OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

116
UNIVERSITATIS OULUENSIS MEDICA ACTA D OULU 2009 D 1018 Olli-Pekka Kämäräinen THE SEARCH FOR SUSCEPTIBILITY GENES IN OSTEOARTHRITIS FACULTY OF MEDICINE INSTITUTE OF BIOMEDICINE, DEPARTMENT OF MEDICAL BIOCHEMISTRY AND MOLECULAR BIOLOGY, UNIVERSITY OF OULU; BIOCENTER OULU, CENTER FOR CELL MATRIX RESEARCH, UNIVERSITY OF OULU

Transcript of OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

Page 1: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9143-2 (Paperback)ISBN 978-951-42-9144-9 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1018

ACTA

Olli-Pekka K

ämäräinen

OULU 2009

D 1018

Olli-Pekka Kämäräinen

THE SEARCH FOR SUSCEPTIBILITY GENESIN OSTEOARTHRITIS

FACULTY OF MEDICINEINSTITUTE OF BIOMEDICINE,DEPARTMENT OF MEDICAL BIOCHEMISTRY AND MOLECULAR BIOLOGY,UNIVERSITY OF OULU;BIOCENTER OULU,CENTER FOR CELL MATRIX RESEARCH,UNIVERSITY OF OULU

Page 2: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …
Page 3: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 1 0 1 8

OLLI-PEKKA KÄMÄRÄINEN

THE SEARCH FOR SUSCEPTIBILITY GENES IN OSTEOARTHRITIS

Academic dissertation to be presented with the assent ofthe Faculty of Medicine of the University of Oulu forpublic defence in Auditorium 101 A of the Faculty ofMedicine (Aapistie 5 A), on 11 June 2009, at 12 noon

OULUN YLIOPISTO, OULU 2009

Page 4: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

Copyright © 2009Acta Univ. Oul. D 1018, 2009

Supervised byDocent Minna Männikkö

Reviewed byDocent Jari ArokoskiDocent Janna Saarela

ISBN 978-951-42-9143-2 (Paperback)ISBN 978-951-42-9144-9 (PDF)http://herkules.oulu.fi/isbn9789514291449/ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)http://herkules.oulu.fi/issn03553221/

Cover designRaimo Ahonen

OULU UNIVERSITY PRESSOULU 2009

Page 5: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

Kämäräinen, Olli-Pekka, The search for susceptibility genes in osteoarthritisFaculty of Medicine, Institute of Biomedicine, Department of Medical Biochemistry andMolecular Biology, University of Oulu, P.O.Box 5000, FI-90014 University of Oulu, Finland;Biocenter Oulu, University of Oulu, P.O. Box 5000, FI-90014 University of Oulu, Finland;Center for Cell Matrix Research, University of Oulu, P.O.Box 5000, FI-90014 University ofOulu, Finland Acta Univ. Oul. D 1018, 2009Oulu, Finland

AbstractThis work engaged Finnish females affected with osteoarthritis (OA) of the hand to define the roleof common sequence variations within the genes of the important structural protein of cartilage,aggrecan (AGC1), and the genes of inflammatory mediators, the interleukin 1 gene cluster andinterleukin 6 (IL6), as possible risk factors for the disease. Also, a genome-wide linkage analysiswas performed in a sample consisting of Finnish families with multiple individuals affected withhip and knee OA in order to reveal new chromosomal areas that are likely to contain diseaseassociated variations.

OA is a chronic disease that leads to the degeneration of articular cartilage in synovial joints.The etiology of the disease is for the most part unknown. Joints of the hand, hip and knee are mostcommonly affected, and obesity, trauma and excess mechanical stress are known risk factors forthe disease. OA also has a significant genetic component.

AGC1 carries a variable number of tandem repeats (VNTR) polymorphism, which may besignificant for the biomechanical properties of cartilage. It was shown that the most common allelewith 27 tandem repeats is protective against hand OA (HOA) (odds ratio 0.46, 95% confidenceinterval 0.27–0.78). Also, carrying two copies of any of the shorter or longer alleles increased therisk of the disease.

