated, FOXO is contained

by 14-3-3 proteins in the cytosol (Nader 2005). However, upon dephosphorylation they

10

readily translocate to the nucleus to influence transcription (Sandri et al. 2004). To date,
three mammalian isoforms have been identified (Nader 2005). These forms include
FOXOl (FKHR), FOXOBa (FKI-IRLl) and FOXO4, all of which are thought to be under
the dominant control of Akt (Nader 2005).

Protein Synthesis

As well as activating means of protein degradation, Akt also controls protein
synthesis. Phosphorylated Akt facilitates the phosphorylation of mTOR (mammalian
target of rapamycin) kinase. This protein has the ability to cause phosphorylation of p70
S6 kinase, which as its name suggests, is able to phosphorylate the p70 subunit of the S6
ribosome (Fingar et al. 2004). S6 ribosome phosphorylation allows protein production to
increase. mTOR is also able to phosphorylate 4E-BP1 (Elongation factor 4E binding
protein 1), which relieves suppression on 4E, encouraging translation (Kandarian et al.
2006). Overall the mTOR pathway is highly active when nutrients are present allowing
muscle hypertrophy to occur (J ackman et al. 2004). Conversely, inactivation of this
pathway leads to decreased protein synthesis, resulting in atrophy.

Studies have Shown that FOXO and mTOR pathways are regulated by Akt
signaling. As a member of the insulin receptor signaling cascade, Akt activation may be
altered by diabetes. Muscle atrophy occurs during uncontrolled diabetes and at different
rates depending on the muscle type used (Lecker et al. 2004; Thomason et al. 1987). The
complete mechanism as to how atrophy occurs is unclear, while the process that makes
atrophy occur at different rates in muscles is particularly vague.

Muscle Composition

ll

Thomason and colleagues (1987) showed that wasting occurs at different in
various muscles; however, it is unclear why this is the case. One explanation may be
substrate preference between the muscle fiber types. Another reason could be that the
functional load between muscles varies so that the muscles can selectively remodel
according to necessity. A third explanation may be that muscle contains different
concentrations of protective proteins which differ between fiber type. Though the
difference in atrophy rates is unclear, the variation between muscle types has been
characterized.

To fully grasp how muscle atrophy may occur at different rates an understanding
of the muscle composition by type in necessary. Muscles are categorized into three
predominant fiber types based on several distinguishing factors. Fibers can be
categorized according to myosin ATPase isoform. As a contractile protein, myosin is a
major contributor to differences in speed and sustainability of contraction. According to
their method of energy production, muscles are classified as oxidative or glycolytic.
Oxidative fibers or slow twitch, have increased ability to perform oxidative respiration,
fatty acid oxidation, increased triglyceride storage, and a lower glycolytic capacity in
comparison to glycolytic muscle fibers (Nemeth et al. 1979). In order to have increased
levels of oxidative respiration, type I fibers contain greater amounts of mitochondria.
Furthermore, the fiber types contain different sets of enzymes and protein isoforms
(Puigserver 2005). Glycolytic muscles rely mainly on glycolysis for their energetic
demands and cannot sustain contraction for extensive time periods. Glycolytic muscle
fibers are further divided into type Ila (oxidative glycolytic) and type IIb (fast glycolytic)

fibers. Type IIa fibers have a greater oxidative capacity and are more fatigue resistant

12

than type IIb glycolytic fibers (Saltin et al. 1983). Variations in muscle fiber proteins
provide the necessary functional capacity.

In experimental diabetes, histochemical analysis of muscles shows that fiber type
switching may begin to occur from a Slow to fast phenotype (Feczko et' al. 1994).
Niederle et al. (1978) demonstrated a Shift from type 2A to 2B in denervated rats. It was
also Shown that 2B fibers underwent atrophy more quickly than 2A fibers (Niederle et al.
1978). In time course study of hindlimb unweighing, Thomason et al. Showed that the
rate and degree of muscle loss differed according to muscle type (1987). In their model,
the soleus muscle was the most rapid to undergo atrophy (Thomason et al. 1987).
Furthermore, phosphorous containing metabolites and regulatory enzyme functions are
also altered in a fiber-specific manner during diabetes (Fewell et al. 1995).

Although skeletal muscles have distinct properties for functionality, all are void of
insulin signaling during type I diabetes. Despite substrate selectivity and load of the
muscle, different fibers have variations in gene activation. The reason for the disparity in
degeneration of muscles is unknown.

Metabolic Control: PGC-la

A primary difference between muscle types is the proportion of glycolytic and
oxidative fibers composing the muscle. The transcriptional coactivator PGC-la is
capable of controlling transcription of metabolic genes. In type I diabetes, loss of insulin
signaling causing metabolic imbalance in the muscle, therefore PGC-la activity may be
altered. PGC-la may also influence FOXO and Akt activity, two key components in

muscle atrophy (Sandri et al. 2006). Much research has characterized the role of PGC-la

13

expression in response to exercise, but less has been revealed as to how PGC-la
expression is influenced by type I diabetes.

Metabolic pathways achieve control partially at the level of gene expression.
Transcription factors exert a level of control on gene expression, but coacti vators are able
to modulate the activity of the transcription factors. In skeletal muscle, PGC-la has been
shown to have an important role in mitochondrial biogenesis and fiber type switching.
Fiber type switching causes muscle fibers to change phenotypes, effectively altering the
exercise capacity of the muscle.

Sandri and colleagues (2006) took many forms of atrophy into account including
cancer, cachexia, diabetes, renal failure, and denervation in order to determine that PGC-
la mRN A levels decreased among the tibialis anterior muscles that were atrophic (Sandri
et al. 2006). This group also created transgenic mice overexpressing PGC-la in Skeletal
muscle. They subjected these mice to denervation and fasting, discovering that the
transgenic mice had less muscle atrophy and decreased induction of atrogenes than the
control mice. They concluded that the ability of FoxO3 to bind at the atrogin-l promoter
was decreased in the PGC-la overexpressing mice (Sandri et al. 2006). This group
suggested that PGC-la physically inhibited the actions of FoxO3 preventing the
induction of atrophy. FOXO controls expression of atrogenes within the insulin signaling
cascade, further implicating PGC-loi alteration in diabetes and muscle atrophy.

Other studies have shown that overexpression of PGC-loi increases mitochondria
and myoglobin content, key components of slow muscle fibers (Lin et al. 2002).

Furthermore, PGC-la is preferentially expressed in type I muscles (Puigserver 2005).

14

PGC-la clearly has a role in metabolism and a connection to the FOXO family, though
these roles during muscle atrophy are in need of clarification.
Type I Diabetes Model: Streptozotocin

Multiple forms of animal models have been developed to emulate the diabetes
mellitus. Genetic and pharmacologic models both exist for further manipulation and
study. One pharmacologic mouse model of type I diabetes mellitus can be created in the
lab using the drug, streptozotocin (Like et a1. 1976).

The chemical structure of streptozotocin is similar to a glucose molecule enabling
it to enter cells bearing the GLUT2 glucose transporter. GLUT2 glucose transporters are
found in the hypothalamus, basolateral side of the small intestine, liver and in a larger
proportion on the B-cells of the pancreatic islets. Other glucose transporters are located
in other tissues; however, STZ is unable to enter. As the drug enters the B-cells, it elicits
a cytotoxic effect by causing damage to the DNA by alkylation (Szkudelski 2001). Using
multiple low doses of STZ, the development of diabetes can be imitated (Like et al.
1976). When diabetes is induced by this method, the animals Show decreased growth,
hyperglycemia, hypoinsulinerrria, and hyperlipidemia (Chen et al. 1982).

Acting as a cytotoxin, the question whether STZ has effects on other cell other
than the B-cells of the pancreas may arise. All tissues rely on glucose uptake for survival,
which hold a potential role for STZ to elicit a negative effect. When comparing the
histology of STZ to genetic models of diabetes, Skeletal muscle morphology appears very
similar (Feczko et al. 1994). This would suggest that the drug does not have a negative
effect directly on the muscle and that the alterations visible in muscle are a consequence

of the metabolic imbalance created by streptozotocin.

15

Current Project

The purpose of this current project is to determine the actions of diabetes on
Skeletal muscle. Understanding the influence of diabetes on skeletal muscle will help to
understand the impact of diabetes in other organ systems. Diabetic models that undergo
rapid muscle degeneration are useful to provide insight into the dynamic nature of
skeletal muscle and their overall role in total body metabolic balance.

This research has investigated a portion of this problem while suggesting further
projects to continue this inquiry. At this point, this research begins to identify a portion
of the body’s mechanism in response to early STZ diabetes. This project examined the
insulin signaling components that control protein catabolism and anabolism without
consideration of converging pathways so that these specific paths may be isolated. Also
considered was PGC-la, an important component to metabolic control and potential
mediator of muscle atrophy. The diabetic model here was examined two days following
the final acute injection of streptozotocin. Along with preliminary data presented, further
experiments are proposed that would investigate a pathway that may explain the
difference in atrophy between muscles. Additional experiments will allow Speculation
about what cellular components are important in the prevention of muscle wasting.
Although research has characterized the effect of Streptozotocin in Skeletal muscle, few
studies have examined different muscle types in parallel and at any early time of STZ
treatment. This project may help to elucidate the role of diabetes in early muscle atrophy
in differing muscle types, which will assist in directing further research on atrophy in

skeletal muscle.

16

 

 

Insulin

 

 

Insulin

L Receptor
"' ' \\

.

 

 

 

(f .
IRS-1

I p85

PISK W

.1 +
..5KT_..
/ \

mTOR FOXO

(protein synthesis) (gene expression)

 

 

 

 

Figure 1. Insulin Receptor Signaling Pathway. This diagram depicts some of the
proteins affected by the binding of insulin to the insulin receptor. When insulin
binds, tyrosine autophosphorylation of the receptor occurs resulting in the
phosphorylation of IRS-1 (Insulin receptor substrate-1). This causes
phosphorylation of PI3K (phosphatidyl inositol-3 kinase) which is able to influence
phosphorylation of Akt. Akt is then able to influence protein synthesis through the
mTOR pathway and gene expression through a FOXO-mediated path. This
diagram has been modified from Kido et al. (2001).

17

Deactivation/

 

 

 

dephosphorylation
® ic-Pisk
/ 4E-BP1 *6.)
f activation , , T
C -3 deactivation Decreased
Increased 339339 (9 ix- AKT —' mTOR *6“) protein
. I I synthesis
apopoSIs Aw"
fragmentation p70 36K *6) J
Ub proteosome!
degradation
Cytoplasm
uni 't' it
Nucleus r / 0334,17,?
(D It FOXO activation AII'O in-1
Increased L , /' g
Proteolysis AITOQ'N'I
L Activation of other atrophy genes (MuRF1)

 

Figure 2. Mechanisms of Atrophy. Putative atrophy signaling pathways associated
with diabetes mellitus. Boxes indicate proteins, phosphorylation is denoted by
circles. With dephosphorylation of PI3K, its activity is decreased. This causes
activation of caspase-3 and deactivation/dephosphorylation of Akt. Akt is able to
influence both catabolic and anabolic pathways. With dephosphorylation of F OX0,
atrogin-l is actively transcribed. When Akt causes dephosphorylation of mTOR,
protein synthesis is decreased by dephosphosphorylation of 4E-BP1 and p70 S6
kinase. Diagram has been modified from Kandarian et al. (2006).

18

SPECIFIC AIM I

INTRODUCTION

The first question that must be considered is whether the STZ diabetic model is an
appropriate model for the determination of early skeletal muscle wasting. In order to
destroy the B cells of the pancreas to ‘prevent insulin production, STZ was administered.
This causes the mouse model to become comparable to type I diabetes mellitus. Other
studies have shown that with administration of STZ, hyperglycemia, hypoinsulinemia,
and rapid weight loss occur in the animal (Cheng et al. 1982).

