溫感性甲殼素/明膠/甘油磷酸水膠做為阿魏酸持續釋放於髓核再生之應用

(1)

國立臺灣大學醫學院暨工學院醫學工程學研究所博士論文

Institute of Biomedical Engineering

College of Medicine and College of Engineering

National Taiwan University Doctoral Dissertation

溫感性甲殼素/明膠/甘油磷酸水膠做為阿魏酸持續釋放於髓核再生之應用

Thermosensitive Chitosan/Gelatin/Glycerol Phosphate Hydrogel as a Sustained Release System of Ferulic Acid

for Nucleus Pulposus Regeneration

鄭詠馨 Yung-Hsin Cheng

指導教授﹕林峯輝博士 Advisor: Feng-Huei Lin, Ph.D.

中華民國 101 年 6 月

(2)

中文摘要

椎間盤退化與椎間盤突出及背痛有高度的相關性，這些病徵增加了健康照護的支出。椎間盤退化過程可分為五個階段，在退化的第一至二階段時，並沒有明顯的病徵出現，但可透過核磁共振或電腦斷層掃描檢查來追蹤，臨床上一般並不會在此階段給予治療。近來的文獻指出，活性氧自由基不僅會加速椎間盤退化的過程，且會造成髓核細胞的凋亡和細胞外基質的降解。阿魏酸是一種抗氧化物並且可以較穩定地存在於空氣中；阿魏酸被證實對於活性氧自由基所引起的相關疾病具有預防的效果。本研究的目的除了評估阿魏酸對於雙氧水所引起的氧化壓力導致髓核細胞傷害的可能治療效果外，並評估利用溫感性甲殼素/明膠/甘油磷酸水膠做為阿魏酸持續釋放早期治療椎間盤退化的可行性。

在本研究中的試驗結果指出，500 μM 為阿魏酸對紐西蘭兔的髓核細胞的安全

閥值，利用阿魏酸治療被雙氧水所引起的氧化壓力所傷害的髓核細胞，其aggrecan,

type II collagen 和 BMP-7 的基因表現可以有顯著的提升，而 MMP-3 的表現量有顯 著的下降，而硫酸化葡萄胺聚醣含量有顯著的上升，細胞凋亡的情形也有顯著的抑制。而以甲殼素/明膠/甘油磷酸水膠做為阿魏酸持續釋放的試驗中，阿魏酸可以從水膠中緩釋，包覆阿魏酸的水膠除了能提升被雙氧水所引起的氧化壓力所傷害的髓核細胞中aggrecan 和 type II collagen 基因的表現量外，並能抑制 MMP-3 的表

現量，而在硫酸化葡萄胺聚醣生成量及alcian blue 的染色的結果指出，包覆阿魏酸

的水膠可以使受傷害的髓核細胞恢復到正常髓核細胞的表現量，另外在 caspase-3

和TUNEL 的染色結果上，也指出其細胞凋亡的情形可以被顯著抑制。在本研究中

證明，阿魏酸可成功藉由 N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC)/

N-hydroxysuccinimide (NHS)固定於甲殼素/明膠/甘油磷酸水膠上；在中性環境下，

該水膠之成膠溫度為31.8C，而以阿魏酸固定於甲殼素/明膠/甘油磷酸上的水膠治

(3)

療被雙氧水所引起的氧化壓力所傷害的髓核細胞，可顯著的提升受傷害細胞中 aggrecan 和 type II collage 的表現量，並抑制其 MMP-3 的表現量；此外，硫酸化葡

萄胺聚醣生成量也能恢復到正常的水平，另外在caspase-3 和 TUNEL 的染色結果

中顯示，細胞凋亡的情形可以被有效的抑制。

綜合上述，本研究的試驗結果指出阿魏酸可做為髓核再生的治療分子，而溫感性甲殼素/明膠/甘油磷酸水膠可做為阿魏酸長期釋放的良好載體，將阿魏酸固定於甲殼素/明膠/甘油磷酸水膠上可有效延長釋放的時間；結合阿魏酸及溫感性甲殼素/明膠/甘油磷酸水膠顯然可以有效治療因氧化壓力所傷害的髓核細胞，在未來更可應用於髓核再生的微創手術中。

關鍵字: 髓核、阿魏酸、氧化壓力、溫感性水膠、抗氧化劑

(4)

ABSTRACT

Disc degeneration is strongly associated with back pain and herniation that increase the costs of health care. The degeneration of intervertebral disc (IVD) could be divided into 5 stages. In the first and second stages, there are no significant symptoms but could be traced by magnetic resonance imaging or computed tomography-scan. Generally, no aggressive treatment would be processed in the clinics. Recent studies indicated that overproduction of reactive oxygen species (ROS) may accelerate the degenerative process of IVD and associate with apoptosis of nucleus pulposus (NP) cells and degradation of extracellular matrix. Ferulic acid (FA) is an excellent antioxidant and relatively stable in air. FA has been proven to have ability to prevent ROS-induced diseases. The object of the study was aimed to evaluate the possible therapeutic effect of FA on hydrogen peroxide (H2O2)-induced oxidative stress NP cells and the feasibility of use the thermosensitive chitosan/gelatin/glycerophosphate (C/G/GP) hydrogel as a sustained release system of FA for early treatment in IVD degeneration.

In the study, NP cells were harvested from the IVD of New Zealand rabbits. The results showed that 500 μM of FA might be the threshold to treat NP cells without cytotoxicity. Post-treatment of FA on H2O2-induced oxidative stress NP cells significantly up regulated the expression of aggrecan, type II collagen and BMP-7 and

(5)

down regulated the expression of MMP-3 in mRNA level. Post-treatment of FA on H2O2-induced oxidative stress NP cells could restore the production of sulfated glycosaminoglycans (GAGs) and inhibit the apoptosis caused by H2O2. The results showed that the release of FA from C/G/GP hydrogel could decrease the H2O2-induced oxidative stress. Post-treatment of FA-incorporated C/G/GP hydrogel on H2O2-induced oxidative stress NP cells showed up-regulation of aggrecan and type II collagen and down-regulation of MMP-3 in mRNA level. The results of sulfated GAGs to DNA ratio and alcian blue staining revealed that the GAGs production of H2O2-induced oxidative stress NP cells could reach to normal level. The results of caspase-3 activity and TUNEL staining indicated that FA-incorporated C/G/GP hydrogel decreased the apoptosis of H2O2-induced oxidative stress NP cells. The results showed that FA was

successfully immobilized on C/G/GP hydrogel by N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC) and N-hydroxysuccinimide (NHS) crosslinking method. The gelation temperature of the FA-immobilized C/G/GP hydrogel was 31.80 C under neutral pH. Post-treatment of FA-immobilized C/G/GP hydrogel on H2O2-induced oxidative stress NP cells showed down-regulation of MMP-3 and up-regulation aggrecan and type II collagen in mRNA level. The sulfated GAGs

(6)

caspase-3 activity and TUNEL staining showed that the apoptosis of H2O2-induced oxidative stress NP cells could be inhibited by post-treatment of FA-immobilized C/G/GP hydrogel.

From the results of the study, FA could be used as a therapeutic molecule for NP regeneration and FA-incorporated C/G/GP hydrogel might be potentially applied as a long-term release system. The immobilization of FA on C/G/GP hydrogel could significantly prolong the release period of FA. These results suggest that combination of FA and thermosensitive C/G/GP hydrogel can treat NP cells from the damage caused by oxidative stress and may apply in minimally invasive surgery for NP regeneration in the future.