Inflammation seems to play a role in the etiology of OA and certain polymorphisms within theinterleukin 1 gene cluster and IL6 have been previously shown to increase the transcription ofthese molecules and to associate with OA. In this study it was shown that the G alleles in threecommon IL6 promoter single nucleotide polymorphism (SNP) sites are associated with the risk ofmore severe forms of HOA (p = 0.001 for GGG haplotype). A SNP in IL1B associated with thebilateral form of the disease (p = 0.006) and two IL1B-IL1RN extended haplotype alleles wereassociated with the same phenotype.

Genome-wide and fine mapping linkage analyses recognized chromosomal locus 2q21 with amultipoint LOD score of 3.96. Despite the association analyses of several candidate genes withinthe locus, no disease-associating sequence variants were identified.

Keywords: inflammation, linkage analysis, Osteoarthritis - genetics

Page 6: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …
Page 7: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

To my family

Page 8: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

6

Page 9: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

7

Acknowledgements

This study was carried out at the Oulu Center for Cell-Matrix Research, at the Department of Medical Biochemistry and Molecular Biology, University of Oulu, during the years 2003–2009.

I am deeply grateful to my supervisor, Docent Minna Männikkö, for guiding me through this work with her expert advice, constant support and never-ending optimism. I wish to thank her for all the time and effort she has put into my work and for being encouraging throughout these years. Professor Leena Ala-Kokko deserves my warmest appreciation for introducing me to the world of medical science.

I am thankful for the excellent facilities and scientific atmosphere created by Professors Taina Pihlajaniemi, Johanna Myllyharju, Seppo Vainio, Docent Peppi Karppinen and Research Professors Emeritus Kari Kivirikko and Ilmo Hassinen.

Docents Jari Arokoski and Janna Saarela are acknowledged for their valuable comments on the thesis. Sandra Hänninen, M.Sc., deserves thanks for the careful revision of the language of the thesis.

I would like to thank Docent Päivi Leino-Arjas and Svetlana Solovieva, Ph.D., for their opinions and essential contributions in preparing the original articles. Svetlana Solovieva is especially acknowledged for her extraordinary statistical work. I appreciate the opportunity of working with Eveliina Jakkula, M.D., Ph.D., and Mari Taipale, M.Sc.. Other collaborators and research group members, both past and present, are also acknowledged.

I am grateful to Aira Harju, Minta Lumme, Helena Satulehto and Irma Vuoti for their expert assistance in the laboratory. Special thanks for the valuable help with practical matters go to Pertti Vuokila, Auli Kinnunen, Marja-Leena Karjalainen and Seppo Lähdesmäki. Risto Helminen is acknowledged for his technical assistance with the computers.

I want to express my warmest gratitude to my parents, Reijo and Vuokko, for setting me an example and providing a loving family throughout the years. I cannot thank you enough for giving me such an excellent basis for my life and always being there when needed. I also value the friendship and joyful memories with Kaisa, Antti and Satu, my dear siblings. Parents-in-law and colleagues Jaakko and Raija are to thank for their support and precious times spent together. I also want to thank my good friend Pasi and other mates who walked beside me during the long and challenging path to becoming a medical doctor. Mia and

Page 10: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

8

Kimmo, my dear friends, are also to be thanked for their friendship and all their help.

Finally, without the love and encouragement of my beloved wife this work would have never been accomplished. Minna, you have lifted me up and made me a better man. I cherish everything we have experienced while going through medical school together and starting a family. I cannot think of a better mother for our precious little son Kalle. You truly are my eternal companion and best friend.

This work was supported by grants from the Finnish Medical Foundation Duodecim and the Oulu University Foundation.

Oulu, May 2009 Olli-Pekka Kämäräinen

Page 11: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

9

Abbreviations

AC articular cartilage ADAMTS a disintegrin and metalloproteinase with thrombospondin motifs BMP bone morphogenetic protein CDMP1 cartilage-derived morphogenetic protein 1 Chr chromosome CI confidence interval CILP cartilage intermediate layer protein cM centimorgan CMC carpometacarpal COL collagenous domain COMP cartilage oligomeric matrix protein CS chondroitin sulfate DIP distal interphalangeal ECM extracellular matrix ERT estrogen replacement therapy FACIT fibril-associated collagen with interruptions in the triple helix FGF fibroblast growth factor G guanine nucleotide base GAG glycosaminoglycan GDF-5 growth/differentiation factor 5 GWA genome-wide association HA hyaluronan HOA hand osteoarthritis hr human recombinant HS heparan sulfate ICD international classification of diseases ICE IL-1β converting enzyme IGD interglobular domain IGF insulin-like growth factor IHH Indian hedgehog IL interleukin JSN joint space narrowing KL Kellgren & Lawrence KS keratan sulfate LD linkage equilibrium

