THE EFFECT OF ACTIN REORGANIZATION IN
INSULIN MEDIATED GLUCOSE TRANSPORT ON L6
RAT SKELETAL MUSCLE CELLS
CHAN Chung Sing
Degree of Master of Philosophy
in
Medical Science
• The Chinese University of Hong Kong
June 2002
The Chinese University of Hong Kong holds the copyright of this thesis.
Any person(s) intending to use a part or whole of the materials in the thesis
in a proposed publication must seek copyright release from the Dean of the
Graduate School. 」 ffo 2厕• I ！^ p! UNIVERSITY Wi^sLIBRARY SYSTEM ^^^^ ACKNOWLEDGEMENT
I would like to express my sincere gratitude to my supervisor Dr. P. Tong for his supportive guidance, constructive criticisms and valuable advice throughout the study.
Moreover, I would like to thank Dr. F. L. Chan in the Department of Anatomy for his valuable advice and technical support in TEM project.
In addition, special thanks go to Mr. Stanley Ho in the Department of Medicine and Therapeutics for his valuable advice and technical support. Also, thanks go to
Ms Gene L. S. Kung and Ms H. L. Choi in the Department of Anatomy for their skillful assistance in TEM project.
Heart!ul thanks go to Ms Minnie Ng, Ms P. K. Ma and the rest of colleagues for
their encouragement and spiritual support, especially during the time when I was
admitted to the hospital and when I became unreasonable.
Finally, dedicated with love to Ms Anne Lee.
i TABLE OF CONTENT
Acknowledgement i
Abstract ix
List of Abbreviations xvii
CHATPER ONE INTRODUCTION
1.1 Glucose Homeostasis 1 1.1.1 Function 1 1.1.2 Origins and regulation of glucose 2 1.1.3 Glucoregulatory factors 4 1.1.4 Insulin 6 1.1.4.1 Function of Insulin 7 1.1.4.2 Discovery and Production of Insulin 7 1.1.4.3 Insulin Signaling Pathway 8 1.1.4.3.1 Insulin Receptor 8 1.1.4.3.2 MAPK Pathway 9 1.1.4.3.3 Phosphatidylinositol 3-kinase (PI3-K) Pathway 10 1.1.5 Glucose Transporters 11 1.1.6 Role of skeletal muscle in glucose homeostasis 13 1.1.7 Insulin Resistance 14 1.1.8 Glucose abnormality and its complications 16
1.2 Actin 19 1.2.1 Function of Actin 20 1.2.2 Actin Accessory Protein 22 1.2.3 Actin Polymerization 23
1.3 Interaction between Insulin, GLUT4 and Actin in Glucose Homeostasis 24 1.3.1 Insulin-Induced Actin Remodeling 25 1.3.2 Actin Remodeling and Insulin-Induced GLUT4 Translocation 26 1.3.3 Involvement of Insulin Signaling Molecules in Actin Remodeling 27 1.3.4 Actin Remodeling and Insulin Resistance 30
1 • • 11 ij 1.4 Hypothesis and Objective 30 1.4.1 Rationale 30 1.4.2 Hypothesis 31 1.4.3 Objective 31
CHAPTER TWO MATERIALS AND METHODS
2.1 Materials 33
2.2 Cell Culture 36 2.2.1 Cell Culture 36
2.2.2 Reagents Preparation and Incubation 39
2.3 2-Deoxyglucose Uptake 39
2.4 Immunofluorescence Microscopy 41 2.4.1 Permeabilized cell staining 41 2.4.2 Membrane-intact cell staining 43
2.4.3 The analysis of actin remodeling reduction 44
2.5 Live Image Microscopy 44
2.6 Transmission Electron Microscope Study 44
2.7 Statistical Analysis 46
CHAPTER THREE RESULTS
3.1 Cell Growth 48 3.2 Acute Effect of Insulin on L6 myotubes 48 3.2.1 Immunofluorescence Microscopy 49 3.2.1.1 The time profile of insulin on actin cytoskeleton in permeabilized L6 myotubes 49 3.2.1.2 The concentration effect of insulin on actin cytoskeleton in permeabilized L6 myotubes 50 3.2.1.3 Relationship between actin cytoskeleton and GLUT4myc in permeabilized L6 myotubes 51
iii 3.2.1.4 Translocation of GLUT4myc in membrane-intact L6 myotubes 51 3.2.1.5 Effect of methyl-P-cyclodextrins, MeOH or EtOH in permeabilized and membrane-intact L6 myotubes 52 3.2.2 2-Deoxyglucose Uptake 52 3.2.2.1 Effects of insulin, methyl-P-cyclodextrins, MeOH and EtOH in L6 myotubes 52 3.2.3 TEM Study 53 3.2.3.1 Effects of insulin on actin cytoskeleton and GLUT4myc in L6 myotubes 53
3.3 Effect of high glucose and high insulin incubation in L6 myotubes 54 3.3.1 Immunofluorescence Microscopy 54 3.3.1.1 High insulin and high glucose preincubation in permeabilized L6 myotubes 55 3.3.1.2 Effect of high insulin and high glucose incubation in membrane-intact L6 myotubes 55 3.3.2 2-Deoxyglucose Uptake 56 3.3.2.1 Effect of high insulin and high glucose incubation in L6 myotubes 56 3.3.3 TEM Study 57 3.3.3.1 Effect of high insulin and high glucose incubation in L6 myotubes 57
3.4 Effect of FFA incubation in L6 myotubes 58 3.4.1 Immunofluorescence Microscopy 58 3.4.1.1 FFA preincubation in permeabilized L6 myotubes 58 3.4.1.2 FFA incubation in membrane-intact L6 myotubes 59 3.4.2 2-Deoxyglucose Uptake 59 3.4.2.1 FFA incubation in L6 myotubes (24 hours) 60 3.4.3 TEM Study 62 3.4.3.1 FFA incubation in L6 myotubes 62
3.5 Effect of CHO incubation in L6 myotubes 62 3.5.1 Immunofluorescence Microscopy 62 3.5.1.1 CHO preincubation in permeabilized L6 myotubes 63 3.5.1.2 CHO incubation in membrane-intact L6 myotubes 63 3.5.2 2-Deoxyglucose Uptake 64
iv 3.5.2.1 CHO incubation in L6 myotubes (24 hours) 64 3.5.3 TEM Study 65 3.5.3.1 CHO incubation in L6 myotubes 65
3.6 Overall changes in glucose uptake after preincubation experiment 65
CHAPTER FOUR DISCUSSION
4.1 Effect of insulin on L6 myotubes 69
4.2 Effect of methyl-P-cyclodextrins, MeOH and EtOH on L6 myotube 75
4.3 Effect of pretreatment of cells in conditions of insulin resistance 76 4.3.1 Effect of high glucose and high insulin preincubation on L6 myotubes 76 4.3.2 Effect of FFA preincubation on L6 myotubes 78 4.3.3 Effect of CHO preincubation on L6 myotubes 82 4.3.4 Effect of cell preincubation in conditions of insulin resistance on L6 myotubes (TEM) 83
4.4 Summary of the effects of cell preincubation in conditions of insulin resistance 84
4.5 Possible mechanisms involved in insulin resistance induction 86 4.5.1 Possible changes in GLUT expression and activities 87 4.5.2 Possible changes in insulin signaling propagation 88
4.5.3 Altered functioning of various actin accessory proteins 89
4.6 Limitation of the study 90
4.7 Conclusion 90
4.8 Future study 91
REFERENCES 93
TABLES
vii 1. Ratio of LR White and EtOH used in TEM study. 45a
2. Effects of preincubations on actin remodeling. 52d 52e
3. Effects of insulin, methyl -/5- cyclodextrins, MeOH and EtOH preincubations on glucose uptake. 53a
4. Effect of Glucose and Insulin preincubation on glucose uptake. 56c
5. Effect of OA preincubation on glucose uptake. 60a
6. Effect of LA preincubation on glucose uptake. 60b
7. Effect of PA preincubation on glucose uptake. 61a
8. Effect of CHO preincubation on glucose uptake. 64a
9. Percentage of increase on glucose uptakes compared with carrier-control cells before and after insulin treatment. 66a
FIGURES
1. Metabolic and mitogenic signaling pathway in insulin action 8a
2. Differentiation of L6 muscle cells from myoblasts to myotubes. 48a
3. Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (time). 49a
3. Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (concentration). 50a
4. Relationship between actin cytoskeleton and GLUT4myc in permeabilized L6 myotubes. 51a i！^I •jII i 5. Translocation of GLUT4myc in membrane-intact L6 myotubes. 52a Ij I i
vi 6. Effect of methyl-P-cyclodextrin, MeOH or EtOH in permeabilized L6 myotubes. 52b
6. Effects of methyl-p-cyclodextrin，MeOH or EtOH in membrane-intact L6 myotubes. 52c
7. Distribution of actin cytoskeleton and GLUT4myc at basal state in L6 myotubes (TEM). 54a
7. Effects of insulin on actin cytoskeleton and GLUT4myc in L6 myotubes (TEM). 54b 8. High insulin and high glucose preincubation on actin cytoskeleton in permeabilized L6 myotubes. 55a
9. High glucose and high insulin and CHO preincubations on GLUT4myc translocation in membrane-intact L6 myotubes. 56a
9. FFA preincubation on GLUT4myc translocation in membrane-intact L6 myotubes. 56b
10. High glucose and high insulin preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes (TEM). 57a
10. CHO and FFA preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes. 57b
10. FFA preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes. 57c
11. OA preincubation on actin cytoskeleton in permeabilized L6 myotubes. 58a
12. LA preincubation on actin cytoskeleton in permeabilized L6 myotubes. 58b
13. PA preincubation on actin cytoskeleton in permeabilized L6 myotubes. 58c
14. CHO preincubation on actin cytoskeleton in permeabilized L6 myotubes. 63a
vii i I 1i GRAPH
1. Example of protein standard curve. 41a
viii
Ii I The Effect of Actin Reorganization in Insulin Mediated Glucose Transport on L6 Rat Skeletal Muscle Cells
Abstract
Maintaining of glucose homeostasis is vital to human life. The human brain, which serves as the central processing unit of the body, utilizes glucose as the primary metabolic fuel. Since the brain cannot store or synthesize glucose, glucose is derived from the circulation continuously. Thus, the prevention or rapid correction of hypoglycemia is critical to our survival. However, sustained hyperglycemia can cause severe tissue damage and lead to diabetic complications such as blindness, renal failure, cardiac and peripheral vascular diseases, neuropathy and sexual dysfunction. As a result, by achieving a balance between the glucose utilization and the sum of endogenous glucose production and dietary glucose intake, blood glucose level can be maintained in a physiological range.
One of the key factors in regulating glucose utilization is the stimulation of glucose uptake into muscles by insulin. Skeletal muscle is the primary site of insulin-mediated glucose uptake. Since the lipid bilayers that make up the cell membranes are impermeable to glucose, facilitated glucose transport down the concentration gradient across the cell membranes relies on transmembrane proteins - Glucose transporters
(GLUT). In muscle, GLUT4 are the major insulin-responsive GLUT. Upon the action of insulin, translocations of GLUT4 from intracellular compartments to the plasma membrane (PM) are enhanced. The net effect is an increase of glucose transport into the cells. This enhancement is facilitated by the reorganization of cortical actin filaments
ix that localizes insulin downstream signaling intermediates to the GLUT-containing
compartments. Proper interactions are essential to the regulation of glucose
homeostasis.
Insulin resistance, the ineffectiveness of insulin to stimulate glucose uptake, is a major
characteristics of Type II Diabetes Mellitus (DM) and obesity. Although insulin
resistance is well characterized, the exact mechanism remains unclear. Patients with
diabetes are normally hyperglycemia and hyperinsulinemia. Higher plasma free fatty
acid (FFA) and total serum cholesterol levels are also common in diabetics. In the
present study, we have examined the role of the actin reorganization in the stimulation of
glucose uptake and the distribution of GLUT4 in L6 myotubes expressing myc-tagged
GLUT4 (GLUT4myc). The study was divided into three parts: (1) the effect of insulin
and glucose; (2) the effect of FFA; (3) the effect of cholesterol. Cells were incubated in
various concentrations of insulin (up to lOOnM) and glucose (up to 25mM), FFA (up to
75^M) or cholesterol (up to 60|iM) for 24 hours and stimulated with insulin (lOOnM, 30
min for glucose uptake and lOOnM, 10 min for actin remodeling). Glucose uptakes
were assessed with 2-deoxy-D-[^H] glucose and actin remodeling was examined by
immunofluorescence microscopy (IM) and transmission electron microscopy (TEM).
Results from present study showed that preincubation with high glucose (25mM) and
high insulin (lOOnM) levels reduced glucose uptake by 74.4 土 10.9%，(p 二 0.00001).
Secondly, high levels of unsaturated fatty acids preincubation (50|iM) - oleic acid (OA)
and linoleic acid (LA) for 24 hours diminished glucose uptake by 45.8 土 8.6%, (p =
xii
I 0.0005) and 57.1 土 8.5%，（p = 0.004) respectively. Meanwhile, exposure of saturated fatty acid (50^M) - palmitic acid (PA) caused a 33.3 土 8.8%, (p = 0.0008) reduction in glucose uptake. Finally, high cholesterol (CHO) preincubation (50|aM) caused a 56.7 土
8.5%, (p = 0.0005) reduction in glucose uptake.
IM was carried out to examine the morphological changes of cortical actin. We
visualize cortical actin and GLUT4myc by double staining of permeabilized cells with
labeled phalloidin and a monoclonal antibody to the myc epitope. Under the basal state,
actin bundles were aligned longitudinally along the cells while GLUT4myc were
concentrated around the myonuclei. The lO-minute insulin stimulation caused actin
filament reorganization as membrane ruffles throughout the cells. Meanwhile,
GLUT4myc colocalized with these actin structures. Preincubation with high glucose
(25mM) and high insulin (lOOnM) levels reduced actin remodeling by 64.7%, (p < 0.001).
In addition, high levels of OA, LA and PA preincubation (50|iM) for 24 hours diminished
actin remodeling by 46.2%, (p < 0.005)，51.0%, (p < 0.0005)，and 37.6%, (p < 0.0001)
respectively. Finally, high cholesterol (50|iM) preincubation caused a 53.0%, (p <
0.0005) reduction in actin remodeling.
Moreover, we have conducted an IM study in membrane-intact cells. Cell surface
immunostaining was carried out for both actin and GLUT4myc. In the basal state, actin
structures were distributed along the surface while GLUT4myc were barely visible.
After the lO-minute insulin treatment, colocalization of actin and GLUT4myc were
observed. L6 myotubes were then incubated in insulin (lOOnM) and glucose (25mM),
xi
I FFA (50|iM) or cholesterol (50|iM) for 24 hours and stimulated with insulin (lOOnM, 10
min). By comparing with their respective controls, there were reductions in
insulin-induced recruitment of GLUT4myc.
In order to have a more accurate picture of insulin-mediated glucose uptake, we
performed TEM in our study. We identified cortical actin and GLUT4myc with
immunogold labeling. The lO-minute insulin stimulation reconfirmed the GLUT4myc
colocalization with actin structures. Additionally, the preincubation experiments also
reduced the degree of GLUT4myc colocalization with actin.
In conclusion, both IM and TEM studies showed a clear spatial relationship between
actin networks and GLUT4 during the course of insulin-mediated glucose uptake in
muscle cells. Their colocalization is vital to the success of glucose transport. Patients
with diabetes are normally have elevated levels of glucose, insulin, FFA and total serum
cholesterol. The preincubation experiments resulted in dose-dependent reductions in
glucose uptake. In addition, there were also parallel dose-dependent reductions in the
actin reorganization. Therefore, disassembly of the actin network may play a role in
glucose transport reduction, which may in turn allow us to have a better understanding of
the mystery of insulin resistance in human.
xii 實驗摘要
保持葡萄糖的體內平衡是人類生活所必需的。而人類身體的腦袋，正是利用葡萄糖作爲其主
要的燃料。因爲腦不能存儲或者合成葡萄糖，它需要從血液循環中連續不斷地得到葡萄糖。
所以，預防血糖過少或對其快速的糾正對我們的生存是非常重要。然而，血糖持續過高卻能
夠引起嚴重的細胞損傷和糖尿病倂發症，如失明，臂衰竭，心臟或外周的血管疾病，神經病
和性機能障礙等。總括來說，透過平衡身體葡萄糖的用量和體內葡萄糖的生產和飲食葡萄糖
的吸入，人類便能夠保持血葡萄糖在生理的水平中。
調節葡萄糖使用的關鍵原素之一就是肌肉裡透過胰島素所產生葡萄糖攝取的剌激作用。骨豁
肌肉是胰島素調節葡萄糖攝取的首要地點。因爲葡萄糖並不能滲透細胞膜，葡萄糖運輸是需
要依靠一種特別的蛋白質：葡萄糖運輸裝置。在肌肉中，葡萄糖運輸裝置一4是主要胰島素
附應的葡萄糖運輸裝置。胰島素的作用是增加葡萄糖運輸裝置一4從細胞內到達細胞膜的次
數，從而增加葡萄糖的運輸。主要是透過皮層肌動蛋白的重組，而將胰島素信號途徑的蛋白
質中間體和葡萄糖運輸裝置-4放在一起。恰當的相互作用對於調節葡萄糖的體內平衡是不
可少0
當胰島素不能刺激葡萄糖的攝取，抗胰島素性便產生。抗胰島素性是糖尿病II型和肥胖症
的主要特性。雖然已經有很多關於抗胰島素性的研究，但是其精確機制仍然是不清楚。糖
尿病人通常都患有血糖過高症和胰島素過高症。除此之外，高於正常的游離脂肪酸和總血清
xiii
I 膽固醇水平在糖尿病人中也很普遍。在目前的硏究中，我們測驗皮層肌動蛋白的重組對葡萄
糖攝取和葡萄糖運輸裝置-4分佈的影響。我們把硏究分成三份：（1)胰島素和葡萄糖的影
響；（2)游離脂肪酸的影響；（3 )膽固醇的影響。我們將細胞培養在不同濃度的胰島素
(100毫微摩爾或以下）和葡萄糖（25毫摩爾或以下)，脂肪酸（75微摩爾或以下）或膽固醇（60
微摩爾或以下）共24小時，並且用胰島素刺激（葡萄糖攝取，100毫微摩爾，30分鐘；肌動
蛋白重組，100毫微摩爾’ 10分鐘）。葡萄糖的攝取是利用去氧超重氫葡萄糖來評定，而肌
動蛋白重組則是用免疫營光顯微鏡方法和傳遞電子顯微鏡方法來觀察。
目前硏究的結果表明具有高葡萄糖（25毫摩爾）和高胰島素（100毫微摩爾）含量的培養減少
了百分之74.4 ±10.9的葡萄糖攝取，（P ： 0.00001)�其次’與24小時不飽和脂肪酸一起培養
下：油酸和亞麻油酸（50微摩爾）分別減少了達百分之45.8 ± 8.6，(P = 0.0005 )和百分之
57.1 ± 8.5，(P = 0.004)葡萄糖攝取。同時’與24小時飽和脂肪酸一起培養下’棕櫚酸（50
微摩爾)減少了葡萄糖攝取百分之33.3 ± 8.8，（P 二 0.0008)�最後’與高膽固醇一起培養（50
微摩爾）葡萄糖的攝取減少了百分之56.7 ± 8.5 ’ (P - 0.0005)�
我們利用免疫蛋光顯微鏡方法檢查皮層肌動蛋白形態的變化。我們用抗體來顯現皮層肌動蛋
白和葡萄糖運輸裝置-4。在基本的狀態下，A沿著細胞縱向排成一線’而葡萄糖運
輸裝置一4則集中在細胞核的周遭�10分鐘的姨島素刺激作用產生了肌動蛋白的重組。同
時’葡萄糖運輸裝置-4亦發現集中在叽動蛋白的架溝中。具有高葡萄糖（25毫摩爾）和高
xiv 胰島素（100毫微摩爾）濃度的培養減少百分之64.7，(P<0.001)的肌動蛋白重組。此外，
24小時高濃度油酸，亞麻油酸和棕櫚酸的培養（50微摩爾）分別減少了百分之46.2，（P <
0.005)，51.0 ‘ (P < 0.0005)，和 37.6, (p < 0.0001)的肌動蛋白重組。最後，高
膽固醇（50微摩爾）減少了百分之53.0，（P < 0.0005)的肌動蛋白重組。
此外，我們再利用免疫螢光顯微鏡方法檢查擁有完整細胞膜的細胞。我們用抗體來顯現細胞
表面的皮層肌動蛋白和葡萄糖運輸裝置—4�在處於基部狀態的細胞中，沿著表面可觀察到
肌動蛋白，而葡萄糖運輸裝置-4卻僅僅可見時。在10分鐘的胰島素處理以後，我們觀察了
葡萄糖運輸裝置—4集中在肌動蛋白的架構中。之後，我們將細胞培養在具有高葡萄糖(25
毫摩爾）和高胰島素（100毫微摩爾），高游離脂肪酸（50微摩爾），或高膽固醇（50微摩爾）
的環境。透過與他們各自的控制比較，由胰島素所弓丨起的從細胞內部到細胞膜葡萄糖運輸裝
置一 4的轉移有重要的減少。
爲了獲得由胰島素所弓丨起的葡萄糖攝取的更精確的圖片，我們完成了電子顯微鏡的硏究。10
分鐘的胰島素刺激作用再次確認了肌動蛋白架構和葡萄糖運輸裝置一4位置重疊的關係。除
此之外，培養試驗也展示出一個減少了葡萄糖運輸裝置-4位置重疊的程度。
在胰島素協助下葡萄糖攝取的過程中，免疫螢光顯微鏡和傳遞電子顯微鏡在肌肉細胞中的硏
究都展示了肌動蛋白網路和葡萄糖運輸裝置一4之間一個清楚的位置關係。他們的位置重疊
XV 是有效的葡萄糖運輸所必需。糖尿病病人通常有高於正常的葡萄糖，胰島素，游離脂肪酸和
總血清膽固醇的水平。我們的培養試驗証明葡萄糖攝取是依劑藥的含量而下降。此外，肌
動蛋白的重組也是同樣地依劑藥的含量而減少了。因此，肌動蛋白網路的拆卸可能影響葡萄
糖運輸。這可能允許我們對抗胰島素性的一個更深的理解。
xvi LIST OF ABBREVIATIONS
AA Arachidonic Acid
AB Aantibiotic-Antimycotic
ADP Adenosine Diphosphate
AGEs Advanced Glycation End Products
Arp Actin-Related Protein
ATP Adenosine Triphosphate
BSA Bovine Serum Albumin
CB Cytochalasin B
CD Cytochalasin D
CHO Cholesterol
DD-H2O Double Distilled Water
DM Diabetes Mellitus
DMSO Dimethyl Sulfoxide
EtOH Ethanol
F-actin Filamentous Actin
FBS Fetal Bovine Serum
FFA Free Fatty Acid
G-6-P Gucose-6-Pospahte
xvii G-actin Globular Actin
GAP GTPase-Activating Proteins
GDI Guanine Nucleotide Dissociation Inhibitors
GDP Guanosine Diphosphate
GEF Guanine Nucleotide Exchange Factors
GK Glucokinase
GLUT Glucose Transporters
GLUT4myc Myc-Tagged GLUT4
GSK-3 Glycogen Synthase Kinase-3
GTP Guanosine Triphosphate
HK Hexokinase
IM Immunofluorescence Microscopy
IR Insulin Receptor
IRS Insulin Receptor Substrates
LA Linoleic Acid
MAPK Mitogen-Activated Protein Kinase
MeOH Methanol
min Minutes
MLC Myosin Light Chain
xviii mRNA Messenger Ribonucleic Acid
mSos mammalian Son of Sevenless
OA Oleic Acid
PA Palmitic Acid
PBS Phosphate-Buffered Saline
PDK PI3-K-Dependent Kinase
PFA Paraformaldehyde
PI3-K Phosphatidylinositol 3-Kinase
PIP2 Phosphatidylinositol 4, 5-bisphosphate
PIP3 Phosphatidylinositol 3, 4, 5-triphosphate
PKC Protein Kinase C
PM Plasma Membrane
ROK Rho-associated Kinases
SEM Standard Error of the Mean
SH2 Src Homology 2
She Src and Collagen Homologous
TBS 0.05M Tris-HCl buffer with 0.15M sodium hydroxide, pH 7.6
TEM Transmission Electron Microscopy
a -MEM a - Eagle's Minimum Essential Medium
xix CHAPTER ONE INTRODUCTION
1.1 Glucose Homeostasis
Glucose is a small and polar monosaccharide. However, its physiological significance to our bodies is indispensable. Since abnormalities in glucose homeostasis threaten our survivals, blood glucose level is maintained within a narrow range despite the wide variations in activities and food intake. There is a collective effort by a controlled balance of hormonal, neural and substrate glucoregulation.