Despite a suspected decrease in total body weight with STZ, it needs to be
confirmed whether this model is losing weight in the form of skeletal muscle. In multiple
causes of atrophy, data has Shown two ubiquitin ligase mRN A levels to be increased
specifically in muscle (Bodine et al. 2001). Atrogin-l and MuRFl can therefore be used
as markers of skeletal muscle atrophy. These genes are capable of increasing in
expression before measurable weight loss occurs (Bodine et al. 2001; Gomes et al. 2001).
This report will focus on the physiologic data presented by the model of diabetes as well
as the genetic markers of atrophy. Data could then be expanded upon to determine the
methods of achieving consistent results with experimental STZ diabetic animals helping
to understand the impact of diabetes on muscle in more complex models.

Specific Aim I investigates the modification by STZ diabetes on body weight,

muscle weight and activity of atrophy markers in the skeletal muscle.

MATERIALS AND METHODS

19

Animals

Adult male (15 week old) BALB/c mice (Harlem Laboratories, Houston, TX)
were kept in 12 hour light-dark cycle at a temperature of 23 C and had access to food
(standard lab chow) and water ad libitum. All mice were housed in the Michigan State
University Biochemistry Animal Facility. Animal studies were performed in agreement
with the Michigan State University All-University Committee on Animal Care and Use.
The animals were examined daily.

Diabetes was induced using intraperitonial injection of streptozotocin (40 ug/ g
body weight in 0.1 M citrate buffer), a pancreatic beta cell cytotoxin, over the course of
five days (Like et al. 1976). The injections occurred during the same time period each
day. Control mice were injected with buffer alone and the treated and control mice were
kept in separate cages with 3 to 4 mice per cage. The tissues were harvested
approximately 24 hours following the last injection for the mice sacrificed days 1 through
5.

The animals were euthanized using a carbon dioxide chamber with the hindlimb
immediately removed following sacrifice. We removed the tibialis anterior, plantaris and
soleus muscles from one of the hindlimb. Blood glucose levels were measured post-
mortem using a glucometer (Accu-Check instant, Roche Molecular Biochemicals Corp.,
Indianapolis, IN). The muscles were immediately frozen in liquid nitrogen and stored at -
80 °C until further use. Muscle weights were taken prior to running an assay.
Experimental Groups

The mouse experimental groups were divided into eight groups to be sacrificed 1,

3, 5 and 7 days following initial injection of streptozotocin, each group containing control

20

and diabetic mice. Six control mice and eight treated mice composed each of the groups.
The focus of this project was the seven day group Since this would allow the mice two
days to recover from the last injection. These two days could allow any stress response
to the injection protocol to subside in the mice. Choosing this time point also provided
time for the pancreatic B cells to fully undergo cell death (Pi ghin et al. 2005).
Furthermore, at this time point the animals reached a heightened blood glucose level of
300 mg/dl, varying greatly from the control. Given the small Size of the muscles and
multiple assays to be run on each muscle sample, a number between 5-6 mice were used
for the controls and between 7-8 mice for the diabetic mice tissues.
Tissue Preparation

In order to utilize the muscles for multiple assays, the muscle was ground or
halved under liquid nitrogen and distributed into multiple vials. This was accomplished
by pestle and a mortar filled with liquid nitrogen and performed in a cold room. Prior to
pulverization or fractionation, the muscles were weighed while frozen and the weight was
recorded. The weight of each tissue distributed to each vial was also measured and
recorded. Different fiber types are distributed throughout the tibialis anterior so in order
to distribute the tissue more uniformly in each vial, it was necessary to crush the tissue.
The soleus and plantaris have more uniform fiber type distribution so could simply be
halved in order to use the tissues for separate assays. Also, as these muscles are much
smaller than the tibialis anterior, it was important to conserve as much tissue as possible.
The tissue was stored at -80 °C until it was used for further analysis.

Reverse Transcription and Quantitative Real Time PCR

21

The total RNA was isolated from the mouse muscles using Trizol reagent
(Invitrogen). The tissues were homogenized in the reagent then placed in a 37° C water
bath to allow for dissociation of nucleoprotein from the nucleic acids. Chloroform and
glycogen were added, the homogenate was placed in the centrifuge and following
centrifugation, the supernatant containing RNA was removed and isopropanol was added.
The RNA containing mixture was placed at 4° C overnight to allow for precipitation of
RNA to occur. The RNA precipitate was washed in ethanol and stored in forrnarnide at -
70° C for further use.

Reverse transcription of the RNA to cDNA was accomplished using a mixture of
Superscript H (Invitrogen), dNTPs, dithiothreitol, DEPC-water, RNAse inhibitor
(Invitrogen), 5x First Strand Buffer (Invitrogen) and primers (Invitrogen) in a 20 til total
reaction volume. A maximum of four primers were mixed in the primer mix at a time
and sequences were compared to discourage interaction of the primers. The mixture was
incubated at 42 °C for 60 minutes, followed by 72 °C for 15 minutes. In order to confirm
that the cDNA preparations were not varying greatly, separate cDNA samples were
prepared for some tissues. These separate cDN A samples were then analyzed for gene
expression using quantitative RT-PCR.

Quantitative real time PCR was carried out using the iQ Sybr Green Supermix
System, the iCycler and iCycler software for analysis (BioRad). Experiments were
performed in duplicate with the following parameters: rrrinutes at 95 °C for 3 minutes
followed by 40 cycles of 95 °C for 30 seconds, 58 °C for 60 seconds and finally with 72

°C for 30 seconds. To ensure the quality of analysis, a negative control (sample replaced

22

with RNAse free water) and a positive control consisting of a pool of positive samples
was used each time the qRT-PCR was run.

B-actin was used as the housekeeping gene and results were expressed using the
comparative threshold cycle of B-actin and the gene of interest. In order to confirm that
B—actin mRN A was acting as an appropriate control, some samples were analyzed for
cyclophilin as well as B-actin. B-actin is a ubiquitous cytoskeletal protein and has been
Shown to be stably expressed following resistance exercise in skeletal muscle (Mahoney
et al. 2004). Cyclophilin is found in the cytosol and functions in cellular protein folding
and interactions. The samples that were chosen for comparison of the housekeeping
genes consisted of both the plantaris and tibialis anterior as well as control and diabetic
muscles. The samples were run from the same cDNA preparations and were run through
quantitative real time PCR at the same time as well to ensure conditions were identical.
Therefore, any differing results could be attributable to the housekeeping gene.
Furthermore, expression levels of two target genes were compared to further assess the
reference genes.

All primers were verified using BLAST which determined that the sequence
homology was specific for the target gene. The primers were purchased from Invitrogen.
Forward and reverse sequences for each primer can be found in Table 1.

Threshold Cycle (CT) values for the Specific and control genes were obtained
automatically by the BioRad software. The data was then analyzed using the 2 ‘AACT
method (Livak et al. 2001). The relative level of the specific gene of interest (PGC-la,
MuRFl or atrogin-l) was determined by substracting the CT values of the housekeeping

gene (B-actin or cyclophilin) from the CT value of the gene of interest.

23

Statistical Analysis

Statistical analyses throughout this project were performed using Microsoft Excel
data analysis program for Student’s t test analysis. The values are expressed as means 1
SE (standard error of the mean). A p value less than 0.05 was considered to be

significant.

24

 

Primer Sequences
Atrogin- l/MAFbx Forward 5 ’ -GCAGAGAGTCGGCAAGTC-3’
Reverse 5 ’ —CAGGTCGGTGATCGTGAG-3 ’

 

 

 

 

MuRF—l Forward 5’-TGTCTGGAGGTCGI I I CCG-3’
Reverse 5 ’ -CAGGTCGGTGATCGTGAG-3 ’

PGC-la Forward 5’ -GCCGTGTGAI I I ACGTTGG-3’
Reverse 5 ’ -GATAGTACACGCTGTCCCATG-3’

B-actin Forward 5 ’ -GTATGCCT CT GGTCGTACCAC-3’

Reverse 5 ’ -CTAGAAGCACTTGCGGTGCA—3 ’

 

 

 

 

Table 1. Primer Sequences. Forward and reverse primer sequences used to
determine mRNA expression with quantitative real time PCR.

25

PRELIMINARY DATA
Blood Glucose

At day seven, the diabetic mice reached significantly elevated blood glucose
levels in comparison to the control mice (Figure 3). The treated mice had a blood
glucose level reaching approximately 300 mg/dL, while the control mice were below 200
mg/dL (McCabe et al., unpublished observation). This data corresponds to blood glucose
levels found in the literature for mice injected acutely with streptozotocin.
Body Weight

Within a time span of 7 days following initiation of drug treatment, the body
weights of the treated mice were less than the control mice, reaching a statistically
significant level. Total body weight decreased approximately 10% in the diabetic mice
by day 7. The body weights of the mice for the length of the time course can be found
graphically in Figure 4.
Muscle Weights

Whether there was a difference in muscle weights varied according to muscle
(Figure 4). The tibialis anterior experienced approximately a 14% decrease in weight in
comparison to the control weight. This weight difference was Significantly different
between diabetic and control. The soleus and plantaris muscles showed no apparent
difference in muscle weight across the entire time course.
cDNA Preparation analysis

Separate cDNA samples were prepared in order to determine variability in mRNA
analysis. The samples that were repeated showed consistency between the trials. This is

depicted in Figure 6.

26

Housekeeping Gene Analysis

The use of cyclophilin confirmed the results obtained from using B-actin as the
housekeeping gene. In Figure 7, the values are expressed as a percent of the overall
average of the expression level. The expression levels of most of the samples
corresponded between B-actin and cyclophilin when examining both at i'ogin-l and PGC-
la genes.
Molecular Markers of Atrophy

FOXO activation was assessed by detemrining mRNA levels of atrogin-l and
MuRF 1. This data is depicted graphically in Figure 8. The mRNA expression levels in
comparison to B-actin were assessed individually for each sample then averaged for the
group. In the tibialis anterior, the mRNA expression level of atrogin-l, but not MuRFl
was increased in comparison to the control muscles. The increase in expression reached
statistical Si gnificance among the atrogin-l mRNA. This suggests that atrophic
mechanisms had been induced in this muscle. The plantaris muscle trended toward an
elevated level in atrogin-l mRNA expression among the diabetic mice,‘ without obtaining
Si gnificance. Again, the MuRFl levels were unaltered between control and diabetic
plantaris muscles. In the soleus muscle, the levels of both atrogin-l and MuRFl
expression were not significantly altered overall at day 7. The expression levels are
compared to the levels of the housekeeping gene, B-actin. In order to assure that B
-actin levels were not changing with drug treatment or atrophy, cyclophilin was also
assessed as a housekeeping gene for some of the samples. Figure 7 displays validation of

B-actin as a housekeeping gene in comparison to cyclophilin.

27

Blood Glucose Levels

350.0
300.0 j ,
250.0 - i.

200.0 - I +Control
150.0 ~ i {-l— Diabetic
100.0 .

50.0 ~
0.0

 

 

 

 

 

Blood Glucose (mgldL)

 

 

 

\I

1 3 5
Time (days)

Figure 3. Blood Glucose Levels. Blood glucose was measured immediately prior to
sacrifice. The blood glucose levels reached statistically different values at day 3. At
day 7, the blood glucose levels reach 300 mg/dL. (McCabe et al., unpublished

observations). (as) Signifies significance.

28

 

 

Body Welght

30.0

*
25.0 - Mi!

20.0 .