Page 12: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

10

LIF leukemia inhibitory factor LOD logarithm of odds LP link protein MED multiple epiphyseal dysplasia MMP matrix metalloproteinase MRI magnetic resonance imaging NC noncollagenous domain NO nitric oxide OR odds ratio PCM pericellular matrix PCR polymerase chain reaction PIP proximal interphalangeal PSACH pseudoachondroplasia RA rheumatoid arthritis ROS reactive oxygen species SLRP small leucine-rich proteoglycan SNP single nucleotide polymorphism TACE TNF-α converting enzyme TGF transforming growth factor TIMP tissue inhibitor of metalloproteinase TNF tumor necrosis factor VNTR variable number of tandem repeats VWA von Willebrand factor A Wnt wingless-related MMTV integration site X any amino acid Y any amino acid

Page 13: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

11

List of original articles

This thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Kämäräinen OP, Solovieva S, Vehmas T, Luoma K, Leino-Arjas P, Riihimäki H, Ala-Kokko L & Männikkö M (2006) Aggrecan core protein of a certain length is protective against hand osteoarthritis. Osteoarthritis Cartilage 14(10): 1075–1080.

II Kämäräinen OP, Solovieva S, Vehmas T, Luoma K, Riihimäki H, Ala-Kokko L, Männikkö M & Leino-Arjas P (2008) Common interleukin-6 promoter variants associate with the more severe forms of distal interphalangeal osteoarthritis. Arthritis Res Ther 10(1): R21.

III Solovieva S, Kämäräinen OP, Hirvonen A, Hämäläinen S, Laitala M, Vehmas T, Luoma K, Näkki A, Riihimäki H, Ala-Kokko L, Männikkö M & Leino-Arjas P (2009) Association between interleukin 1 gene cluster polymorphisms and bilateral dip osteoarthritis. J Rheumatol, in press.

IV Jakkula E, Taipale M*, Kämäräinen OP*, Barral S, Kiviranta I, Kröger H, Ott J, Ala-Kokko L & Männikkö M (2009) Primary osteoarthritis of the hip and knee shows linkage to Chromosome 2q21 in Finnish families. Manuscript.

*These authors contributed equally to this study.

Page 14: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

12

Page 15: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

13

Contents

Abstract Acknowledgements 7 Abbreviations 9 List of original articles 11 Contents 13 1 Introduction 15 2 Review of the literature 17

2.1 The structure and formation of the synovial joints.................................. 17 2.2 Cartilage.................................................................................................. 20

2.2.1 General structure of articular cartilage ......................................... 20 2.2.2 The chondrocyte ........................................................................... 22 2.2.3 The extracellular matrix (ECM) ................................................... 23

2.3 Osteoarthritis (OA) ................................................................................. 35 2.3.1 Symptoms and clinical characteristics of osteoarthritis................ 35 2.3.2 Prevalence of osteoarthritis .......................................................... 39 2.3.3 Etiopathogenesis of osteoarthritis................................................. 41 2.3.4 Relevance of inflammation in osteoarthritis................................. 44

2.4 Inheritance of osteoarthritis..................................................................... 47 2.4.1 Association analyses of candidate genes ...................................... 49 2.4.2 Genome-wide studies ................................................................... 53

3 Outlines of the present study 59 4 Materials and methods 61

4.1 Study populations.................................................................................... 61 4.2 Southern hybridization............................................................................ 62 4.3 Linkage analysis...................................................................................... 63

5 Results 65 5.1 Aggrecan VNTR polymorphism and hand osteoarthritis (I) ................... 65

5.1.1 VNTR allele frequency................................................................. 65 5.1.2 The most common aggrecan VNTR allele, A27, is

protective against hand osteoarthritis ........................................... 66 5.2 Interleukin 6 promoter polymorphisms and symptomatic distal

interphalangeal osteoarthritis (II) ............................................................ 67 5.2.1 G-alleles of the promoter polymorphisms associate with

symptomatic distal interphalangeal osteoarthritis ........................ 67 5.2.2 Haplotype analysis of the promoter polymorphisms .................... 68