One of the pivotal glucoregulatory components is the insulin-mediated glucose uptake
in skeletal muscle. The transportation of glucose across the PM is dependent on the
functioning of GLUT. Hence, aberrant GLUT action may disrupt glucose transport
and affect normal glucose homeostasis.
1.1.1 Function
Glucose is in essential for life. First of all, glucose serves as metabolic fuel. For
example, the brain and skeletal cells utilize glucose for the production of adenosine
triphosphate (ATP) which is required for various cellular activities. In fact, glucose
1 is the primary fuel for brain. Inadequate supply of glucose can cause severe neural dysfunction. Secondly, glucose is converted to glycogen through the action of enzyme glycogen synthase {Halse, Bonavaud, et al. 2001}. The exchange between glucose and glycogen allows cells to build up glucose storage for future use. It
provides a reserve pool by which tissues can interact with the surrounding and alter
physiological glucose concentration. Thirdly, glucose can act as the precursor for
FFA, protein and cholesterol production. The extensive interaction among glucose,
EFA, protein and cholesterol metabolism allows the body to maximize the use of
existing resources {Cryer 1997}.
1.1.2 Origins and regulation of glucose
Glucose is derived from three sources: dietary intake of carbohydrates; endogenous
production by gluconeogenesis; and glycogen breakdown by glycogenolysis. The
digestion of carbohydrates takes place along the gastrointestinal tract. Glucose is
actively taken up into the cells by sodium-linked GLUT in the intestinal lumen.
These transporters actively transport glucose against a glucose-concentration gradient
using the energy provided from the electrochemical potential difference of sodium
ions across the PM {Kahn 1992}. Once it enters the target tissues through the
2 circulation, glucose is rapidly phosphorylated to gIucose-6-phospahte (G-6-P).
Intracellular G-6-P can be metabolized through glycolysis, stored as glycogen or converted back to glucose if suitable enzymes and precursors are available. Through the process of glycolysis, G-6-P is converted to pyruvate and ATP is also generated.
Pyruvate can be further reduced to lactate, transaminated to form alanine, or
converted to acetyl coenzyme and generate more ATP via the Krebs cycle {Cryer
1997}.
Endogenous glucose production relies on gluconeogenesis and glycogenolysis.
Gluconeogenesis can be viewed as the reversal of glycolysis. It involves the
formation of glucose from precursors such as lactate, glycerol, and amino acids.
Glycogenolysis involves the breakdown of glycogen to G-6-P with its subsequent
hydrolysis to free glucose. Both the liver and kidney contain the enzymes required
for gluconeogenesis and glycogenolysis while muscle contains only the enzymes for
glycogenolysis {Cryer 1997}.
Maintenance of the normal blood glucose concentration requires a precise matching
of endogenous glucose availability and glucose utilization. The metabolic shift of
glucose depends on the current body nutritional status. During the fasting phase,
3 peripheral glucose utilization is reduced while gluconeogenesis and glycogenolysis are enhanced. The body compensates the lack of glucose by switching to other possible metabolic fuels such as ketone body or FFA. In the fed state, elevated glucose concentration promotes the utilization of glucose, reduction of endogenous glucose production and increased formation of glycogen, fat and protein {Frier &
Fisher 1999}.
1.1.3 Glucoregulatory factors
Glucose homeostasis requires close interactions between different hormones,
neurotransmitters and substrate. Hormones are the most important glucoregulatory
factors and their secretions are normally guided by plasma glucose concentration.
Glucoregulatory hormones are divided into two groups: insulin and counterregulatory
hormones including glucagons, epinephrine, growth hormone, and Cortisol. Insulin
is a glucose-lowering hormone that suppresses endogenous glucose production and
enhances glucose utilization. It acts on liver, kidney and peripheral tissues. It
suppresses hepatic glycogenolysis and gluconeogenesis, and promotes hepatic
glycogen storage. Besides, it also stimulates glucose uptake and utilization in
muscle and fat cells. The net result of insulin action is to reduce plasma glucose
4 level by enhancing its metabolic use and storage. On the other hand, suppressing insulin secretion causes the reverse effect by increasing glucose production and decreasing glucose utilization {Wilson 1998}. As a result, by controlling the insulin secretion, the blood glucose level can be adjusted according to the physiological need.
The counterregulatory hormones include glucagons, epinephrine, growth hormone, and Cortisol. In response of hypoglycemia they are secreted into the circulation to raise the glucose level. For instance, glucagon is secreted from pancreatic a cells.
It activates hepatic glycogenolysis and gluconeogenesis to promote glucose production. Moreover, epinephrine suppresses insulin secretion and therefore
increases blood glucose levels. It also activates hepatic glycogenolysis and
gluconeogenesis, and limits glucose utilization in muscle and fat cells. Furthermore,
long-term elevations of growth hormone and Cortisol reduce glucose utilization and
stimulate glucose production {Wilson 1998}. Different types of counterregulatory
hormones work together and produce the glucose-raising effect.
The neural glucoregulatory factor includes norepinephrine, a neurotransmitter
released from terminals of sympathetic postganglionic neurons. Sympathetic
stimulation through the hepatic sympathetic nerves suppresses glycogen formation
5 and enhances glucose release. On the other hand, parasympathetic stimulation increases glycogen formation and reduces glucose release. As a result, like insulin， the neural glucoregulatory factor can regulate the glucose level in both directions
{Wilson 1998}.
The substrate glucoregulatory factors include glucose and fatty acid. Glucose acts as a feedback signal that induces glucose storage and reduces glucose production. In addition, fatty acid promotes glucose production and limit glucose oxidation probably through the glucose-fatty acid cycle {Boden，Chen, et al. 1994}.
1.1.4 Insulin
Since the discovery of insulin in the early 1920s, elucidating insulin intracellular
signaling pathway and its action on target cells has been the primary focus for a large
number of investigators for decades. We will put our main focus on the ability of
insulin to stimulate glucose uptake since it is more relevant to the clinical symptoms
of DM. Defining the insulin signaling mechanism could lead to a better
understanding of insulin resistance associated with obesity and Type 2 diabetes.
6 1.1.4.1 Function of Insulin
Insulin is an anabolic hormone with powerful metabolic and mitogenic effects.
Insulin exerts its effect predominately on three main target tissues: liver, fat and
skeletal muscle cells. Insulin regulates a variety of metabolic and mitogenic effects
in target tissues, including stimulation of glucose transport and glycogen synthesis,
inhibition of hepatic gluconeogenesis and glycogenolysis, stimulation of lipogenesis,
gene transcription and translation, and cell replication {Pessin & Saltiel 2000}.
1.1.4.2 Discovery and Production of Insulin
The road of insulin discovery started in Toronto. In the early 1920s, Banting, Best
‘ and Macleod began the work by succeeded in preparing extracts of pancreas that had j i )
a potential to lower blood glucose level in patients with diabetes. This discovery
leaded to the crystallization of insulin and its eventual identification as a peptide
hormone. Methods for the preparation of insulin from animal pancreas were
developed by commercial firms, which became available for both the treatment of
DM and the study of protein biochemistry {Bliss 1987}.
7
J The P cells of the pancreatic islets are responsible for insulin production. The removal of a 23-amino-acid leader sequence from the preproinsulin produces proinsulin. Proinsulin is then converted to insulin and C-peptide through the action of endoproteases in secretary vesicles {Hunter & Garvey 1998}. With the aid of the cytoskeleton, secretary vesicles fuse with the PM of |3 cell and release their contents to the circulation.
The secretion of insulin is determined by many regulatory factors. For example, protein kinase C (PKC) in insulin signaling pathway has been identified in the pancreatic islet to stimulate insulin secretion. Besides, glucagons, glucose and amino acid have also been found to enhance the secretion. On the other hand,
somatostatin has been reported to inhibit insulin secretion {Leslie & Robbins 1995}
1.1.4.3 Insulin Signaling Pathway
1.1.4.3.1 Insulin Receptor
The insulin intracellular signaling pathway (fig. 1) is initiated with the specific
binding of insulin to insulin receptor (IR) on the PM of target cells. The IR consists
8 Fig.l Metabolic and mitogenic signaling pathway in insulin action IR = insulin receptor; IRS = IR substrate; She = Src and Collagen Homologous; mSos =mammalian Son of Sevenless; MAPK = Mitogen-Activated Protein Kinase; PI3-K = Phosphatidylinositol 3-Kinase; PIP2 = Phosphatidylinositol 4， 5-bisphosphate; PIP3 = Phosphatidylinositol 3，4，5-triphosphate; PDK = PI3-K-dependent kinase; PKC = Protein Kinase C; GSK-3 = glycogen synthase kinase-3 Figure 1. Insulin Signaling Pathway
IR PM GLUT4 II 丄
fY\ PI3-K ^~乂 ^Klh^ I^I / GLUT- \ pllO p85 ” containing MAPK \ Vesicles J n GLUT4 遍 a \j PDK p90 So kinase A ，r • T • METABOLIC ^_ Akt PRC GSK 3 MITOGENIC EFFECTS EFFECTS
D00 ) of two a and two P subunits that are disulfide linked into a a2p2 tetrameric complex
{Pessin & Saltiel 2000}. The a-subunit is located entirely at the extracellular face of the PM and contains the insulin-binding site while the P-subunit is a transmembrane peptide. The binding of insulin to the a-subunit activates the tyrosine kinase and leads to autophosphorylation of tyrosine residues of P-subunit. Once activated, the
IR phosphorylates a number of important proximal substrates on tyrosine.
Molecules with phosphate-tyrosine binding domains or with Src homology 2 (SH2) domains are targets for IR phosphorylation. IR substrate (IRS) and Src and collagen
homologous (She) engage the IR directly and serve as docking sites for downstream
signaling substrates {Le Roith & Zick 2001}.
1.1.4.3.2 MAPK Pathway
Divergence of insulin signaling pathways occurs at the level of IRS and She docking
proteins. The binding of Grb-2 adaptor protein with IRS or She via their SH2
domains triggers mitogen-activated protein kinase (MAPK) pathway. Grb-2 then
binds with the mammalian Son of Sevenless (mSos), a nucleotide exchange protein
that catalyzes the exchange of guanosine diphosphate (GDP) for guanosine
triphosphate (GTP) on Ras. The GTP bound Ras activates Raf kinase and initiates a
9 cascade of phosphorylations leading to the activation of MAPK kinase, MAPK and pp90 S6 kinase {Hunter & Garvey 1998}. In addition, activated forms of Ras also activate Rac, which in turn induces membrane ruffles {Tsakiridis, Tong, et al. 1999}.
The MAPK pathway mediates the mitogenic and growth-promoting effects of insulin by phosphorylating transcription factors, leading to the induction of gene.
1.1.4.3.3 Phosphatidylinositol 3-kinase (PI3-K) Pathway
The activation of PI3-K pathway is initiated from the phosphorylation of IRS. PI3-K,
a lipid kinase consisting of a p85 regulatory subunit and a pi 10 catalytic subunit,
binds to IRS through the SH2 domains of the p85 subunit. This results in the
activation of pi 10 subunit to produce phosphatidylinositol 3，4，5-phosphate (PIP3).
PIP3 binds to PI3-K-dependent kinase (PDK) and leads to the activations of
downstream kinases including the Akt (also known as protein kinase B), PKC ； and
PKC X and inactivation of glycogen synthase kinase-3 (GSK-3). The PI3-K pathway
is mainly responsible for the metabolic actions of insulin including glucose uptake,
glycogen and protein synthesis.
The PI3-K pathway is essential for insulin-induced glucose uptake. It activates the
10 translocation of intracellular GLUT4 to the PM and thereby enhances the glucose transport across the membrane. Besides, the PI3-K pathway is also mediating the glycogen and protein synthesis. GSK-3 mediates the inactivation of glycogen synthesis and protein synthesis eukaryotic factor. In response to insulin, activated
Akt phosphorylates and inactivates GSK-3. This directly results in enhancement in glycogen synthesis and protein synthesis. In addition, insulin also activates protein synthesis through activation of proteins downstream of Akt in the kinase mammalian
Target of Rapamycin pathway {Le Roith & Zick 2001}. As a result, successful sequentially activations along the insulin-signaling pathway are essential for the homeostasis of glucose and protein metabolism.
1.1.5 Glucose Transporters
Since the lipid bilayers that make up the cell membranes are impermeable to glucose,
two families of GLUT are responsible to the transport of glucose across the cell.
The sodium-linked GLUT is an active transport transporter located in the luminal
epithelial surface of cells of small intestine and proximal convoluted tubule of the
kidney. These transporters actively transport glucose against a
glucose-concentration gradient using the energy provided from the electrochemical
11 potential difference of sodium ions across the PM {Kahn 1992}.
The second family of GLUT is a facilitative transporter. This type of transporters facilitates the diffusion of glucose down the glucose-concentration gradients. Five glucose transporters genes have been identified according to their chronological order of discovery. Most facilitative transporters are saturable and are mostly specific for the transport of D-glucose. GLUTl is present in most tissues，particularly in the PM of blood-brain barrier, erythrocyte, placenta, kidney and perineurial sheath of muscle.
GLUTl is largely a basal transporter and is responsible for transporting glucose across the blood-brain barrier; GLUT2 is predominately found in the PM of kidney, small intestine, liver and pancreatic beta cell. GLUT2 is insulin independent and functions to facilitate rapid glucose equilibrium across the membrane. Its ability of glucose exchange is so high that it can cope with a wide range of physiological I ！ glucose concentration across the membrane where transport does not become rate limiting in GLUT2 predominating tissues {Kahn 1992}. This is especially important
to liver and kidney as it provides a mechanism to regulate physiological glucose
levels by shifting between glucose production and glucose utilization; GLUTS is
mostly found in all tissues but is predominately expressed in brain and placenta.
GLUTS interacts with GLUTl and is responsible for transporting glucose to neurons
12 {Frier & Fisher 1999}. Once again GLUT3 is a basal transporter and provide cells with basic glucose metabolic need; GLUT4 is the major insulin responsive transporter that is mostly found in intracellular compartments of adipose cells, cardiac and skeletal muscle cells. Upon stimuli such as insulin or exercise, the intracellular
GLUT4 is recruited to the PM that can produce up to a ten-fold increase in glucose transport over the basal rate. The intracellular location of GLUT4 at basal state provides a reserve pool by which GLUT4-containing tissues can rapidly modify the rate of glucose transport. This transporter can be saturated in vivo. Thus, transport can be rate limiting for tissue where GLUT4 is the major glucose transporter; GLUT5 is a fructose transporter that is located in small intestine, sperm, kidney, brain, adipose cells and muscle {Shepherd & Kahn 1999}.
1.1.6 Role of skeletal muscle in glucose homeostasis
Skeletal muscle is an important site for glucose utilization. It accounts for
approximately 70 to 80 % {Jackson, Bagstaff, et al. 2000} or even 85% {DeFronzo,
Jacot, et al. 1981} of total glucose metabolism. Another study from Meyer et al
suggested that after a carbohydrate meal, liver took up about 25 to 35% of ingested
carbohydrate. Of the remaining 65 to 75% that entered the circulation, about 40%
13 would be taken up by skeletal muscle and about 10% by kidney. The brain would take up another 15 to 20% of the ingested glucose. Therefore, uptake of glucose by liver, kidney, skeletal muscle and brain would be accounted for almost 90% of the ingested glucose {Meyer, Dostou, et al. 2002}. Despite there are numbering differences between different studies, skeletal muscle is still the predominately site for peripheral glucose utilization. Since there is a lack of intracellular free glucose accumulation, it is widely believed that glucose transport is the rate-limiting step for glucose metabolism in skeletal muscle {Ziel, Venkatesan, et al. 1988}. Thus, normal functioning of GLUT4 in skeletal muscle is important to regulate the glucose utilization.
1.1.7 Insulin Resistance . i[
The concept of insulin resistance arose from the insulin therapy to patients with ''
* diabetes. In the late 1930s, it became evident that patients with diabetes could be
divided into two groups based on their relative sensitivity to exogenous insulin
therapy {Moller 1993}. The two patterns of insulin sensitivity correspond to the
modem classification of Type I and Type II diabetes. The response of insulin
treatment in patients with Type II diabetes is far less effective than the one from
14 patients with Type I diabetes. Despite the multiple functional roles of insulin, the term insulin resistance is generally referred as the ineffectiveness of insulin to stimulate glucose uptake. It contributes to clinical symptoms of diabetes: hyperglycemia and impaired glucose tolerance.
In the case of insulin resistance, insulin levels are elevated to compensate for the loss of insulin effectiveness. This compensatory insulin secretion is the result of expanded p cell mass and altered enzyme expression in pancreas {Cavaghan,
Ehrmann, et al. 2000}. Evidences for these two correctional mechanisms come from studies of Zucker fatty rats，which are obese and become insulin resistant at the later stages of their lives. They are found to contain expanded pancreatic P cell mass and do not develop DM. In addition, they are also found to have altered activities in two glucose-phosphorylating enzymes: glucokinase (GK) and hexokinase (HK).
Conversion of glucose to G-6-P is normally controlled by GK in pancreatic P cells, ’
i
which is less glucose-sensitive than HK. Insulin resistance causes an increase in HK
expression. The overexpression of HK allows the pancreatic p cells to induce a
higher glucose-stimulated insulin release and overall glucose metabolism {Cavaghan,
Ehrmann, et al. 2000}. Insulin secretion is initially able to compensate for insulin
resistance to produce a normal glucose tolerance. Thus, insulin resistance may occur
15 for many years before the diagnosis of Type II diabetes is made. However, the p cells may be exhausted by the overproduction of insulin or genetic and inherited defect. It appeared that the defect is specific for glucose-dependent insulin release
{Leslie & Robbins 1995}. Once the p cells fail to produce sufficient insulin, hyperglycemia occurs and may lead to the development of Type II DM. As a result, insulin resistance could be the main cause of DM. Thus, by understanding the nature
( of insulin resistance, it could lead us to a better understanding of the pathology of
DM.
1.1.8 Glucose abnormality and its complications
Glucose abnormality posts a severe threat to our survivals. On one end of the spectrum, hypoglycemia predominantly affects the human brain. Normal functioning of the brain requires a continuous supply of glucose as its primary
metabolic fuel. Since the brain cannot synthesize or store glucose, inadequate
supply of glucose could impair neural metabolism. In addition, glucose can be
converted into other biochemical compounds such as neurotransmitters. Lack of
glucose precursors may affect their productions {Frier & F 1993} and influence brain
function. Taken together, glucose deprivation can cause severe brain dysfunctions
16 and even brain death.