Keyword: nucleus pulposus; ferulic acid; oxidative stress; thermosensitive hydrogel;

antioxidant

(7)

TABLE OF CONTENTS

中文摘要 ... i

ABSTRACT ... iii

TABLE OF CONTENTS ... vi

CHAPTER 1 INTRODUCTION ... 1

1.1 Structure and function of intervertebral disc ... 1

1.2 Intervertebral disc degeneration ... 5

1.3 Current treatment options for disc degeneration ... 7

CHAPTER 2 THEORETICAL BASIS ... 10

2.1 The pathophysiology of disc degeneration ... 10

2.2 Reactive oxygen species and oxidative stress ... 14

2.3 Polyphenol: ferulic acid ... 19

2.4 In situ forming hydrogel ... 22

2.5 The purpose of study ... 27

CHAPTER 3 MATERIALS AND METHODS ... 29

3.1 Isolation of nucleus pulposus cells ... 29

3.2 Cytotoxicity of ferulic acid on nucleus pulposus cells ... 30

3.3 Chemiluminescence assay for reactive oxygen species production ... 32

3.4 The effects of ferulic acid on hydrogen peroxide-induced oxidative stress nucleus pulposus cells ... 33

3.4.1 Induction of oxidative stress and ferulic acid treatment ... 33

3.4.2 RNA extraction and gene expression... 33

(8)

3.4.4 Caspase-3 activity ... 35

3.4.5 TUNEL staining ... 36

3.5 Thermosensitive chitosan/gelatin/β-glycerol phosphate (C/G/GP) hydrogel as a controlled release system of ferulic acid (FA) for nucleus pulposus regeneration ... 37

3.5.1 Preparation of thermosensitive C/G/GP hydrogel ... 37

3.5.2 Reheological characterization... 38

3.5.3 In vitro FA release study ... 38

3.5.4 Induction of oxidative stress and FA treatment ... 39

3.5.5 RNA extraction and gene expression... 40

3.5.6 Total DNA quantification ... 41

3.5.7 Sulfated glycosaminoglycan content ... 41

3.5.8 Alcian blue staining ... 42

3.6 The effects of thermosensitive ferulic acid-immobilized

chitosan/gelatin/β-glycerol phosphate (FA-immobilized C/G/GP) hydrogel

(9)

on nucleus pulposus cells under hydrogen peroxide-induced oxidative

stress ... 44

3.6.1 Preparation of thermosensitive FA-immobilized C/G/GP hydrogel44 3.6.2 Characterization of FA-immobilized gelatin ... 45

3.6.4 Cytotoxicity of thermosensitive FA-immobilized C/G/GP hydrogel on NP cells ... 46

3.6.5 In vitro FA release study ... 47

3.6.6 Induction of oxidative stress and FA-immobilized C/G/GP hydrogel treatment 48 3.6.7 RNA extraction and gene expression... 49

3.6.8 Analysis of cell numbers ... 50

3.6.9 Sulfated glycosaminoglycan content ... 50

(10)

4.1 Cytotoxicity of ferulic acid on nucleus pulposus cells ... 54

4.2 Reactive oxygen species scavenging effect ... 57

4.3.1 Gene expression ... 59

4.3.2 Sulfated glycosaminoglycan production ... 63

4.4 Thermosensitive chitosan/gelatin/β-glycerol phosphate (C/G/GP) hydrogel as a controlled release system of ferulic acid for nucleus pulposus regeneration ... 67

4.4.1 The release of FA from C/G/GP hydrogel ... 67

4.4.3 Sulfated glycosaminoglycan production ... 71

4.4.4 Alcian blue staining ... 72

4.5 The effects of thermosensitive ferulic acid-immobilized

(11)

chitosan/gelatin/β-glycerol phosphate (FA-immobilized C/G/GP) hydrogel on nucleus pulposus cells under hydrogen peroxide-induced oxidative

stress ... 76

4.5.1 TNBS assay ... 76

4.5.3 Cytotoxicity of thermosensitive FA-immobilized C/G/GP hydrogel on NP cells ... 80

4.5.4 The release of FA from FA-immobilized C/G/GP hydrogel ... 82

4.5.6 Sulfated glycosamninoglycan production ... 89

CHAPTER 5 DISCUSSIONS ... 93

5.2 Thermosensitive chitosan/gelatin/β-glycerol phosphate hydrogel as a controlled release system of ferulic acid for nucleus pulposus regeneration ... 99

(12)

chitosan/gelatin/β-glycerol phosphate hydrogel on nucleus pulposus cells

under hydrogen peroxide-induced oxidative stress ... 104

CHAPTER 6 CONCLUSSION ... 110

References ... 113

Curriculum Vitae ... 125

(13)

LIST OF CHARTS

CHAPTER 1

Fig. 1. 1 A schematic diagram of a spinal segment and the intervertebral disc ... 4

Fig. 1. 2 Schematic view of nutrient gradients across the disc ... 4

Fig. 1. 3 The pathophysiology of disc degeneration ... 6

Fig. 1. 4 The normal IVD (left) and degenerative IVD (right) . ... 6

CHAPTER 2 Fig. 2. 1 Major signaling pathways activated in response to oxidative stress ... 17

Fig. 2.2 Components of caspase-dependent intrinsic and extrinsic apoptosis pathways ... 18

Fig. 2. 3 The hypothetical model for NP cell senescence in degenerative disc ... 18

Fig. 2. 4 Resonance structure of FA ... 20

Fig. 2. 5 A schematic model for polyphenols and flavonoids mediated modulation of cell signaling ... 21

Fig. 2. 6 The structure of chitosan ... 25

Fig. 2. 7 The structure of β-glycerol phosphate disodium salt hydrate ... 25

Fig. 2. 8 Three major approaches of drug loading for chitosan hydrogel ... 26

(14)

CHAPTER 4

Fig. 4. 1 Cytotoxicity of FA on NP cells was measured on day 1 and day 3. ... 56 Fig. 4. 2 The ROS scavenging effect of different concentration of FA ... 58 Fig. 4. 3 Gene expression of normal NP cells (without treatment), H2O2-induced NP

cells (H100) and post-treatment of FA on H2O2-induced NP cells (H100FA500) ... 62 Fig. 4. 4 The ratio of sulfated GAGs to DNA of normal NP cells (without treatment,

Control), H2O2-induced NP cells (H100) and post-treatment of FA on

H2O2-induced NP cells (H100FA500) ... 63 Fig. 4. 6 The caspase-3 activity of normal NP cells (without treatment, Control),

H2O2-induced NP cells (H100) and post-treatment of FA on H2O2-induced NP cells (H100FA500) ... 64 Fig. 4. 7 TUNEL staining of (a) Normal NP cells (without treatment, Control), (b)

H2O2-induced NP cells (H100) and (c) post-treatment of FA on H2O2-induced NP cells (H100FA500). ... 66 Fig. 4. 8 The release profile of 500 μM FA from C/G/GP hydrogels ... 67 Fig. 4. 9 Gene expression of normal NP cells (without treatment), post-treatment of

C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel) and post-treatment of FA-incorporated C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel-FA)... 70 Fig. 4. 10 The ratio of sulfated GAGs to DNA of normal NP cells (without treatment,

Control), post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel) and post-treatment of FA-incorporated C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (Gel-FA) ... 71

(15)

treatment, Control), (b) post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel) and (c) post-treatment of

FA-incorporated C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (Gel-FA) ... 72 Fig. 4. 12 The caspase-3 activity of normal NP cells (without treatment, Control),

post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel) and post-treatment of FA-incorporated C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (Gel-FA) ... 73 Fig. 4. 13 TUNEL staining of (a) Normal NP cells (without treatment, Control), (b)

post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (Gel) and (c) post-treatment of FA-incorporated C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (Gel-FA). ... 75 Fig. 4. 14 The percentage of residual amino groups in the gelatin (without

immobilization of FA) and FA-immobilized gelatin group ... 77 Fig. 4. 15 (a) Temperature dependence of storage modulus (G’) and loss modulus

(G”), (b) time dependence of G’ and G” at 37 C and (c) time dependence of G’

and G” at 25 C ... 79 Fig. 4. 16 Cytotoxicity of FA-immobilized C/G/GP to NP cells: (a) WST-1 assay and

(b) LDH assay ... 81 Fig. 4. 17 The release profile of FA from FA-immobilized C/G/GP hydrogel ... 83 Fig. 4. 18 Gene expression of the normal NP cells (without treatment, control group),

post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (C/G/GP group) and post-treatment of FA-immobilized C/G/GP

(16)

Fig. 4. 19 The sulfated GAGs production per cell in normal NP cells (without treatment, control group), post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (C/G/GP group) and post-treatment of FA-immobilized C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (FA-immobilized C/G/GP group) ... 89 Fig. 4. 20 The caspase-3 activity in normal NP cells (without treatment, control

group), post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (C/G/GP group) and post-treatment of FA-immobilized C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells (FA-immobilized C/G/GP group) ... 90 Fig. 4. 21 TUNEL staining of (a) Normal NP cells (without treatment, control group),