Page 16: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

14

5.3 Interleukin 1 gene cluster polymorphisms and bilateral distal interphalangeal osteoarthritis (III)........................................................... 69 5.3.1 Frequency of the alleles, genotypes and carriage rates of

the studied polymorphisms........................................................... 70 5.3.2 Analysis of the extended haplotypes ............................................ 72

5.4 Linkage analysis of primary hip and knee osteoarthritis (IV) ................. 73 5.4.1 Genome-wide scan ....................................................................... 74 5.4.2 Fine mapping ................................................................................ 74 5.4.3 Association analyses of potential candidate genes ....................... 75

6 Discussion 77 6.1 The complex genetics of osteoarthritis.................................................... 77 6.2 Characteristics of the study populations.................................................. 78 6.3 Is the effect of the aggrecan VNTR polymorphism significant in

the development of osteoarthritis? (I) ..................................................... 79 6.4 Genetic factors affect the inflammation reaction observed in

osteoarthritis cartilage (II, III) ................................................................. 81 6.5 Linkage analysis, a way to identify new osteoarthritis-associated

chromosomal loci (IV) ............................................................................ 83 7 Conclusions 85 References 87 Original articles 111

Page 17: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

15

1 Introduction

The synovial joint is macroscopically a rather simple structure that connects two bones together, allowing almost frictionless motion between the skeletal surfaces. The development of joints is thought to arise from the mesenchymal cells in a process that is still rather unclear and leads to the formation of the fluid-filled synovial cavity, with articular cartilage (AC) covering the ends of the bones and the surrounding synovial capsule that protects the structure. Joint cartilage is composed of a small amount of chondrocytes that synthesize the rest of the large surrounding extracellular matrix (ECM), consisting mainly of collagen II fibrils and tens of different proteoglycans and other macromolecules. The homeostasis of AC is maintained by chondrocytes, but affected also by synovial cells and osteocytes of the subchondral bone via secretion of different regulatory and signaling molecules. The structure of the collagen-proteoglycan network in AC gives the joint its biomechanical properties by allowing hydration of glycosylated macromolecules to create osmotic pressure within the tissue, empowering it to sustain continuous mechanical pressure.

The most common disease affecting the synovial joints is osteoarthritis (OA). It is a chronic, often slowly developing process that leads to the degeneration and erosion of AC, causing pain and the inability to use the affected joint. OA affects a substantial proportion of the Western populations causing personal suffering as well as substantial financial burden. The breakdown of cartilage in OA is a result of an imbalance between catabolic and anabolic processes within the AC involving the structural molecules as well as various regulatory elements. Despite the fact that many individual pathological processes taking place in OA-affected cartilage have been described, the etiopathogenesis of the disease is still, in many respects, under debate. OA can be caused by obesity, continuous excess of mechanical stress or joint trauma and it affects most often the joints of the hand, hip and knee. OA is more prevalent among women, and the overall prevalence of the disease is highly associated with aging. In a substantial amount of OA cases, no specific etiologic factor can be identified. However, the disease clearly accumulates in certain families and has been shown to involve a significant heritable component. There is currently no cure available for this condition and the process of joint degeneration cannot be stopped once it has begun. Treatment is mostly based on relieving the pain symptoms and ultimately on joint replacement therapy.

Page 18: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

16

Genes are believed to be the answer to the unknown etiology of OA. To date, dozens of genetic variations across the genome have been reported to associate with the disease. Yet, the results have been modest and no major disease-causing factor has been identified. Association analyses have not supported the idea that the genes of the main structural proteins of AC are major players in OA. More convincing associations have been reported concerning the genes of different regulatory and signaling molecules present in cartilage. Besides association analyses of the suspectibility genes, information about the genetics of OA has been sought from genome-wide linkage analyses of disease-affected families. This strategy has resulted in the identification of many chromosomal regions that are likely to harbor previously unknown OA predisposing variations. Overall, the genetics of OA have proven to be complex and the contribution of independent genetic factors may differ depending on the ethnicity and phenotype of the disease.

In this study, sequence variations in the genes of one major structural protein of AC and several significant inflammatory mediators were analyzed in order to evaluate their possible contribution to the development of OA of the hand. Also, a genome-wide linkage analysis was perfomed using families with multiple members affected with hip and knee OA. The purpose was to identify new chromosomal loci that are likely to contain previously unidentified disease-associated variants.