 

 

 

—o— Ctrl

15.0 ~
-—I—— STZ

 

 

 

BOW Weight (9)

10.0 «

5.0 J

 

 

 

0.0

Time (Days)

 

 

 

Figure 4. Total body Weight. The mice were weighed prior to sacrifice. Each group
is representative of the average of 6 control mice and 8 diabetic mice. At day 7, the
two groups reach statistically different body weights. (McCabe et al., unpublished
observations)

29

Average Muscle Weights

0.06

 

0.05 -
0.04 —

0.03 ~

1 Control
LI Diabetic:

Weight (grams)

0.02 -

0.01 4

 

 

 

Tibialis Anterior Plantaris Soleus

Figure 5. Average Muscle Weight. The weights included are not tendon-free. The
control and diabetic samples of the soleus and plantaris muscles were not
statistically Significant. The tibialis anterior reached a statistically different weight
by day 7 between the two groups. (t) signifies significant difference (ps0.05)

30

Comparison of Separate cDNA Preparations

 

_L
ix)

.4
i

9
on
J_

E First Prep
I Second Prep ‘

 

Atrogin-I mRNA expression (A U)
C
O)

 

0.4 «
0.2
0 J
1 2 3 4 5 6 7
Samples

Figure 6. Comparison of cDNA Preparations. Expression levels of mRNA prepared
in separate cDNA reactions. The same RNA mixture was used to prepare separate
cDNA samples. The samples were then run in the same quantitative real time PCR
reactions. Both atrogin-l and B-actin were analyzed separately for each sample.
Both diabetic and control muscles were used. Arbitrary units (A U)

31

Comparison of Housekeeping Genes using Atrogin-I as
Target Gene

 

 
  
 

01
O
I

.— 250

.é

or

2 200 ~

5 s

E’ 3 150 « B-Actin
o o ..
3, E 100 1 ICyclophilin
‘5 I.IJ

E

§

0

n.

 

 

 

O
M

1 2 3 4 5 6 7
Samples

Comparison of Housekeeping Genes using PGC-1 as
target gene

_A
CD
0

 

160 -

_‘ A _L

co 0 N Ii

0 O O O
l l l

B-Actin
I Cyclophilin

60*
40‘
20~ 1"
0—

Percent of Average PGC-1
Expressmn

 

 

 

1 2 3 4 5 6 7 8
Samples

Figure 7. Comparison of Housekeeping Genes. Atrogin-l and PGC-la were both
used as target genes. This graph seeks to validate the use of B-actin as a
housekeeping gene in comparison to cyclophilin. Both control and diabetic as well
as various muscle samples were utilized. The sample numbers on the two graphs do
not correspond to each other.

32

Atrogin-1 mRNA Expression

 

b
I

I Control
I Diabetic

AtrogIn-t expresslon (A U)
N (a)

 

 

   

Tibialis Anterior Plantaris Soleus

Figure 8. Atrogin-l mRNA Expression. There was a significant difference among
the expression levels of atrogin-l in the tibialis anterior muscle. Although the
plantaris trended toward an increase in expression, the difference was not
statistically significant. Arbitrary Units (A U) (*) signifies significant difference
(p50.05)

33

MuRF1 mRNA Expression

 

2.5 ‘
A 2
D
S,
C
.3 1.5 —
3 IControl
a IDiabetic
K
O 1 I
E
I:
:I
E 0.5

0 -

 

 

 

Tibialis Anterior Plantaris Soleus
Figure 9. MuRFl mRN A Expression. There was no significant difference between

the control and diabetic expression levels for this gene among all three muscles.
Arbitrary units (A U)

34

PRELIMINARY CONCLUSIONS

Blood glucose levels were significantly elevated in the diabetic mice in
comparison to the control (McCabe et al., unpublished observations). Since heightened
blood glucose levels are characteristic of diabetes mellitus, it would indicate that these
mice are diabetic. Furthermore, the streptozotocin protocol used in this experiment has
been utilized frequently as a model of type I diabetes. Since the seven day time point was
chosen, it enabled the drug to be given for all of the proposed days, so that destruction of
the pancreatic B cells could be achieved fully. Since use of this method before has
confirm loss of insulin production, the insulin measurements were not assessed in this
study.

Body weights indicate a shift towards a decrease in weight, though not to the
extent that was anticipated. In unpublished data from the McCabe laboratory, extensive
weight loss was visible within a shorter time period when using the $31116 protocol
(McCabe et al. 2006). This could be attributable to the demeanor of the handler towards
the mice. Nonetheless, approximately 6% of total body weight was decreased in 7 days
(Figure 2). The weight loss exhibited by this model confirms results of weight loss
following STZ induced diabetes found in the literature.

In this study, only one of the muscle displayed a decrease in weight following
treatment. The tibialis anterior experienced visible weight loss. The plantaris trended
toward a weight decrease while the soleus showed no Sign of weight change. Weight
measurements were not tendon-free, which could have created an error in muscle
weights. Due to the time constraint during the muscle harvest, the tendons could not be

removed. To do so would have compromised the integrity of the tissues.

35

Atrogin-l was induced transcriptionally, indicating that atrophy is or is about to
take place. It is interesting that MuRFl expression was unaltered. Bodine et al. (2001)
as well as Gomes et el. (2001) determined that these atrophy markers were increased
prior to loss in body weight in multiple models of muscle wasting. The atrogin-l data
acquired in our study corresponds to their findings. This data may indicate that MuRFl
is not under regulation of the FOXO family of transcription factors as atrogin-l has been
Shown to be. This explanation is supported by Kamei and colleagues (2004) who showed
that overexpression of FoxOl caused atrogin-l expression to rise without a concomitant
increase in MuRFl expression. Alternatively, MuRFl upregulation may be induced at a
later time period than atrogin-l. Activation of atrogin—l expression corresponded to the
decrease in individual muscle weights exhibited by the muscles, suggesting that
upregulation of this ubiquitin ligase is occurring. Since atrogin-l is under the dominant
control of FOXO, these results suggest that FOXO has been dephosphorylated and
translocated to the nucleus in order to influence transcription.

According to both muscle weights and activation of atrogin-l, the results were not
consistent across the three muscles. The preliminary data suggests that the plantaris may
be in the beginning stages of atrophy. Since the tibialis anterior and the plantaris are
mainly comprised of glycolytic fibers, it is interesting that they were the first to trend
towards muscle atrophy. The data exhibited here point towards a protective role of
oxidative muscle during atrophy. Further analysis of the Specific mechanisms causing
delay in atrophy of the soleus needs to be assessed.

Also in need of consideration is the anatomical location of the muscle, which

provides differential stimulation patterns that may contribute to varied results among

36

muscles. The tibialis anterior displayed the greatest increase in activation of atrogin-l,
whereas the soleus and plantaris did not show a significant increase. Both of the former
muscles are located in the posterior compartment of the mouse hindlimb whereas the
tibialis anterior is located in the anterior position. This could indicate that there is a
selective loss of muscle fiber during diabetic muscle atrophy according to placement of
the muscle in the mouse hindlimb, which may be dependent upon usage.

The conclusion, based on preliminary data, is that this model does cause skeletal
muscle atrophy within one week of beginning STZ treatment. With this model, atrophy
occurs in different muscles at different rates. An understanding of the mechanisms

affected by STZ diabetes will facilitate further understanding of muscle atrophy.

CONTINUED RESEARCH

The mechanism leading to atrophy during diabetes has not been fully elucidated.
This study showed that the glucose levels were gradually increasing in the bloodstream of
the treated mice. Other research has shown that with administration of STZ diabetes in
conjunction with amelioration of hyperglycemia, considerable weight loss still occurs
(Cheng et a1. 2006). Further research is necessary to fully characterize the role of glucose
and insulin in atrophy. Blood levels of insulin should also be determined in continued
research. This could elucidate loss of insulin as a director of muscle atrophy at this time
point.

It is also interesting that when the same procedure was followed previously, the
mouse weight did not decrease to such an extent (McCabe 2006). Continued research

could include reproducing the animal trials multiple times. More tissue allotted for

37

mRNA would be beneficial to indicating the levels of the target gene. This would require
the use of many more animals.

Finally, animals could be analyzed at an earlier time point to determine when
atrogin-l is first activated in accordance with the loss of insulin Signaling. It would be
expected that further animal trials would supply more data showing earlier activation of
atrogin-l prior to a decrease in body weight. Further time points could also allow insight
into transcriptional upregulation of MuRFl in different muscle types. One way to test
whether hyperglycemia is at fault would be to treat animals as in the STZ protocol with
metformin to ameliorate hyperglycemia. Although metformin studies with streptozotocin
thus far have revealed that the animals still experience weight loss, it has not been
examined at such an early time point as the 7 day period (Suwa et al. 2006). This
research would be able to tell us whether hyperglycemia impacts the insulin signaling
pathway components directly or if hyperglycemia is the primary reason for Skeletal

muscle atrophy in the early stages of atrophy.

38

SPECIFIC AIM IIError! Bookmark not defined.

INTRODUCTION

When protein degradation exceeds protein synthesis, muscle wasting occurs.
Animal measurements and molecular markers of atrophy provided clues that STZ
diabetes is an activating response of muscle degradation. Since insulin depletion has
been demonstrated to occur in this model, components of the insulin signaling pathway
that can influence muscle wasting need to be examined. Western blot analysis is a
technique that allows for the differentiation between phosphorylation levels of proteins
involved in the insulin Signaling pathway.

Si gnaled through Akt, there are two major pathways that contribute to protein
synthesis and degradation. These pathways include FOXO and mTOR. Downregulation
of Akt causes dephosphorylation and activation of FOXO. Transcriptional upregulation
of atrogin-l occurs when FOXO transcription factors are active in the nucleus of the cell.
Atrogin-l is able to cause ubiquitination of proteins, tagging them for degradation.
Regulation of MuRFl is controversial, but it may also be upregulated by the FOXO
family of transcriptional activators (Kandarian et al. 2006). Atrogin-l and MuRFl were
investigated in Specific Aim 1. Collectively, these factors promoting muscle degradation
are upregulated in many forms of atrophy, however, their activation in early models of
streptozotocin induced diabetes have not been completely explored.

Also signaled through Akt, mTOR influences rates of protein synthesis through
p70 S6 kinase. Decreased phosphorylation of Akt consequently causes a decrease in the

phosphorylation levels of mTOR. P70 S6 kinase causes phosphorylation and activation

39

of the S6 ribosome increasing protein synthesis. Current research puts a heavier weight
on increased protein degradation over decreased protein synthesis during muscle wasting.
Specific Aim II will focus on key components of the insulin signaling cascade
during STZ diabetes. Components dealing with protein synthesis and degradation will be
investigated in this portion of the project. Although signaling pathways are able to
interact, it is important to isolate components for inspection, yet consider alternative
influences. Furthermore, oxidative and glycolytic muscles will once again be examined

in order to provide insight into signaling mechanisms of different muscles.

MATERIALS AND METHODS
Animals

Animals used were the same as utilized in the Materials and Methods of Specific
Aim I.
Tissue Preparation

The muscles were prepared as in Specific Aim 1. The same tissues were used for
comparison in this section of the study.
SDS-polyacrylamide gel electrophoresis

The tibialis anterior, soleus and plantaris muscles were homogenized in a solution
containing 10X TBS, glycerol, 0.1 M dithiothreitol, 0.5 M sodium fluoride, 10 mM
sodium orthovanadate, 20 mM sodium pyrophosphate, 0.25 M beta-glycerophosphate and
double-deionized water. 20% w/v sodium dodecyl sulfate was added to the homogenate
allowing cell lysis to occur. After centrifugation, the supernatant was acquired and

subjected to electrophoresis.