On the other end of the spectrum, sustained hyperglycemia causes severe tissue damage and leads to diabetic complications. The processes of tissue damage involve protein glycation, excessive production of polyol compounds from glucose and increase in blood flow. First of all, in the presence of high glucose concentration, glucose can be incorporated into proteins by unregulated nonenzymatic glycation reaction. The glucose incorporation may modify the protein structure and thereby impair the protein function. Besides, the glycated proteins also form advanced glycation end products (AGEs) that further disrupt and cause additional deleterious effects {Wilson 1998}. The second potential cause of tissue damage lies in the polyol pathway. Glucose is reduced to sorbitol by aldose reductase and then
.. I' oxidized to fructose by sorbitol dehydrogenase. Increased activation of polyol
i) pathway inactivates 3-phosphate dehydrogenase that converts glyceraldehydes 丨
3-phospahte to pyruvate. Excess glyceraldehydes 3-phospahte induce nonenzymatic
protein glycation. In addition to defected protein structure and function such as in
the case of basement membrane of blood vessel, there is also an alteration in the
sensitivity of intracellular receptor, which stimulates tissue thickening in the
endothelial surface through the release of cytokines and growth factors. Besides,
17
』 sorbitol accumulation also leads to a decrease in intracellular myoinositol and results in damages of tissues {Nussey & Saffron Whitehead 2001}. Furthermore, tissue-damaging reactive oxygen species may be produced through the formation of
AGEs and altered polyol pathway activity {Inoguchi, Li, et al. 2000}. Since aldose reductase is present in the retina, nerve, lens, glomerulus and blood vessel wall, these are also the sites that mostly be affected by tissue disruption. The third potential cause of tissue damage lies in the increase of blood flow. The arteriolar pressure was increased despite of a normal blood pressure. The elevated blood flow may increase the entry of potentially damaging proteins into the wall of blood vessels and cause tissue devastation {Wilson 1998}.
Tissue damaged by these three factors can be illustrated in diabetic retinopathy. In a normal situation, the capillary blood vessel of retina is wrapped by pericytes that may ii ！HI help to control blood flow. Early features of diabetic retinopathy include increase | blood flow, thickening of basement membrane and loss of endothelial cells and pericytes. High blood pressure may rupture the weakening capillary and cause hemorrhages and accumulation of serum as hard exudates. It may result in capillary closure and ischemia. In addition, activated leukocytes and the production of growth factors may induce the growth of new vessels. Unfortunately, these new-formed
18
A vessels are fragile and can rupture easily. Thus, tissue damages caused by
hyperglycemia are deleterious and may result in vision loss {Nussey & Saffron
Whitehead 2001}.
Tissue damages caused by sustained hyperglycemia result in diabetic complications
including retinopathy with risk of vision loss; nephropathy resulting renal failure;
peripheral neuropathy with foot ulcers and the potential amputation; autonomic
neuropathy with gastrointestinal, cardiovascular and sexual dysfunction. Moreover,
patients with diabetes have increased risks of hypertension, abnormalities in
lipoprotein metabolism, and periodontal diseases. They are also in high risk of
having atherosclerotic, cardiovascular, peripheral vascular, and cerebrovascular ii disease {The Expert Committee on the Diagnosis and Classification of Diabetes 丨!I
Mellitus 2001}. Since the resulting symptoms affect the whole body and require ‘
chronic treatment, management of hyperglycemia is critical to our health. I
1.2 Actin
Actin, one of the most versatile and abundant proteins in eucaryotic cells {Bray 2001},
plays an important role in maintaining normal body activities. Actin filaments,
19
A which are made up of actin molecules and a number of actin-binding proteins, are highly dynamic structures. They normally exist in bundles or networks and are usually found in the cell cortex just beneath the PM. The organizational flexibility of the actin filaments permits a cell to interact with the continuous changes in the cellular environment.
1.2.1 Function of Actin
The diverse roles of actin filaments are essential to maintain a normal cellular functioning. First of all, actin filaments define organelle position by connecting jt » organelles and protein complexes with cell cortex. Secondly, they serve as tracks for j intracellular transportation. For instance, the Listeria monocytogenes, a ！I microorganism that causes severe food poisoning in humans, propels within and ； spreads between the host cells by controlling the host's actin-based motility system 'i
{Goldberg 2001}. Listeria forms an actin tail to propel the organism through the cell cytoplasm. To spread to the adjacent cell，Listeria pushes against the membrane to form protrusion with the organism at the tip. The adjacent cell then takes up both the protrusion and the organism by the process of endocytosis. Thirdly, actin filaments promote cell locomotion by the formation of actin structures such as
20
A lamellipodium. Lamellipodium is a board membrane protrusion that projects
outward from leading end of the cell membrane and attaches to the nearby substratum.
This attachment acts as anchor and pulls the tail end of the cell toward the cell body.
Fourthly, actin filaments provide structural support and define cell shape by forming a
scaffold underlying the cell body. The actin network helps support and stiffen the
PM. This structural enhancement allows the cell to perform many different tasks.
On one hand, they can push the PM outward to form membrane projections such as
microvilli. On the other hand, during cell division, PM is pushed inward by the
formation of the contractile ring composed mainly of actin filament and myosin-II
filaments. Finally, actin is the main component of the thin filament in sarcomere,
} which in turn is the basic unit of a muscle fiber. The ability of muscle contraction j
generates the force required for body movement. i ‘j
11 1),
One interesting example of actin functioning is the involvement of actin in the ？|
transport of synaptic vesicles in neurons. The neurotransmitter-containing synaptic
vesicles are responsible for the transmission of nerve impulse between neurons.
Movement of synaptic vesicles to the PM is facilitated by cytoskeleton. The arrival
of action potential at the nerve terminal triggers the exocytosis of synaptic vesicles
and therefore the release of neurotransmitter into the synaptic cleft. To avoid the
21
• failure of synaptic transmission, rapid replenishment of synaptic vesicles must occur.
Indeed, there is a reserve pool of synaptic vesicles locating within the presynpatic terminal {Greengard, Valtorta, et al. 1993}. A family of phosphoproteins called synapsin reversibly tether synaptic vesicles with the actin cytoskeleton in the reserve pool {Hilfiker, Pieribone, et al. 1999}. In response to the physiological requirement, these replacement vesicles are recruited to the releasable pool for later neurotransmitter release.
1.2.2 Actin Accessory Protein
More than 60 different accessory proteins have been identified and their bindings with j 丨I ii the actin filaments provide structural and functional advantages. For example, cross-linking proteins such as a-actinin and filamin control the tertiary structures of I actin filaments into bundles or networks respectively; monomer-binding proteins 丨丨! including thymosin and profilin affect the rate of actin polymerization; filament-capping proteins such as CapZ prevent the monomer exchange at the capped
end; actin-depolymerizing proteins such as cofilin accelerate minus end
depolymerization； and force-generating proteins such as myosin produce force and
movement at the expense of an ATP molecule {Southwick, Purich, et al. 2000}. The
22
J interaction of the accessory proteins allows the cells to perform various functions and tackle cellular problems.
1.2.3 Actin Polymerization
Actin filament is highly dynamic as it reorganizes continuously in response to cellular and environmental changes. The basic structural subunit of an actin filament is globular actin (G-actin) {Bray 2001}. Each G-actin equips with either an ATP or adenosine diphosphate (ADP). G-actin monomer interacts with each other and assembles into a two-stranded helical filamentous actin (F-actin) polymer. The
I assembly is accompanied by the hydrolysis of ATP to ADR ATP hydrolysis
ii I, increases the rate of polymerization but is not essential for filament formation. ii I • i Instead，the stored energy released from ATP hydrolysis provides a driving force for
‘ actin depolymerization {Lodish 1995}. 丨丨
The polymerization of actin filament occurs in three phases. There is an initial lag phase as G-actins begin to aggregate into short fragments. A rapid polymerization phase follows in which short fragments elongate by adding G-actins on both ends.
The steady state phase is accomplished when the concentration of G-actin is in
23
J equilibrium with the filaments. This particular concentration of G-actin during steady state is called the critical concentration. Moreover, the polymerization rate is different at the two ends of the actin filament� In fact, the plus end of the actin filament grows five to ten times faster than the minus end {Lodish 2000}. As a result, if the G-actin concentration falls between the critical concentrations of plus and
minus ends, a treadmilling type of mechanism is employed. Without changing the
length of the filament, actin monomer is continuously added to the plus end and
moved toward the minus end. Eventually, it will be dissociated from the minus end.
Treadmilling is an important phenomenon as it can generate force and movement
resulted from the site-specific actin polymerization. Lamellipodia extension is an
example of treadmilling involvement {Pantaloni, Le Clainche, et al. 2001}. i
！ M
I ,1 1.3 Interaction between Insulin, GLUT4 and Actin in Glucose Homeostasis
M
A number of studies have identified the ability of insulin to induce actin remodeling.
It is also widely accepted that actin is involved in the cellular transport of organelles
and vesicles including the synaptic vesicles transport in neurons {Hilfiker, Pieribone,
et al. 1999}. Thus, it is conceivable that actin reorganization may also involve in the
translocation of GLUT4 to the PM. The regulation of glucose uptake by insulin in
24
J skeletal muscle is critical for the maintenance of glucose homeostasis {Jackson,
Bagstaff, et al. 2000}. Since glucose transport is the rate-limiting step for skeletal muscle in glucose metabolism, aberrant insulin-induced GLUT4 translocation may affect the glucose homeostasis and contribute to insulin resistance.
1.3.1 Insulin-Induced Actin Remodeling
Insulin-induced actin remodeling is ubiquitous in a variety of cell types including isolated rat adipocytes {Omata, Shibata，et al. 2000}, 3T3-L1 adipocytes {Wang,
Bilan, et al. 1998} and L6 myotubes {Tsakiridis, Vranic, et al. 1994}. In unstimulated cells, actin filaments are emerged as stress fibers. This long
>,1 microfilament bundles span the cell longitudinally. Insulin stimulation elicits a ！I dramatic reorganization of actin filaments into radially oriented actin aggregations. ‘ 1*
These structures projects from the dorsal surface and forms membrane ruffles. 丨,
These actin filaments cross-link with each other and form mesh-like structures with interdigitating branches. They are variable in length and size and are located on dorsal surface of PM. Under the view of the inverted microscope, these aggregations are mostly found near the nuclei of the cells. They appear as early as 3 min after the insulin stimulation. In addition, the length of the actin aggregations
25
J increases through time. By 10 min of insulin treatment, the aggregations are the
most pronounced. These actin structures will begin to condense and break into small
fragments. They appear as dense spots and only traces of actin structures remain
after 30 min of insulin treatment. Throughout the stimulation, insulin does not alter
the structure of long actin stress fibers.
1.3.2 Actin Remodeling and Insulin-Induced GLUT4 Translocation
Insulin-mediated glucose transport can be described as the translocation of GLUT4
1 from the intracellular compartment to the PM of target cells. At the basal state, there > »j I are about 90% of GLUT4 located in the intracellular compartments. Insulin | ‘ ii mobilizes GLUT4 to attain a new equilibrium state that about 30% is now inserted on j I i PM {Shepherd & Kahn 1999}. Besides, insulin also promotes the PI3-K sensitive ‘ ‘ J *i
extemalization of GLUT4 by accelerating the movement of GLUT4 from early 'fi
endosome to the PM {Foster, Li, et al. 2001}. In neuron, synaptic vesicles are
transported and accumulated at the PM using the actin transport system. Thus, it is
convincible that actin reorganization may also involve in the translocation of
GLUT4-containing vesicles to the PM. Indeed，study from Tong et al has revealed
the colocalization of GLUT4 with actin structures at the dorsal surface after insulin 26
A stimulation. Jasplakinolide, drug that affect actin filament stability, reduced GLUT4
appearance at the cell surface after insulin treatment {Tong, Khayat, et al. 2001}. In
addition, intact actin network is necessary for the exocytotic recruitment of GLUT4
{Omata, Shibata, et al. 2000}. Actin disassembly has also been shown to reduce
glucose uptake {Tsakiridis, Vranic, et al. 1994}. As a result, actin filaments may
directly participate in controlling the GLUT4 vesicle trafficking. Completed intact
actin reorganization may play a critical role in insulin-mediated GLUT4 translocation
[ and thereby control the glucose uptake. ！
1 1.3.3 Involvement of Insulin Signaling Molecules in Actin Remodeling
I- I,'
i>' I ..5 . ( The membrane-ruffling phenomenon is mainly mediated by the activation of PI3-K 丨
！ pathway. PI3-K is essential to actin remodeling and glucose uptake. PI3-K 'i ‘ ？丨
inhibitor such as wortmannin completely blocked the formation of insulin-induced
actin structures and glucose uptake {Somwar，Niu, et al. 2001}. In fact, the p85
subunit of PI3-K was found to colocalize with actin structures in L6 myotubes
{Khayat, Tong, et al. 2000} and IRS-1, PI3-K and cytoskeletal elements in 3T3-L1
adipocytes {Clark, Martin, et al. 1998}. It is believed that insulin causes the
translocation of PI3-K to GLUT-containing vesicles and sequentially propagates the
27
A insulin-signaling pathway and stimulates the translocation of GLUT vesicles to the
cell surface. Rac, a GTP-binding protein that is activated by Ras, is also a critical
protein leading to actin remodeling. Transient transfection of dominant inhibitory
Racl prevents insulin-induced dorsal actin reorganization and blocks GLUT4
translocation to the cell surface {Khayat, Tong, et al. 2000}. Rac appears to locate
downstream of PI3-K in the pathway regulating membrane ruffles {Tsakiridis, Vranic，
et al. 1994 }.
f
！
Cytochalasin D (CD)，a widely used inhibitor of actin filament formation, inhibited
actin reorganization and glucose uptake. The inhibitory effect of CD has no effect '丨
t' •身 I衫 1 on tyrosine phosphorylation of IRS-1, binding of PI3-K to ERS-l {Tsakiridis, Vranic, 1 - J, et al.‘ 1994} and stimulation of PI3-K {Wang, Bilan, et al. 1998}. . However, CD 丨�
！I
significantly diminishes the insulin-mediated association of IRS-1, p85a and pilOp of
•j ！丨
PB-K with the GLUT-containing vesicles {Wang, Bilan, et al. 1998}. Thus, the ？i
cellular insulin-induced actin reorganization is not required for the activation of
insulin signaling molecules but instead provides a site for communication of signaling
molecules.
Intact actin filaments are essential for the propagation of insulin signals leading to the
28
A activation of the MAPK pathway {Tsakiridis, Wang, et al. 1997}. The treatment with CD in insulin-stimulated cells abolishes DNA synthesis. Moreover, activation of Ras, tyrosine phosphorylation of p38 MAPK and She, and the binding of Grb2 are also affected {Tsakiridis, Bergman, et al. 1998}. Taken together, very early events of insulin action are not dependent on the actin remodeling. The possible actin-associated step may locate between She phosphorylation and Ras activation.
Therefore, actin reorganization governs the mitogenic action of insulin probably through tethering the signaling molecules together for communication.
PI3-K pathway is responsible for GLUT4 translocation. In addition, it has also been reported that the MAPK pathway controls its activation. There is a differential sensitivity of GLUT4 translocation and glucose uptake to wortmannin: in a low dose of wortmannin. GLUT4 translocation is unaffected while glucose uptake is abolished. Thus, a successful glucose uptake cannot be relied on GLUT4 translocation alone. On the other hand, P38 MAPK phosphorylation is also reduced by a low wortmannin concentration through the inhibition of an upstream insulin-signaling target. Unfortunately, the identity of this wortmannin-sensitive target remains to be defined {Somwar, Niu, et al. 2001}. However, it is convincible that full stimulation of insulin-stimulated glucose uptake may require corresponding
29 GLUT4 translocation and activation.
1.3.4 Actin Remodeling and Insulin Resistance
Taken together, the insulin-induced GLUT4 translocation and activation are depended on an intact actin network and insulin-stimulated actin reorganization. The newly formed actin network may allow correct sequential activations in insulin signaling pathway by distributing the signaling molecules to proper locations. In addition, the actin reorganization and the formation of membrane ruffles may also directly participate in the translocation of GLUT4. Nowadays, despite of numerous
researches, the pathology of insulin resistance is still unclear. Since actin
disassembly abolishes GLUT4 translocation, aberrant GLUT4 translocation caused by
actin disruption affects glucose homeostasis and may contribute to insulin resistance.
1.4 Hypothesis and Objective
1.4.1 Rationale
Insulin resistance, the ineffectiveness of insulin to stimulate glucose uptake, is a major
30 characteristics of Type II DM and obesity. Although insulin resistance is well characterized, the exact mechanism remains unclear. Pancreatic B cells compensate for insulin resistance by over-secreting insulin, leading to hperinsulinaemia. Other metabolic disturbances including hyperglycemia, hypercholesterolaemia and elevation in FFA levels are also common in patients with DM. Previous studies demonstrate the participation of the actin cytoskeleton in the translocation of GLUT4 following insulin stimulation. It is conceivable that these metabolic derangements may further perpetuate insulin resistance, by interfering with the actin reorganization.
1.4.2 Hypothesis
Preincubation of L6 myotubes in conditions of insulin resistance would suppress actin remodeling and the association of GLUT4 with actin filaments, leading to impairment of insulin-medicated glucose uptake. It is plausible that interference of actin dynamics may contribute to insulin resistance in vivo.
1.4.3 Objective
1. To demonstrate insulin can stimulate actin reorganization and GLUT4
31 translocation in L6 myotubes
2. To examine the effect of preincubation in conditions of insulin resistance on
actin reorganization, GLUT4 translocation and glucose uptake in L6 myotubes
3. To elucidate the ultra-structural relationship between actin and GLUT4 using
TEM.
32
A CHAPTER TWO MATERIALS AND METHODS
2.1 Materials
Tissue culture medium, serum, and other tissue culture regents were purchased from
Gibco™ Invitrogen Corporation (Grand Island, NY，USA). Humulin R insulin was
purchased from Lilly France S.A. (Fegersheim, France). ProLong Antifade
mounting solution, Oregon Green-phalloidin, rhodamine-phalloidin and Alexa Fluor
anti-rabbit or anti-mouse secondary antibodies were purchased from Molecular
Probes Inc. (Eugene, Oregon, USA). Polyclonal rabbit c-myc antibody was
purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Monoclonal
anti-P-actin antibody, OA, LA, PA, CHO hemisuccinate, cytochalasin B,
methyl-p-cyclodextrins, and bovine serum albumin (BSA) were purchased from
Sigma-Aldrich (St Louis, MO, USA). Bio-Rad Protein Assay was purchased from
Bio-Rad Laboratories (Hercules, CA, USA). 2-deoxy-[^H]D-glucose was purchased
from Amersham Pharmacia Biotech UK Limited (Little Chalfont, Buckinghamshire,
England). OptiPhase "HiSafe" 2 Scintillation Solution was purchased from Wallac
Scintillation Product (Wallac Oy, Turku, Finland). LR White Hard Grade
Embedding Resin was purchased from London Resin Company Ltd. (Reading,
33 Berkshire, England). 18nm Colloidal Gold-AffiniPure donkey anti-rabbit and 12nm
Colloidal Gold-AffiniPure goat anti-rabbit antibodies were purchased from Jackson
ImmunoResearch Laboratories, Inc (West Grove, PA, USA). Chemicals for the
TEM study were kindly provided by Dr. F. L. Chan (The Chinese University of Hong
Kong, Hong Kong, China). Unless otherwise specified, chemicals used in our study
were provided by Lee Hysan Clinical Research Laboratories (Shatin, Hong Kong,
China).
The L6 myotubes, which differentiate in vitro from myoblasts into multinucleated
myotubes by multiple cell fusions, was derived from rat thigh skeletal muscle by
Yaffe et al in 1968 {Yaffe 1968}. GLUTl, GLUTS, and GLUT4 are all found in L6
cells. GLUTS is largely found on the cell surface, whereas half of the GLUTl is
intracellular and respond to insulin. GLUTl is believed to be responsible for the
basal glucose transport and is found during all stages of cell development. The
amount of GLUTl is very high in the myoblast stage. Its amount decreases sharply
during alignment and reaches the lowest amount in myotubes stage {Mitsumoto,
Burdett, et al. 1991}. In addition, there is only a marginally increase of GLUTl in
PM after insulin stimulation {Mitsumoto & Klip 1992}. Meanwhile, in contrast to
GLUTl, the amount of GLUT4 is the lowest in myoblast and alignment stage and
34 sharply increases in the myotubes stage {Mitsumoto, Burdett，et al. 1991}. The inulsin-mediated glucose uptake is negligible in myoblasts, and increases with cell alignment and fusion {Mitsumoto & Klip 1992}.
Numerous studies have been established to show that L6 cell line is a suitable cell culture system to study glucose uptake. First of all, the development of L6 myoblasts to myotubes closely resembles to that of embryonic muscle in vivo. They both multiply, align and fuse to form multinucleated myotubes. Secondly, L6 cells are responsive to insulin and serum treatments. The insulin-induced glucose uptake is about 200% in serum-deprived L6 myotubes with respect to basal rate {Klip, Li, et al. 1984}. Thirdly, glucose uptake is rate limiting and GLUT activity is not affected by changes of cell shape and substratum after trypsinization and subculture {Klip,
Logan, et al. 1982}. Fourthly, human muscle cells have been used to study their insulin-mediated glucose uptake. The results are highly comparable with those of rat
L6 myotubes. In fact, they both show similar dose-dependent response and Km values for glucose uptake {Sarabia, Ramlal, et al. 1990}. In addition, cytochalasin B
(CB), a fungal metabolite that specifically binds to GLUT, is used to measure non-specific glucose uptake. This inhibitor results in inhibition of transport activity
by binding with GLUT in a one-to-one ratio. The use of CB provides a more
35 accurate measurement of glucose uptake {Klip, Logan, et al. 1982}. Thus, L6 myotubes are indeed an accurate tool to investigate glucose transport.