Page 19: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

17

2 Review of the literature

2.1 The structure and formation of the synovial joints

Synovial joints are essential for organism function, movement and quality of life. Early investigators identified four different types of joints. The synovial joint was described as eudiarthrosis, having separate articulating elements with a joint cavity limited by synovial tissues. The other three types of joints included synarthrosis, where cartilage or other connective tissue binds skeletal elements together, schizarthrosis, where the space between the skeletal elements consists of connective tissue and a number of usually small cavities, and periarthrosis, where skeletal elements unite around a single joint cavity (Haines 1941, 1942). Joints can also be classified in terms of the movement which they allow. There is no movement, for example, between cranial bones because they are connected by synarthrosis. A limited movement is possible between the vertebrae provided by the flattened discs of fibrocartilage (amphiarthroses). Only the synovial joints can permit a wide range of movement. They are also referred to as diarthroses in the current literature. The function of a synovial joint is to allow smooth articulation between opposing skeletal elements. Biomechanical loads are transmitted through this structure comprised of articular cartilage (AC) that covers the ends of the bones, synovial fluid that lubricates the cartilaginous surfaces and nourishes the avascular tissue, ligaments that bind the skeletal elements together, and a fibrous capsule that protectively encapsulates the joint (Figure 1). The synovial joint may contain meniscal structures internally. The AC is a fibrous structure and is continuous with the periosteum. It is important in stabilizing the joint. The capsule is lined by the synovial membrane, which secretes the synovial fluid required for low-friction articulation. Covering the epiphyses of the bones, there is mostly cartilage, which is responsible for maintaining the functional ability of the joint and providing a durable, load-bearing surface against extensive biomechanical strain. (For a review, see Khan et al. 2007.)

The formation of joints is a complex and highly organized process in which the synovial joint cavities must eventually generate two opposing, non-adherent surfaces facilitating painless and almost frictionless articulation. This process also involves the creation of a cell-free, fluid-filled separation (Lamb et al. 2003). The main phases of joint development are presented schematically in Figure 1 and described in more detail below.

Page 20: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

18

According to current understanding, the formation of joints begins with the migration of undifferentiated mesenchymal cells to areas destined to become bone and joints. First, in the developing embryo, the condensation of mesenchymal prechondrogenic cells in the early limb manifests no signs of joint formation. The first recognized event in joint formation is the condensation of mesenchymal cells to form the so-called interzone at the future location of the joint (Holder 1977, Mitrovic 1978). The interzone varies between species, being a thicker three-layered structure in chick bone elements (Craig et al. 1987), but rather thin in developing mammalian joints. In humans this cell layer is flattened, consisting of two to three cells (Edwards et al. 1994). The interzone can be ultrastructurally distinguished as two outer layers that are contiguous to the epiphyseal ends and a thin central intermediate zone. Articular cartilage that covers the epiphyses of the opposing bones has been thought to derive from the cells of the intermediate layer (Archer et al. 1994, Ito & Kida 2000, Pacifici et al. 2006).

Following the interzone formation and chondrogenic differentiation, the joint cavity begins to form between two cartilaginous elements that are growing against each other through forces largely generated by hypertrophy and ECM secretion within the elements. This force is generated through the upregulation of hyaluronan (HA) in the cells of the interzone, articular surface and synovium and is dependent on the movement of the embryo (Khan et al. 2007). At present, it is a widely accepted theory that the accumulation of HA contributes to the loss of tissue integrity, causing the separation of the opposing joint sides and formation of the fluid-filled cavity (Pacifi et al. 2005). Another, older theory of joint cavitation is based on selective cell degeneration and increased apoptosis within the interzone (Mitrovic 1978). After the recognition of necrotic features in the cells of the middle interzone, this theory has been widely studied. Despite several more recent reports of observed cell death in the intermediate zone of the interzone in avian and mammalian embryos (Nalin et al. 1995, Kimura & Shiota 1996, Abu-Hijleh et al. 1997), it is still under debate whether or not cell death is an important contributor to joint formation or if the local changes in the ECM and mechanical factors are independently responsible for cavitation. Details of the three-dimensional configuration, shape and organization of the joints are the least understood aspects of joint formation, likewise is the morphogenesis of the whole embryo. (For reviews, see Lamb et al. 2003, Pacifici et al. 2005 and Khan et al. 2007.)