(b) post-treatment of C/G/GP hydrogel on 100 μM H2O2-induced oxidative stress NP cells (C/G/GP group) and (c) post-treatment of FA-immobilized C/G/GP hydrogel on 100 μM H2O2 induced oxidative stress NP cells

(FA-immobilized C/G/GP group). ... 92

(17)

LIST OF TABLES

CHAPTER 1

Table 1. 1 Thompson grade in the degenerative disc ... 9

CHAPTER 2

Table 2. 1 The effect of growth factors on disc cells ... 13

(18)

LIST OF ABBREVIATIONS

AF Annulus Fibrosus

BMP Bone Morphogenic Protein

CEP Cartilage Endplate

CL Chemiluminescence C/G/GP Chitosan/Gelatin/β-glycerophosphate C/GP Chitosan/β-glycerophosphate

cDNA Complementary DNA

c-Apaf-1 Cytochrome Apoptotic Protease Activating Factor-1 DISC Death-inducing Signaling Complex

DNA Deoxyribonucleic Acid

DMMB Dimethylmethylene Blue

DMSO Dimethyl Sulfoxide

DMEM-F12 Dulbecco’s Modified Eagle’s Medium-nutrient Mixture F-12 Ham Medium

EDC N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide

ECM Extracellular Matrix

ELISA Enzyme-linked Immunosorbent Assay FADD Fas-associated Protein with Death Domain

FA Ferulic Acid

β-GP Glycerol 2-phosphate Disodium Salt Hydrate GAG Glycosaminoglycan

HSF Heat-shock Transcription Factor

H2O2 Hydrogen Peroxide

iNOS Inducible Nitric Oxide Synthase

IGF Insulin-like Growth Factor

IL Interleukin

IVD Intervertebral Disc

LDH Lactate Dehydrogenase

MMP Matrix Metalloproteinase

MAP Mitogen-activated Protein

NF-κB nuclear factor-kappa B

NHS N-hydroxysuccinimide

NO Nitric Oxide

NP Nucleus Pulposus

(19)

PBS Phosphate Buffered Saline ROS Reactive Oxygen Species

pRB Retinoblastoma Protein

RT-PCR Reverse Transcription Polymerase Chain Reaction

RNA Ribonucleic Acid

TNBS 2, 4, 6-trinitrobenzenesulfonic Acid Solution

TUNEL Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling

TIMP Tissue Inhibitors of Metalloproteinase TGF Transforming Growth Factor

TNF Tumor Necrosis Factor

t-Bid Truncated-Bid UV Ultraviolet UV-VIS-NIR Ultraviolet-Visible-Near Infrared

VB Vertebral Body

WST Water Soluble Tetrazolium Salt

(20)

CHAPTER 1 INTRODUCTION

1.1 Structure and function of intervertebral disc

Intervertebral disc (IVD) lies between vertebral bodies (VBs) and consists of central nucleus pulposus (NP), outer annulus fibrosus (AF), and upper and lower cartilage endplates (CEPs) (Fig. 1). The main components of IVD are water, proteoglycans and collagen, which provide adequate mechanical strength to resist the external stress. Collagen network provides the tensile strength to the IVD and anchors the tissue to the bone; proteoglycans attract and retain water in the IVD that serve to resist compressive loads [1].

The IVD possesses a variety of collagen in the extracellular matrix (ECM). Type I and II collagen (approximately 80%) are the most abundant collagen in the disc matrix.

The AF consists mostly of Type I collagen with small amounts of Type III, V (3%), VI (10%) and IX (1%-2%) collagens. The NP consists mainly of Type II collagen and only small amounts of Type VI (15%-20%), IX (1%-2%), XI (3%) collagens [2-3].

The IVD contains a variety of proteoglycans, including aggrecan, versican, hyaluronan, decorin, biglycan and lumican. Proteoglycan is constructed of glycosaminoglycans (GAGs) which are attached to the protein core of aggrecan.

(21)

Aggrecan is the most abundant component of the proteoglycans in the disc matrix and possesses chondroitin sulfate and keratan sulfate. Proteoglycans are interacted to hyaluronic acid to form the large aggregates. [2-3]

The NP is a gelatinous structure and composed mainly of type II collagen and proteoglycans which are organized randomly. NP contains the chondrocyte-like cell and the cell density of NP is approximately 4 x 10⁶ per cm³. AF is a lamellae structure which composed mainly of type I collagen and elastin fibers. AF contains fibroblast-like cell and the cell density of AF is about 9 x 10⁶ per cm³. In the normal disc, the percentage of type I and type II collagen is about 70% and 20% of the dry weight in the AF and NP, respectively. The proteoglycans and water content are both higher in the NP (approximately 50% and 80% of wet weight, respectively) than in the AF (around 20% and 70% of wet weight, respectively) [1-3]. The CEP is a hyaline cartilage which is an interface between the VB and IVD. The IVD is avascular, the exchange of nutrients and metabolic waste products of disc is diffused mainly through the CEP.

Although small blood vessels can be found in the outer AF, they do not penetrate into inner AF and NP. The CEP serves as a selectively permeable barrier to nutrients and solutes. Small molecules such as water, oxygen and amino acids can directly penetrate

(22)

diffusion gradient of oxygen, glucose and lactate from upper to lower CEP has been observed in the recent years (Fig. 2). The center of the disc is at low glucose and oxygen and lactic acid concentration is highest in the center of the disc. The normal range for the pH of the disc is 6.9 to 7.2 [3-4].

(23)

Fig.

Nucl (VB)

1. 1 A schem leus pulposu ); spinal cor

matic diagr us (NP); ann rd (SC); ner

ram of a spin nulus fibros rve root (NR

nal segmen sus (AF); ca R); apophys

nt and the in artilage endp seal joints (A

tervertebral plate (CEP) AJ)

l disc [1].

); vertebral body

(24)

1.2 Intervertebral disc degeneration

The IVD degeneration could cause to low back pain that trouble lots of the aged people and those who are in intensive labor [3]. There are several factors have been implicated in disc degeneration (Fig. 2). Disc degeneration can be defined as an age-related, cell-mediated molecular degradation process under genetic factor that is accelerated by nutrition, mechanical factors, toxic or metabolic influence [4].

In the degenerative disc, the boundary between NP and AF becomes less clear, NP becomes less gel-like and more fibrous and AF becomes disorganized (Fig. 3). With disc degeneration, proteoglycans of the NP are progressively lost with poor hydrodynamic transfer. Simultaneously, the integrity of the AF is degraded and radial fissures are generated [3]. The CEPs are also affected by the degenerative process accompanying with ossification. As known, the NP is an avascular tissue in the body;

the nutrient supply depends on the capillaries of the surrounding tissue, and then diffuses through CEPs into the inner cells. Once degeneration, the regenerative ability of the NP is limited due to the low cell proliferation rate and insufficient nutrient supply [5]. Recent studies indicated that the disc degeneration originates in the NP with loss of cell number, increase in type I collagen, denaturation of type II collagen and loss of proteoglycans. In the degenerated NP, the anabolic and catabolic metabolism of the

(25)

ECMM could not

Fig.

be kept in b

Fig. 1. 3 T

1. 4 The no

balance and

The pathoph

ormal IVD

d altered the

hysiology of

(left) and d

response to

f disc degen

egenerative

o mechanica

neration [4]

e IVD (right

al loading [

t) [1]

6-8].