In our study a modified family of L6，the L6 GLUT4myc myotubes, was used. The presence of an exofacial c-myc epitope tag on GLUT4 enables us to identify its
intracellular position via immunofluorescence method. It allows us to estimate the
proportion of GLUT4 at the cell surface or the total content in either intact or
permeabilized cells, respectively. They also express more than ten times higher level
of GLUT4myc than do any of the endogenous transporters {Tong, Khayat, et al.
2001}. As a result, the use of L6 GLUTmyc myotubes can be an effective tool to
study insulin action {Ueyama, Yaworsky, et al. 1999).
2.2 Cell culture
2.2.1 Cell Culture
Mediums used in our study:
1. Growth medium: a - Eagle's minimum essential medium (a -MEM) with
10% (v/v) fetal bovine serum (FBS) and 1% (v/v) antibiotic-antimycotic
36 (AB) solution (10,000U/ml penicillin Q lOmg/ml streptomycin and 25mg/inl
amphotericin B)
2. Differentiation medium: a -MEM with 2% (v/v) FBS and 1% (v/v) AB
3. Freezing medium: a -MEM with 10% (v/v) FBS, 1% (v/v) AB and 10%
dimethyl sulfoxide (DMSO)
All medium and chemicals used for cell culture were temperature pre-equilibrated at a
37°C water bath. At the beginning of study, ampoules of L6 skeletal muscle cells expressing GLUT4inyc were kept in liquid nitrogen. At the start of cell culture, L6
ampoule was thawed in a 37°C water bath. Sterile pipette was used to resuspend the
cells three to five times. Cells were added slowly and directly to 10ml growth
medium in a culture flask and incubated in an atmosphere of 5% CO2 at 37®C.
Medium was changed with fresh growth medium four hours after the subculture to
remove DMSO. Thereafter, growth medium was changed at least every 48 hours.
Cells were maintained for 10 to 15 passages before a new batch of L6 cells would be
started.
Trypsinization of cells was done at less than 60 to 80% confluency to maintain
passage and seeding for experiments. Cells subculturing was done as the following:
37 the medium was decanted and cells were rinsed with phosphate-buffered saline (PBS) twice. Then, cells were trypsinizated with 3ml 0.25% (w/v) trypsin for approximately 5 to 10 seconds. Flask was rocked side-to-side to ensure the cell
surface was covered with trypsin. Trypsin was decanted and the flask was placed
in 37� Cincubator for 30 seconds. Differentiation medium (6ml) was added and the
flask was tapped firmly. Cells were observed under a microscope to ensure that cells
had detached from the flask. To maintain passage, approximately 1 X lO^ cells/ml
cells were added to a flask with 10ml growth medium.
For myotubes differentiation, myoblasts were plated in differentiation medium at
approximately 4 X lO* cells/ml, fed fresh medium every 48 hours and used six to
eight days after seeding. For glucose uptake, cells were grown to the stage of
myotubes in 12-well plates. Meanwhile, for IM study, cells were grown on
25-mm-diameter glass coverslips placed in 6-well plates. Finally, for TEM study,
cells were grown in 100 X 20nmi culture dish.
For freezing down of cells, trypsinizated cells were diluted with freezing medium at a
final concentration approximately 2X1 (Z cells/ml. Diluted cells were then
transferred 1.0 to 1.5ml per cryotube. They were frozen at once in acetone with dry
38 ice or placed in the -80®C freezer overnight and then into liquid nitrogen for storage.
Cell culture was maintained as described previously {Khayat, Tong, et al. 2000}.
2.2.2 Reagents Preparation and Incubation
Glucose and insulin were diluted in double distilled water (DD-H2O) to the desired concentrations. For the FFA incubation experiments, OA, LA and PA were selected.
Both OA and LA are water-soluble as they are pre-equipped with methyl-p-cyclodextrins to facilitate their dissolutions in culture medium. Stock solutions of PA (30mM) and CHO (lOOmM) were dissolved in 100% methanol
(MeOH) and 100% ethanol (EtOH) respectively and were then diluted with DD-H2O to the desired concentration. For the incubation experiments, cells were incubated in various concentrations of insulin (up to lOOjjM) and glucose (up to 25mM), FFA (up to 75jjM) or CHO (up to 60^lM) for up to 24 hours. Same amount of reagents were added to the medium during the serum-deprived period.
2.3 2-Deoxyglucose Uptake
39 L6 GLUT4inyc muscles cells were grown to the stage of myotubes in 12-well plates.
Myotubes were deprived of serum for 5 hours and treated with lOOnM insulin for 30
min at After discarding the medium, cell monolayers were rinsed twice with
HEPES-bufFered saline (140mM sodium chloride, 2.5mM magnesium sulphate, ImM
calcium chloride, 5mM potassium chloride, pH 7.4) and the remaining medium was
aspirated. Glucose uptake was quantitated by exposing each well with 0.5ml lO^iM
2-deoxy-[^H]D-glucose (IjiCi/ml). Nonspecific uptake was determined by
quantitating cell-associated radioactivity in the presence of 0.5ml lOpM CB per well,
which blocks transporter-mediated uptake. At the end of 5-minute period, the uptake
solution was rapidly aspirated and the cells were washed three times with 1ml
ice-cold isotonic saline. Cells were allowed to air-dry and were lysed in 300|il
0.05N sodium hydroxide overnight.
The associated radioactivity was determined by liquid scintillation counting.
Samples were neutralized in 30jil 5N hydrochloric acid. Sample aliquot (250jil) was
added to scintillation solution (3ml) and was counted using the Beckman LS 5801
Liquid Scintillation Counter (Irvine, CA, USA). Bio-Rad Protein Assay was used to
determine the protein concentration in each well. Absorbance was measured by a
TECAN SpectraFluor Plus Microplate Reader from TECAN Austria
40 Gesellschaft.M.B.H. (Untersbergstrasse, Grodig, Austria). BSA was used as the standard of the assay. The absorbance of protein was measured at wavelength
590nm with the Magellan Reader Software Version 2. Protein concentrations were calculated, using its absorbance and compared it to the standard curve (graph. 1).
The final radioactivity level in each well was calculated by subtracting the glucose uptake level from the non-specific uptake and adjusted with the protein concentration.
{Tong, Khayat, et al. 2001}.
2.4 Immunofluorescence Microscopy
2.4.1 Permeabilized Cell Staining
L6 GLUT4myc muscles cells were grown to the stage of myotubes on
25-mm-diameter glass coverslips placed in 6-well plates. Myotubes were deprived of serum for 3 hours and treated with lOOnM insulin for up to 30 min at 37®C. After discarding the medium, myotubes were fixed with ice-cold 3% (v/v) paraformaldehyde (FFA) in PBS for 5 min on ice and the plate was then transferred to room temperature for another 30 min fixation. Fixative was discarded and the cells were incubated with 0.05M ammonium chloride for 10 min to neutralize excessive
41 Graph. 1 Example of protein standard curve. Known concentrations of BSA were used as protein standards and the absorbances were measured as described in Ch. 3 Materials and Methods. The linear regression equation obtained from the standard curve was used to estimate the protein concentration of the sample. Each concentration was done in duplicate. Graph 1. Example of protein standard curve
Protein Standard Curve
y = 1.4579X + 0.32766 r 二 0.99785 ^^
1 0.8 ,0.6 —— I 0.4 0.2
0 ‘ I I ‘ I
0 0.1 0.2 0.3 0.4 0.5 0.6
Concentration (// g/ml)
H-' QJ aldehyde groups. Cells were washed with PBS three times and were permeabilized with 0.1% Trition X-100 for 3 min. Fixed and permeabilized cells were washed with
PBS and blocked with 0.1% BSA for 3 min.
Labelings of actin filament and GLUT4 in fixed and permeabilized myotubes were
carried out as described previously {Khayat, Tong, et al. 2000}. Briefly, for labeling
of actin filaments, cells were incubated for 30 min with either rhodamine- (1:1000
dilution), Oregon green-phalloidin (1:1000 dilution) or monoclonal anti-p-actin
antibody (1:250 dilution). For labeling of GLUT4myc, cells were incubated for 30
min at room temperature with polyclonal rabbit c-myc antibody (1:50 dilution).
Following the incubation of primary antibodies the cells were washed with PBS and
incubated in dark with Alexa Fluor anti-rabbit or anti-mouse secondary antibodies
respectively, at a dilution factors of 1:500 for 30 min at room temperature if needed.
For double staining of actin and GLUT4, cells would be incubated with polyclonal
rabbit c-myc antibody as primary antibody for 30 min. Then the anti-actin antibody
would be mixed with Alexa Fluor secondary antibodies and incubated in dark for
another 30 min. After the antibodies incubation, cell monolayers were washed with
PBS and mounted in ProLong Antifade solution onto glass slides.
42 For fluorescence microscopy, cells were examined with a Nikon Eclipse TE 300
Inverted Microscope with a Nikon Plan Fluor ELWD 400X/0.45 objective from
Nikon Corporation (Shinagawa-ku，Tokyo, Japan). Digital images were acquired using a Quantix Digital Camera from Photometries (Tucson, AZ, USA). The software used was the MetaMorph Imaging System Version 4.01 from University
Imaging Corporation (West Chester, PA, USA).
2.4.2 Membrane-Intact Cell Staining
Cell surface immunostaining was carried out as described above for L6 myotubes with the following modifications: the 6-well plate was placed on ice immediately after 10-minute insulin treatment. Cells were washed in ice-cold PBS and were fixed with ice-cold 3% (v/v) PFA in PBS for 5 min. After washing with ice-cold
PBS, cells were incubated in primary antibody for 1 hour at 4°C, followed by
secondary antibodies for 30 min at 4� Cin the dark. Cells were washed thoroughly
with PBS three times and were fixed with 4% (v/v) PFA for 30 min at room
temperature. Finally, cells were washed, mounted and examined as described above.
43 2.4.3 The Analysis of Actin Remodeling Reduction
Images acquired from various conditions mimicking insulin resistance were enlarged with locked aspect ratio to the standard size of 85mm * 65mm. Grids with 5mm^ grid sizes were applied to the images. The numbers of squares containing new actin structures were counted against the total number of squares occupied by the myotubes. By comparing with the control，we estimated the reduction of actin remodeling in the stimulated cells.
2.5 Live Image Microscopy
For live images, cells were examined with the Leica DM-IRB Inverted Research
Microscope with a Leica 200X/0.4 objective from Leica Microsystem Wetzlar GmbH
(Emst-Leitz-StraBe, Wetzlar, Germany). Images were captured with either the Leica
DC Digital Camera through Leica DC Viewer Software Version 3.2.0.0 or the Ricoh
XR-X3000 camera from Ricoh Company, Ltd from (Shinagawa-ku, Tokyo, Japan)
2.6 Transmission Electron Microscope Study
44 L6 GLUT4myc muscles cells were grown to the stage of myotubes in 100 X 20mm culture dish. Cells were deprived of serum for 5 hours and treated with lOOnM insulin for 10 min at 37®C. At the end of 5-hour period, cells were washed briefly with PBS and fixed with 4% (v/v) PFA and 0.1% (v/v) glutaraldehyde in O.IM cacodylate buffer (pH 7.3) for 45 min. Fixative was discarded and the cells were incubated with O.IM ammonium chloride for 1 hour. Cells were washed with cacodylate buffer three times for 15 min or stored in cacodylate buffer containing 4% sucrose overnight at 4®C.
Cells were dehydrated in graded EtOH from 30% to 95% at 15-minute interval.
Finally, the cells were dehydrated in three 15-minute intervals of 100% EtOH. After the dehydration, cells were detached from the plastic dish with propylene oxide and stored with 100% EtOH in 1.5ml Eppendorf tubes. Cells were then infiltrated with a series of LR White embedding resin and 100% EtOH mixtures (table 1) with a gentle agitation on a rotor. The resulting mixture was stored in 0.5ml Eppendorf tube and centrifliged before they were placed in an oven to polymerize at 55°C for 24 hours.
The embedding blocks were first sliced with a glass knife using the Leica Ultracut R
Ultramicrotome from Leica Microsystem Wetzlar GmbH (Erast-Leitz-StraBe,
45 Table.l Ratio of LR White and EtOH used in TEM study. Different combinations of LR White and EtOH were used for the embedding of cell samples in TEM study. Cells were prepared for TEM as described in Ch. 3 Materials and Methods. Table 1. Ratio of LR White and EtOH used in TEM study
step LR White Ratio 100% EtOH Ratio Time
1 1 1 30-60 min
2 2 1 30-60 min
3 3 1 30-60 min
4 Pure resin None Overnight
5 Pure resin None 4 hours
•Pa. Ul QJ Wetzlar, Germany). Ultrathin sections (100-120nm thick) were then sliced with a
Diatome diamond knife (Diatome Ltd, Bid，Switzerland) from the resulting blocks.
These sections were mounted on 0.2% formver in chloroform coated nickel grids (200 meshes). Formver-coated grids with sections side down were put on the droplets of
0.05M Tris-HCl buffer with 0.15M sodium chloride, pH 7.6 (TBS) for 5 min. Cells were blocked with BSA in 0.05M TBS containing 0.1% Tween 20 for 15 min. For immunogold labeling, cells were incubated with the primary antibodies for 1 to 2 hours at room temperature or overnight at 4°C. Following this incubation the cells were washed three times with 0.05M TBS and subsequently incubated with secondary
antibodies conjugated to gold for 1 hour at room temperature. Cells sections were
further washed three times with 0.05M TBS with 0.1% Tween 20 for 15 min. For
double staining, this staining procedure would be repeated. Grids were allowed to
air-dry and counter-stain with 4% uranyl acetate for 4 min and lead citrate or lead
acetate for 2 min. Sections were examined with Hitachi H-7100FA Electron
Microscope (Tokyo, Japan). Actin and GLUT4 ultrastructures were observed under
magnifications of 15000X and 30000X.
2.7 Statistical Analysis
46 Data are expressed in mean 士 SEM. Significance statistical analysis was carried out using the paired Student's T-test for comparisons between basal and insulin-mediated uptake at the same concentration. In addition, unpaired Student's T-test was used to
compare glucose uptakes and actin reduction between different concentrations or between different conditions.
47 CHAPTER THREE RESULTS
3.1 Cell Growth
L6 muscle cells differentiate through multiple cell fusions of myoblasts into myotubes, which are large, multinucleated cells. Fig. 2a showed myoblasts attached on the surface of the culture plate 4 hours after subculture. On day 1 and 2, cells multiplied
(Fig. 2b) and began to align (Fig. 2c). On day 4-5, fusions began as myoblasts merged into myotubes (Fig. 2d). On day 6-8, approximately 80% of the cells were in myotubes state and were ready for experiment (Fig. 2e).
3.2 Acute Effect of Insulin on L6 myotubes
To investigate the effect of insulin-mediated actin remodeling and glucose uptake, we performed IM, TEM and 2-Deoxyglucose Uptake studies on L6 myotubes. By performing the IM study, the spatial relationships of actin and GLUT4myc in permeabilized and membrane-intact cells can be revealed before and after insulin stimulation. Similarly, the TEM study allows us to have a more concreted evidence of actin and GLUT4myc colocalization in an ultra-structural level. In addition, we
48 Fig.2 Differentiation of L6 muscle cells from myoblasts to myotubes. L6 cells were viewed at 4 hours (a), 1 to 2 days (b and c)，4 to 5 days (d)，and 6 to 8 days (e) after subculture. The cells were examined in 200X magnification. : =霊疆靈 F,。^^s
4 8 a have measured glucose uptake under the stimulation of insulin. The degree of
glucose uptake can be served as an index of insulin responsiveness in L6 myotubes.
3.2.1 Immunofluorescence Microscopy
The IM study was performed in permeabilized and membrane-intact cells. Time and
concentration profiles were first constructed and the spatial relationships of actin and
GLUT4myc were examined.
3.2.1.1 The time profile of insulin on actin cytoskeleton in permeabilized L6
myotubes
Fig. 3 showed the morphological characteristics of actin cytoskeleton of L6 myotubes.
Labeling of myotubes with Oregon Green-phalloidin revealed the existence of
organized stress fiber network (fig. 3a). These networks were consisted of long actin
microfilament bundles of variable length and thickness, spanning the cell
longitudinally. They were detected throughout the cell. Stimulation of
serum-deprived L6 myotubes with lOOnm insulin elicited a reorganization of actin
filaments into radially oriented actin aggregations. These actin filaments
49 Fig.3 Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (time). Serum-deprived L6 GLUT4myc myotubes were left untreated (a) or stimulated with lOOnM insulin for 3 min (c)，10 min (b and d)，20 min (e) and 30 min (f) at 37�C . The cells were examined in 400X magniHcation and the white arrows in the images indicated the newly formed actin structure. The images are representatives of at least three experiments. Fig 3. Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (time).
0 min 10 min
3 min 10 min • • • 20 min 30 min •
49a cross-linked with each other and formed mesh-like structures with interdigitating branches. They were variable in length and size and were located on top of the
stress fibers. Under the view of the inverted microscope, these aggregations were
mostly found near the nuclei of the cells (fig. 3b). They appeared as early as 3 min
after the insulin stimulation (fig. 3c). In addition, the length of the actin
aggregations increased through time. The aggregations were the most pronounced
(fig. 3d) at 10 min of insulin treatment. These actin structures would begin to
condense and appear as dense spots in the photos. Only traces of new actin
structures remained after 20 min (fig. 3e) and 30 min (fig. 3f) of insulin treatment.
3.2.1.2 The concentration effects of insulin on actin cytoskeleton in
permeabilized L6 myotubes
Fig. 3 showed the morphological characteristics of actin cytoskeleton of L6 myotubes
in a concentration gradient. A concentration profile was constructed from OnM (fig.
3g), InM (fig. 3h), lOnM (fig. 3i), lOOnM (fig. 3j), lOOOnM (fig. 3k) and 6000nM (fig.
31) of insulin. Increased insulin concentration caused an increase of the degree of
actin remodeling. An inhibiting effect was observed above lOOOnM of insulin.
The newly formed insulin-induced actin structures appeared to scatter into small
50 Fig.3 Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (concentration). Serum-deprived L6 GLUT4myc myotubes were left untreated (g) or stimulated with InM insulin (h), lOnM (i) lOOnM (j), lOOOnM (k) and 6000nM (1) for 10 min at 37T. The cells were examined in 400X magnification and the white arrows in the images indicated the newly formed actin structure. The images are representatives of at least three experiments. Fig. 3 Effects of insulin on actin cytoskeleton in permeabilized L6 myotubes (concentration)
OnM li^
mm^^^lOnN^^^^ lOOnM • • lOOOnM 6000nM
c: rv _ fragments and distribute unevenly on the surface of the myotubes.
3.2.1.3 Relationship between actin cytoskeleton and GLUT4myc in
permeabilized L6 myotubes
To determine the spatial relationship between actin structures and GLUT4myc after insulin stimulation, permeabilized L6 myotubes were stained for actin and
GLUT4myc. In unstimulated L6 myotubes, actins were appeared as stress fiber networks (fig. 4a) while GLUT4myc were located near the nuclei (fig. 4b). After a
10-minute stimulation of insulin, F-actins aggregated mostly near the nuclei and formed membrane ruffles (fig. 4c). Meanwhile, GLUT4myc translocated and colocalized with these actin structures (fig. 4d). Figures e and f were images merged
from the corresponding photos from basal and insulin-stimulated L6 myotubes stained
with actin and GLUT4myc.
3.2.1.4 Translocation of GLUT4myc in membrane-intact L6 myotubes
To examine the translocation of GLUT4myc after insulin stimulation, cell surface
immunostaining was carried out on L6 myotubes. In the basal state, GLUT4myc
51 Fig.4 Relationship between actin cytoskeleton and GLUT4myc in permeabilized L6 myotubes. Serum-deprived L6 GLUT4myc myotubes were left untreated (a and c) or stimulated (b and d) with lOOnM insulin for 10 min at 37°C，followed by fixation and permeabilization. Cells were stained for actin (a and b) and GLUT4myc (c and d) as described in Ch. 3 Materials and Methods. Figures e and f are images merged from the corresponding photos from basal and insulin-stimulated L6 myotubes stained with actin and GLUT4myc. The cells were examined in 400X magnification. The images are representatives of at least three experiments. Fig. 4 Relationship between actin cytoskeleton and GLUT4myc in permeabilized L6 myotubes Basal Insulin-stimulated
Actin Actin • • • GLUT4 ^UT4 • •Acti n & GLUT4 Actin & GLUT•4
51a was not visible (fig. 5a). After the 10-minute insulin treatment, mesh-liked
GLUT4myc structures were observed on PM (fig. 5b). The newly formed
GLUT4myc structures were similar in morphology to the actin structures,
3.2.1.5 Effect of methyl-p-cyclodextrins, MeOH or EtOH in permeabilized
and membrane-intact L6 myotubes
The addition of methyl-P-cyclodextrins, MeOH or EtOH alone had no effect on the morphology of the myotubes at baseline (fig. 6). The percentage of actin structure in insulin-stimulated control myotubes was 26.9 土 0.8% (table. 2). When preincubated in cyclodextrin, EtOH or MeOH, the respective insulin-stimulated actin percentages were 24.1 土 0.7%, 29.3 土 0.4% or 28.8 ± 0.4% of total myotubes. Thus, the presence of these compounds did not affect actin reorganization and GLUT4myc translocations following insulin treatment.