40

For all of the protein preparations, an 8% Tris-glycine SDS-polyacrylamide gel
were used. The gel sizes were determined according to the molecular weight of the
protein. IRS-1 has a molecular weight of 180 kDa, Akt-2 is 60 kDa, p70S6K is 70 kDa
and the full version of caspase-3 is 35 kDa in weight. The percentages of gels optimized
band separation among the proteins, while allowing the membrane to be used for multiple
Western blots. The gel was poured to a thickness of 1.5mm and 10 wells were created
within each of the gels. A 10X SDS-PAGE loading buffer was added to each of the
samples of protein that consisted of 100 mM Tris-base (pH=6.8), 1% SDS, 10 mM DTT,
50% glycerol, and deionized water. To determine how much sample should be loaded
into each well, protein concentrations were determined spectrophotometrically using the
BioRad protein identification reagent system (BioRad). The protein concentration of the
samples was based on a standard curve produced. The volume of each sample amounting
to 100 pg of protein was loaded into each of the wells for electrophoresis. A
kaleidoscope prestained protein standard (BioRad) was loaded into one of the wells on
each gel. The gels were run at 200 volts until the bands showed distinct separation.
Western Blot

Rabbit antibodies against phosphorylated-Akt—Z, Akt-2, IRS-1, and p7OS6K and
phosphorylated-p70 S6K were obtained from Cell Signaling Technologies. Positive
control cell extracts for Akt, phosphorylated-Akt, p70 S6K, and phosphorylated-p70 S6K
were used and obtained from Cell Signaling Technology. These were also loaded into
the wells of the gel with the samples. The controls were used not only to validate

antibody detection, but were also used to normalize detection levels among the proteins.

41

In order to prepare for the Western blot, the SDS-PAGE gel was transferred
electrophoretically to a nitrocellulose membrane. The transfer was run in a buffer
consisting of 25 mM Tris-base, 192 mM glycine and 20% weight to volume of methanol.
Following transfer, protein bands were visualized using Ponceau S stain. The gels were
stained with Commassie Blue to visualize and confirm the transfer of protein. The
membrane was incubated in blocking buffer that was constructed according to the
manufacturers’ instructions for each individual antibody. The primary antibodies were
diluted then applied to the membrane to incubate for 18 to 24 hours. Following
incubation, the membranes were washed in 1X TBS/T then the secondary antibody was
applied. The secondary antibody used was ImmunoPure peroxidase conjugated goat anti-
rabbit IgG (Pierce) and was diluted 1:2000. After an hour of incubation with the
secondary antibody, a detection system was applied (Amersham). The membrane was
then exposed to hyperfilm and developed in order to be analyzed. Quantification of the
Si gnal was accomplished by Quantity One software (BioRad) using densitometric scans
from autoradiograms. This system detected the density of the signal over a given number
of pixels. The number of pixels analyzed for each sample on each gel was identical. The
overall density displayed was determined using a standard curve with a predetermined
density scale. The values obtained from the standard curve for each Western blot were
normalized to the control value in Microsoft Excel and the phosphorylated to total protein
levels were determined.

Blocking buffers and dilutions were constructed for each of the antibodies as
follows. The primary antibodies for all of the antibodies used were diluted 1:500 in

blocking buffer. The blocking buffer utilized for Akt-2, p70 S6 kinase and IRS-1

42

antibodies consisted of 5% BSA, 0.1% Tween, and 1X TBS, as suggested by the
manufacturer. The blocking buffer used for phospho-p70 86K and phospho-Akt-2
antibodies consisted of 50mM NaF, 5% BSA, 0.1% Tween, and 1X TBS. The dilutions
for primary antibodies were determined by running a sample Western blot that was
divided into multiple segments and diluted with different concentrations of antibody.
The dilution ratio used was necessary to optimize the protein signal without overly
increasing the background on the blot.

In order to use a membrane for multiple blots, the phosphorylated antibody was
assessed first, followed by the antibody for total protein. In order to prevent cross
reactions following Western blot analysis of the first antibody, the membrane was
stripped of the first antibody. The membrane was stripped using a solution comprised of
0.2M NaOH incubated for 10 minutes, then rinsed in TBS/T. The respective blocking
buffer was then applied for the antibody and Western blot procedure was followed. This

procedure was followed for both p70 S6 kinase and Akt-2.

PRELIMINARY DATA
Western blot analysis

Following Western blot analysis for protein, it was seen that Akt-2
phosphorylation levels varied between the muscle types. In the tibialis anterior and
plantaris, the phosphorylated Akt-2 ratio decreased in the diabetic muscle. Even after the
treatment to destroy insulin production, Akt-2 phosphorylation was still apparent in the

soleus. This muscle had a variable amount of phosphorylation among the samples. This

43

data is displayed in Figure 10 with protein bands displayed that are representative of each
group.

IRS-1 level were also assessed in this study. Using Western blot analysis, it was
determined in the tibialis anterior that IRS-1 protein expression levels were not altered
between the control and diabetic mice (Figure 11). Four mice from each group were used
to obtain these results. Representative bands from the Western blot are displayed in
Figure 11 with the IRS-1 quantification depicted in the graph.

Since the tibialis anterior is a larger muscle and more protein could be extracted,
it was the only muscle used to assess p70 S6 kinase. Even with the use of 150 pg protein,
phosphorylated p70 S6 kinase was un detected to a quantifiable extent in the control or
diabetic muscles. Total p70 S6 kinase was readily detectable and did not vary between

the two groups.

Tibialis Anterior Plantaris Soleus
Ctrl STZ Ctrl STZ Ctrl STZ

p-Akt —-> w “mm iriii‘lm “ m w M

Akt —» mm», mm M M W W

Ratio of phosphorylated Akt-2 to Total Akt-2

 

25
Q: 20 -
32
<
i
'6 15-
E IControl
3' 10 IDiabetic
A.
.2
g 5 *
o .i

 

 

   

Tibialis Anterior Plantaris Soleus

Figure 10. Western Blot Analysis: Akt-2. The average ratio of phosphorylated Akt-
2 to the total Akt-2 in each muscle is expressed in the graph The Western blot
displayed is representative of the average for each group. The tibialis anterior and
plantaris reached statistically significant levels between the control and diabetic (*)
signifies statistical significance (p50.05).

45

Control Diabetic

 
    

IRS-1 Protein Levels

35

 

30—

 

25-

20'

IRS-1 Expression (Scan Units)

 

 

   

Control Diabetic

Figure 11. Western Blot Analysis: IRS-1. There was no significant difference
between the control and diabetic muscles when assessed in the tibialis anterior. The
plantaris and soleus muscles were not assessed. The Western blot depicted above
the graph is representative of the results depicted graphically. The graph consisted
of four muscles per group.

46

Tibialis Anterior

Ct r| STZ Ctrl STZ

 

p70 36k ——* ,n we” 3;;- :3

Figure 12. Western Blot Analysis: p70 S6 kinase. The phosphorylated signals were
too low for quantification (top). The expression of the total p70 S6 kinase is visible
in all of the samples (bottom).

47

PRELIMINARY CONCLUSIONS

Specific Aim 11 contains preliminary data for Akt-2, IRS-l and p70 S6 kinase.
Completion of these experiments will suggest whether critical components of the insulin
signaling pathway are affected by STZ diabetes at an early time period. It would also
suggest that although insulin signaling may be compromised with STZ, components of
the pathway may be influenced by other proteins converging on the cascade.
Furthermore, completion of these experiments would suggest whether there is a
difference between signaling protein activation among muscle types during STZ diabetes.

When Akt-2 is activated by phosphorylation, it is able to modulate the activity of
a multitude of signaling components. The muscles used in this study displayed
variability in the Akt-2 phosphorylation ratio. The dramatic increase in atrogin-l mRNA
expression, as shown by the tibialis anterior in Specific Aim I was complimented by the
decrease in Akt-2 phosphorylation. The plantaris muscle also followed this trend. Since
the soleus showed a greater biological variability in the phosphorylation level of Akt-2,
the results were more difficult to decipher. This muscle did not display a significant
difference in phosphorylation from the control to the diabetic samples. Likewise in
Specific Aim 1, atrogin-l was not significantly induced in the soleus. Atrogin-l
expression corresponded to levels of Akt-2 phosphorylation which is consistent with the
literature.

As a pivotal component in the insulin signaling cascade, one explanation for the
decrease in Akt could be attributable to the decrease in insulin levels that occurs with
STZ diabetes. Although the insulin levels were not evaluated in this study, this STZ

protocol has been shown to eliminate B cell insulin production within four days of drug

48

injection (Pighin et al. 2005). Therefore if insulin signaling is depleted, the signaling
cascade beginning with the insulin receptor may be downregulated in activity. The IGF-l
(Insulin-like growth factor-1) receptor also converges onto the PI3K/Akt pathway.
Decreases in local or systemic levels of IGF-I may influence the phosphorylation state
and activity of Akt. Since the IGF-l signaling pathway is dramatically interconnected
with the insulin signaling pathway, the possibility that IGF-l is contributing, at least in
part, to muscle atrophy cannot be excluded. The role of insulin and IGF-1 receptor
signaling is need of further study to determine the initial signal in the development of
muscle atrophy in diabetes. Pathways such as the NF-KB path also converge onto Akt,
however, this is not implicated in the onset of atrophy via diabetes in skeletal muscle
(Kandarian et al. 2006). Considering the data presented here, Akt phosphorylation states
are clearly altered in the plantaris and tibialis anterior as a result of STZ diabetes.
Conversely, the Akt-2 phosphorylation ratio was unaltered in the soleus muscle.
The difference in Akt-2 phosphorylation between the muscles presented here could be
attributable to many sources. One possibility for the difference in Akt-2 phosphorylation
between the soleus and plantaris is that it is a result of the fiber composition of the
muscle. The soleus is comprised of predominantly slow twitch fibers, while the plantaris
is mostly fast twitch. Since both of these muscles are located in the posterior
compartment of the hindlimb, they should experience sirrrilar patterns of contractial
stimulation. Thus fiber type seems to be a greater contributor to the difference in Akt-2
phosphorylation and atrogin-l expression. A major difference between glycolytic and
oxidative muscles is the ability to use fatty acids as a source of energy. As an oxidative

muscle, the soleus is capable of using fatty acids for energy more readily than the

49

plantaris (Fewell et al. 1995). Finally, the difference in results between muscle types
could be reflective of intrinsic concentration of specific signaling proteins. The soleus
muscle has a higher intrinsic expression of IRS-1, Akt, and PIBK than the extensor
digitorum longus or the epitrochearis, two predominantly glycolytic muscles (Song et al.
1999 Mar). It is unknown how much a signal must decrease before influencing its
downstream target. It is plausible that the soleus is able to sustain signaling longer than
the tibialis anterior and plantaris muscles due to the inherent content of Akt. At a further
time point in this model, the phosphorylation ratio of Akt in the soleus may also decrease
with a concomitant increase in atrogin-l mRNA expression.

Furthermore, different isoforms of Akt exist. The isoform used in this study was
Akt-2, however, the specificity of the antibody for this isoform may have interactions
with the Akt-l isoform. More research is necessary to determine the absolute difference
among the Akt isoforms and their antibody specificity.