(26)

1.3 Current treatment options for disc degeneration

Disc degeneration is strongly associated with back pain and herniation which results in an increase in health care costs. Disc degeneration could be divided into 5 stages according to the Thompson’s classification (Table 1.1). In the first and second stage, there are no significant symptoms but can be traced by magnetic resonance imaging (MRI). Because of the decreased water content in the NP, the T2-weighted MRI signal of the disc changes dramatically from a bright to a gray signal. Generally, no aggressive treatment will be processed in the clinics [9]. In the later stage, disc degeneration may cause low back pain because herniated NP may compress the nerve root. Current clinical treatments for disc degeneration include medication, physical therapy, fusion, artificial disc replacement and discectomy. Whenever possible, doctors prefer treatment other than surgery. For patients who serve pain despite prolonged non-surgical treatment, surgical intervention will be considered. However, these treatments attempt to relieve pain rather than repair the degenerative disc [10-11]. Novel biological treatments are under investigation to treat degenerative disc in the early stages of degenerative process by promoting synthesis or inhibiting degradation of ECM, which have gained more attention in recent years [12-13]. Biological therapy strategies for disc degeneration include protein injection, therapeutic gene transfer and cell-based

(27)

therapy. Direct injection of growth factors or cytokine inhibitors may present for a short period of time that may only treat degenerative disc for short term. Therapeutic gene transfer is an approach to maintain high-level expression of growth factor in the target tissue [14-15]. In gene therapy, therapeutic gene is introduced into target cells which can continue to produce the desired ECM. There are two common methods of gene therapy which are in-vivo (direct) and ex-vivo (indirect) gene transfer. In vivo gene therapy introduces vectors containing therapeutic gene directly into the degenerative disc. In the ex-vivo method, therapeutic gene is transferred into cells which are harvested from body and then inject these cells into the target tissue [16]. However, the cell density in the normal disc is low that result in the low efficiency of gene therapy for disc degeneration. To overcome this limitation, cell-based therapy is an approach to transplant new cells in the degenerative disc that can combine with growth factor and scaffold to provide the long-term treatment for disc degeneration [14-15].

(28)

Table 1. 1 Thompson grade in the degenerative disc [17]

(29)

CHAPTER 2 THEORETICAL BASIS

2.1 The pathophysiology of disc degeneration

In the normal disc, there is a balance between matrix synthesis and degradation. In the degenerative disc, the homeostasis balance becomes disrupted, leading to diminished synthesis of disc matrix protein and increased expression of catabolic enzymes and inflammatory mediators [18]. Disc degeneration is generally believed that originate in the NP with decrease of cell number, decrease of type II collagen and loss of proteoglycans. In the degenerative NP, cells may lose their phenotype and change extracellular matrix (ECM) composition by decreasing anabolism or increasing catabolism [3, 8]. It has been suggested that degenerative process is accelerated by catabolic factors such as matrix metalloproteinases (MMPs), pro-inflammatory mediators and apoptotic factors. The MMP family degrades at least one component of ECM and can be divided into four main subfamilies: collagenase, stromelysins, gelatinases and membrane metalloproteinases. Collagenases (MMP-1, MMP-8 and MMP-13) degrade the collagen type I, II and III. Stromelysins (MMP-3) degrade many components including proteoglycans, laminin, fibronectin, gelatin, and Type II, III, IV

(30)

Tissue inhibitors of metalloproteinases (TIMPs) specifically inhibit active forms of MMPs. TIMP-1 specifically inhibits the activity MMP-3. The imbalance between MMP-3 and TIMP-1 is strongly associated with disc degeneration that is attracting research attention [2, 19-20].

It is well known that IVDs which cause low back pain secrete high level of pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), inducible nitric oxide synthase (iNOS) and reactive oxygen species (ROS) that promote the production of MMPs and then inhibit the synthesis of the matrix []. Disc cells can be stimulated by IL-1 into increasing their synthesis of MMPs, NO and IL-6 [1997]. The activity of MMP-3 can be enhanced by IL-1 and TIMP-1 production can be decreased by IL-6 that can accelerate degenerative process [19-20].

Anabolic factors, such as insulin-like growth factor (IGF), bone morphogenic protein (BMP) and transforming growth factor (TGF) have been widely studied in recent years (Table 2. 1). Recent studies show that NP cells are capable of producing IGF-1, BMP-7 and TGF-β to promote matrix synthesis. IGF-1 is a single chain polypeptide that can enhance proteoglycan synthesis. BMPs are subfamily of the TGF-β superfamily. BMP-7 that can stimulate NP cells to produce proteoglycans and type II collagen. TGF-β is a multifunctional growth factor that regulates cell growth and matrix

(31)

formation [21]. However, the role of TGF-β in the NP remains controversial. More recent studies show that TGF-β is highly expressed in the matrix of degenerated disc and the up-regulation of TGF-β might relate to inflammatory reaction. They suggested that the strong expression of TGF-β reflects that the disc cells attempt to repair the degraded ECM [22-24]. In addition, it has been shown that ROS can stimulate the expression of TGF-β in different types of cells via activation of MMPs [25-28].

(32)

Table 2. 1 The effect of growth factors on disc cells [21]

(33)

2.2 Reactive oxygen species and oxidative stress

Reactive oxygen species (ROS) are generated by all aerobic cells and is known to participate in a wide variety of molecular and biochemical processes and to directly cause some of the changes observed in cells during differentiation, aging, and transformation [29]. The overproduction of ROS leads to oxidative stress which activates numerous signaling pathways and modulates the function of many enzymes and transcription factors. As known, ROS could be beneficial or detrimental to cell proliferation and differentiation [30]. ROS might be generated as a result of normal intracellular metabolic activity in mitochondria and peroxisomes, as well as from a series of cytosolic enzyme systems. An intricate enzymatic and non-enzymatic antioxidant defense system such as catalase, superoxide mutase, and glutathione peroxidase counteracts regulates overall ROS levels to keep cell in homeostasis. If ROS below the homeostatic set point, it may interrupt the cellular proliferation [31].

Similarly, increased ROS may lead to cell death or to senescence. Generally, the impairment caused by increased ROS is thought to result from random damage to proteins, lipids and nucleus acids. In addition, a rise in ROS levels might induce a stress signal that activates specific redox-sensitive signaling pathways [32-33]. Major

(34)

factor-kappa B (NF-κB), heat-shock transcription factor 1 (HSF1), PI(3)K/Akt, mitogen-activated protein (MAP) kinase and p53 (Fig. 2. 1) [31-33].

The overproduction of ROS led to oxidative stress are commonly associated with apoptosis and senescence of the NP cells that plays an important role to the decrease in number of NP cell during degeneration; moreover, senescent cells may lose their phenotype to produce inadequate ECM in degenerated disc [34-36].

Apoptosis can be initiated via extrinsic (mitochondrial-independent) or intrinsic (mitochondrial-dependent) pathway which depends on apoptotic stimuli [37]. Extrinsic pathway is initiated by binding of specific protein ligand (TNF-α superfamily) to the cell surface receptor that can induce the recruitment of fas-associated protein with death domain (FADD) and procaspase-8 into death-inducing signaling complex (DISC).

Activation of caspase-8 leads to activation of caspase-3 which results in the apoptosis.

The ROS-mediated apoptosis is through the intrinsic pathway, which results in the release of cytochrome c from mitochondria. The complex of cytochrome apoptotic protease activating factor-1 (c-Apaf-1) activates the procaspase-9 which leads to activation of caspase-3 (Fig 2.2). The intrinsic pathway can also be activated via the extrinsic Type II pathway through caspase 8-mediated cleavage of the inactive cytosolic protein Bid. Once activated, truncated-Bid (t-Bid) translocates to the mitochondria

(35)

where it stimulates cytochrome c release. The activation of caspase-3 in both intrinsic and extrinsic pathways cleaves various cellular substrates that results in morphologic changes in cells and nuclei associated with apoptosis. Both apoptotic pathways have been found to occur in degenerative disc, with static mechanical loading-induced degeneration being mediated via intrinsic pathway and disc herniation being mediated via extrinsic pathway [38-41].

Cell senescence occurs when normal cells stop dividing. It has been suggested that cell senescence plays a role in disc degeneration. The senescence state is a response to specific stimulus or signaling pathway, including oxidative stress, DNA damage, telomere uncapping, and oncogene activation etc. There are two common signaling pathways of senescence. In the p16-pRB pathway, senescence stimuli activate p16 which lead to activation of pRB. In the p53-p21-pRB pathway, senescence stimuli activate p53, which then induce senescence by activation of pRB through p21.