3.2.2 2-Deoxyglucose Uptake
3.2.2.1 Effects of insulin, methyl-p-cyclodextrins, MeOH and EtOH in L6
myotubes
52 Fig.5 Translocation of GLUT4myc in membrane-intact L6 myotubes. Serum-deprived L6 GLUT4myc myotubes were left untreated (a) or stimulated (b) with lOOnM insulin for 10 min at 37�C . Cells were fixed and stained for GLUT4myc as described in Ch. 3 Materials and Methods. The cells were examined in 400X magm行cation and the white arrows in the images indicated the GLUT4myc. The images are representatives of at least three experiments. Fig. 5 Translocation of GLUT4myc in membrane- intact L6 myotubes
Basal
Insulin-stimulated
•
52a Fig.6 Effect of methyl-p-cyclodextrin, MeOH or EtOH in pernieahili/ecJ L6 myotubes. L6 myotubes were preincubated with methyl-p-cyclodextrin (a and b), MeOH (c and d) or EtOH (e and f) for 24hr. Serum-deprived L6 GHJT4niyc myotubes were left untreated (a, c and e) or stimulated (b，d and f) with lOOnM insulin for 10 min at 37°C, followed by fixation and pernieabilization. Cells were stained for actin as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the newly formed actin structure. The images are representatives of at least three experiments. Fig 6. Effect of methyl- ^S-cyclodextrins，MeOH or EtOH in permeabilized L6 myotubes
Basal Insulin-stimulated • BM MeOH MeOH • • •EtOH • EtOH
52b Fig.6 Effects of methyl-p-cyclodextrin, MeOH or EtOH in membrane-intact L6 myotubes. L6 myotubes were preincubated with methyl-p-cyclodextrin (g)， MeOH (h) or EtOH (i) for 24hr. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37°C, followed by fixation. Cells were stained for GLUT4myc as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the GLUT4myc. The images are representatives of at least three experiments. Fig 6. Effect of methyl- y9-cyclodextrins，MeOH or EtOH in membrane-intact L6 myotubes Insulin-stimulated
cyclodextrins MeOH •• EtOH •
52c Table.2 Effects of preincubations on actin remodeling. L6 myotubes were preincubated for 24hr and actin immunostaining was carried as described in Ch. 3 Materials and Methods. Results shown are combined means 土 SEM from 3 to 5 experiments. p < 0.001 preincubation conditions compared with respective control actin reduction.
i Table 2. Effect of Preincubation on Actin Remodeling
Table 2a 一 Glucose and Insulin preincubation Agent Control Glu/INS n 4 3 No. of grids with actin 58.0 土 1.9 21.0 土 1.3 No. of grids with myotubes 215.0 土 1.2 221.0 土 0.0 % Actin in myotubes 26.9 土 0.8 9.5 土 1.0 % of Actin Reduction - 64.7 * Table 2b • - CHO preincubation Agent I EtOH CHO 50/z M n 5 5 No, of grids with actin 64.4 ±0.8 29.0 土 2.7 No. of grids with myotubes 219.6 土 0.6 208.4 土 1.9 % Actin in myotubes 29.3 土 0.4 13.8 土 1.2 % of Actin Reduction - 53 0 * *，p < 0.001 preincubation conditions compared with respective control actin reduction.
Ln ro a Table 2. Effect of Preincubation on Actin Remodeling
Table 2c - - OA and LA preincubation Agent Cyclodextrin OA 50 // M LA 50 // M n 5 3 5 No, of grids with actin 51.8 ± 1.7 28.7 ± 1.6 26.0 土 0.9 No. of grids with myotubes 214.6 土 1.8 221 ± 0.0 220.2 土 0.4 % Actin in myotubes 24.1 土 0.7 13.0 土 0.7 11.8 土 0.4 % of Actin Reduction - 46.2 * 51.0 *
Table 2d - - PA preincubation Agent MeOH PA 50 /z M — n 5 5 No. of grids with actin 61.8 ±0.6 39.6 ± 1.3 No. of grids with myotubes 214.6 ±1.4 一 220.0 土 0.4 % Actin in myotubes ± 0.4 18.0 ±0.6 % of Actin Reduction - 37 6 * *，p < 0.001 preincubation conditions compared with respective control actin reduction.
Ln NJ CD Insulin-mediated glucose uptake study was performed in unstimulated L6 myotubes and the fold of increase in glucose uptake was served as reference for later uptake comparisons. We also investigated the effect of methyl-P-cyclodextrins, MeOH and
EtOH preincubations on insulin-stimulated glucose uptake. Table 3 showed the effect of insulin on 2-Deoxyglucose uptakes in L6 myotubes. Insulin caused a 189
±2% increase in glucose uptakes relative to the basal values in the control cells.
Moreover, the preincubations of methyl-P-cyclodextrins, MeOH or EtOH increased insulin-stimulated glucose uptake to 180 土5%, 175 ±2% or 197 土9% respectively.
Importantly, the presence of methyl-P-cyclodextrins, MeOH or EtOH did not alter the insulin-mediated glucose uptake (p = 0.238) when compared with basal control uptakes.
3.2.3 TEM study
3.2.3.1 Effects of insulin on actin cytoskeleton and GLUT4myc in L6 myotubes
From the IM results, insulin stimulated the colocalization of actin and GLUT4myc in
L6 myotubes. To examine the relationship between actin and GLUT4myc at an
53 Table.3 Effects of insulin, methyl - P - cyclodextrins, MeOH and EtOH preincubations on glucose uptake. L6 myotubes were preincubated with methyl-p- cyclodextrin, MeOH or EtOH for 24hr. Glucose uptake were carried out as described in Ch. 3 Materials and Methods. All glucose uptakes values were expressed in relative units relative to the basal control conditions. Results shown are combined means 土 SEM from 5 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *，p < 0.001 compared with respective basal uptake values.
[i Table 3. Effects of insulin，methy卜 cyclodextrins, MeOH and EtOH Preincubations on Glucose Uptake
^ Control Cyclodextrin EtOH MeOH Basal 1 00+000 097+002 1 06+005 090+001 iiVv — \J*\J\J vi/ — \Ji\JLi 丄yJtJxJ — \Jt\Ji Glucose Uptake Insulin . 1.89iQ.Q2 1.80i0.05 1.97iQ.Q9 1.75 土 0.02 Ratio 1.89 1.秘 1.86
2.50 r
OJ ^ ^ — — *
Control Cyclo500 MeOH 0.165% EtOH 0.5%
p < 0.001 compared with respective basal uptake values.
c； -3 = ultra-structural level, TEM was carried out. Actin was decorated with 12nm gold particle whereas GLUT4myc was identified with 18nm. In basal L6 myotubes,
actins were mostly found along the stress fibers while GLUT4myc were located near
the nuclei (fig. 7a). Very few GLUT4myc particles were found near the PM (fig. 7b).
Insulin stimulation translocated GLUT4myc to the PM and colocalized them with
actins (fig. 7d). Importantly, a finger-liked protrusion with both actin and
GLUT4myc gold particles from PM was observed. Despite the translocation of
GLUT4myc to the PM, the majority of GLUT4myc remained in the perinuclear
intracellular compartment. Hence, labeling of GLUT4myc near the nuclei remained
abundant following insulin stimulation (fig. 7c).
3.3 Effect of high glucose and high insulin incubation in L6 myotubes
Hyperglycemia and hyperinsulinemia are common in patients with diabetes. To
understanding the effect of hyperglycemia and hyperinsulinemia on L6 myotubes,
cells were preincubated with high concentration of insulin and glucose for 24 hours.
3.3.1 Immunofluorescence Microscopy
54 Fig.7 Distribution of actin cytoskeleton and GLUT4myc at basal state in L6 myotubes (TEM). Serum-deprived L6 GLUT4myc myotubes were left untreated, followed by fixation and permeabilization. Cells were stained for actin (12nm) and GLUT4myc (18nm) as described in Ch. 3 Materials and Methods. The cells were examined in 30000X magnification and images around the nucleus and PM were taken. The t indicated the actin structure 本 I whereas the 丁 represented the GLUT4myc. Fig. 7 Distribution of actin cytoskeleton and GLUT4myc at basal state in L6 myotubes (TEM) Nucleus
,:、”…：厂rf^ . - • I
PM F"— 1 PM I
> .•^气：‘缚�•^^ ^ • I
’...•押.
‘•^‘鸿》憾、；暴、 I
��‘�^^^^^：…：“：麵 々 ^ I Fig.7 Effects of insulin on actin cytoskeleton and GLUT4myc in L6 myotubes (TEM). Serum-deprived L6 GLUT4myc myotubes were stimulated (c and d) with lOOnM insulin for 10 min at 37°C，followed by fixation and permeabilization. Cells were stained for actin (12nm) and GLUT4myc (18nm) as described in Ch. 3 Materials and Methods. Cells were examined in 30000X magnincation and images around the nucleus and PM were taken. The •‘ represented the actin structure whereas the 令 indicated the GLUT4myc. Fig. 7 Distribution of actin cytoskeleton and GLUT4myc after insulin stimulation (TEM) Nucleus
I V �,,,會 t Q • 7 • I ^ 卜 y ,久輪.,、‘‘^M Actin LiP•^�.. > ’.£/•'•�. I GLUT4 ^
I # /ft 眷、—々於
卜;德書“身;"、 I 嘛 ‘ • j
r ^ ^'Imm ‘ 1 ™
秘麵纖J
Villus .J PM I
54b 3.3.1.1 High insulin and high glucose preincubation in permeabilized L6
myotubes
The effect of insulin and glucose preincubation were shown in Fig. 8. At low concentration of glucose and insulin, the growth of cells was normal. However, as
the concentration of insulin increased above 50nM, cells appeared to be narrower and
fusions to myotubes were also reduced. When basal cells preincubated with glucose
and insulin were viewed under IM, the stimulated myotubes were shorter and slender
compared with control myotubes. Aberrant insulin-stimulated actin mesh formation
was observed in cells preincubated with glucose and insulin. At 25mM glucose, the
degree of actin remodeling diminished with increasing insulin concentrations (fig 8a
to 8h). At lOOnM insulin and 25mM glucose, the reduction was the most noticeable
and was significantly reduced to 64.7%, (p < 0.001) when compared with control cells
(table. 2a). The actin aggregations in the stimulated cells were shorten in length and
were broken into small fragments.
3.3.1.2 Effect of high insulin and high glucose incubation in membrane-intact
L6 myotubes
55 Fig.8 High insulin and high glucose preincubation on actin cytoskeleton in permeabilized L6 myotubes. L6 myotubes were left untreated (a) or preincubated with 25mM glucose and OnM (b)，5nM (c)，50nM (d) or lOOnM (e) insulin for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37°C, followed by fixation and permeabilization. Cells were stained for actin as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the actin structures. The images are representatives of at least three experiments. Fig.8 High glucose and high insulin preincubat jon on actin cy toskeleton in permeabttized L6 myotubes
control 25mM gtacose & &nM
25mM glacose & SOiiM
25mM glucose & lOOnM
55a To determine translocation of GLUT4myc after preincubation, L6 myotubes were left untreated (fig. 9a) or incubated with glucose and insulin (fig. 9b and 9c). The morphological changes during cell growth were similar to basal permeabilized cells preincubated with glucose and insulin. When basal cells preincubated with glucose and insulin were viewed under IM, no GLUT4myc was visible. Insulin stimulation caused the translocation of GLUT4myc to the PM. By comparing with control, there were reductions in insulin-induced recruitment of GLUT4myc.
3.3.2 2-Deoxyglucose Uptake
3.3.2.1 Effect of high insulin and high glucose incubation in L6 myotubes
To determine the effect of hyperglycemia and hyperinsulinemia on glucose uptake, L6
myotubes were exposed to high insulin and high glucose for 24 hours. All uptake
values were compared with basal control glucose uptake. Their effects on glucose
uptake were dose-dependent reductions at 25mM glucose and between 25 to lOOnM
insulin (table. 4). Insulin caused in increase in glucose uptake from 100 ±0% to 204
土 3% in control L6 cells. A 24-hour 25mM glucose and 25nM insulin preincubation
increased basal and insulin-stimulated uptake to 140 士 6% and 225 土 17%
56 Fig.9 High glucose and high insulin and CHO preincubations on GLUT4myc translocation in membrane-intact L6 myotubes. L6 myotubes were left untreated (a) or preincubated with 25mM glucose and 5nM (b)，lOOnM (c) insulin or with EtOH and OjLiM (d)，lOjLiM (e), or 50|iM (f) CHO for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37"C, followed by fixation. Cells were stained for GLUT4myc as described in Ch. 3 Materials and Methods. Cells were examined in 400X magnification and the white arrows in the images indicated the GLUT4myc. The images are representatives of at least three experiments. Fig. 9 Effect of Glucose and Insulin，CHO preincubations in insulin-stimulated membrane- intact L6 myotubes ••Control G25mM I5nM •G25m M IlOOnM ••EtOH 0.5% CHO lO^IM •CH O 50|liM Fig.9 FFA preincubation on GLUT4myc translocation in membrane-intact L6 myotubes. L6 myotubes were preincubated with cyclodextrin and OfxM (g), lOnM (h)，50nM � OA or OfxM (j), lOnM (k)，50nM � LA or with MeOH and OPM (m)，10\xM (n), or 50|liM (o) PA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37'C, followed by fixation. Cells were stained for GLUT4myc as described in Ch. 3 Materials and Methods. Cells were examined in 400X magnification and the white arrows in the images indicated the GLUT4myc. The images are representatives of at least three experiments. Fig. 9 Effect of OA, LA and PA preincubations in insulin-stimulated membrane-intact L6 myotubes
Cyclodextrin 500 _ OA 10^lM OA 50|LIM
Cyclodextrin 500 uM LA lOuM LA 50uM
___
MeOHOJ65% ^P^OuJ^^ P^OuM^^
56b Table.4 Effect of Glucose and Insulin preincubation on glucose uptake. L6 myotubes were preincubated with glucose and insulin for 24 hours. Serum- deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 30 min at 37°C and were rinsed with HEPES-buffered saline solution. Glucose uptake was determined as described in Ch. 3 Materials and Methods. All glucose uptake values were expressed in relative units relative to the basal carrier-control conditions. Results shown are combined means ± SEM from 3 to 9 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *, p < 0.005; **，p < 0.0001 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in glucose and insulin compared with control. Table 4. Effect of Glucose and Insulin Preincubation on Glucose Uptake
n Basal Insulin Ratio % Reduction — INS/Ba^ Control 9.00 1.00 ±0.00 2.04±0"!i 2.04 - G25Id25 4.00 1.40 ±0.06 2.25 ±0.17 1.61 41.83 一 G25In50 5.00 2.08 ±0.l5" 1.55 46.89 ~ G25In75 3.00 T^±0.02 2.37 ±0.^ 1.42 59.52 — G25Inl00 4.00 11.74 ±0.0912.21 ±0.13 1.27 • 74.40
^ 3.50 #-
^ CD I
口 c 3.00 一 ~ —-—— s I I ^ 2.50 ——^ ^ ^ **一―
miMmControl G25In25 G25In50 G25In75 G25Inl00 *，p < 0.005; **，p < 0.0001 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in glucose and insulin compared with control. respectively. The overall glucose uptake after preincubations was 1.61-fold compared with 2.04-fold in control cells. Preincubation with high insulin and glucose mostly affects the basal glucose uptakes. The basal uptakes at 25mM glucose and 50nM, 75nM or lOOnM insulin were 134 士 5%，167 土 2% or 174 士 9%.
Meanwhile, the respective insulin-stimulated uptakes were 208 士 10%，237 土 9% or
221 土 13%. Consequently, insulin caused a 1.27-fold increase of glucose uptake in cells in high glucose and high insulin relative to the basal value of the stimulated cells compared with 2.04-fold in control cells. Hence, the high glucose (25mM) and high insulin (lOOnM) preincubation diminished insulin-mediated glucose uptake by 74.4 土
10.9%, (p = 0.00001).
3.3.3 TEM study
3.3.3.1 Effect of high insulin and high glucose incubation in L6 myotubes
L6 myotubes were preincubated with insulin (lOOnM) and glucose (25mM) for 24 hours and stimulated with insulin (lOOnM, 10 min). By comparing with controls,
there were reductions in insulin-induced colocalization of GLUT4myc with actins (fig.
10b).
57 Fig.lO High glucose and high insulin preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes (TEM). L6 myotubes were left untreated (a) or preincubated with 25mM glucose and lOOnM (b) or with EtOH for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37°C，followed by fixation and permeabilization. Cells were stained for actin (12nm immunogold) and GLUT4myc (18nm immunogold) as described in Ch. 3 Materials and Methods. Images were examined in 30000X magnification. The “ indicated the actin structure whereas the 牛 represented the GLUT4myc. Fig.10 Effect of preincubation on actin and GLUT4myc translocation in insulin-stimualted L6 myotubes (TEM) Control
I — I 25mM glucose and lOOnM l b Stress Fiber ‘ •
57a Fig.lO CHO and FFA preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes. L6 myotubes were preincubated with EtOH and SOfiM CHO (c) or with cyclodextrin and SOfiM OA (d) for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37°C, followed by fixation and permeabilization. Cells were stained for actin (12nm immunogold) and GLUT4myc (18nm immunogold) as described in Ch. 3 Materials and Methods. Images were examined in 30000X magnMcation. The “ indicated the actin structure whereas the 令 represented the GLUT4myc. Fig.10 Effect of preincubation on actin and GLUT4myc translocation in insulin-stimualted L6 myotubes (TEM) 50^M CHO
赞： ./f 难丨1
彻、.资 jl SO^iM OA 叫广，�.— —J �-•.. d \ ‘ • 二/ •、、二 於 / ^ 。
- Stress Fiber
57b Fig.lO FFA preincubation on actin cytoskeleton and GLUT4 translocation in insulin-stimualted L6 myotubes. L6 myotubes were preincubated with cyclodextrin and 50[iM LA (e) or with MeOH and 50[iM PA (f) for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37°C, followed by tixation and permeabilization. Cells were stained for actin (12nm immunogold) and GLUT4myc (18nm immunogold) as described in Ch. 3 Materials and Methods. Images were examined in 30000X magnification. The “ indicated the actin structure whereas the • represented the GLUT4myc. Fig.10 Effect of preincubation on actin and GLUT4myc translocation in insulin-stimualted L6 myotubes (TEM)
50|LIM LA
50^M PA
57c ibW•恭：華r L .^” ^ ‘》:a^i ‘‘ y- ‘：〜 l V jA我 i…]、 � ^• 气I I 3.4 Effect of FFA incubation in L6 myotubes
Increased plasma FFA concentrations are associated with many insulin resistant conditions including obesity and Type II DM. Thus, we incubated FFA in L6 myotubes and examined its effect on actin remodeling and glucose uptake.
3.4.1 Immunofluorescence Microscopy
3.4.1.1 FFA incubation in permeabilized L6 myotubes
To determine the relationship between actin and GLUT4myc after preincuabtion, L6 myotubes were left untreated, or preincubated with OA (fig. 11), LA (fig. 12) or PA
(fig. 13) (up to 75|iM) for 24 hours. There were dose-dependent reductions in insulin-stimulated actin reorganization when compared with their respective controls.
The cell growth in low FFA was normal. However, as the FFA concentration increased above 20|iM for OA and LA and above 40|iM for PA, cells appeared to be narrower and fusions to myotubes were also reduced. When basal cells preincubated with FFA were viewed under IM, the myotubes were shorter and slender compared
58 Fig. 11 OA preincubation on actin cytoskeleton in permeabilized L6 myotubes. L6 myotubes were preincubated with cyclodextrin and OpAI (a), 20|iM (b), 30^M (c), 40^M (d) or SOjiM (e) OA for 24 hours. Serum-deprived 1.6 GIA T4niyc myotubes were stimulated with lOOn.M iasulin for 10 min at 37*C\ followed by fixation and permeabilization. Cells were stained for actin a^s described in Ch. 3 Materials and Methods. The cells wen? examined in 4(M)\ magnificatiun and the white arrows in the images indicated the actin structures. I he iunices an* representatives* of at least thrci* expcriineiiLs. Fig.11 OA preincubation on actin cytoskeleton in permeabilized L6 myotubes
O^M OA 20^M OA MM IBB
HBI30\iM OA 4(HiHM OA • • •SO^i M OA
58a Fig.l2 LA preincubation on actin cytoskeleton in permeabilized L6 myotubes. L6 myotubes were preincubated with cyclodextrin and O^M (a), 20|iM (b)，30\xM (c), 40jxM (d)，or SO^iM (e) LA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37�C，followe dby fixation and permeabilization. Cells were stained for actin as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the actin structures. The images are representatives of at least three experiments. Fig.l2 LA prelnciibation on actin cytoskeleton in permeabifized LB myotitbes
OjiM LA 2(HiMLA
i^SlH
30nMLA 40jiMLA • • SO^iM LA
58b Fig.l3 PA preincubation on actin cytoskeleton in permeabilized L6 myotubes. L6 myotubes were preincubated with MeOH and 0|liM (a), 20[iM (b)，50[iM (c)， 60mM (d)，or 75fiM (e) PA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37�C，followe dby fixation and permeabilization. Cells were stained for actin as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the actin structures. The images are representatives of at least three experiments. Fig.13 PA preincubation on actin cytoskeleton in permeabilized L6 myotubes
O^iMPA 20^M PA
mmSO^iM PA _M PA •SBlll• B
75^M PA •
5 8n with control myotubes. Aberrant insulin-stimulated actin mesh formation was
observed in cells preincubated with FFA. These actin aggregations in the stimulated
cells were shorten in length and were broken into small fragments. At the high
levels of OA, LA and PA preincubations (SOjiM), actin remodeling was significantly
diminished by 46.2%, (p < 0.005), 51.0%, (p < 0.0005), and 37.6%, (p < 0.0001)，
respectively (table. 2c and 2d).