IRS-1 has been identified as a critical signaling component at the top of the
PI3K/Akt cascade. In the preliminary data, phosphorylation levels of Akt-2 decreased
among the plantaris and tibialis anterior muscles. Since IRS—1 is a key regulator of signal
transduction to Akt-2, this protein was assessed. The Western blot exhibited an
abundance of nonspecific binding, preventing complete analysis of this protein.
Information can be divulged by considering the total IRS-1. Exposure to free fatty acids
and hyperosmotic stress, such as hyperglycemia, causes degradation of IRS—1 (Gual et al.
2005). Therefore, an apparent decrease in IRS-1 protein could lead to a decrease in
phosphorylation of Akt-2. It is unknown to what degree IRS—1 levels would need to

decrease to negatively influence signal transduction to Akt-2. In this preliminary data,

50

IRS-l levels were not altered between the diabetic and control tibialis anterior muscle
samples. However, Akt-2 phosphorylation levels decreased in this muscle. Several
explanations count account for this data.

One explanation for the results obtained is that the time period examined in this
study may not be long enough for degradation to occur. Like Akt-2, IRS-l is a main
component of the IGF-l receptor signaling cascade. This pathway could continue to
allow stimulation and prevent degradation of IRS-l. This explanation would also require
that another signal is causing a decrease in phosphorylation of Akt-2. Once again,
assessment of IGF—1 during STZ diabetes would provide further answers. Another
interpretation of these results is that although the degradation is unaffected is this
preliminary data, perhaps the protein is sterically inhibited preventing the signal to be
tranduced along the cascade to Akt. O-linked glycosylation by hyperglycemia is one
mechanism that causes inhibition of IRS-1 (Gual et al. 2005). In this study, the diabetic
mice exhibit hyperglycemia making it conceivable that hyperglycemia is causing
inhibition of IRS-1 in the diabetic mice preventing phosphorylation of Akt while
maintaining protein expression. Degradation and steric hindrance may take a longer time
period to influence IRS-1 than the 7 day time point.

Taking these possibilities into account, perhaps Western blot analysis is not an
appropriate measure of IRS-1. Measuring IRS-l activity may offer more of an
explanation into the control of this molecule in STZ diabetes. Again, experimenting with
amelioration of hyperglycemia would offer further assessment of the roles of insulin and

glucose in this signaling cascade.

51

Another complication with this preliminary data could be attributable to the
specificity of the antibody. Even with repetition of the Western blot, IRS-l exhibited an
abundance of unspecific binding. It would be appropriate the use IRS-l antibodies from
other companies or different lot numbers to determine the actual levels of IRS-1
expression. Another possibility is that the antibody is identifying IRS-1 degradation
products. Furthermore, the Western blot displayed in Figure 11 assesses only four tissue
samples. Using a greater number of samples may show that there is a difference between
the two experimental groups. Further experimentation is necessary to determine the
complete effect of STZ diabetes on this signaling molecule.

Considering the data from the p70 S6 kinase phosphorylation levels would allow
protein synthesis rates to be evaluated. Since p70 S6 kinase is partially regulated by Akt-
2 phosphorylation, it would be expected that p70 S6 kinase phosphorylation would
decrease when phosphorylation ratio of Akt-2 also decreases. In this preliminary data,
however, the phosphorylation levels were unable to be assessed. Since the protein
concentration of the tibialis anterior extract was larger than the soleus and plantaris, a
greater quantity of protein could be loaded to the gel. This problem could be solved by
concentration of the protein samples. Even after increasing the amount of protein loaded
onto the gel and increasing the dilution ratio of the primary antibody, the signal produced
on the blot was not quantifiable (Figure 12). The binding affinity of the antibody could
also account for the inability to quantify the Western blot signal. The plantaris would be
expected to follow the same trend as the tibialis anterior, as it has for Akt

phosphorylation and atrogin-l mRNA expression. Total p70 S6 kinase protein

52

expression was apparent and seemingly similar between the two groups. To fully assess
this data, a loading control would need to be evaluated.

According to the literature, p70 S6 kinase phosphorylation levels are increased
following exercise regimens. The phosphorylation levels of this protein are able to be
assessed in rats following exercise protocols, however, are not as easily determined in
mice. Since the rrrice in this study remained at normal activity level and were not being
overly stimulated by exercise, it is a possibility that p70 S6 kinase would be at a resting
level. Furthermore, determination of p70 S6 kinase phosphorylation has not been fully
characterized in muscle during early administration of STZ diabetes in mice. An
explanation for the inability to identify a signal could be due to a basally low level of
phosphorylated p70 S6 kinase in addition to our search for a diminishing signal.

If p70 S6 kinase phosphorylation would have been determinable in this study, it
would have lead to several possibilities. Under the assumption that p70 S6 kinase
phosphorylation was able to be assessed, like Akt-2, the phosphorylation may differ
among the muscles. If the phosphorylation was increased in the control over the diabetic,
it would appear that p70 S6 kinase would follow the phosphorylation of Akt. One
interpretation of these results would be that p70 S6 kinase is being activated by an
alternate path. This would be consistent with MAPK activation of p70 S6 kinase, under
the assumption that MAPK is activated differently during STZ diabetes (Markuns et a1.
1999). Another interpretation would be that although Akt is decreased in
phosphorylation, it may need a further reduction in phosphorylation before impacting

mTOR signaling. Since mTOR is in direct regulation of p70 S6 kinase, any effector

53

modulating mTOR could impact the phosphorylation levels of p70 S6 kinase. The
phosphorylation levels of mTOR would be helpful in assisting this analysis.

As determined by the preliminary data in Specific Aim 1, atrogin-l is upregulated
with type I diabetes. Since it is necessary for FOXO transcription factors to bind to the
atrogin-l promotor in order to increase mRNA expression, this data suggests that FOXO
transcription factors have been activated with early STZ diabetes.

Collectively, these results indicate the complexity of the insulin signaling
pathway. Based on the preliminary data and anticipated results of continued research, it
may indicate that there are many other factors contributing to overall regulation of the
insulin signaling cascade. Each of these pathway constituents are influenced by other
signaling molecules within the cell. In order to examine a single pathway, the
components must be examined individually, however, consideration of other effectors
must be acknowledged. More research is necessary to fully determine which signals are

altered in STZ diabetes and are important to the onset of muscle atrophy.

CONTINUED RESEARCH
Insights into how muscle atrophy signals are initiated by signaling components in
STZ diabetes will be obtained from the conclusions of the aforementioned experiments.
In order to fully reveal the impact of STZ diabetes on skeletal muscle, further study is
necessary. In this project, mRNA levels of 4E-BP1 and caspase-3 protein activation
could be determined, as well as assessing IRS-1 and p70 S6 kinase to completion.
Myofibrillar breakdown could be measured as activation levels of capase-3.

Caspase-3 is responsible for degradation of myofibrillar protein which enables the

54

fragments to be further degraded by ubiquitin li gases. We would anticipate that Western
blot analysis of caspase-3 and cleaved caspase-3 show in the tibialis anterior that cleaved
caspase-3 is abundant. This would correspond to the increase in muscle atrophy that has
been depicted in this study. The soleus and plantaris muscles would potentially display
decreased cleavage of caspase-3 in comparison to the tibialis anterior. The plantaris,
however, would show more activation than the soleus muscle. Overall, this data would
indicate whether breakdown of myofibrillar protein was occurring early in STZ diabetes.

In this project, mRNA levels of 4E-BP1 could also be determined. It has been
shown that mRN A levels of 4E-BP1 rise with chronic atrophy and is another factor
controlled by the Akt/mTOR signaling pathway (Kandarian et al. 2006). This would
provide further insight into the activity of protein synthesis during loss of insulin
signaling.

Overall it is apparent that further efforts to understand the mechanism behind the
response elicited by STZ diabetes are needed. These should entail further investigation
with components that impact the insulin receptor signaling pathway. As in Specific Aim
1, earlier time periods could be examined to determine when signaling components are
altered during atrophy. Some of these factors could include determining phosphorylation
of the insulin receptor itself, PIBK, FOXO, or mTOR. Further investigation of the FOXO
family also needs to be examined in order to determine precisely what each isoform
regulates.

Continued repetition of PCR and Western blots should be run as additional animal

trials with STZ diabetes conclude. More samples would provide more statistically

55

relevant data. These data will increase our understanding of the mechanism diabetes

might use to provoke muscle atrophy and further complications.

56

SPECIFIC AIM III

INTRODUCTION

Thus far, this project has shown that disruption in glucose homeostasis by STZ
causes skeletal muscle atrophy to ensue. PGC-la is a transcriptional coactivator that
responds to physiologic stressors, allowing the cell to meet the changing energy demands.
This component involved in metabolic regulation may also play a role in muscle atrophy
by interacting with components of the insulin signaling cascade. Therefore, this specific
aim seeks to determine if PGC-la expression is altered with STZ diabetes. It can be
hypothesized that when diabetes is induced pharmacologically by STZ, PGC-la
expression levels decrease. Furthermore, PGC—la expression levels will vary between
muscle types since different amounts of weight loss were visible depending on the
muscle.

Support for this hypothesis is gained from the report by Patti et al. (2003) that
reveals decreased PGC-la expression in type H diabetes mellitus. This provides
evidence for altered control of PGC-la due to metabolic imbalance in the diabetic state.
In this study, the authors did not address multiple muscle types. This is of importance
since overall PGC—la expression may vary between muscles. Mootha et al. (2003) also
showed during type H diabetes that PGC-la regulated genes were coordinately
downregulated in muscle. These reports did not investigate PGC-la using multiple
muscle types nOr did they inspect PGC-la expression during a period of early atrophy.

Since muscle wasting decreases force production and increases the vulnerability

of muscle to injury, other proteins may act to prevent the production of muscle wasting.

57

Sandri et al. (2006) showed prevention of atrogene upregulation with overexpression of
PGC-la. They determined this was attributable to PGC-la interaction with FoxO3.
Their study paints a protective role for PGC-la in muscle atrophy. PGC—lu has been
well investigated following exercise, but its expression during early STZ induced muscle
atrophy has not been thoroughly investigated.

This section seeks to explore whether there is altered control of the energetic
regulator, PGC-la, during metabolic imbalance and muscle atrophy induced by STZ

diabetes.

MATERIALS AND METHODS
Animals

For all proposed experiments, the mice should be treated as they were in the
materials and methods of Specific Aims I and II.
Quantitative Real Time PCR

For this experiment, quantitative real time PCR was carried out as described in
Specific Aim 1. The primer to be used for PGC-la was obtained from Invitrogen. The
forward and reverse sequences are listed in Table 1.
Enzyme Assay

Mitochondrial integrity was assessed using a citrate synthase enzyme assay. For
this assay, the muscle portions were homogenized in a 0.05 M Tris HCl buffer (pH=8.0).
The fractions were weighed and 1 part muscle for every 20 parts buffer was added to the
homogenization via]. The tissue was homogenized for 30 seconds with the Ultraterrax

homogenizing tool at a setting of 3. The homogenate was then centrifuged for 3 minutes

58

at a speed of 900g at room temperature, followed by removal of the supernatant.
Homogenizing buffer was added to reach a final dilution of 1:400. 100 mM Tris buffer
(pH=8.0), 0.1 mM DTNB, and 0.5 mM oxaloacetate were mixed and allowed to
equilibriate for 5 minutes. The homogenate (20 ul) was added to the mixture as well as
0.3 mM acetyl CoA whereby the rate of mercaptide ion production was measured
spectrophotometrically at 412 nm. This method of citrate synthase activity was measured
according to the method of Srere (1966). The protein concentration was used to

normalize the activity of the citrate synthase enzyme.

PRELIMINARY DATA

The basal mRNA expression levels of PGC—la in the controls did not vary
between muscle types (Figure 13). The tibialis anterior and plantaris muscles did not
differ in PGC-la expression between diabetic and control samples. The slow twitch
soleus muscle exhibited significantly decreased levels of PGC-la in comparison to its
control. The tibialis anterior showed a large degree of variability in PGC-la expression.
The expression of PGC-la was expressed as relative to the expression of B—actin.