Page 21: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

19

Fig. 1. Scheme depicting the major steps in digit synovial joint formation. (From Pacifici et al. 2005. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.)

The interzone expresses a wide variety of regulatory and signaling molecules. Wingless-related MMTV integration site (Wnt) 14, a secreted growth factor of the Wnt-gene family, is highly expressed in joint-forming regions in the interzone and is essential in initiating synovial joint formation (Hartmann & Tabin 2001). Localized production of Wnt-14 induces the expression of growth/diffentiation factor 5 (GDF-5), a mouse homolog of cartilage-derived morphogenetic protein 1 (CDMP1). It is expressed in the developing joints throughout the cavitation process and is believed to promote interzone cell function and joint development (Storm & Kingsley 1999, Shum & Nuckolls 2002). Also, bone morphogenetic protein (BMP) antagonists are considered to be important in the regulation of chondrogenesis and joint formation (Pacifici et al. 2005). In addition, Indian hedgehog (IHH), a signaling factor produced by maturing chondrocytes, is a critical and possibly direct regulator of joint development (Koyama et al. 2007).

In summary, many of the molecules that contribute to joint formation have been identified and the main phases in of the process described; however the

Page 22: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

20

detailed biology of joint formation and morphogenesis remains to be uncovered by future studies. (For more detailed reviews, see Lamb et al. 2003, Pacifici et al. 2005 and Khan et al. 2007.)

2.2 Cartilage

Three different types of specialized connective tissue, known as cartilage, exist in the human body: hyaline cartilage, elastic cartilage and fibrocartilage. Hyaline cartilage forms the cartilaginous model of the developing skeleton. It is replaced by bone in a process known as endochondral ossification. In adults, hyaline cartilage exists in the AC of joints and the cartilage of the respiratory tract (nasal, laryngeal, tracheobronchial and costal cartilage). Hyaline cartilage is composed of a small amount of chondrocytes that synthesize the rest of the large surrounding ECM, consisting mainly of collagen II fibrils and proteoglycans. (The structure of ECM and cartilaginous collagens are later reviewed in more detail.) Elastic cartilage is distinguished from hyaline cartilage by the elastic fibers it contains in the ECM. Due to elastin secreted by chondrocytes, elastic cartilage has a remarkable ability to regain its original shape after deformation. Elastic cartilage can be found in the pinna of the ear, epiglottis and in the arytenoid cartilage of the larynx. Fibrocartilage consists of chondrocytes and fibroblasts surrounded by type I collagen. It provides great tensile strength and is found in the intervertebral disks, tendinous and ligamentous insertions, menisci and the symphysis pubis. Characteristic to all these cartilage types is the lack of blood vessels, low amount of cells and a substantial ECM. (Kierszenbaum 2007.)

2.2.1 General structure of articular cartilage

The structure of articular cartilage has fascinated scientists for centuries. Considering the contemporary technology of the time, the structure of AC was described amazingly adequately as early as the mid-eighteenth century, by the famous London anatomist and surgeon William Hunter (1743). He, as well as other former scientists, had been intrigued especially by the fine composition, nourishment and stability of AC, which holds such unique biomechanical properties (Benedek 2005). Today, the general structure of this tissue is known to consist of four different horizontal layers that vary in size among different joints. The thinnest of the layers is the superficial zone beginning from the articular surface bathed by synovial fluid. Here, the density of chondrocytes is highest, and

Page 23: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

21

cells are flattened and situated parallel to the articular surface. Densely packed bundles of collagen II fibrils run between the chondrocytes. Collagen fibrils are also parallel close to the surface, contributing to the mechanical properties. Deeper in the superficial zone, collagen fibrils run vertically. The second layer is the midzone, where cells are more rounded and situated more broadly. The ECM is more extensive in the midzone and collagen fibrils are arranged randomly. Cell volume is at its lowest in the deep zone, where collagen content is minimal, but the fibril diameter is maximal. Situated between the deep zone and underlying subchondral bone is the calcified zone. The deep zone is separated from the calcified zone by a wavy, irregular line known as the tide mark. The outline of the tide mark is highly variable. Structurally, it may prevent collagen fibrils in the noncalcified zone from anchoring to the calcified zone. The calcified zone provides a buffer with intermediate mechanical properties between those of the uncalcified cartilage and the subchondral bone. Here, the chondrocytes are located in uncalcified lacunae. The collagen fibrils, large in diameter, are arranged perpendicular to the articular surface and are anchored in a calcified matrix. From the calcified layer, the dynamic forces are transmitted to the underlying subchondral bone. A schematic drawing of the orientation of collagen fibrils and chondrocytes in AC is presented in Figure 2. (For reviews, see Huber et al. 2000 and Poole et al. 2001.)