Senescent NP cells may alter gene expression and decrease cell renewal that compromise tissue homeostasis and function (Fig. 2. 3). Recent studies indicated that p53-p21-pRB pathway play a more important role than p16-pRB pathway; moreover, the p53-p21-pRB pathway is reversible. Therefore, prevention of senescence of NP cells

(36)

Fig. 2. 1 Major signaling pathways activated in response to oxidative stress [31]

(37)

Fig. 2 [31]

Fig. 2

2. 2 Compo

2. 3 The hy

onents of ca

ypothetical m

aspase-depen

model for N

ndent intrin

NP cell senes

nsic and extr

scence in de

rinsic apopt

egenerative

tosis pathwa

disc [34]

ays

(38)

2.3 Polyphenol: ferulic acid

Polyphenols are naturally occurring compounds found largely in the fruits and vegetables. Polyphenols are classified on the basis of the number of phenol rings they contained and of the structural elements binding these rings to one another. They have several hydroxyl groups on aromatic ring(s) and are commonly dived in four classes including phenolic acids, flavonoids, stilbenes and lignans. Ferulic acid (4-hydroxy-3-methoxy cinnamic acid) (FA) is a phenolic acid most abundant in vegetables, especially in eggplants and maize bran [42]. FA possesses the resonance structure and has been found to be an excellent antioxidant (Fig. 2. 4). The mechanism of antioxidant action of FA is that phenolic nucleus and unsaturated side chain of FA can form a resonance stabilized phenoxy radical. Phenoxy radical is highly resonance stabilized since the unpaired electron may be present not only on the oxygen but also it can be delocalized across the entire molecule. This phenoxy radical is unable to initiate a radical chain reaction, and the most probable fate of phenoxy radical is a condensation with another ferulate radical to yield the dimmer curcumin (Fig. 2. 5). Such coupling may lead to a host product, all of which still contain phenolic hydroxyl groups capable of radical scavenging [43-44].

The anti-inflammatory effect of FA has been demonstrated in recent years. FA

(39)

decreases the level of inflammatory mediators including ROS, NO and prostaglandin E2. Oxidative stress induced inflammation is mediated by NF-κB activation, MAP kinases and affect a wide variety of cellular signaling processes leading to generation of inflammatory mediators such as IL-1b, IL-8, TNFα and iNOS and chromatin remodeling (Fig. 2. 6). Polyphenol family inhibits pro-inflammatory gene expression via inhibition of IκB, thus inhibiting NF-κB transactivation, as well as restoring

transrepressive pathways through the activation of histone deacetylases [45].

FA has been proven to have ability to prevent ROS-induced diseases and to have protective effect on hydroxyl radical oxidation in neuronal cells [46], nicotine-induced DNA damage in blood lymphocytes [47], UV-induced oxidative stress in human lymphocytes [48], γ-radiation induced DNA damage [49] and lipid peroxidation in rat hepatocytes [50].

Fig. 2. 4 Resonance structure of FA [43]

(40)

Fig. 2. 5 A schematic model for polyphenols and flavonoids mediated modulation of cell signaling [45]

(41)

2.4 In situ forming hydrogel

In the past few years, in situ-forming hydrogel have be widely studied for biomedical application. The mechanism of in situ-forming hydrogel formation includes solvent-exchange, UV-irradiation, ionic cross-linkage, pH change and temperature response. Thermosensitive hydrogel formation by simple sol-gel transition and without organic crosslinking agents has been increasing interest in a wide range of biomedical and pharmaceutical applications [51-52].

Chitosan-based thermosensitive hydrogel is currently a great deal of interest for drug and protein delivery [53-54]. Chitosan is a linear copolymer which composed of D-glucosamine and N-acetyl-D-glucosamine by (1, 4)-linkage (Fig. 2. 6). It is soluble in acid solution and contains free amino groups. Chenite et al. [55-57] developed thermosensitive chitosan/β-glycerophosphate (C/GP) system that has been widely used in drug and protein delivery. In the system, the gelation time, gel strength and gelation temperature can be controlled by the GP concentration, chitosan concentration and deacetylation degrees of chitosan [56]. The mechanism of sol/gel transition in the C/GP system includes hydrophobic interaction, hydrogen bonding, electrostatic interaction and molecular chain movement. The effective interactions responsible for the sol/gel

(42)

interchain hydrogen bonding as a consequence of the reduction of electrostatic repulsion due to the basic GP addition, (2) the electrostatic attraction increased between chitosan and GP via the amino- and the phosphate groups, respectively, and (3) the hydrophobic interactions between chitosan and chitosan molecules should be enhanced by the structuring action of glycerol on water [55-57]. Roughley et al. [58] suggested that lack for firm structure of the C/GP hydrogel may not be suitable for cell culture. The gel strength of the C/GP hydrogel could be increased by added other polymers. In our previous study [59], the thermosensitive chitosan/gelatin/glycerol phosphate hydrogel has successfully developed as a cell carrier for nucleus pulposus regeneration. The gelatin molecules were added into the C/GP solution to increase the molecular interactions through which to improve the gel strength and shorten the gelation time.

When the gelatin is added in the C/GP solution at low temperature, hydrogen bonds exist not only between the OH group of gelatin and the OH and NH2 groups of chitosan but also between gelatin and water due to the high hydrophilicity of gelatin. At the same time, the low temperature can also reduce the mobility of chitosan molecules, which further prevents the association of chitosan chains. It is thus a poor conformation or shape to build up a 3D network structure because of the difficulty of creating contacts between the junction chains. When temperature is elevated, the intermolecular hydrogen

(43)

polymer are removed. The dewatered hydrophobic chitosan chains and gelatin molecules entangled one another. As a result, a gel is formed. This type of thermosensitive gelation has also been observed in other cases [58, 60].Therefore, hydrophobic interactions are assumed to be the main driving force to form the gel consisting of chitosan and gelatin at high temperature.

There are several different approaches to drug incorporation have been developed.

Direct addition of drugs to hydrogels can be accomplished by encapsulation. The therapeutic molecules can be allowed to release from hydrogel by diffusion. This method show the easiest way to add drug to the hydrogel, however, typical release profiles show the burst release of the drug. The addition of therapeutic molecules before or after cross-linking of polymer can influence the release profile [61]. Incorporation of drug in another release system within hydrogel provides the way to prolong the period of release. Therapeutic molecules can be encapsulated in the microsphere within hydrogel. After hydrogel degradation, the therapeutic molecules are sustained release from hydrogel. Drug delivery system via covalent bond between drug and polymer can significantly prolong the period of release compared with the direct incorporation [61].

Covalent attached therapeutic molecules are released from hydrogels due to degradation

(44)

chito Both hydro the g bond Cova of re

osan-based h permeation

ogel via dif gradient con ding betwee alent bondin

lease for se

Fig. 2. 7

hydrogel: p n and entra ffusion. Th ncentration en therapeu ng method veral month

Fi

The structu

permeation, apment meth

e initial bur formed bet utic molecu

is suitable hs [61-63].

g. 2. 6 The

ure of β-gly

, entrapmen hod allow t urst generall tween hydr ules and hy for hydroph

structure of

ycerol phosp

nt and cova the therapeu ly occurs in rogel and su ydrogels ma hobic drug

f chitosan [6

phate disodi

alent bondi utic molecu n these two

urrounding ay overcom

and can pro

61]

ium salt hyd

ing (Fig. 2 ules release o methods d tissue. Cov me this prob

olong the p

drate [64]

. 10).

from due to valent blem.

period

(45)

Fig. 2. 8 TThree majorr approachees of drug looading for chitosan hyddrogel [61]

(46)

2.5 The purpose of study

The object of the study was to evaluate the effects of FA on H2O2-induced oxidative stress NP cells. The safety concentration of FA on NP cells was used for further study. The feasibility of use the thermosensitive chitosan/gelatin/

β-glycerophosphate (C/G/GP) hydrogel as a controlled release system of FA for NP regeneration was elucidated. Then we developed the thermosensitive FA-immobilized C/G/GP hydrogel via N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC) and N-hydroxysuccinimide (NHS) crosslinking method. The feasibility of use the

thermosensitive FA-immobilized C/G/GP hydrogel as a sustained delivery system for NP regeneration was evaluated.