Mitochondrial activity was evaluated by enzymatic analysis of citrate synthase.
Results from the tibialis anterior at 5 and 17 days showed no difference inactivity levels.

Due to experimental design, the soleus and plantaris were unable to be assessed.

59

PGC-1 mRNA Expression

 

01

h

I Control
I Diabetic

PGC-1 expression (A U)
M co

_A

 

 

 

0
%

Tibialis Anterior Plantaris Soleus

Figure 13. PGC-la mRNA Expression. The diabetic tibialis anterior reached a
statistically different level of PGC-lo. expression. There was no difference between
the plantaris and the soleus muscles. PGC-la mRNA expression is in arbitrary units
(A U), (*) indicates statistical significance.

60

PRELIMINARY CONCLUSIONS

The framework for Specific Aim HI was built on the decreased expression of
PGC-la evident in type II diabetes and the potential ability of this protein to act as a
deterrent to muscle atrophy (Patti et al. 2003; Sandri et al. 2006). Diabetes mellitus
insti gates an imbalance in metabolic control, pointing towards an influence on PGC-la
regulation. The preliminary data presented in this Specific Aim show that PGC-la
expression levels are altered by STZ diabetes in the soleus muscle, but not in the tibialis
anterior or plantaris. The slow twitch soleus muscle was shown to have a decrease in
levels of PGC-la mRNA expression, but not measurable weight loss in the muscle
(Figures 11, 3). The tibialis anterior, however, did show measurable weight loss and
activation of atrogin-l without a change in PGC-la (Figures 3, 8, 11). This preliminary
data proved to be contradictory to our original hypothesis. There are several explanations
for the results that have been obtained from this preliminary data.

In previous studies using the same streptozotocin protocol, citrate synthase
enzymatic activity was assessed in order to determine whether the capacity of the
mitochondria was affected. This previous study revealed in the tibialis anterior that
citrate synthase activity was unaltered 5 and 17 days following STZ treatment. This
indicates that the density of the mitochondria was not altered in this diabetic model,
validating the results establish here with PGC—la expression. Oxidative muscle,
however, was not examined in the previous citrate synthase study. From the PGC—la
data divulged above, it seems that the soleus would ultimately acquire a decreased ability

for mitochondrial biogenesis and lessened oxidative capacity in comparison to the

61

control. The glycolytic muscles, however, do not seem to shift towards a decrease in
ability for oxidative capacity at this time point.

In response to exercise-induced upregulation of oxidative genes by PGC-la,
muscle type specificity is seen (Ikeda et al. 2006). This is partially consistent with the
preliminary data presented here in the display of varied expression of PGC-la by muscle
type. Exercise notoriously increases the expression levels of PGC-la in skeletal muscle
(Handschin et al. 2003). With consideration of this present study, the activity levels of
the mice were not greatly altered among the control and diabetic mice. Therefore, the
major difference appears to be the metabolic imbalance caused by diabetes which
includes increased blood levels of fatty acids and glucose, with decreased insulin. This
suggests that one or more of these factors is ultimately causing the expression of PGC-la
to become altered. Metformin treatment, a drug that decreases hyperglycemia, caused an
increase in PGC-la levels among diabetic models (Kumagai et al. 2006). This increases
the likelihood of heightened glucose as a signal influencing PGC-la expression. The
muscles presented here displayed differences in mRNA expression despite encountering
identical levels of circulating blood glucose.

The difference among the muscles examined here could be attributable to
oxidative and glycolytic phenotypes. As an oxidative muscle, the soleus showed altered
PGC-la expression, while the predominantly glycolytic muscles did not. As a potential
reason for this occurrence, the soleus relies on fatty acids as a source of energy more than
the glycolytic muscles. Therefore, the increase in fatty acids that is elicited by the
diabetic condition (Chen g et al. 2006) influences the soleus more so than the other

muscles.

62

Since the soleus requires more constant energy supply than the primarily
glycolytic muscles presented here, the decrease in PGC-la expression could indicate the
transition of metabolic phenotypes. Feczko et al (1994) showed a fiber type transition
from type 2A to 2B fibers over the course of 42 days after induction of diabetes by STZ.
This demonstrates the ability of muscle to change in response to metabolic imbalance.
This study does not take into account slow oxidative muscles nor an early time point.
Our preliminary data presented here shows that downregulation of a transcriptional
regulator of metabolic genes is altered preferentially in oxidative muscle. This may be an
early indication that the soleus is transitioning from a slow to a faster phenotype. This
result corresponds to fiber type switching in humans with diabetes (Oberbach et al.
2006). This preliminary data provides just a small glimpse into alterations in metabolic
control during diabetes.

The data presented here introduce a variety of potential connections of PGC-la
with muscle atrophy and insulin signaling components. Important to consider are the
possible influences that PGC-la may have over skeletal muscle atrophy. One possibility
is that interaction of PGC-la and the insulin signaling pathway involves regulation
through Akt. As a coactivator of transcription, PGC-la controls transcription of genes
involved in glucose metabolism and fatty acid oxidation. One of the genes controlled by
PGC-lu is TRB-3, a mammalian homolog of the drosophila tribbles gene (Mortensen et
al. 2006). This protein is capable of binding and suppressing Akt activity in the liver and
myotubes (Du et al. 2003). Although PGC-la levels were decreased in the soleus,
atrophy was not displayed in this muscle by weight or molecular markers of atrophy.

Also in the soleus, Akt-2 phosphorylation was not modified, as determined in Specific

63

Aim H. In accordance with the preliminary data presented here, it is plausible that PGC-
la downregulation in the soleus causes downregulation of TRB-3, which eliminates the
negative regulation imposed on Akt. Meanwhile, in the tibialis anterior and plantaris,
PGC-la levels allow continued expression of TRB-3, allowing Akt phosphorylation to be
suppressed. This scenario would require further characterization of TRB-3 as it has not
been examined using skeletal muscle from a STZ diabetic model.

Further interaction of PGC-la and muscle atrophy is implicated by Sandri et al
(2006). Overexpression of PGC-la prevented a rise in atrogene expression during
denervation. They suggest that this occurred as a result of PGC-la interaction with
FoxO3. This suggests that muscles preferentially expressing higher protein levels of
PGC-la may have a greater ability to resist atrophy. Puigserver (2005) suggests that
PGC-la is preferentially expressed in oxidative muscle, such as the soleus. Our data
displayed a decrease in PGC-la mRNA in the soleus, without indication of atrophy.
Examination of the protein content of PGC-la would be helpful in characterization of
this affect on atrophy. Much further research is necessary to determine the interplay of
atrophy and mitochondrial biogenesis and the mechanisms that employ the components.

In conclusion, STZ diabetes causes metabolic imbalance with a differential
influence according to muscle type. The reasons for this effect are still in need of
characterization. The oxidative phenotype shows a resistance to muscle atrophy, which
may be contributed by PGC-la. More precise studies of the involvement of PGC-la in

metabolic control and muscle atrophy during STZ diabetes will be required.

CONTINUED RESEARCH

64

Insights into mechanisms that allow skeletal muscle to atrophy at different rates
can be partially divulged from the preliminary data above. Research that builds further
on that data may ask how various transcriptional activators interact in the nucleus of the
muscle cell. Mutants using gene silencing techniques with various components that
affect PGC-la and atrogenes could be used. Physiologic data including that of fatty acid
and insulin measurements would be helpful to further identify metabolic alterations in
STZ diabetes.

Further investigation into earlier and later periods within the diabetic time course
may provide further indications of fiber type switching. Histological data would also
provide helpful to determining whether a difference in fiber type was apparent between
diabetic and control muscle types. Since histology may take a prolonged period to show
a visible difference in muscle fibers, mRNA expression is helpful in assessing whether
fiber type switching is occurring at this early period. If this is verified even further, it
may exemplify the remarkable plasticity of muscle.

Instead of measuring transcriptionally activated genes of FOXO in order to
determine its activation, phosphorylation of FOXO by Western blot analysis could be
analyzed. This would determine the direct activity of FOXO instead of indirect
assumption. Furthermore, the protein levels of FOXO according to tissue type would be
helpful to determine. This could be one of the contributors to differences in muscle
atrophy between muscle types. Advances in characterization of the FOXO family as a
whole to determine which activators may be in direct connection with PGC-la activation

is definitely needed for research in the future. Further research could also include

65

looking into the role of TRB-3 in STZ diabetes as it has not been investigated in this
model to any extent.

Continued research could also include taking a look into the impact of diabetes on
more muscles, including cardiac muscle. As the heart is extremely oxidative, it could
provide further insight into the actions of PGC—la in muscle. Beyond atrophy initiation,
the relationship between diabetes and skeletal muscle atrophy mechanisms may be
uncovered through understanding of how the cell guards against muscle atrophy to begin
with. This understanding may further enhance our understanding of the mechanisms of
atrophy.

More data and research is needed to determine what controls PGC-la expression
during diabetes. Hypoinsulinemia, hyperglycemia, or the increase in fatty acids within
the bloodstream could potentially be the culprit of deactivation. It would be helpful to

examine each to these effectors on muscle during diabetes individually.

66

DISCUSSION

This work aims to clarify the relationship between STZ diabetes and muscle
atrophy. The areas examined are: 1) occurrence of skeletal muscle atrophy in a model of
type I diabetes 2) determinants of the insulin signaling pathway that may lead to muscle
wasting and 3) impact of STZ diabetes on PGC-la expression according to muscle type
during early STZ diabetes. As previously stated, the answers to these questions will shed
light on skeletal muscle atrophy induced by diabetes as well as offer clues for a
mechanism of atrophy in other sources. Original preliminary data along with data
reported so far by other research groups support our hypothesis that metabolic disruption
by STZ diabetes plays a critical role in the onset of muscle atrophy. Continued research
in this area is important because it will shed light on the complications of diabetes in
muscle degeneration that may be applicable to atrophy induced by other means.

In specific aim I, atrophy was observed in experimental diabetes mouse models.
Diabetes was induced with a single pharmacologic agent, streptozotocin. The mice that
received the drug treatment had a decrease in body weight and heighten blood glucose
levels. Three different muscle groups were examined for the affects of muscle wasting.
Skeletal muscle atrophy was assessed by increased mRNA expression of genes known to
be activated in all forms of atrophy. Figure 3 shows that the muscle weight loss occurred
differentially among the muscle types. The tibialis anterior showed decrease in weight as
well as activation of atrogin-l. The plantaris muscle trended toward a decrease in weight
with an corresponding trend toward increased atrogin-l gene expression. The soleus did

not indicate a decrease in weight or an increase in atrogin-l mRNA. MuRFl expression

67

 

was unaltered in all of the muscle types suggesting that it may take a longer time period
to be activated or that it may be under the control of other factors. Thus, the atrophy
process had been initiated in the diabetic animals, which was reflected in the decrease in
total body weight. Since hyperglycemia and hypoinsulemia were the predominant
alterations from homeostasis in these early animal models of type I diabetes and even
when hyperglycemia is ameliorated during streptozotocin induced diabetes the animals
still experience a drastic decrease in weight loss, hypoinsulinemia would seem to be the
culprit causing the increase in muscle degeneration.

Specific components of the insulin signaling cascade were examined in the next
section. Since a vast amount of proteins have been implicated in the insulin cascade, it is
imperative that research use simplified models. By using the STZ diabetic model and
isolating the components associated with insulin and atrophy, a simplified model was
utilized in this project. Preliminary Western blot analysis show altered phosphorylation
levels of Akt with STZ diabetes. Interestingly there are also differences in
phosphorylation levels between different muscle types. IRS-1 levels were unaltered in
the tibialis anterior suggesting inhibition of this molecule or alterate control of Akt
phosphorylation. One area of continued research here is to discern the differences
between the muscles that allow for atrophy to occur at different rates. This may prove
difficult since there are a multitude of differences in protein isoforms and concentrations
among the muscles. The main focus of this continued research should be to gain more
detail about the insulin response through quantitative real time PCR and Western blot

analysis.