Page 24: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

22

Fig. 2. Schematic drawing of collagen fibrils and chondrocytes in cartilage. In the superficial zone, the collagen fibrils run parallel to the articular surface and lie close to each other in a dense arrangement. The collagen fibrils in the middle zone are randomly oriented and are more loosely packed. In the deep zone, the collagen fibrils orient themselves perpendicular to the subchondral bone surface. Percentages describe the estimated volume of each zone of the total cartilage volume. (From Ge et al. 2006. Reprinted with permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.)

The dry weight of adult AC is two-thirds collagen (Eyre et al. 2006). Another major structural component of AC is a noncollagenous large aggregating proteoglycan, aggrecan (van der Kraan et al. 2002). Collagen II is the major collagen and is crosslinked together with minor collagens IX and XI, thus forming the framework of the ECM (van der Rest & Mayne 1987). The collagen construct embedded with glycoprotein aggregates gives the tissue its form and enables tensile strength and the ability to endure compression.

2.2.2 The chondrocyte

Arising from the mesenchymal stem cells, the chondrocyte is a unique cell type in AC tissue. In adult humans, chondrocytes account for approximately 1–5% of the AC tissue volume. Chondrocytes are non-uniform and their number and shape vary in different zones of the AC as described above. Chondrocytes are the only cell type found in the AC and are responsible for production of the surrounding network of ECM proteins and their organization into a functional structure.

Page 25: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

23

Chondrocytes control the turnover of the ECM proteins and their metabolic activity varies between the different layers of AC (Huber et al. 2000). There are no blood vessels in AC, thus the metabolism of chondrocytes is essentially anaerobic as the nutrients are received by diffusion from the synovial fluid and, in a very small measure, from the subchondral bone (Corvol 2000). Chondrocyte metabolism operates at low oxygen levels that range from 1% (deep zones) to 10% at the surface. Nevertheless, chondrocytes can respond to mechanical stimuli, growth factors and cytokines that influence normal homeostasis (Goldring 2006). Glucose is the major energy source for these cells and an essential precursor for matrix glycosaminoglycan (GAG) synthesis, whose importance for the biomechanical properties of AC will be reviewed later. In adult joint cartilage, chondrocytes undergo a slow maturation that slows with age. At the same time, the turnover rate of matrix proteins decreases gradually, slowly weakening the AC. (Corvol 2000.)

2.2.3 The extracellular matrix (ECM)

The structure that enables the joint to endure an extreme amount of pressure (up to ten times the body weight) and continuous mechanical stress is the solid organic matrix, commonly referred to as the ECM. It consists of three principal phases: (i) a solid phase composed mainly of a densely woven collagen fibrillar network enmeshed with a high concentration of charged proteoglycan aggregates, (ii) a fluid phase of mainly water, gases, small proteins and metabolites, and (iii) an ion phase with many ionic species of dissolved electrolytes required to neutralize the charges fixed to the solid matrix (Mow et al.1999). From the cell outwards, the ECM is divided in to the pericellular, territorial and interterritorial regions (Figure 3), each region having its own structure and specific distribution of proteoglycans (Poole et al.1982, Aigner & Stöve 2003).

Chondrocytes are surrounded singly or in groups by a narrow region called the pericellular matrix (PCM), which together with the cell has been termed the chondron. It is a 3-μm-thick pericellular microenvironment rich in collagen types II, IX, XI and especially type VI, which distinguishes the area from the ECM. The PCM is also rich in hyaluronan, glycoproteins such as fibronectin, and different proteoglycans. (Ross et al. 2006.) It is believed that the biochemical and biophysical signals received by chondrocytes are influenced by the surrounding PCM, which may also play an important role in the response of cells to mechanical compression (Guilak et al. 2006). In addition, the PCM may serve as

Page 26: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

24

a protective layer for the chondrocytes (Choi et al. 2007). A number of experimental and theoretical studies indicate that the PCM is important in regulating the microenvironment of the chondrocyte; still, further research is needed to improve our understanding of its structure and function in cartilage physiology. (For a more detailed review, see Guilak et al. 2006.)