In the study, the cytotoxicity was evaluated by the water-soluble tetrazolium salt (WST-1), crystal violet and lactate dehydrogenase (LDH) assays. The release profile of FA was performed by ultra violet-visible-near infrared (UV-VIS-NIR) spectrophotometer. The characterization of FA-immobilized C/G/GP hydrogel was analyzed by 2, 4, 6-Trinitrobenzenesulfonic acid solution (TNBS) assay and rheometer.

The expression of ECM related gene (type I collagen, type II collagen and aggrecan), catabolic gene (MMP-3), anti-catabolic gene (TIMP-1), anabolic gene (TGF-, BMP-7 and IGF-1) and pro-inflammatory gene (IL-1 and iNOS) were selected to examine the

(47)

effect of FA on H2O2-damaged NP cells. The cell viability was evaluated by total DNA assay and crystal violet assay. The sulfated glycosaminoglycans (GAGs) content was performed by dimethylmethylene blue (DMMB) assay and alcian blue staining. The apoptosis analysis of the NP cells was evaluated by caspase-3 activity and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining.

(48)

CHAPTER 3 MATERIALS AND METHODS

3.1 Isolation of nucleus pulposus cells

The animal study was approved by the Animal Experimentation Ethics Committee of National Taiwan University Hospital. New Zealand rabbits breed from the Animal Research Center, College of Medicine, National Taiwan University were used in this study and maintained in accordance with the guidelines for the care and use of laboratory animals. The nucleus pulposus (NP) cells were harvested from the IVD of 4 month olds New Zealand rabbits with body weight approximately 2 kg, which were scarified by overdose ketamine hydrochloride injection. The IVD were separated from the rest of tissue and directly dissected to harvest NP. The NP were firstly treated with the 10% penicillin of phosphate buffered saline (PBS) at 37 C for 10 minutes and and then resuspended in Dulbecco’s modified eagle’s medium-nutrient mixture F-12 ham medium (DMEM-F12, D8900, Sigma, USA) containing 10% fetal bovine serum (100-106, Gemini Bio-products, USA), 1% penicillin and 0.05% L-Ascorbic acid (A5960, Sigma, USA) with 0.2% collagenase (C0130, Sigma, USA) at 37 C for 18 hours. NP cells were collected and cultured in DMEM-F12.

(49)

3.2 Cytotoxicity of ferulic acid on nucleus pulposus cells

Cytotoxicity of ferulic acid (FA) on nucleus pulposus (NP) cells was performed by the WST-1 (Cell Proliferation Reagent WST-1, Roche, Germany), crystal violet (C3886, Sigma, USA) and lactate dehydrogenase assay (LDH, CytoTox96 Non-Radioactive Cytotoxicity Assay, Promega, USA). The procedures were briefly described as follows.

FA was dissolved in dimethyl sulfoxide (DMSO, D2438, Sigma, USA) and added to DMEM-F12 until final concentration of 0.1 M; that would be used as a stock solution.

NP cells were seeded on the 96-well cell culture plates (92096, TPP, USA) with the density of 5000 cells per well and cultured in DMEM-F12. After incubated for 18 hours, cells were washed with PBS; the 5 μM, 50 μM, 500 μM and 5000 μM of FA (46280, Fluka, USA) in DMEM-F12 were then separately added to the culture well (200 μl per well) as culture medium throughout the culture. The groups of NP cells cultured in 5 μM, 50 μM, 500 μM and 5000 μM of FA would be abbreviated as FA5, FA50, FA500, and FA5000, respectively. WST-1 was measured on day 1 and day 3 to evaluate the cell viability. The absorbance was measured at 450 nm by the enzyme-linked immunosorbent assay reader (ELISA, Sunrise remote, TECAN, USA).

Crystal violet assay was used to evaluate the cell number. Crystal violet was

(50)

3, cells were washed with PBS, and 50 μl of 0.2% (weight percent) crystal solution was added in the culture well for 10 minutes. The crystal violet dye was carefully washed in running water, and then the 100 μl of 33% (volume percent) acetic acid was added in the dry well. The absorbance was measured by the ELISA reader at the wavelength of 570 nm.

LDH assay was used to evaluate the cytotoxicity of FA on NP cells. Released LDH in culture supernatant was measured with a 30-minute coupled enzymatic assay and measured by the ELISA reader at the wavelength of 490 nm. The percentage of cytotoxicity was calculated by the following equation:

……….. (1)

(51)

3.3 Chemiluminescence assay for reactive oxygen species production

The effective concentration of FA for 100 μM H2O2 induced oxidative stress was determined by chemiluminescence (CL) assay using 0.2 mM of luminol (09253, Fluka, USA). The CL measurement was performed by Multi Luminescence Spectrometer (Tohoku Electronic Industrial Co., Ltd, Japan). Oxidative stress was induced by 100 μM H2O2 and the 0.5 μM, 1 μM, 5 μM, 50 μM and 500 μM of FA were then separately added to the 35 mm culture dish (T2881-6, Greiner, USA). The sample without treatment was abbreviated as Control group. 100 μM H2O2 induced oxidative stress without FA further treatment was designed and abbreviated as H100 group. Oxidative stress was induced by 100 μM H2O2 and then further treated with 0.5 μM, 1 μM, 5 μM, 50 μM and 500 μM of FA would be abbreviated as H100FA0.5, H100FA1, H100FA5, H100FA50 and H100FA500, respectively. The culture dish was placed in chamber of spectrometer for background measurement for 50 sec and 1 ml of luminol was then added. The result of reactive oxygen species (ROS) production was expressed as CL counts.

(52)

3.4 The effects of ferulic acid on hydrogen peroxide-induced oxidative stress nucleus pulposus cells

3.4.1 Induction of oxidative stress and ferulic acid treatment

The NP cells were seeded in the 6-well cell culture plates with the density of 10⁵ cells per well and cultured in DMEM-F12. After incubated for 18 hours, cells were washed with PBS, oxidative stress on NP cells was induced by H2O2 (1275, RDH, USA).

The NP cells treated with 100 μM H2O2 for 30 minutes without FA further treatment was designed and abbreviated as H100 group. The NP cells with 100 μM H2O2 for 30 minutes and then further treated with 500 μM FA for 2 hours was designed and abbreviated as H100FA500 group.

3.4.2 RNA extraction and gene expression

The NP cells were collected and total RNA was extracted by RNeasy Protect Mini kit (74104, QIAGEN, Germany). Total RNA yield was quantified by the spectrophotometer at the wavelength of 260 and 280 nm. The ratio of 260 to 280 nm was between 1.6 and 2.0. RNA was treated in RNase-free water and stored in -80 C for RT-PCR assay. The complementary DNA (cDNA) was synthesized from RNA and

(53)

SuperScript^TM III First-Strand Synthesis System (18080-051 Invitrogen, USA) for reverse transcription polymerase chain reaction (RT-PCR, PTC-200, MJ Research, USA). In the first-Strand complementary DNA Synthesis, the RNA mixture of 1 μl of 50 ng per μl random hexamers, 1 μl of 10 mM dNTP mix and 8 μl of total RNA was incubated to 65 C for 5 minutes. The mixture was then kept in ice for 2 minutes. The primer mixture was composed of 2 μl of 10X RT buffer, 4 μl of 25 mM MgCl2, 2 μl of 0.1 M DTT, 1 μl of 40 U per μl RNaseOUT and 1 μl of 200 U per μl SuperScript III.

The primer mixture was added to 10 μl of RNA mixture. The RNA-primer mixture was incubated at 25 C for 10 minutes and then incubated at 50C for 50 minutes. The reaction was terminated at 85C for 5 minutes. The RNA-primer mixture was kept in ice for 4 minutes, and 1μl of RNase H was added and incubated at 37C for 20 minutes.

The volume of the PCR mixture of single reaction was 20 μl, which included of 1 μl of primer solution, 9 μl of cDNA and 10 μl of 2X TaqMan universal PCR master mix (4304437, ABI, USA). The target genes of real-time PCR are listed in Table 1. Reaction was performed by ABI PRISM 7700 sequence detection system and ABI PRISM 7700 sequence detection software 1.9.1. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as endogenous housekeeping gene. The relative expression of each

(54)

3.4.3 Total DNA quantification

At the end of 3-day culture, NP cells were collected and total DNA was purified by use of the DNeasy Blood and Tissue kit (69504, QIAGEN, Germany). DNA yield was quantified by ultra violet-visible-near infrared (UV-VIS-NIR) spectrophotometer (DU 7500, Beckman, USA) at the wavelength of 260 nm.