3.4.1.2 FFA incubation in membrane-intact L6 myotubes
To determine translocation of GLUT4myc after preincubation, L6 myotubes were
incubated with cyclodextrin and OA, LA or MeOH and PA for 24 hours (fig. 9). The
morphological changes during cell growth were similar to basal permeabilized cells
preincubated with FFA. When basal cells preincubated in FFA were viewed under
IM, no GLUT4myc was visible. Insulin stimulation caused the translocation of
GLUT4myc to the PM. By comparing with their respective controls, there were
reductions in insulin-induced recruitment of GLUT4myc.
3.4.2 2-Deoxyglucose Uptake
59 3.4.2.1 FFA incubation in L6 myotubes (24 hours)
To determine the effect of preincubation on glucose uptake, L6 myotubes were exposed to FFA for 24 hours. The reductions in insulin-mediated glucose uptake were dose-dependent at concentrations between 20 to 75^iM of FFA. All uptake values were compared with basal control glucose uptake.
The responses of unsaturated fatty acids - OA preincubation was shown on table 5.
Insulin stimulated glucose uptake from 100 ± 0% to 185 士 1% in methyl-P-cyclodextrins-control L6 myotubes. The basal uptakes at 20|^M, 30|iM,
40|iM or 50jiM OA were 140 土 14%, 150 土 11%, 143 土 7% or 168 土 4%.
Meanwhile, the respective insulin-stimulated uptakes were 261 土 23%, 259 士 19%,
224 土 9% or 228 土 5%. A 24 hours OA preincubation (50|iM) increased basal and insulin-stimulated uptake to 168 土 4% and 228 土 5% respectively. High levels of
OA (50|iM) for 24 hours diminished glucose uptake by 45.8 土 8.6%，(p = 0.0005).
Insulin caused 1.36-fold increase in glucose uptakes relative to the basal values of the stimulated cells compared with 1.85-fold in methyl-P-cyclodextrins-control.
The responses of unsaturated fatty acids - LA preincubation was shown on table 6.
60 Table.5 Effect of OA preincubation on glucose uptake. L6 myotubes were preincubated with cyclodextrin and OA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 30 min at 37°C and were rinsed with HEPES-buffered saline solution. Glucose uptake was immediately determined as described in Ch. 3 Materials and Methods. All glucose uptake values were expressed in relative units relative to the basal cyclodextrin-control conditions. Results shown are combined means 土 SEM from 5 to 13 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *，p < 0.01; **, p < 0.001 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in OA compared with cyclodextrin-control. Table 8. Effect ofCHO Preincubatio n on Glucose Uptake
n Basal Insulin Ratio % Reduction
- — INS/B^ CycloSOO 13.00 1.00 ±0.00 T^5±0.01 1.85 - OA20 5.00 1.40 ±0.14 2.61 ±0.23 1.87 2.59 _ OA30 6.00 1.50 ±0.11 2.59 ±0.19 1.72 14.88 OA40 6.00 一1.43±0.07 了24±0.09 1.57 33.14 OA50 9.00 I 1.68 ±0.04 12.36 ±0.05 1.41 51.83
3.50 1 TT # ^ 3.00 * M
*，p < 0.01; **，p < 0.001 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in OA compared with cyclodextrin-control.
5a Table.6 Effect of LA preincubation on glucose uptake. L6 myotubes were preincubated with cyclodextrin and LA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 30 min at 37�C and were rinsed with HEPES-buffered saline solution. Glucose uptake was determined as described in Ch. 3 Materials and Methods. All glucose uptake values were expressed in relative units relative to the basal cyclodextrin-control conditions. Results shown are combined means ± SEM from 4 to 9 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *，p < 0.01; **, p < 0.0001 insulin-treated compared with respective basal uptake values. #，p < 0.001 folds of increase of glucose uptake preincubated in LA compared with cyclodextrin-control. Table 6. Effect of LA Preincubation on Glucose Uptake
n Basal Insulin Ratio % Reduction INS/Basal CycloSOO 9.00 1.00 ±0.00 1.87 ±0.03 1.87 “ - LA20 — 6.00 1.18 ±0.04 2.25 ±0^ 1.90 3.16 LA30 5.00 1.54 ±0.11 2.26 ±0.10 1.47 — 45.84 _ LA40 - 4.00 172 ±0.16 2.00 1.41 53.54 一 LA50 8.00 I 1.64 ±0.04 I 1.39 57.09
3.50 n —. 二它 f-H 3»00 ~ ------._._.— _—._ — — — - — — --.-. — — —•— Q 2 50 ** -*—* _ ••
miMwCyclo500 LA20 LA30 LA40 LA50
*, p < 0.01; **，p < 0.0001 insulin-treated compared with respective basal uptake values. #, p < 0.001 folds of increase of glucose uptake preincubated in LA compared with cyclodextrin-control.
60b Insulin stimulated glucose uptake from 100 ±0% to 187 土 3% in methyl-p-cyclodextrins-control L6 myotubes. The basal uptakes at 20|aM, 30|iM,
40jiM or 50|iM LA were 118 土 4%，154 土 11%，142 土 16% or 164 土 4%.
Meanwhile, the respective insulin-stimulated uptakes were 214 士 6%，226 ± 10%, 200
土 20% or 228 ±5%. A 24 hours LA preincubation (50|aM) increased basal and
insulin-stimulated uptake to 164 土 4% and 228 士 5% respectively. High levels of LA
(50|iM) for 24 hours diminished glucose uptake by 57.1 土 8.5%, (p = 0.004).
Insulin caused 1.39-fold increases in glucose uptakes relative to the basal values of
the stimulated cells compared with 1.87-fold in methyl-P-cyclodextrins-control.
The responses of saturated fatty acid - PA preincubation was shown on table 7.
Insulin stimulated glucose uptake from 100 土 0% to 196 土 1% in MeOH-control L6
myotubes. The basal uptakes at 20|iM, 40jaM, 50|iM, 60j^M or 75|iM PA were 98 土
4%, 100 土 2%, 124 土 2%，136 土 3% or 148 士 12%. Meanwhile, the respective
insulin-stimulated uptakes were 196 土 7%, 193 土 5，206 土 3%，221 土 7% or 242 土
24%. A 24 hours PA preincubation (50|iM) increased basal and insulin-stimulated
uptake to 124 土 2% and 206 士 3% respectively. Exposure of 50|iM PA caused a 33.3
土 8.8%，(p = 0.0008) reduction in glucose uptake. Insulin stimulated glucose uptake
was 1.63-fold higher than the basal values of the stimulated cells compared with
61 Table.7 Effect of PA preincubation on glucose uptake. L6 myotubes were preincubated with MeOH and PA for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 30 min at 37°C and were rinsed with HEPES-buffered saline solution. Glucose uptake was determined as described in Ch. 3 Materials and Methods. All glucose uptake values were expressed in relative units relative to the basal MeOH-control conditions. Results shown are combined means ± SEM from 3 to 13 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *, p < 0.05; **, p < 0,001 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in PA compared with MeOH-control.
/ Table 8. Effect ofCHO Preincubatio n on Glucose Uptake
n Basal Insulin Ratio % Reduction INS/Basal MeOH0.16^ 13.0 “ 1.00±0.00 _ 1.96±0.01 1.96 — - PA 20 5.0 0.98 ±0.04 1.96 ±0.07 — 2.01 4.65 PA 40 6.0 0.97 ±0.02 1.87 ±0.04 1.94 _ 2.74 PA 50 10.0 - 1.24 ±0.02 - 2.06 ±0.03 1.67 33.30 PA 60 5.0 “ 1.31 ±0.03 -2.21 ±0.07 1.63 — 34.44 PA 75 3.0 1.36 ±0.06 2.42 ±0.24 1.65 32.19
O M s S3.50 ——— . o f 3.00 - ^― 2 50 — — “— — ~———— _|誦:_ 一 MeOH 0.165% PA 20 PA 40 PA 50 PA 60 PA 75
*, p < 0.05; **，p < 0.001 insulin-treated compared with respective basal uptake values.
#，p < 0.0001 folds of increase of glucose uptake preincubated in PA compared with MeOH-control.
61a 1.96-fold in MeOH-control.
3.4.3 TEM study
3.4.3.1 FFA incubation in L6 myotubes
L6 myotubes were preincubated with cyclodextrins and SO^M OA (fig. lOd), 50jiM
LA (fig. lOe) or with MeOH and 50|LIM PA (fig, lOf) for 24 hours. By comparing with controls, there were reductions in insulin-induced colocalization of GLUT4myc with actins.
3.5 Effect of CHO incubation in L6 myotubes
Elevated serum total CHO level is commonly found in patients with diabetes. Thus,
we preincubated CHO in L6 myotubes and examined its effect on actin remodeling
and glucose uptake.
3.5.1 Immunofluorescence Microscopy
62 3.5.1.1 CHO incubation in permeabilized L6 myotubes
To determine the relationship between actin and GLUT4myc after preincuabtion, L6 myotubes were left untreated, or preincubated with CHO (up to 50|liM) for 24 hours
(fig. 14). There were dose-dependent reductions in insulin-stimulated actin reorganization when compared with EtOH controls. The cell growth in low CHO was normal. However, as the CHO concentration increased above 20|iM, cells appeared to be narrower and fusions to myotubes were also reduced. When basal cells preincubated with CHO were viewed under IM, the myotubes were shorter and slender compared with control myotubes. Aberrant insulin-stimulated actin mesh formation was observed in cells preincubated in CHO. These actin aggregations in the stimulated cells were shorten in length and were broken into small fragments.
High levels of CHO preincubation (50)iM) caused a 53.0%, (p < 0.0005) reduction in
actin remodeling (table. 2b).
3.5.1.2 CHO incubation in membrane-intact L6 myotubes
To determine translocation of GLUT4myc after preincubation, L6 myotubes were left
untreated, or treated with EtOH and CHO at 0)iM (fig. 9d), \0[iM (fig. 9e), or 50|iM
63 Fig.l4 CHO preincubation on actin cytoskeleton in permeabilized L6 myotubes. L6 myotubes were preincubated with EtOH and OjnM (a), lO^iM (b)， 2Q\m (c)，30^M (d)，40\M (e)，or 50|jM (f) CHO for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 10 min at 37�C， followed by fixation and permeabilization. Cells were stained for actin as described in Ch. 3 Materials and Methods. The cells were examined in 400X magnification and the white arrows in the images indicated the actin structures. The images are representatives of at least three experiments. Fig.14 CHO preincubation on actin cytoskeleton in permeabilized L6 myotubes
O^M CHO lO^M CHO • •
20nM CHO 30^M CHO
40^M CHO 50nM CHO
mm 63a (fig. 9f) for 24 hours. The morphological changes during cell growth were similar to basal permeabilized cells preincubated with CHO. When basal cells preincubated with CHO were viewed under IM, no GLUT4myc was visible. Insulin stimulation caused the translocation of GLUT4myc to the PM. By comparing with their respective controls, there were reductions in insulin-induced recruitment of
GLUT4myc.
3.5.2 2-Deoxyglucose Uptake
3.5.2.1 CHO incubation in L6 myotubes (24 hours)
To determine the effect of preincubation on glucose uptake, L6 myotubes were
exposed to CHO for 24 hours. All uptake values were relative to basal EtOH-control
glucose uptake. Chronic exposure of L6 myotubes to CHO reduced the insulin
stimulated glucose uptake significantly. The effects of insulin on glucose uptake
were dose-dependent at concentrations between 10 to 60|iM of CHO (table 8).
Insulin stimulated glucose uptake from 100 土 0% to 193 土 1%. The basal uptakes
at 20|iM, 30|^M, 40}iM, 50|iM or 60|iM CHO were 70 土 0%, 82 士 6%, 143 土 5%, 141
± 3% or 200 ±25%. Meanwhile, the respective insulin-stimulated uptakes were 127
64 Table.8 Effect of CHO preincubation on glucose uptake. L6 myotubes were preincubated with EtOH and CHO for 24 hours. Serum-deprived L6 GLUT4myc myotubes were stimulated with lOOnM insulin for 30 min at 37"C and were rinsed with HEPES-buffered saline solution. Glucose uptake was determined as described in Ch. 3 Materials and Methods. All glucose uptake values were expressed in relative units relative to the basal EtOH-control conditions. Results shown are combined means ± SEM from 3 to 7 experiments. A graphic presentation of glucose uptake was shown on the bottom of the table. *，p < 0.01; **，p < 0.005 insulin- treated compared with respective basal uptake values. #, p < 0.0001 folds of increase of glucose uptake preincubated in CHO compared with EtOH-control. Table 8. Effect of CHO Preincubation on Glucose Uptake
n Basal Insulin Ratio % Reduction INS/Basal ~ EtOH 0.5% 7.00 1.00 ±0.00 —1.93 ±0.01 1.93 — - CH020 一 3.00 0.70 ±0.00 _1.27±0.00 — 1.82 11.53 CH030_ 3.00 0.82 ±0.06 1.47 ±0.10 — 1.80 — 14.30 CH040 4.00 1.43 ±0.05 2.04 ±0.05 — 1.43 53.69 CH050_ 5.00 l41±0.03 _1.97±0.04 1.40 ~ 56.70 CH060 3.00 2.00 ±0.25 2.72 ±0.40 1.36 61.72
3.50 ^ # 3.00 - T D 8 2 I 2.50 - •
mMMWEtOH 0.5% CH020 CH030 CH040 CH050 CH060 *, p < 0.01; **，p < 0.005 insulin-treated compared with respective basal uptake values. #，p < 0.0001 folds of increase of glucose uptake preincubated in CHO compared with EtOH-control.
64a 土 0%，147 土 10%，204 土 5%，197 士 4% or 272 土 40%. A 24 hours CHO preincubation (SOjiM) increased basal and insulin-stimulated uptake to 141 ±3% and
197 士 4% respectively. Preincubation with high CHO (50jiM) reduced the glucose uptake by 56.7 ±8.5%, (p = 0.0005). Insulin caused a 1.36-fold increase in glucose uptake relative to the basal value of the stimulated cells compared with 1.93-fold in
EtOH-control.
3.5.3 TEM study
3.5.3.1 CHO incubation in L6 myotubes
L6 myotubes were incubated with CHO (50|iM) for 24 hours and stimulated with
insulin (lOOnM, 10 min). By comparing with controls, there were reductions in
insulin-induced colocalization of GLUT4myc with actins (fig. 10c)
3.6 Overall changes in glucose uptake after preincubation experiment
From the experiments of preincubating L6 myotubes in conditions stimulating insulin
resistance, a common characteristic of increasing basal glucose uptake was observed
65 (table 9). Basal uptake values from stimulated cells were compared with values from control or carrier-control cells. In the high glucose and high insulin study,
25mM glucose and lOOnM insulin preincubation increased the basal glucose uptake by 73.9 士 3.0% when compared with control cells. Besides, the basal glucose uptakes for OA, LA and PA (50jiM) compared with their respective basal carrier-control values increased by 64.4 土 5.6%, 62.1 土 2.0% and 28.1 土 2.3% respectively. In addition, high CHO (50|aM) preincubation caused a 40.7 土 3.2% increase in basal glucose uptake.
Following incubations with various conditions mimicking insulin resistance, the
insulin-stimulated glucose uptake was also increase, but to a lesser extent as
compared to the increase in basal glucose uptake (table 9). Insulin-stimulated uptake
values from stimulated cells were compared with values from control or
carrier-control cells. In the 25mM glucose and lOOnM insulin experiment, the
insulin-stimulated glucose uptake increased by 5.82 土 2.2% when compared with
control L6 myotubes. Besides, the insulin-stimulated glucose uptakes for OA, LA
and PA (50}iM) compared with insulin-stimulated carrier-control values increased
29.7 土 3.0%, 19.3 土 3.1% and 7.47 土 2.5% respectively. In addition, high CHO
(50|iM) preincubation increased 1.87 土 1.5% in insulin-stimulated glucose uptake.
66 Table.9 Percentage of increase on glucose uptakes compared with carrier- control cells before and after insulin treatment. Cells were preincubated with glucose and insulin, FFA or CHO and glucose uptakes were determined as described in Ch. 3 Materials and Methods. Results shown are combined means ± SEM. All values were determined by comparing the basal or insulin-stimulated uptake of the preincubated cells with their respective basal or insulin-stimulated uptake of the carrier-control cells. Table 9. Effect of Preincubation on Basal and Insulin- Stimulated Glucose Uptake
,• % of increase in basal glucose % of increase in insulin-stimulated Agents uptake over control cells glucose uptake over control cells Glucose 25mM，Ins—lOOnM 73.9 + 3.0% 5.82 + 2.2% OA 50//M — 64.4 + 5.6% 29.7 + 3.0% FFA LA50//M — 62.1 + 2.0% 19.3 + 3.1% — 一 PA50//M 28.1 + 2.3% — 7.47 + 15% CHO 50//M 40.7 士 3.2% 1.87 + 1.5% —
J Consequently, the folds of increase in insulin-mediated glucose uptake were significantly reduced following preincubation with various FFA, CHO or high insulin and glucose conditions.
67 CHAPTER FOUR DISCUSSION
The purpose of our study is to elucidate the effect of insulin-stimulated actin reorganization on insulin-mediated GLUT4 translocation and glucose uptake.
Previous studies have shown the importance of actin remodeling on these two issues.
Disrupting of actin ruffles induces a form of insulin resistance to L6 myotubes and the same mechanism may also apply to the state of insulin resistance in general. Thus, we preincubated cells in conditions associated with insulin resistance and examined their effects on insulin-mediated actin reorganization and glucose uptake. We hypothesize that the defect of glucose uptake in the state of insulin resistance is a combined effects of abnormal GLUT expression and activities, aberrant insulin signal propagation and interference with actin accessory proteins. Preincubating cells in the conditions of insulin resistance may affect the normal signal propagation and avoid the initiation of actin reorganization. There are increasing evidences showing that a correlative relationship between actin reorganization and GLUT4 translocation exists. The insulin-induced GLUT4 translocation and activation may be depended on an intact actin network and insulin-stimulated actin reorganization. The newly formed actin network may allow correct sequential activations in insulin signaling pathway by distributing the signaling molecules to proper locations. In addition,
68 actin reorganization and membrane ruffling formation may also directly participate in
the translocation of GLUT4. Nowadays, despite of numerous researches, the
pathology of insulin resistance is still unclear. Since actin disassembly abolishes
GLUT4 translocation and creates a form of resistance to insulin, aberrant GLUT4
translocation caused by actin disruption affects glucose homeostasis and may
contribute to insulin resistance in general.
4.1 Effect of insulin on L6 myotubes
The ability of insulin to stimulate glucose uptake in L6 myotubes has been
demonstrated for many years. In 2-Deoxyglucose uptake experiments, insulin
causes a 2.00-fold increase in glucose uptakes relative to the basal values of the
control cells {Klip, Logan, et al. 1982}. This is similar to the insulin-stimulated
glucose uptake value of 1.89-fold reported from our study. In addition, stimulation
of glucose uptake by insulin is observed within a few minutes and remains constant
for at least an hour {Klip, Li, et al. 1984}. We have demonstrated a similar effect by
incubating cells in 5 and 10 min of 2-Deoxyglucose. Both treatments yield similar
ratios between insulin-stimulated and basal glucose uptakes.
69 In addition to act as an anabolic hormone with powerful metabolic and mitogenic effects, insulin also induces actin reorganization that leads to the formation of membrane ruffles {Tsakiridis, Vranic, et al. 1994}. We have examined the time
(fig. 2) and concentration (fig. 3) course of insulin addition to L6 myotubes. Under the basal state, actin filaments were expressed as long actin microfilament bundles of variable length and thickness that could be detected throughout the cell. Insulin stimulation elicited a dramatic reorganization of actin filaments in serum-deprived L6 myotubes. The length of the actin aggregations increased through time and was the most pronounced at 10 min of insulin treatment. On the other hand, a concentration profile was constructed from 0 to 6000nM of insulin. Results showed that there was a positive correlation between concentration of insulin and degree of actin reorganization. Noticed the increased insulin concentration caused an increase in the magnitude of actin remodeling. However, an inhibiting effect was observed above lOOOnM of insulin. The newly formed insulin-induced actin structure appeared to scatter into small fragments and distribute unevenly on the surface of the myotubes.
Nonetheless, the addition of lOOnM insulin produced a good effect on actin remodeling, and is same concentration used in other studies used {Khayat, Tong, et al.
2000}. Thus, we have chosen the optimal time and concentration of insulin usage for our study of actin reorganization to be 10 min and lOOnM.
70 Upon the stimulation of insulin, GLUT4 located in the intracellular compartment translocate to the PM and promote glucose uptake. Several studies have shown the importance of actin remodeling in the translocation of GLUT4. First of all, CD, a widely used inhibitor of actin filament formation, inhibits actin reorganization and glucose uptake {Tsakiridis, Vranic, et al. 1994}. Another study from Tong has revealed the colocalization of GLUT4 with actin structures from the ventral to the dorsal surface after insulin stimulation. Besides, Jasplakinolide, drug that affects actin filament stability, reduces GLUT4 appearance at the cell surface after insulin treatment {Tong, Khayat, et al. 2001}. Indeed, our findings are consistent with their observations (fig. 4). The presence of an exofacial c-myc epitope tag on GLUT4 allows us to track the cellular GLUT4myc in permeabilized and membrane-intact cells. Under the basal state of permeabilized L6 myotubes, actin bundles were aligned longitudinally along the cells. Meanwhile, GLUT4myc appeared in clusters were mostly concentrated around the myonuclei. Insulin stimulation caused actin filament reorganization throughout the cells. GLUT4myc were also found to begin localizing with the actin structure. GLUT4myc have been reported to become more condensed and colocalized in the ruffles after 30 min of insulin treatment {Khayat,
Tong, et al. 2000}. Interestingly, only a small portion of GLUT4myc translocated
71 with the actin structures where there was still a large portion of GLUT4myc remained in the intracellular compartments. Previous study from Shepherd et al supported our observation {Shepherd & Kahn 1999}. They suggested that there are about 10% of
GLUT4 located on the PM at the basal state. Insulin mobilizes GLUT4 to attain a new equilibrium state that about 30% is now inserted on PM. Cell surface immunostainings were carried out in membrane-intact L6 myotubes (fig. 5). Under the basal state, no GLUT4myc was visible. After the 10-minute insulin treatment,
GLUT4myc were observed on PM.