68

A main difference between the muscle types is their ability to rely on glycolytic or
oxidative energy production. In order to investigate one protein that is be responsible for
controlling cellular metabolism and may have a role in atrophy, PGC- l or expression was
determined. Other components may converge on the insulin signaling pathway to
determine overall atrophy in muscle. PGC-la expression has been shown to decrease
with type H diabetes, suggesting altered regulation of this protein (Patti et al. 2003). It
has been proposed by Speigelman et al (2006) that PGC-la inhibits FoxO3 upregulation
of atrogin-l expression during contractial stimulation. Similar experimentation using
type I diabetes as a model when insulin is obliterated has not been examined. As well as
FOXO interaction, PGC-la is a major contributor to the expression of genes involved in
oxidative metabolism. It causes upregulation of TRB-3, a protein that is able to suppress
Akt activity. These studies offer further support for a role of PGC-lor in skeletal muscle
atrophy via the insulin signaling components.

As in type II diabetes, PGC-l levels were altered with STZ diabetes. In Specific
Aim III, PGC-la levels decreased only in the soleus following induction of diabetes. The
preliminary data looking at PGC-la may indicate a trend towards fiber type switching at
an early period of diabetes. Feczko et al. (1994) have shown with STZ diabetes that fiber
type switching does occur over an extensive period of time. These preliminary results
may indicate that gene expression is apparent before visible and functional changes in
fiber types can be determined. Repeated animal models should be set up and PCR run at
earlier time points in the atrophy process to determine whether fiber type switching is in

the beginning stages at this early time point.

69

 

The preliminary data indicates the intricacy of muscle atrophy in early STZ
diabetes. Many other factors still need to be examined. This study indicated the
importance of the muscle model used when experimenting with muscle atrophy as each
muscle has distinct characteristics and reactions that need to be considered.
CONCLUSIONS

Insulin signaling is essential to normal homeostasis of skeletal muscle and
is disrupted during diabetes. Insulin regulates not only uptake of glucose as energy for
the muscles, but impacts protein synthesis and degradation pathways in the muscle.
Overall, the preliminary data collected here support the hypothesis that STZ diabetes
upregulates atrophy processes in skeletal muscle. Our preliminary data also shows that
the different muscles experience different rates with this diabetic model. Components of
the insulin signaling pathway are implicated in causing protein degradation and synthesis.
In this project, phosphorylated Akt, a major signaling protein, was altered with STZ
diabetes. The literature establishes that PGC-la may play a preventative role in muscle
atrophy and also that PGC-la expression levels decrease during type II diabetes (Sandri
et al. 2006; Patti et al. 2003). The preliminary data show that STZ diabetes influences
components of the insulin signaling pathway and expression of P601, but further
research is needed to determine the regulatory mechanism. Oxidative and glycolytic
muscles respond differentially to the induction of diabetes by STZ. PGC—la levels may
control the expression of genes necessary for metabolism and consequently cause skeletal
muscle to atrophy at difference rates due to interaction with the insulin signaling cascade.
Further research is necessary to elucidate the complete mechanism of muscle atrophy

during STZ diabetes.

70

CLINICAL SIGNIFICANCE

The results of this work hold clinical value in determining how the loss of insulin
signaling affects skeletal muscle. Since atrophy is also implicated in many other forms of
disease as well as inactivity, this research holds answers to many unanswered questions
in atrophy. Many forms of atrophy are characterized by similar biochemical changes
(Lecker et al. 2004). Therefore using this model, advances applicable to muscle wasting
induced by other means can also be elucidated.

Further clinical benefits will be obtained if the mechanism of atrophy due to loss
of insulin signaling such as in type I diabetes becomes clearer. Prevention and treatments

will follow providing an understanding of the disease mechanism.

71

BIBLIOGRAPHY

Antonetti DA, Reynet C, and Kahn CR. (1995) Increased expression of mitochondrial-
encoded genes in skeletal muscle of human with diabetes mellitus. Journal of Clinical
Investigation. 95:1383-1388.

Armstrong RB and Ianuzzo CD. (1977) Compensatory hypertrophy of skeletal muscle
fibers in streptozotocin-diabetic rats. Cell Tissue Res. Jul;181(2):255-66.

Aughsteen AA, Billah Khair A, and Suleiman AA. (2006) Quantitative morphometric
study of the skeletal muscles of normal and streptozotocin—diabetic rats. J Pancreas
7(4):382-389.

Baldwin KH, Herrick RE, Ilyina-Kakueva E and Oganov VS. (1990) Effects of zero
gravity on myofibril content and isomyosin distribution in rodent skeletal muscle.
FASEB J 4:79-83.

Beck- Nielson H, Vaag A, and Damsbo P. (1992) Insulin resistance in skeletal muscle in
patients with NIDDM. Diabetes Care 15:418-429.

Bodine SC, Latres E, Baumhueter S, Lai VK, Nunez L, Clarke BA, Poueymirou WT,
Panaro FJ, Na E, Dharrnarajan K, Pan ZQ, Valenzuela DM, DeChiara TM, Stitt TN,
Yancopoulos GD, and Glass DJ. (2001) Identification of ubiquitin ligases required for
skeletal muscle atrophy. Science. Nov;294(5547):1704-8.

Bouillon R, Bex M, Van Herck E, Laureys J, Dooms L, Lesaffre E, and Ravussin E.
(1995) Influence of age, sex, and insulin in osteoblast function: osteoblast dysfunction in
diabetes mellitus. J Clin Endocrinol Metab. 80:1194-1202.

Bouzakri K, Zachrisson A, Al-Khalili L, Zhang BB, Koistinen HA, Krook A, and Zierath
JR. (2006) siRNA-based gene silencing reveals specialized roles of [RS-1/Akt2 and IRS-
2/Akt1 in glucose and lipid metabolism in human skeletal muscle. Cell Metab.
Jul;4(1):89-96.

Cao PR, Kim HJ, and Lecker SH. (2005) Ubiquitin-protein ligases in muscle wasting. Int.
J. Biochem. Cell Biol. 37:2088—2097.

Cardozo T and Pagano M. (2004) The SCF ubiquitin ligase: Insights into a molecular
machine. Nature Reviews Molecular Cell Biology. 5:739—751.

Charron MJ, Brosius FC, Alper SL, and Lodish HF. (1989) A glucose transport protein

expressed predominately in insulin-responsive tissues. Proc Natl Acad Sci U S A.
Apr;86(8):2535-9.

72

Chen V and Ianuzzo CD. (1982) Dosage effect of streptozotocin on rat tissue enzyme
activities and glycogen concentration. Can J Physiol Pharmacol. Oct;60(10): 1251-6.

Center for Disease Control, (2005) http://www.cdc.gov/diabetes/pubs/estimates.htm#prev
Centner T, Yano J, Kimura E, McElhinny AS, Pelin K, Witt CC et a]. (2001)
Identification of muscle specific ring finger proteins as potential regulators of the kinase
domain. J Mol Bio 306:717-726.

Cheng JT, Huang CC, Liu 1M, Tzeng TF, and Chang CJ. (2006) Novel Mechanism for
Plasma Glucose—Lowering Action of Metformin in Streptozotocin-Induced Diabetic Rats
Diabetes. Mar;55(3):819-25.

Chibalin AV, Yu M, Ryder JW, Song XM, Galuska D, Krook A, Wallberg-Henriksson
H, and Zierath JR. (2000) Exercise-induced changes in expression and activity of proteins
involved in insulin signal transduction in skeletal muscle: differential effects on insulin-
receptor substrates 1 and 2. Proc Natl Acad Sci U S A. Jan;97(1):38-43.

Criswell DS, Booth FW, DeMayo F, Schwartz RJ, Gordon SE, and Fioretto ML. (1998)
Overexpression of IGF-I in skeletal muscle of transgenic mice does not prevent
unloading-induced atrophy. Am J Physio]. 275(3 Pt l):E373-9.

Crozier SJ, Anthony JC, Schworer CM, Reiter AK, Anthony TG, Kimball SR, and
Jefferson LS. (2003) Tissue-specific regulation of protein synthesis by insulin and free
fatty acids. Am J Physiol Endocrinol Metab. Oct;285(4):E754-62.

Du J, Wang X, Miereles C, Bailey JL, Debi gare R, Zheng B, Price SR, Mitch WE. (2004)
Activation of caspase-3 is an initial step triggering accelerated muscle proteolysis in
catabolic conditions. J Clin Invest. J an;l 13(1):l 15-23.

Du K, Herzip S, Kulkami RN, Montminy M. (2003) TRB3:a tribbles homolog that
inhibits Akt/PKB activation by insulin in liver. Science 300: 1574-1577.

Fewell JG and Moerland TS. (1995) Responses of mouse fast and slow skeletal muscle to
streptozotocin diabetes: myosin isoenzymes and phosphorous metabolites.

Mo] Cell Biochem. Jul;148(2):147-54.

Feczko JD and Kleuber. (1994) Anatomical Record 239:18-34.

Finck BN and Kelly DP. (2006) PGC-l coactivators: inducible regulators of energy
metabolism in health and disease. J Clin Invest. 116(3):615-22.

Fingar DC and Blenis J. (2004) Target of raparnycin (TOR): an integrator of

nutrient and growth factor signals and coordinator of cell growth and cell
cycle progression. Oncogene. 23:3151—3 171.

73

Gual P, Le Marchand-Brustel Y, and Tanti JF. (2005) Positive and negative regulation of
insulin signaling through IRS-l phosphorylation. Biochimie 87: 99—109.

Gomes MI), Lecker SH, J agoe RT, Navon A, and Goldberg AL. (2001) Atrogin-l, a
muscle specific F—box protein highly expressed during muscle atrophy. PNAS. 98
(25): 14440-14445.

Handschin C, Rhee J, Lin J, Tarr PT, and Spiegelman BM. (2003). An‘autoregulatory
loop controls peroxisome proliferator-activated receptor 7 coactivator la expression in
muscle. PNAS 100(12):7111-7116.

Henriksen EJ, Bourey RE, Rodnick KJ, Koranyi L, Perrnutt MA, and Holloszy JO.
(1990) Glucose transporter protein content and glucose transport capacity in rat skeletal
muscles. Am J Physiol. Oct;259(4 Pt l):E593-8.

Hershko A and Ciechanover A. (1998) The ubiquitin system. Annu Rev Biochem.
67:425-79.

Ikeda S, Kawamoto H, Kasaoka K, Hitomi Y, Kizaki T, Sankai Y, Ohno H, Haga S, and
Takemasa T. (2006). Muscle type-specific response of PGC-la and oxidative enzymes
during voluntary whee] running in mouse skeletal muscle. Acta Physiol. 188:217-223.

J ackman RW and Kandarian SC. (2004) The molecular basis of skeletal muscle atrophy.
Am J Physiol Cell Physiol. Oct;287(4):C834-43.

James DE, Zorzano A, Boni-Schnetzler M, Nemenoff RA, Powers A, Pilch PF, and
Ruderrnan NB. (1986) Intrinsic differences of insulin receptor kinase activity in red and
white muscle. J Biol Chem. Nov;26l(32):l4939-44.

Kahn R. (1994) Insulin action, diabetogenes, and the cause of type II diabetes.
Diabetes. 43(8): 1066-84.