3.4.4 Caspase-3 activity

CaspACE assay (G7220, Promega, USA) was used to evaluate the caspase-3 activity. At the end of 1-day culture, 30 μg of total protein of each sample was added to reaction buffer (32 μl of caspase assay buffer, 2 μl of DMSO and 10 μl of 100 mM DTT), and 2 μl of DEVD-pNA was then added in a 96-well microplate. After reacted for 6 hours at 37 ºC, the buffer was analyzed by ELISA reader at 405 nm for pNA measurement. The amount of pNA is directly proportional to the concentration of caspase-3 which was calculated by the calibration curve of pNA.

(55)

3.4.5 TUNEL staining

TUNEL staining (S7111, Chemicon, USA) was used to detect apoptotic nuclei. NP cells was seeded in the 4-well chamber slides (154526, Lab-Tek II, USA) at the density of 5 x 10⁴cells per well for further induction of oxidative stress and post-treatment of FA. After incubation for 1 day, the slides were washed in 2 changes of PBS and fixed in 4% paraformaldehyde solution for 10 minutes and washed in 2 changes of PBS. The equilibration buffer was added for 10 seconds and working strength TdT enzyme solution (77 μl reaction buffer and 33 μl TdT enzyme) was added for 1 hour at 37 ºC.

The slides were immersed in stop buffer (1 ml stop buffer and 34 ml distilled de-ionized water) for 10 minutes at room temperature and washed in 3 changes of PBS and then incubated with working strength anti-digoxigenin conjugate solution (68 μl blocking solution and 62 μl anti-digoxigenin conjugate solution) for 30 minutes at room temperature and washed in 4 changes of PBS.

(56)

3.5 Thermosensitive chitosan/gelatin/β-glycerol phosphate (C/G/GP) hydrogel as a controlled release system of ferulic acid (FA) for nucleus pulposus regeneration

3.5.1 Preparation of thermosensitive C/G/GP hydrogel

2.5% chitosan (degree of deacetylation > 95%, molecular weights ≒ 340,000, Kiotek, Taiwan) with 1%, 1.5% or 2% gelatin (G1890, Sigma, USA) were dissolved in 0.1 M acetic acid (242853, Sigma, USA) and sterilized by autoclave. Glycerol 2-phosphate disodium salt hydrate (β-GP, G6251, Sigma, USA) was dissolved in deionized water (0.8 w/v) and filtered by 0.22 μm filter (Millex-GV, Millipore, USA) for sterilization. The β-GP solution was added into the chitosan/gelatin solution dropwise under stirring and adjusted the pH value to 7.4. The C/G/GP solution was stored at 4C and utilized as a cell carrier for the further study.

(57)

3.5.2 Reheological characterization

Gelation temperature, gelation time and gel strength were measured by HAAKE RheoStress 600 rheometer equipped with parallel plate sensor (PP35 Ti) in oscillatory mode. The storage (elastic) modulus G’ and loss (viscous) modulus G’’ versus temperature were measured at a gap of 0.105 mm and the frequency of 1.0 Hz. The temperature at the cross point of G’ and G’’ was defined as gelation temperature.

3.5.3 In vitro FA release study

Five hundred μM of FA-incorporated C/G/GP solution were added to the transwell (200 μl per well) mounted on 24-well plates, and 1.5 ml of PBS was added in each well and then incubated at 37 C. The 1.5 ml of PBS was collected and 1.5 ml of fresh PBS was then replenished at each time (0.5, 1, 2, 6, 24 and 48 hours). According to the absorption curve of FA, the FA content of the each sample was analyzed by ultra violet-visible-near infrared (UV-VIS-NIR) spectrophotometer (DU 7500, Beckman, USA) at the wavelength of 343 nm. The standard curve of FA was constructed at the wavelength of 343 nm. The FA concentration of each sample was determined by using a

(58)

3.5.4 Induction of oxidative stress and FA treatment

The NP cells were seeded in the 24-well cell culture plates with the density of 5 x 10⁴cells per well and cultured in DMEM-F12. After incubated for 18 hours, cells were washed with PBS, oxidative stress on NP cells was induced by 100 μM H2O2 (1275, RDH, USA) for 30 minutes. FA (46280, Fluka, USA) was dissolved in dimethyl sulfoxide (DMSO) (D2438, Sigma, USA) with final concentration of 0.5 M that would be used as a stock solution. The C/G/GP solution was added to the transwell (200 μl per well) mounted on 24-well plates (3413, Corning, USA) incubated with 100 μM H2O2-induced oxidative stressNP cells and 1.5 ml of DMEM-F12 was added in each well and then cultured at 37 C. Based on the cytotoxicity test of FA on NP cells, we have found that the 500 μM of FA might be the threshold to treat NP cells without cytotoxicity. The 100 μM H2O2-induced oxidative stressNP cells without and with 500 μM of FA in the C/G/GP hydrogel were abbreviated as Gel and Gel-FA group, respectively.

(59)

The NP cells were collected and total RNA was extracted using RNeasy protect mini kit (74104, QIAGEN, Germany). Total RNA yield was detected by the spectrophotometer at 260 and 280 nm. RNA was stored in -80 C for reverse transcription polymerase chain reaction (RT-PCR). The first strand complementary DNA (cDNA) was synthesized from RNA and SuperScript^TM III first-strand synthesis system (18080-051 Invitrogen, USA) for RT-PCR (PTC-200, MJ Research, USA) according to the instructions provided by the manufacturer. The volume of the PCR mixture of single reaction was 20 μl containing 1 μl of primer solution, 9 μl of cDNA and 10 μl of 2X TaqMan universal PCR master mix (4304437, ABI, USA). The target genes of real-time PCR are summarized in Table 1. Reaction was performed by ABI PRISM 7700 Sequence detection system and ABI PRISM 7700 sequence detection software 1.9.1. The target genes were normalized to the glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The relative mRNA expression of each target gene was determined using ^ᇞᇞCt method.

(60)

3.5.6 Total DNA quantification

At the end of 1-day culture, NP cells were collected and total DNA was purified by use of the DNeasy blood and tissue kit (69504, QIAGEN, Germany) following the instructions provided by the manufacturer. Total DNA yield was quantified by ultra violet/visible/near infrared (UV/VIS/NIR) spectrophotometer (DU 7500, Beckman, USA) at the wavelength of 260 and 280 nm. The ratio of 260 to 280 nm was between 1.6 and 2.0.

3.5.7 Sulfated glycosaminoglycan content

The sulfated-glycosaminoglycans (GAGs) production was evaluated by DMMB (341088, Sigma, USA) assay. At the end of 1-day culture, the culture medium of each sample was collected and transferred 40 μl of the supernatant of each sample to a 96-well microplate, and 250 μl of DMMB solution was then added. The DMMB-sulfated GAGs complex product was examined by ELISA reader at the wavelength of 595 nm. The sulfated GAGs content of each sample was determined by using a calibration curve which was performed by condroitin-6-sulfate (C4384, Sigma, USA).

(61)

3.5.8 Alcian blue staining

At the end of 1-day culture, cells were washed twice with PBS and fix in 10%

neutral buffered formalin (H121-08, Mallinckrodt Analytical, USA) for 30 minutes and then washed twice with PBS. 3% acetic acid was added and then washed in running water for 1 minute. Alcian blue (pH 1.0, Muto pure chemicals, Japan) was added for 30 minutes and cells were then washed in running water for 1 minute. Kernechtrot (Muto pure chemicals, Japan) was added for 5 minutes and then washed in running water for 1 minute. The cells were dehydrated in 2 changes of 95% alcohol and absolute alcohol (459844, Sigma, USA) for 1 minute each.

3.5.9 Caspase-3 activity

At the end of 1-day culture, caspase-3 activity was analyzed by CaspACE assay system (G7220, Promega, USA) according to the instructions provided by the manufacturer. Briefly, 30 μg of total protein of each sample was added to reaction buffer (32 μl of caspase assay buffer, 2 μl of DMSO and 10 μl of 100 mM DTT), and 2 μl of DEVD-pNA was then added in a 96-well microplate and incubated at 37 ºC for 4

(62)

pNA and the concentration of caspase-3 was obtained.