In conclusion, through the results of our IM study, insulin stimulates the reorganization of actin filaments into actin aggregations. These actin networks may facilitate the insertion of GLUT4 to the PM and thereby induce glucose uptake. The actin structures may tether the necessary signaling molecules in proper positions for effective phosphorylation and signal transduction down the pathway.
In order to obtain a more accurate idea of insulin-mediated glucose uptake in relation to actin and GLUT4, we performed TEM using immunogold labeling of actin and
GLUT4myc. In our knowledge we are the first group to investigate the translocation of GLUT4 with actin using the technique of TEM. Indeed, we have achieved direct
72 evidences of actin remodeling and GLUT4 colocalization (fig. 7). In basal condition of L6 myotubes, actins were mostly found along the stress fibers whereas GLUT4myc were located near the nuclei. It was rare for GLUT4myc to locate near the PM or colocalize with actin particles. The 10-minute insulin stimulation revealed the colocalization of GLUT4myc with actin structures. In some events we were able to directly observe the structures of actin remodeling as a finger-liked protrusion from
PM. Noticed both actin and GLUT4myc gold particles were found in the protrusion.
Most importantly, these structures can only be found in the insulin-stimulated L6 myotubes.
Our TEM procedure is a novel technique for sample grown from cell culture. We have to modify the standard procedure to suit the nature of L6 myotubes and our study purpose. The problem of TEM is that the final sample condition is quite unpredictable. After insulin stimulation of L6 myotubes, samples have undergone numerous tentative procedures including cell fixation, dehydration, scrapping, slicing and staining before they are ready to be observed under the microscope. Hence, there are a lot of possible sites for discrepancy to occur. The most probably site will be the slicing process. Since the sample is sliced, cellular structures are unpreventably damaged or even destroyed. This may seriously affect the illustration
73 of fragile structures such as the insulin-induced actin protrusion. In addition, cells are scrapped and centrifuged before they are embedded for polymerization. Thus, it is highly unlikely that the cell surfaces in the embedded block are facing the same
direction, which is found in the case of IM. However, TEM cell samples may be
facing all possible directions when they are sliced. As a result, we have actually no
control on how cells are located on the ultrathin section. The unpredictable slicing
may seriously affect their cellular outlooks in TEM photos.
Another problem we encounter with TEM is the low antibody staining. Unlike in
the case of IM where samples are seeded on the surface of a cover slide and have
equal assess to antibody staining, TEM sections are mounted on formver-coated
nickel grids. The formver mounting may limit the antibody contact with the cells.
Besides, tentative TEM procedure may also affect the conditions of antigen sites.
There are several possible explanations for the loss of staining: (1) Cross-linking
fixation by PFA and glutaraldehyde may mask or even destroy the necessary antigen
sites for antibody binding; (2) Resin polymerization at high temperature or the use of
propylene oxide may damage the antigen sites; (3) Secondary immunogold binding to
antigen sites may be blocked by primary immunogold particles during sequential
double staining. Hence, we carried out numerous experiments to optimize the
74 labeling procedure. With all these conditions, the interpretation of TEM imaging should be cautious. The images presented in this thesis represent the typical findings of at least 3 to 5 experiments. This procedure was adopted to ensure consistency in the spotting of the TEM results.
4.2 Effect of methyl-p-cyclodextrins, MeOH and EtOH on L6 myotubes
Due to the low solubilities in aqueous solutions, lipids and CHO are either predissolved in organic solvents or accompanied with carrier molecules to facilitate their dissolutions. Both OA and LA purchased were accompanied with methyl-P-cyclodextrins whereas PA and CHO were predissolved in MeOH and EtOH respectively. The effect of methyl-P-cyclodextrins, MeOH or EtOH alone on actin and GLUT4 was negligible for permeabilized or membrane-intact L6 myotubes (table
2 and fig. 6). Besides, the addition of methyl-p-cyclodextrins, MeOH or EtOH alone
with insulin stimulation caused an insignificant reduction in glucose uptake (table 3).
Thus, the incubation of methyl-P-cyclodextrins, MeOH or EtOH had no effect on
actin reorganization and glucose uptake in L6 myotubes. For accurate comparison,
all glucose uptakes in control and preincubation experiments are expressed in relative
units relative to the basal control or their respective basal carrier-containing glucose
75 uptake values.
4.3 Effect of pretreatment of cells in conditions of insulin resistance
Insulin resistance, the ineffectiveness of insulin to stimulate glucose uptake, is a major characteristics of Type II DM and obesity. Although insulin resistance is well characterized, the exact mechanism remains unclear. Patients with diabetes normally have hyperglycemia and hyperinsulinemia. Increased plasma FFA and total serum cholesterol levels are also common in diabetics. As a result, we preincubated cells in the conditions similar to those found in insulin resistance and studied their effects on actin reorganization and glucose uptake.
4.3.1 Effect of high glucose and high insulin preincubation on L6 myotubes
To understanding the effect of hyperglycemia and hyperinsulinemia on glucose uptake, cells were incubated with insulin and glucose for 24 hours. Dose-dependent reductions of actin remodeling and glucose uptake were observed. Results from our study showed that preincubation with high glucose (25mM) and high insulin (lOOnM) levels significantly reduced glucose uptake (table 4). Our results are consistent with
76 previous observations. A 10-hour preincubation with high insulin alone has been reported to increase basal and decrease acute insulin-stimulated glucose uptake in
3T3-L1 adipocytes {Ricort, Tanti, et al. 1995}. Besides, 3T3-L1 adipocytes treated with insulin for 18 hours demonstrated elevated basal but minimal insulin-stimulated glucose uptake {Knutson, Donnelly, et al. 1995}. Thus, it is not surprising that our
24-hour insulin incubation yielded the same conclusion. Meanwhile, preincubation with high glucose and low insulin has been found to inhibit basal and acute insulin-stimulated glucose uptake relative to control {Nelson, Robinson, et al. 2000}.
In addition, high glucose and high insulin incubation has been reported to reduce the insulin-stimulated glucose uptake by 56% in L6 skeletal muscle cells {Tong, Khayat, et al. 2001}. Taken together, chronic glucose and insulin exposure may exert an additive effect on glucose uptake.
IM was carried out to examine the locations of actin reorganization (fig. 8).
Preincubation with high glucose (25mM) and high insulin (lOOnM) levels reduced actin remodeling by 64.7% (table. 2). Moreover, in membrane-intact L6 myotubes, by comparing with their respective controls, there were noticable reductions in insulin-induced GLUT4myc recruitment in glucose and insulin stimulated cells (fig.
9). Previous studies have confirmed the same observation. For instance, a similar
77 study of high glucose and high insulin preincubation in L6 myotubes for 24 hours shows similar reduction in actin reorganization {Tong, Khayat, et al. 2001}. Besides, high glucose and high insulin preincubation also causes a noticeable reduction of
GLUT4myc recruitment to PM {Tong, Khayat, et al. 2001}. Furthermore, high glucose and high insulin preincubation reduces the total amount of myotubes with insulin-stimulated actin remodeling by 50%. These actin structures are discrete and scattered and are similar to our own findings. Interestingly, the appearance of actin structure and reduction in GLUT4myc PM insertions in high glucose and high insulin preincubation resembles to those resulted from actin inhibitors such as swinholide-A.
As a result, both swinholide-A and high glucose and high insulin preincubation may affect a similar mechanism.
4.3.2 Effect of FFA preincubation on L6 myotubes
Increased plasma FFA concentrations are associated with many insulin resistant conditions including obesity and Type II DM {Dresner, Laurent, et al. 1999}. Thus, we incubated L6 myotubes with FFA to induce insulin resistance. Preincubation of
OA (table 5)，LA (table 6) and PA (table 7) for 24 hours all significantly diminished glucose uptake. Again, dose-dependent reductions of insulin-mediated glucose
78 uptake were observed. Numerous studies have demonstrated a negative relationship between FFA and insulin-stimulated glucose uptake. For example, elevated FFA levels caused by a high fat diet {Oakes, Cooney, et al. 1997} or Intralipid and heparin infusion {Kim, Seo, et al. 2000} in rats have been shown to decrease insulin-stimulated glucose transport. In addition, OA, LA and PA preincubations have also been demonstrated separately to reduce insulin-stimulated glucose transport in C2C12 skeletal muscle cells {Schmitz-Peiffer, Craig, et al. 1999}. Furthermore,
Storz and coworkers preincubate pmi28 myotubes with PA and produce the same result of decreased insulin-stimulated glucose uptake {Storz, Doppler, et al. 1999}.
As a result, our findings once again confirmed with their observations.
Preincubation of OA (fig. 11), LA (fig. 12) and PA (fig. 13) for 24 hours diminished insulin-stimulated actin remodeling (table. 2) and GLUT4myc recruitment on PM (fig.
9) in a dose-dependent manner. Given that unsaturated FFA is able to cause abnormal distribution pattern of spectrin organization {Stephen, Yokota, et al. 1990}, it is not surprising that FFA preincubation may also alter the actin remodeling.
Besides, previous study has incubated radiolabeled arachidonic acid (AA), an example of polyunsaturated fatty acids and PA on blood platelets. Tight bindings between FFA to cytoskeletal structures and a-actinin were found {Bum 1988}. In
79 addition, there is a difference in cytoskeletal binding between the saturated and unsaturated fatty acids. Thus, FFA may have the ability to interfere with the functioning of actin accessory proteins. Given that actin accessory proteins can regulate actin dynamics, together with our results they propose a possible involvement of FFA in the assembly of actin filaments.
Surprisingly, our PA preincubation yielded the lowest reductions in glucose uptake and actin remodeling among the choices of FFAs. Previous studies have been demonstrated that PA preincubations reduced insulin-stimulated glucose transport in
C2C12 skeletal muscle cells {Schmitz-Peiffer, Craig, et al. 1999} and pmi28 myotubes {Storz, Doppler, et al. 1999}. Thus, it has been previously confirmed that chronic PA preincubation can reduce glucose uptake. The discrepancy in the degree of glucose uptake reduction between saturated and unsaturated FFA may be due to altered membrane fluidity, aberrant insulin signal propagation or abnormal actin accessory protein function. For example, membrane lipid profile of cultured cells
can be changed by fatty acid supplements and each FFA has its specific rate of
membrane incorporation {Spector & Yorek 1985}. These differences may result in
various membrane lipid profiles that may have distinct actions on glucose transport.
Meanwhile, prolonged treatment of rat adipocytes with PA has been shown to reduce
80 glucose uptake with only a slight decrease in insulin binding {Hunnicutt, Hardy, et al.
1994}. A post-insulin binding defect including modified activities in components of insulin signal pathway may be presented in these PA-pretreated cells. Furthermore, there are also different regulatory actions on actin accessory proteins between various
FFAs {Bum 1988}. As a result, these could be accounted for the minimal effect of
PA preincubation on glucose uptake and actin remodeling on L6 myotubes.
Supplementing the culture medium with specific lipids and CHO can modify the lipid content of cultured cell membranes {Spector & Yorek 1985}. The cultured cells tend to take up the lipid and CHO available in the medium and cease its own endogenous production. As a result, exogenous lipids and CHO will be able to incorporate into the cell membrane. However, the incorporation of lipids and CHO are different for various lipids and CHO. For example, data from Spector et al has shown the different incorporation rates for four different fatty acid supplements
{Spector & Yorek 1985}. PA supplement only causes a small amount of PA and total saturated fatty acid incorporation in cell membrane. This could be the reason that PA has less effect on insulin-mediated glucose uptake and actin remodeling in L6 myotubes. Moreover, supplementing with OA increases the OA content by 60% and also a large decrease in the saturated fatty acid content. Another study has proved
81 that there is a difference in lipid incorporation from various lipid supplements in
HepG2 hepatoma cell. Thus, it is conceivable that the unique response in lipid and
CHO supplements may induce different membrane incorporation patterns that produce various effects on cellular metabolism. This could be the reason for different results in the FFA preincubation experiments. Despite of short period of incubation time, the preincubation experiments may insert exogenous FFA into the membrane and alter normal membrane functioning. For example, actin binding has been illustrated to increase membrane permeability and the increased membrane permeability is reduced by CHO introduction {Okimasu, Nobori, et al. 1987}. Thus, the change in membrane fluidity may affect insulin-mediated glucose uptake.
4.3.3 Effect of CHO preincubation on L6 myotubes
Elevated serum total CHO level is commonly found in patients with diabetes. Our
CHO preincubation experiment may provide insight in explaining the adverse effect
of CHO on glucose uptake. The CHO preincubation caused dose-dependent
reductions in insulin-induced glucose uptake (table 8)，actin remodeling (table. 2 and
fig. 14) and PM recruitment of GLUT4myc (fig. 9). Our results have provided direct
evidences that high levels of CHO impaired insulin action in skeletal muscle. Indeed,
82 chronic CHO incubation has been demonstrated to alter the organization of actin filament in human renal carcinoma cells {Ludes, Schmit, et al. 1993}. In addition, introduction of CHO into liposomal membrane causes a decrease in membrane permeability and inhibits the binding of actin and other cytoplasmic proteins
{Okimasu, Nobori, et al. 1987}.
4.3.4 Effect of cell preincubation in conditions of insulin resistance on L6
myotubes (TEM)
TEM study was performed on cells preincubated in condition of insulin resistance including glucose and insulin, FFA or CHO. Unfortunately, due to the low antibodies staining in TEM, we cannot conclude a dose-dependent reduction similar to the IM study. Instead, unlike in the control cells where GLUT4myc colocalized
with actin structures upon insulin stimulation (fig. 7), insulin failed to induce the
translocation of GLUT4 with actin (fig. 10). The dissociation of actin and GLUT4
supports our IM observations that pretreatment of cells in condition of insulin
resistance reduces actin modeling and GLUT4 translocation. In the TEM study,
most actin and GLUT4 stainings from insulin-stimulated cells preincubated in
conditions of insulin resistance remained in their non insulin-stimulated location.
83 Given that glucose uptakes were also reduced, the TEM results confirmed the importance of GLUT4 translocation with actin in relation to glucose transport.
4.4 Summary of the effects of cell preincubation in conditions of insulin
resistance
Taken together, the preincubation of cells in insulin resistant condition yields reductions in actin reorganization. By comparing with all three types of
preincubations, the high glucose (25mM) and high insulin (lOOnM) incubation
reduced the highest degree in actin reorganization, followed by LA, CHO, OA and PA.
Surprisingly, the saturated FFA — PA produced the least effect on actin reorganization
reduction despite of the adverse effect on saturated FFA on increasing blood CHO
levels and CHD risk.
Meanwhile, the obvious reduction in glucose uptake after incubation in conditions of
insulin resistance was the combination of elevated basal values and suppressed
insulin-stimulated glucose uptake values (table 9). The degree of increase in basal
uptake was much greater than the insulin-stimulated state. Hence, the net effect was
a reduction in glucose uptake following insulin stimulation. High glucose and high
84 insulin preincubation produced the highest reduction in glucose uptake followed by preincubation with unsaturated FFA and CHO. Finally, saturated PA incubation produced the least effect on glucose uptake reduction. Interestingly, there is a concordant in the reduction in glucose uptake and actin reorganization following preincubation of cells in conditions of insulin resistance.
By comparing the data from insulin-mediated actin reorganization and glucose uptake, a striking conclusion can be made. The preincubation of cells in conditions of insulin resistance reduced both the actin reorganization and glucose uptake in a similar magnitude. The cell surface immunostaining also confirmed that there was a
similar reduction of GLUT4 insertion on PM after the preincubation experiment.
Furthermore, the colocalization of actin and GLUT4 was also abolished in the TEM
study. Therefore, it is possible that a direct relationship between insulin-mediated
actin reorganization and glucose uptake exists. There are previous studies supported
our conclusion. An intact actin network is necessary for the exocytotic recruitment
of GLUT4 {Omata, Shibata, et al. 2000}. Furthermore, actin disassembly has also
been shown to cause reduction in glucose uptake {Tsakiridis, Vranic, et al. 1994}. In
addition, insulin-stimulated GLUT4 appearance at the cell surface is reduced after
Jasplakinolide treatment {Tong, Khayat, et al. 2001}. In the same study,
85 preincubation of high insulin and high glucose also diminishes glucose uptake and
GLUT4 translocation. In fact, our results reconfirmed their findings. Therefore, it is conceivable that actin filaments may directly participate in controlling the GLUT4 vesicle trafficking. Completed intact actin reorganization may play a critical role in insulin-mediated GLUT4 translocation and thereby control the glucose uptake.
Several lines of evidences have shown the adverse effects of glucose, insulin, FFA and
CHO preincubation in vivo on glucose metabolism. For example, high glucose perfusion in male Sprague-Dawley rats decreases both the glucose clearance rate and
PM GLUT4 content {Mathoo, Shi, et al. 1999}. Meanwhile, elevated FFA levels caused by high fat diets in rats {Oakes, Cooney, et al. 1997} or triglyceride emulsion
combined with heparin in humans {Roden, Price, et al. 1996} have been illustrated to
decrease insulin-stimulated glucose transport. Indeed, our in vitro results confirmed
these in vivo observations.
4.5 Possible mechanisms involved in insulin resistance induction
Given preincubating cells in conditions of insulin resistance abolishes both actin
reorganization and GLUT4 translocation in L6 myotubes, several possible
86 mechanisms may be involved in the induction of insulin resistance: (1) Changes in
GLUT expression and activities; (2) Changes in insulin signaling propagation; (3)
Altered functioning of actin accessory proteins.
4.5.1 Possible changes in GLUT expression and activities
The effectiveness of glucose transport in muscle cells is largely depended on the expression and activities of GLUT. Preincubations with high glucose and insulin, elevated FFA or CHO caused an elevated basal glucose uptake out of proportion with the insulin-stimulated glucose uptake (table 9). Elevated basal uptake may be secondary to the mass-effect of high glucose concentration in the culture medium.
Since GLUTl and GLUT3 are the primary GLUT at the basal state, insulin resistant
preincubation in L6 myotubes may increase glucose uptake in basal state by
increasing the GLUTl and GLUTS insertions on the PM. In fact, chronic insulin
incubation has been observed to increase GLUTl and GLUTS protein expressions
{Taha, Mitsumoto, et al. 1995}. On the other hand, GLUT4 is mainly responsible
for the insulin-stimulated glucose uptake. High glucose incubation has been found
to down-regulate GLUT4 translocation that leads to a reduction in muscle glucose
uptake {Mathoo, Shi, et al. 1999}. In addition, elevated FFA inhibits glucose
87 transport in skeletal muscle probably through a decrease in GLUT-4 translocation
{Shulman 2000}. Together with our own finding of decreased insulin-stimulated
GLUT4 translocation, the decrease in GLUT4 insertion on PM may be responsible for the decrease in insulin-stimulated glucose uptake. Hence, the combination of
increased expression in GLUTl and GLUT3, and the reduced translocation of GLUT4,
may explain the induction of insulin resistance in L6 myotubes following high insulin
and glucose preincuabtion.
4.5.2 Possible changes in insulin signaling propagation
Insulin signaling pathway is a multi-step pathway and requires numerous substrates
and proteins to work in concert. Spatial arrangement and sequential phosphorylation
of each intermediate are important for normal signal propagation. Study from
Knutson has demonstrated that chronic insulin treatment results in an increased rate of
IR degradation {Knutson, Donnelly et. 1995}. Furthermore, chronic insulin
treatment has been reported to cause a reduction of IRS-1 and IRS-2 association with
the cytoskeleton underlying the PM in 3T3-L1 fibroblasts {Clark, Molero, et al.
2000}. Besides, OA and PA have been shown to decrease insulin binding without
changing the receptor affinity {Svedberg, Bjomtorp, et al. 1990}. Taken together,
88 these results suggest that aberrant insulin signaling transudation may occur in cells maintained in high insulin or high fat conditions. The inhibitory effects can be additive causing derangement in actin reorganization and GLUT4 translocation.
4.5.3 Altered functioning of various actin accessory proteins
Our findings have demonstrated a reduction of actin remodeling in the preincubation experiments. It is plausible that the preincubation may influence the functioning of actin accessory proteins and cause aberrant actin remodeling. Previous study has been shown that unsaturated FFA alters the normal distribution pattern of spectrin
organization {Stephen, Yokota, et al. 1990}. In addition, lipid-specific differences
in bindings between FFA to cytoskeletal structures and a-actinin have been
established {Burn 1988}. PA incubation shows a six time higher labeling than those
in AA incubation. Furthermore, chronic CHO incubation has been demonstrated to
alter the organization of actin filament in human renal carcinoma cells {Ludes, Schmit,
et al. 1993}. Introduction of CHO into liposomal membrane inhibits the binding of
actin and other cytoplasmic proteins (Okimasu, Nobori, et al. 1987). Hence, by
interfering the functioning of actin accessory proteins, preincubations may affect the
insulin-mediated actin remodeling and thereby reduce glucose uptake.