Kainulainen H, Schurmann A, Vilja P, and Joost HG. (1993) In-vivo glucose uptake and
glucose transporter proteins GLUTl and GLUT3 in brain tissue from streptozotocin-
diabetic rats. Acta Physiol Scand. Oct;l49(2):221-5.

Kamei Y, Miura S, Suzuki M, Kai Y, Mizukami J, Tani guchi T, et a]. (2004) Skeletal
muscle FOXOl (FKHR) transgenic mice have less skeletal muscle mass, down-regulated
Type I (slow twitch/red muscle) fiber genes, and impaired glycemic control. J Biol Chem
2004;279:41114—41123.

Kamura T, Koepp DM, Conrad MN, Skowrya D, Moreland RJ, Illiopuolos O et a].

(1999) bel, a component of the VHL tumor suppressor complex and SCF ubiquitin
ligase. Science 284:657-661.

74

Kandarian SC and Jackman RW. (2006) Intracellular signaling during skeletal muscle
atrophy. Muscle & Nerve. Feb;33: 155-165.

Kedar V, McDonough H, Arya K, Li HI-I, Rockman HA, and Patterson C. (2004) Muscle
specific RING finger I is a bona fide ubiquitin ligase that degrades cardiac troponin 1.
Proc Nat] Acad Sci USA 52: 18135-18140.

Kido Y, Nakae J, and Accili D. (2001) The insulin receptor and its cellular targets.
Journal of Clinical Endocrinology. 86(3):972-979.

Lecker SH, J agoe RT, Gilbert A, Gomes M, Baracos V, Bailey J, Price SR, Mitch WE,
and Goldberg AL. (2004) Multiple types of skeletal muscle atrophy involve a common
program of changes in gene expression. FASEB J. Jan;18(1):39-51.

Lee SW, Dai G, Hu Z, Wang X, and Mitch WE. (2004) Regulation of muscle protein
degradation: coordinated control of apoptotic and ubiquitin-proteosome systems by
phosphatidylinositol 3 kinase. J Am Soc Nephrol. Jun;15(6):1537-45.

Like AA and Rossini AA. (1976) Streptozotocin-induced pancreatic insulitis: new model
of diabetes mellitus. Science. 193:415-417.

Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss 0, Michael LF, Puigserver P, Isotani E,
Olson EN, Lowell BB, Bassel-Duby R, Spiegelman BM. (2002) Nature. Transcriptional
activator PGC-la drives formation of slow twitch fibers. 418(6899):797-801.

Livak KJ and Schmittgen TD. Analysis of relative gene expression data using real-time
quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001 Dec;25(4):402-8.

Marette A, Richardson JM, Ramlal T, Balon TW, Vranic M, Pessin JE, and Klip A.
(1992) Abundance, localization, and insulin-induced translocation of glucose transporters
in red and white muscle. Am J Physiol. Aug;263(2 Pt 1):C443-52.

Markuns JF, Napoli R, Hirshman MF, Davalli AM, Cheatham B, and Goodyear LJ.
(1999). Effects of streptozotocin-induced diabetes and islet cell transplantation on insulin
signaling in rat skeletal muscle. Endocrinology. 140(1): 106-1 1 1.

Mas A, Montane J, Anguela XM, Munoz S, Douar AM, Riu E, Otaegui P, and Bosch F.
(2006) Reversal of type I diabetes by engineering a glucose sensor in skeletal muscle.
Diabetes. Jun;55:1546-1553.

McCabe et a]. (2007) Unpublished observations via personal communications.

McElhinny AS, Kakinuma K, Sorimachi H, Labeit S, and Gregorio CC. (2002) Muscle-
specific RING finger-1 interacts with titin to regulate sarcomeric M-line and thick
filament structure and may have nuclear functions via its interaction with glucocorticoid
modulatory element binding protein-1.'Joumal of Cell Biology. 157: 125—136.

75

Moretti, A., Weig, H. J ., Ott, T., Seyfarth, M., Holthoff, H. P., Grewe, D., et a]. (2002).
Essential myosin light chain as a target for caspase-3 in failing myocardium. PNAS.
99:11860—11865.

Morino K, Peterson KF, and Shulman GI. (2006) Molecular mechanisms of insulin
resistance in humans and their potential links with mitochondrial dysfunction. Diabetes.
Dec;55:S9-S15.

Mortensen OH, Frandsen L, Schjerling P, Nishimura E, and Grunnet N. (2006) PGC-la
and PGC-IB have both similar and distinct effects on myofiber switching towards an
oxidative phenotype. Am J Physio Endocrinol Metab 291:E807-E816.

Nader GA. (2005) Molecular determinants of skeletal muscle mass: getting the “AKT”
together. Journal of Biochemistry and Cell Biology. Oct;37(10):1985-96.

National Institutes of Health: N ationa] Institutes of Diabetes and Digestive and Kidney
Diseases. (2006) www.diabetes.niddk.nih.gov/dm/pubs/statistics

Nemeth P and Pette D. (1993) Succinate dehydrogenase activity in fibers classified by
myosin ATPase in three hindlimb muscles of rat. J Physio (Lond). 320273-80.

Niederle B and Mayr R. (1978) Course of denervation atrophy in type I and type H fibres
of rat extensor digitorum longus muscle. Anat Embryol (Berl). May; 153(1):9-21.

Nolte LA, Galuska D, Martin IK, Zierath JR, and Wallberg-Henriksson H. (1994)
Elevated free fatty acid levels inhibit glucose phosphorylation in slow-twitch rat skeletal
muscle. Acta Physiol Scand. May; 15 l (1)25 1-9.

Oberbach A, Bossenz Y, Lehmann S, Niebauer J, Adams V, Paschke R, Schon MR,
Bluher M, and Punkt K. (2006) Altered fiber distribution and fiber-specific glycolytic and
oxidative enzyme activity in skeletal muscle of patients with type 2 diabetes. Diabetes
Care. Apr;29(4):895-900.

Oku A, N awano M, Ueta K, Fujita T, Umebayashi I, Arakawa K, Kano-Ishihara T, Saito
A, Anai M, Funaki M, Kikuchi M, Oka Y, and Asano T. (2001) Inhibitory effect of

hyperglycemia on insulin-induced Akt/protein kinase B activation in skeletal muscle.
Am J Physiol Endocrinol Metab. May;280(5):E816-24.

Passmore LA and Barford D. (2004) Getting into position: the catalytic mechanisms
of protein ubiquitylation. Biochem. J. 379:513-525.

Patti ME, Butte AJ, Crunkhom S, Cusi K, Berria R, Kashyap S, Miyazaki Y, Kohane I,

Costello M, Saccone R, Landaker EJ, Goldfine AB, Mun E, DeFronzo R, Finlayson J,
Kahn CR, Mandarino LJ. (2003) Coordinated reduction of genes of oxidative metabolism

76

in humans with insulin resistance and diabetes: Potential role of PGCl and NRF 1. Proc
Natl Acad Sci U S A. Jul;100(14):8466—7l.

Pi ghin D, Karabatas L, Pastorale C, Dascal E, Carbone C, Chicco A, Lombardo YB and
Basabe JC. (2005) Role of lipids in the early developmental stages of experimental
immune diabetes induced by multiple low-dose streptozotocin. J Appl Physiol.
Mar;98(3):1064-9.

Puigserver P. (2005) Tissue-specific regulation of metabolic pathways through the
transcriptional coactivator PGCl-alpha. Int J Obes (Lond). Mar;29(Suppl l):SS-9.

Puigserver P and Spiegelman BM. (2003) Peroxisome proliferator-activated receptor-
gamma coactivator 1 alpha (PGC-l ALPHA alpha): transcriptional coactivator and
metabolic regulator. Endocr Rev. 24:78—90.

Puigserver P, Wu Z, Park CW, Graves R, Wright M, and Spiegelman BM. (1998) A cold
inducible coactivator of nuclear receptors linked to adaptive therrnogenesis. Cell.
Mar;92(6):829-39.

Phillips PJ, and Jeffries B. (2006) Gestational diabetes--worth finding and actively
treating. Aust Fam Physician. Sep;35(9):701-3.

Saltin B and Gollnick PD. (1983) Skeletal muscle adaptability: significance for
metabolism and performance. Handbook of Physiology. Peachy LD, Ed. Oxford,
Oxford University, p 555-631.

Sandri, M., Sandri, C., Gilbert, A., Skurk, C., Calabria, E., Picard, A., et al. (2004). Foxo
transcription factors induce the atrophy related ubiquitin ligase atrogin-l and cause
skeletal muscle atrophy. Cell. 117:399—412.

Sandri M, Lin J, Handschin C, Yang W, Arany ZP, Lecker SH, Goldberg AL, and
Schwerzmann K, Hoppeler H, Kayar SR, and Weibe] ER. (1989) Oxidative capacity of
muscle and mitochondria: correlation of physiological, biochemical, and morphometric
characteristics. Proc Natl Acad Sci U S A. Mar;86(5):1583-7.

Spiegelman BM. (2006) PGC-l ALPHA protects skeletal muscle from atrophy by
suppressing FoxO3 action and atrophy-specific gene transcription. PNAS.

Oct; 103(44): 16260-16265.

Song XM, Kawano Y, Krook A, Ryder JW, Chibalin AV, and Zierath JR. ( 1999) Muscle
fiber type specificity in insulin signal transduction. Am Joum Physiology. R1690-R1695.

Srere PA. (1966) Citrate-condensing enzyme oxaloacetate binary complex. Journal of
Biological Chemistry. 241:2157-2165.

77

 

Stitt TN, Drujan D, Clarke BA, Panaro F, Timofeyva Y, Kline WO, Gonzalez M,
Yancopoulos GD, Glass DJ. (2004) The IGF-l/PI3K/Akt pathway prevents expression of
muscle atrophy-induced ubiquitin li gases by inhibiting FOXO transcription factors. Mol
Cell. 14(3):395-403

Suwa M, Egashira T, N akano H, Sasaki H, Kumagai S. (2006) Metformin increases the
PGC-lalpha protein and oxidative enzyme activities possibly via AMPK phosphorylation
in skeletal muscle in vivo. J Appl Physiol. 101(6):]685-92.

Szkudelski T. (2001) The mechanism of alloxan and streptozotocin action in B cells of
the rat pancreas. Physio Res. 50:537-546.

Thomason DB, Herrick RE, Surdyka D, and Baldwin KM. (1987) Time course of soleus
muscle myosin expression during hindlimb suspension and recovery. J Appl Physiol.
Jul;63(1):l30-7.

Tintignac LA, Lagirand J, Batonnet S, Sirri V, Leibovitch MP, and Leibovitch SA.
(2005) Degradation of MyoD mediated by the SCF (MAFbx) ubiquitin ligase. J Bio
Chem. 280(4):2847-2856.

Virkamaki A, Ueki K, and Kahn CR. (1999) Protein-protein interaction in insulin
signaling and the molecular mechanisms of insulin resistance. J Clin Invest.
Apr; 103(7):93 1-43.

Weinert S, Bergmann N, Luo X, Erdmann B, and Gotthardt M. (2006) M line-deficient
titin causes cardiac lethality through impaired maturation of the sarcomere. J Cell Biol.
May;173(4):559-70.

Zierath JR, Krook A, and Wallberg-Henriksson H. (2000) Insulin action and insulin
resistance in human skeletal muscle. Diabetologia. Jul;43(7):821-35.

Zong H, Ren JM, Young LH, Pypaert M, Mu J, Bimbaum MJ, and Shulman GI. 2002
(Dec 10) PNAS Vol 99 no 25. AMP kinase is required for rrritochondria] biogenesis in
skeletal muscle in response to chronic energy deprivation.

78

 

 

uiijijrjjijjjriimini

 

..A. ‘1“‘