3.5.10 TUNEL staining

Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) staining was performed by ApopTag® plus fluorescein in situ apoptosis detection kit (S7111, Chemicon, USA). At the end of 1-day culture, cells were washed twice with PBS and fix in 10% neutral buffered formalin (H121-08, Mallinckrodt Analytical, USA) for 30 minutes and then washed twice with PBS. Cells were post-fixed in precooled ethanol/acetic acid (2:1) solution for 5 minutes at -20 °C and then washed twice with PBS. The equilibration buffer was added for 10 seconds and working strength TdT enzyme solution (70% reaction buffer and 30% TdT enzyme) was added for 1 hour at 37 ºC. The stop buffer (1 ml stop buffer and 34 ml distilled de-ionized water) was added for 10 minutes at room temperature and washed in 3 changes of PBS and then incubated with working strength anti-digoxigenin conjugate solution (53% blocking solution and 47% anti-digoxigenin conjugate solution) for 30 minutes at room temperature and washed in 4 changes of PBS.

(63)

3.6 The effects of thermosensitive ferulic acid-immobilized chitosan/gelatin/β-glycerol phosphate (FA-immobilized C/G/GP) hydrogel on nucleus pulposus cells under hydrogen peroxide-induced oxidative stress

3.6.1 Preparation of thermosensitive FA-immobilized C/G/GP hydrogel

Ferulic acid (46280, Fluka, USA) was dissolved in dimethyl sulfoxide (DMSO, D2438, Sigma, USA) with concentration of 0.1 M.

N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC, E1769, Sigma, USA) and

N-hydroxysuccinimide (NHS, 56480, Fluka, USA) were dissolved in water with

concentration of 1 M and 0.25 M respectively. 8% gelatin (G1890, Sigma, USA) and 2.5% chitosan (degree of deacetylation > 95%, molecular weights ≒ 340,000, Kiotek, Taiwan) were dissolved in water and 0.1 M acetic acid (242853, Sigma, USA) respectively. Both gelatin and chitosan solution were sterilized by autoclaving at 121 C for 30 minutes. FA-immobilized gelatin solution was prepared by mixing 100 μl of 1 M EDC and 400 μl of 0.25 M NHS with 1 ml of 0.01 M FA and vortexed for 1 hour. The mixture was filtered by 0.22 μm filter (Millex-GV, Millipore, USA) for sterilization and then mixed with 1ml of 8% gelatin solution and vortexed for 1 hour. FA-immobilized gelatin solution was then mixed with 2.5% chitosan solution to form the

(64)

hydrate (β-GP, G5422, Sigma, USA) solution was sterilized by passing through a 0.22 μm filter (Millex-GV, Millipore, USA) and added drop by drop to the FA-immobilized chitosan-gelatin solution under stirring and adjusted the pH value to 7.4. The FA-immobilized C/G/GP solution was stored at 4 C until further use.

3.6.2 Characterization of FA-immobilized gelatin

The 2, 4, 6-trinitrobenzenesulfonic acid solution (TNBS, P2297, Sigma, USA) assay was used for the detection of primary amino groups. The residual amino group of both gelatin (without immobilization of FA) and FA-immobilized gelatin were analyzed by TNBS assay. 10 μl of sample was mixed with 90 μl of 0.1 M sodium hydroxide (221465, Sigma, USA). The mixture was transferred into a 96-well microplate and reacted with 50 μl of 0.1% TNBS for 2 hours at 37 C. After 2 hours, 75 μl of stop solution containing 50 μl of 10% SDS (L4522, Sigma, USA) and 25 μl of 1 N HCl was added to terminate the reaction. The optical density (OD) value was measured at 420 nm using an enzyme-linked immunosorbent assay reader (ELISA, Sunrise remote, TECAN, USA). The residual amino group content of sample was determined by using a linear standard curve which was constructed by glycine (G7126, Sigma, USA).

(65)

3.6.3 Reheological characterization

Gelation temperature and gelation time of the FA-immobilized C/G/GP hydrogel were measured by HAAKE RheoStress 600 rheometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) equipped with parallel plate sensor (PP35 Ti) in oscillatory mode.

The elastic modulus (G’) and viscous modulus (G’’) versus temperature were measured at a gap of 0.105 mm and the frequency of 1.0 Hz. The gelation temperature and gelation time were defined at which the G’ becomes larger than the G”.

3.6.4 Cytotoxicity of thermosensitive FA-immobilized C/G/GP hydrogel on NP cells

Cytotoxicity of FA-immobilized C/G/GP hydrogel on NP cells was performed by extraction method. The 0.1 g FA-immobilized C/G/GP hydrogel was immersed in 1 ml DMEM-F12 in a 48-well culture plate at 37 C. The supernatant from each well was collected on day 3 for cytotoxicity test. NP cells were seeded in 96-well cell culture plates at a density of 5,000 cells per well and cultured in DMEM-F12 for 18 hours.

Cells were then cultured in the extraction medium obtained from the developed hydrogel. WST-1 (Cell Proliferation Reagent WST-1, Roche, Germany) and lactate

(66)

NP cells on day 1 and day 3. The OD value of WST-1 and LDH assay were measured at 450 and 490 nm with an ELISA reader respectively. The percentage of cytotoxicity was calculated by the following equation:

100

(%) ^exp 



 

medium lysis

total

medium

OD OD

OD ty OD

Cytotoxici

3.6.5 In vitro FA release study

Two hundred μl of FA-immobilized C/G/GP solution or C/G/GP (without immobilization of FA) was added to the transwell mounted on 24-well plates (3413, Corning, USA) and 1.5 ml of PBS was then added in each well and incubated at 37 C.

The 1.5 ml of PBS was collected and 1.5 ml of fresh PBS was then added at each time (0.5, 1, 2, 6, 24 and 48 hours). The content of FA was evaluated by ultra violet-visible-near infrared (UV-VIS-NIR) spectrophotometer (DU 7500, Beckman, USA) at the wavelength of 343 nm according to the absorption spectrum of FA. The FA concentration of each sample was calculated from the linear standard curve of FA.

(67)

3.6.6 Induction of oxidative stress and FA-immobilized C/G/GP hydrogel treatment

The NP cells were seeded in the 24-well cell culture plates with the density of 5 x 10⁴cells per well and cultured in DMEM-F12. After 18 hours, cells were washed with PBS and 1.5 ml of DMEM-F12 was then added. Oxidative stress on NP cells was induced by 100 μM H2O2 (1275, RDH, USA) for 30 minutes. The 200 μl of FA-immobilized C/G/GP or C/G/GP (without immobilization of FA) solution were added to the transwell mounted on 24-well plates incubated with 100 μM H2O2-induced oxidative stress NP cells and then cultured at 37 C. The 100 μM H2O2-induced oxidative stress NP cells with C/G/GP hydrogel and with FA-immobilized C/G/GP hydrogel were abbreviated as C/G/GP and FA-immobilized C/G/GP group respectively.

(68)

The total RNA was extracted from NP cells using RNeasy protect mini kit (74104, QIAGEN, Germany) after 2.5 hour. The OD ratio (260/280 nm) of each RNA sample was between 1.8 and 2.0. RNA was stored in -80 C for reverse transcription polymerase chain reaction (RT-PCR). The first strand complementary DNA (cDNA) was synthesized from RNA and SuperScript^TM III First-Strand Synthesis System (18080-051, Invitrogen, USA) for RT-PCR (PTC-200, MJ Research, USA) in accordance with the manufacturer’s instruction. 1 μl of primer, 9 μl of cDNA and 10 μl of 2X TaqMan universal PCR master mix (4304437, ABI, USA) were mixed in a final volume of 20 μl for single reaction. Reaction was performed by ABI PRISM 7700 sequence detection system and ABI PRISM 7700 sequence detection software 1.9.1.

The target genes of real-time PCR reaction were summarized in Table 1 and each target gene was calibrated for glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The relative quantitation value of gene expression was determined using ^ᇞᇞCt method.