89 4.6 Limitation of the study
There are several limitations in our study. First of all, L6 skeletal cell is an in vitro
cell line model and may not truly be representative in in vivo situation. Secondly, L6
myotubes were separately preincubated in FFA, CHO or glucose and insulin in our
experiments. However, in in vivo situation, the body may be constantly influenced
by a combination of conditions of insulin resistance and the resulting effects may be
different from our own findings. Thirdly, the introduction of agents mimicking the
conditions of insulin resistance may be different between in vivo and in vitro
situations. For example, free CHO and FFA were introduced to the cell culture while
in human body CHO and FFA bind to carrier molecules. Thus, our experimental
results may not be directly related to the in vivo situation, as the artificial conditions
created in our studies may not reflect the true cellular environment in diabetes.
Further studies are therefore required to overcome these limitations.
4.7 Conclusion
In conclusion, insulin stimulates glucose transport in skeletal muscle cells by inducing
90 actin reorganization and GLUT4 translocation. The results of TEM provide concreted evidence on the colocalization of actin and GLUT4 at an ultra-structural level. Hyperglycemia and hyperinsulinemia are common in patients with diabetes.
They usually have higher plasma FFA and total serum CHO levels than normal subjects. Thus, these metabolic derangements act synergistically to exacerbate insulin resistance in patients with Type II diabetes. Preincubation of cells in the conditions of insulin resistance creates a form of resistance to insulin treatment by suppressing actin reorganization and abolishing glucose uptake. Importantly, the novel knowledge of actin reorganization and glucose uptake suppression by FFA and
CHO preincubations provides new sight on cellular mechanism of insulin resistance.
4.8 Future study
The experimental findings of the present study demonstrated reductions of actin remodeling and glucose uptake following preincubation of L6 myotubes in conditions of insulin resistance. The aberrant actin remodeling and glucose uptake may be resulted from the changes in GLUT expression and activities, changes in insulin signal propagation and interference of the functioning of various actin accessory proteins. To understand how these are attained will be an important challenge for
91 the future. Future studies will be aimed at identifying the exact intermediates involved in abnormal mechanisms. For instance, we can identify the activities of different insulin signaling molecules before and after preincubation. Besides, the investigation of GLUTl, GLUTS and GLUT4 expression and activities also allow us to elucidate the effect of GLUT on glucose transport. Furthermore, transient transfection of various insulin signaling molecules or actin accessory proteins involved in actin remodeling may provide further insights of the basis of insulin resistance. Finally, this study demonstrates that L6 myotubes is a good in vitro cellular model for the investigation of insulin resistance. Hence, the use of L6 myotubes may allow us to screen for pharmacological agents that are effective in the management of insulin resistance and related conditions.
92 REFERENCES
1. Albu, J. 1998. Obesity and Diabetes. In Handbook of obesity / edited by George A. Bray, Claude Bouchard, W.P.T. James. G.A. Bray and C. Bouchard, editors. M. Dekker, New York. 697-708.
2. Bliss, M. 1987. The discovery of insulin / Michael Bliss. Macmillan, London.
3. Boden, G.，X. Chen, J. Ruiz, J.V. White, and L. Rossetti. 1994. Mechanisms of fatty acid-induced inhibition of glucose uptake. J Clin Invest 93:2438-2446.
4. Bray, D. 2001. Cell movements : from molecules to motility / Dennis Bray. Garland Pub, New York.
5. Bum, P. 1988. Phosphatidylinositol cycle and its possible involvement in the regulation of cytoskeleton-membrane interactions. Journal of Cellular Biochemistry 36:15-24.
6. Cavaghan, M.K., D.A. Ehrmann, and K.S. Polonsky. 2000. Interactions between insulin resistance and insulin secretion in the development of glucose intolerance. [Review] [38 refs]. J Clin Invest 106:329-333.
7. Clark, S.R, S. Martin, A.J. Carozzi, M.M. Hill, and D.E. James. 1998. Intracellular localization of phosphatidylinositide 3-kinase and insulin receptor substrate-1 in adipocytes: potential involvement of a membrane skeleton. J.Cell.Biol. 140:1211-1225.
8. Clark, S.F., J.C. Molero, and D.E. James. 2000. Release of insulin receptor substrate proteins from an intracellular complex coincides with the development of insulin resistance. J.Biol Chem 275:3819-3826.
9. Coghlan, M.P., M.M. Chou, and C.L. Carpenter. 2000. Atypical protein kinases Clambda and -zeta associate with the GTP-binding protein Cdc42 and mediate stress fiber loss. Molecular & Cellular Biology 20:2880-2889.
10. Cryer, P.E. 1997. Hypoglycemia : pathophysiology, diagnosis, and management. Oxford University Press, New York.
93 11. Dan, C.，A. Kelly, O. Bernard, and A. Minden. 2001. Cytoskeletal changes regulated by the PAK4 serine/threonine kinase are mediated by LIM kinase 1 and cofilin. J.Biol Chem 276:32115-32121.
12. DeFronzo, R.A., E. Jacot, E. Jequier, E. Maeder, J. Wahren, and J.P. Felber. 1981. The effect of insulin on the disposal of intravenous glucose. Results from indirect calorimetry and hepatic and femoral venous catheterization. Diabetes 30:1000-1007.
13. Denke, M.A., C.T. Sempos, and S.M. Grundy. 1993. Excess body weight. An underrecognized contributor to high blood cholesterol levels in white American men. [see comments.]. Archives of Internal Medicine 153:1093-1103.
14. Dresner, A., D. Laurent, M. Marcucci, M.E. Griffin, S. Dufour, G.W. Cline, L.A. Slezak, D.K. Andersen, R.S. Hundal, D.L. Rothman, K.F. Petersen, and G.I. Shulman. 1999. Effects of free fatty acids on glucose transport and IRS-1-associated phosphatidylinositol 3-kinase activity. J Clin Invest 103:253-259.
15. Felber, J.P., E. Ferrannini, A. Golay, H.U. Meyer, D. Theibaud, B. Curchod, E. Maeder, E. Jequier, and R.A. DeFronzo. 1987. Role of lipid oxidation in pathogenesis of insulin resistance of obesity and type II diabetes. Diabetes 36:1341-1350.
16. Foster, LJ., D. Li, V.K. Randhawa, and A. Klip. 2001. Insulin accelerates inter-endosomal GLUT4 traffic via phosphatidylinositol 3-kinase and protein kinase B. J.Biol Chem 276:44212-44221.
17. Frier, B.M. and F. 1993. Hypoglycaemia and diabetes : clinical and physiological aspects. E. Arnold, London.
18. Frier, B.M. and B. Fisher. 1999. Hypoglycaemia in clinical diabetes / edited by Brian M. Frier and B. Miles Fisher. John Wiley & Sons, LTD, Chichester ； New York.
19. Garvey, W.T., L. Maianu, J.A. Hancock, A.M. Golichowski, and A. Baron. 1992. Gene expression of GLUT4 in skeletal muscle from insulin-resistant patients with obesity, IGT, GDM, and NIDDM. Diabetes A\'A65-A15.
94 20. Gaziano, J.M., P.R. Hebert, and C.H. Hennekens. 1996. Cholesterol reduction: weighing the benefits and risks, [see comments.]. [Review] [29 refs]. Annals of Internal Medicine 124:914-918.
21. Goldberg, M.B. 2001. Actin-based motility of intracellular microbial pathogens. [Review] [199 refs]. Microbiology & Molecular Biology Reviews 65:595-626.
22. Greengard, P., F. Valtorta, A.J. Czemik, and F. Benfenati. 1993. Synaptic vesicle phosphoproteins and regulation of synaptic function. Science JID - 0404511 259:780-785.
23. Halse, R.，S.M. Bonavaud, J.L. Armstrong, J.G. McCormack，and S.J. Yeaman. 2001. Control of glycogen synthesis by glucose, glycogen, and insulin in cultured human muscle cells. Diabetes 50:720-726.
24. Hilfiker, S., V.A. Pieribone, A.J. Czemik, H.T. Kao, G.J. Augustine, and P. Greengard. 1999. Synapsins as regulators of neurotransmitter release. [Review] [88 refs]. Philosophical Transactions of the Royal Society of London - Series B: Biological Sciences 354:269-279.
25. Hunnicutt, J.W., R.W. Hardy, J. Williford, and J.M. McDonald. 1994. Saturated fatty acid-induced insulin resistance in rat adipocytes. Diabetes 43:540-545.
26. Hunter, S.J. and W.T. Garvey. 1998. Insulin action and insulin resistance: diseases involving defects in insulin receptors, signal transduction, and the glucose transport effector system. [Review] [100 refs]. American Journal of Medicine 105:331-345.
27. Inoguchi, T.’ P. Li, F. Umeda, H.Y. Yu, M. Kakimoto, M. Imamura, T. Aoki, T. Etoh, T. Hashimoto, M. Naruse, H. Sano, H. Utsumi, and H. Nawata. 2000. High glucose level and free fatty acid stimulate reactive oxygen species production through protein kinase C-dependent activation of NAD(P)H oxidase in cultured vascular cells. Diabetes 49:1939-1945.
28. Jackson, S.，S.M. Bagstaff, S. Lynn, S.J. Yeaman, D.M. Tumbull, and M. Walker. 2000. Decreased insulin responsiveness of glucose uptake in cultured human skeletal muscle cells from insulin-resistant nondiabetic relatives of type 2 diabetic families. Diabetes 49:1169-1177.
95 29. Kahn, B.B. 1992. Facilitative glucose transporters: regulatory mechanisms and dysregulation in diabetes. [Review] [76 refs]. J Clin Invest 89:1367-1374.
30. Khayat, Z.A., P. Tong, K. Yaworsky, R.J. Bloch, and A. Klip. 2000. Insulin-induced actin filament remodeling colocalizes actin with phosphatidylinositol 3-kinase and GLUT4 in L6 myotubes. Journal of Cell Science 113 Pt 2:279-290.
31. Kim, S.C., K.K. Seo, H.W. Kim, and M.Y. Lee. 2000. The effects of isolated lipoproteins and triglyceride, combined oxidized low density lipoprotein (LDL) plus triglyceride, and combined oxidized LDL plus high density lipoprotein on the contractile and relaxation response of rabbit cavernous smooth muscle. International Journal of Andrology 23 Suppl 2:26-29.
32. Klip, A., G. Li, and W.J. Logan. 1984. Induction of sugar uptake response to insulin by serum depletion in fusing L6 myoblasts. American Journal of Physiology 247:E291-E296
33. Klip, A., W.J. Logan, and G. Li. 1982. Hexose transport in L6 muscle cells. Kinetic properties and the number of [3H]cytochalasin B binding sites. Biochimica et Biophysica Acta 687:265-280.
34. Knutson, V.P., P.V. Donnelly, Y. Balba, and M. Lopez-Reyes. 1995. Insulin resistance is mediated by a proteolytic fragment of the insulin receptor. J.Biol Chem 270:24972-24981.
35. Le Roith, D. and Y. Zick. 2001. Recent Advances in Our Understanding of Insulin Action and Insulin Resistance. Diabetes Care 24:588-597.
36. Leslie, R.D.G. and D.C. Robbins. 1995. Diabetes: Clinical Science in Practice / edited by R.D.G. Leslie and David C. Robbins. Cambridge Unviersity Press, Cambridge.
37. Lodish, H.F. 1995. Molecular cell biology / Lodish, Harvey ... [et al.]. In H.F. Lodish, editor. Scientific American Books, New York.
38. Lodish, H.F. 2000. Molecular cell biology / Harvey Lodish ... [et al.]. In H.F. Lodish, editor. W.H. Freeman, New York.
96 39. Ludes，B.，A.C. Schmit, G. Cremel, A.M. Lambert, P. Hubert, D. Jacqmin, C. Bollack，and C. Staedel. 1993. Influence of cholesterol derivatives on cytoskeletal organization of human carcinoma cells. European Urology 23:490-501.
40. Maddala, R.L., V.N. Reddy, and P.V. Rao. 2001. Lovastatin-induced cytoskeletal reorganization in lens epithelial cells: role of Rho GTPases. Investigative Ophthalmology & Visual Science 42:2610-2615.
41. Mathoo, J.M., Z.Q. Shi, A. Klip, and M. Vranic. 1999. Opposite effects of acute hypoglycemia and acute hyperglycemia on glucose transport and glucose transporters in perfused rat skeletal muscle. Diabetes 48:1281-1288.
42. Meyer, C.，J.M. Dostou，S.L. Welle, and J.E. Gerich. 2002. Role of human liver, kidney, and skeletal muscle in postprandial glucose homeostasis. American Journal of Physiology - Endocrinology & Metabolism 282:E419-E427
43. Mitsumoto, Y., E. Burdett, A. Grant, and A. Klip. 1991. Differential expression of the GLUTl and GLUT4 glucose transporters during differentiation of L6 muscle cells. Biochem.Biophys.Res. Commun. 175:652-659.
44. Mitsumoto, Y. and A. Klip. 1992. Development regulation of the subcellular distribution and glycosylation of GLUTl and GLUT4 glucose transporters during myogenesis of L6 muscle cells. J.Biol Chem 267:4957-4962.
45. Moller, D.E. 1993. Insulin resistance. John Wiley & Sons, Chichester ； New York.
46. Nelson, B.A.，K.A. Robinson, and M.G. Buse. 2000. High glucose and glucosamine induce insulin resistance via different mechanisms in 3T3-L1 adipocytes. Diabetes 49:981-991.
47. Nussey, S. and Saffron Whitehead. 2001. Endocrinology : an integrated approach / Stephen Nussey and Saffron Whitehead. Bios, Oxford.
97 48. Oakes, N.D., GJ. Cooney, S. Camilleri, D.J. Chisholm, and E.W. Kraegen. 1997. Mechanisms of liver and muscle insulin resistance induced by chronic high-fat feeding. Diabetes 46:1768-1774.
49. Okimasu, E., K. Nobori, S. Kobayashi, E. Suzaki, S. Terada, and K. Utsumi. 1987. Inhibitory effect of cholesterol on interaction between cytoplasmic actin and liposomes, and restorative effect of high osmotic pressure. Cell Structure & Function 12:187-195.
50. Omata, W., H. Shibata, L. Li, K. Takata, and I. Kojima. 2000. Actin filaments play a critical role in insulin-induced exocytotic recruitment but not in endocytosis of GLUT4 in isolated rat adipocytes. Biochem J 346 Pt 2:321-328.
51. Pantaloni, D., C. Le Clainche, and M.F. Carlier. 2001. Mechanism of actin-based motility, [erratum appears in Science 2001 Jun 15;292(5524):2012.]. [Review] [70 refs]. Science 292:1502-1506.
52. Pessin, J.E. and A.R. Saltiel. 2000. Signaling pathways in insulin action: molecular targets of insulin resistance. [Review] [38 refs]. J Clin Invest 106:165-169.
53. Ricort, J.M., J.F. Tanti, E. Van Obberghen, and Y. Le Marchand-Brustel. 1995. Alterations in insulin signalling pathway induced by prolonged insulin treatment of 3T3-L1 adipocytes. Diabetologia 38:1148-1156.
54. Roden, M., T.B. Price, G. Perseghin, K.F. Petersen, D.L. Rothman, G.W. Cline, and G.I. Shulman. 1996. Mechanism of free fatty acid-induced insulin resistance in humans. J Clin Invest 97:2859-2865.
55. Sarabia, V., T. Ramlal, and A. Klip. 1990. Glucose uptake in human and animal muscle cells in culture. Biochemistry & Cell Biology 68:536-542.
56. Schmitz-Peiffer, C, D.L. Craig, and T.J. Biden. 1999. Ceramide generation is sufficient to account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal muscle cells pretreated with palmitate. J.Biol Chem 274:24202-24210.
57 Shepherd, P.R. and B.B. Kahn. 1999. Glucose transporters and insulin action—implications for insulin resistance and diabetes mellitus. [letter;
98 comment]. [Review] [103 refs]. New England Journal of Medicine 341:248-257.
58 Shulman G.I. 2000: Cellular mechanisms of insulin resistance. [Review] [46 refs]. J Clin Invest; 106: 171-176.
59. Somwar, R., W. Niu, D.Y. Kim, G. Sweeney, V.K. Randhawa, C. Huang, T. Ramlal, and A. Klip. 2001. Differential effects of phosphatidylinositol 3-kinase inhibition on intracellular signals regulating GLUT4 translocation and glucose transport. J.Biol Chem 276:46079-46087.
60. South wick, F.，D. Purich, and P. Cossart. 2000. Actin Filaments: Self-Assembly and Regulatory Interactions. In Cellular microbiology. P. Cossart, P. Boquet, S. Normark，and R. Rappuoli, editors. ASM Press, Washington, DC. 153-170.
61. Spector, A.A. and M.A. Yorek. 1985. Membrane lipid composition and cellular function. [Review] [172 refs]. Journal of Lipid Research 26:1015-1035.
62. Stephen, F.D., S.J. Yokota, and E.A. Repasky. 1990. The effect of free fatty acids on spectrin organization in lymphocytes. Cell Biophysics 17:269-282.
63. Storz, P., H. Doppler, A. Wemig, K. Pfizenmaier, and G. Muller. 1999. Cross-talk mechanisms in the development of insulin resistance of skeletal muscle cells palmitate rather than tumour necrosis factor inhibits insulin-dependent protein kinase B (PKB)/Akt stimulation and glucose uptake. European Journal of Biochemistry 266:17-25.
64. Sullivan, R., L.S. Price, and A. Koffer. 1999. Rho controls cortical F-actin disassembly in addition to, but independently of, secretion in mast cells. J.Biol Chem 274:38140-38146.
65. Svedberg, J., P. Bjomtorp, U. Smith, and P. Lonnroth. 1990. Free-fatty acid inhibition of insulin binding, degradation, and action in isolated rat hepatocytes. Diabetes 39:570-574.
66. Taha, C., Y. Mitsumoto, Z. Liu, E.Y. Skolnik, and A. Klip. 1995. The insulin-dependent biosynthesis of GLUTl and GLUTS glucose
99 transporters in L6 muscle cells is mediated by distinct pathways. Roles ofp21ras and pp70 S6 kinase. J.Biol Chem 270:24678-24681.
67. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 24 Supplement 1， S5-S20. 2001. (GENERIC) Ref Type: Report
68. Tong, P.C., Z.A. Khayat, C.S. Chan, and A. Klip. 2001. A Mechanism for the Impairment of Insulin Action by High Levels of Insulin and Glucose in L6 Rat Skeletal Muscle Cells. Diabetes 50 SUPPLEMENT 2:A507(Abstr.)
69. Tong, P., Z.A. Khayat, C. Huang, N. Patel, A. Ueyama, and A. Klip. 2001. Insulin-induced cortical actin remodeling promotes GLUT4 insertion at muscle cell membrane ruffles. J Clin Invest 108:371-381.
70. Tsakiridis, T., A. Bergman, R. Somwar, C. Taha, K. Aktories, T.F. Cruz, A. Klip, and G.P. Downey. 1998. Actin filaments facilitate insulin activation of the src and collagen homologous/mitogen-activated protein kinase pathway leading to DNA synthesis and c-fos expression. J.Biol Chem 273:28322-28331.
71. Tsakiridis, T., P. Tong, B. Matthews, E. Tsiani, P.J. Bilan, A. Klip, and G.P. Downey. 1999. Role of the actin cytoskeleton in insulin action. [Review] [125 refs]. Microscopy Research & Technique 47:79-92.
72. Tsakiridis, T.’ M. Vranic, and A. Klip. 1994. Disassembly of the actin network inhibits insulin-dependent stimulation of glucose transport and prevents recruitment of glucose transporters to the plasma membrane. J.Biol Chem 269:29934-29942.
73. Tsakiridis, T.’ Q. Wang, C. Taha, S. Grinstein, G. Downey, and A. Klip. 1997. Involvement of the actin network in insulin signalling. Society of General Physiologists Series 52:257-271.
74. Ueyama, A., K.L. Yaworsky, Q. Wang, Y. Ebina, and A. Klip. 1999. GLUT-4myc ectopic expression in L6 myoblasts generates a GLUT-4-specific pool conferring insulin sensitivity. American Journal of Physiology 277:E572-E578
100 75. Wang, Q.，P.J. Bilan, and A. Klip. 1998. Opposite effects of insulin on focal adhesion proteins in 3T3-L1 adipocytes and in cells overexpressing the insulin receptor. Molecular Biology of the Cell 9:3057-3069.
76. Wang, Q., P.J. Bilan, T. Tsakiridis, A. Hinek，and A. Klip. 1998. Actin filaments participate in the relocalization of phosphatidylinositolB-kinase to glucose transporter-containing compartments and in the stimulation of glucose uptake in 3T3-L1 adipocytes [published erratum appears in Biochem J 1999 Aug l;341(Pt 3):861]. Biochem J 331:911-92^.
77. Weed, S.A. and J.T. Parsons. 2001. Cortactin: coupling membrane dynamics to cortical actin assembly. [Review] [210 refs]. Oncogene 20:6418-6434.
78. Wilson, J. 1998. Williams textbook of endocrinology. W.B. Saunders, Philadelphia.
79. Yaffe, D. 1968. Retention of differentiation potentialities during prolonged cultivation of myogenic cells. Proceedings of the National Academy of Sciences of the United States of America 61:477-483.
80. Ziel, F.H., N. Venkatesan, and M.B. Davidson. 1988. Glucose transport is rate limiting for skeletal muscle glucose metabolism in normal and STZ-induced diabetic rats. Diabetes 37:885-890.
101 - ；fl
4..., • i
CUHK Libraries mmMim \ ••a^issbso