The interterritorial matrix compartment represents the major portion of the cartilage matrix far from the chondrocytes and consists of the collagen II/IX/XI fibril, aggrecan and various other proteoglycans (Aigner & Stöve 2003). When carilage ECM is discussed in the literature, it usually refers to this interterritorial matrix. The territorial matrix lies between the interterritorial and pericellular matrix and is under the metabolic control of the chondrocyte (Ross et al. 2006). So far, the cell-associated territorial matrix region lacks a specific biochemical characterization (Aigner & Stöve 2003). The properties of different collagen types, proteoglycans and other significant matrix molecules as well as their known functions in the AC are described in more detail in the following sections. The implications of the varying genetic characteristics of these molecules in OA are reviewed later, in section 2.4.

Page 27: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

25

Fig. 3. Electron microscopic image of the ECM compartments. The extracellular matrix of the AC can be subdivided into the pericellular, territorial and interterritorial compartments (a); at the ultrastructural level, the pericellular matrix appears as fine filamental material (partly type VI collagen) (c,d); the interterritorial cartilage matrix shows the typical cross-striated type II collagen fibrils in a network-like arrangement and the amorphous non-fibrillar matrix in between (b). (Adapted from Aigner & Stöve 2003, with permission from Elsevier.)

Collagenous components of the ECM

The superfamily of collagens. At present, 29 different triple-helical ECM proteins have been named as collagens. These proteins form a superfamily that constitutes about 30% of the total protein mass in the human body. Collagens have a plethora of essential biological functions including maintaining of the structure of various tissues, such as AC. A structure consisting of three polypeptide chains, called α chains, is distinctive to all collagens. Each α chain contains a repeating Gly-X-Y sequence (X and Y meaning any of the 20 amino acids in proteins), where proline is often found in the X position. Forty-two different polypeptide chains are known

Page 28: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

26

to date, and the three α chains in the mature protein can be identical or vary depending on the type of collagen. Being in every third position, the smallest amino acid, glycine, enables the coiling of the α chains into a triple-helical structure, important for the functioning of these molecules. Posttranslational modification by specific enzymes is typical for collagens. This often includes the hydroxylation of proline in the Y position, which is important for the stability of the triple helix.

In the ECM, most collagens form supramolecular assemblies and can be divided into several subfamilies according to the three-dimensional construction of the aggregate. In cartilage tissue, a major structural role is played by fibril-forming collagens (collagens II and XI) and fibril-associated collagen with interruptions in the triple helix (FACITs) (collagen IX). The tissue distribution of collagen varies substantially, as the foregoing collagen types are found almost exclusively in cartilage, whereas e.g. collagen VI, which forms beaded filaments, occurs in most connective tissues including cartilage. (For reviews, see Myllyharju & Kivirikko 2001, Myllyharju & Kivirikko 2004 and Heino 2007.)

Collagens II, IX and XI are referred to as cartilage-specific collagens, and together they form a copolymer that is indispensable to the structure of cartilage. Collagen II represents over 90%, collagen IX about 1% and collagen XI about 3% of the total collagen in the cartilage matrix (Eyre et al. 2006). Mutations in any of the genes coding for these collagens have been observed to cause diseases affecting connective tissue that vary from barely detectable mild conditions to lethal (for reviews, see Myllyharju & Kivirikko 2001 and 2004).

Collagen II. The fibril-forming collagen II is the foundation of the hyaline cartilage, representing about 80% of the tissue’s collagen content. It is a homotrimer consisting of three identical α1 (II) chains coiled into triple-helical fibrils, which are covalently crosslinked together to form a polymer of axially staggered periodic fibrils, providing collagen its tensile strength (Eyre 1991). The crosslinked polymers further assemble into supramolecular bundles forming a framework for the ECM. The deployment of the fibrillar network varies between the different layers of AC, as discussed previously. Collagen II is also found in the vitreous humor of the eye, in the nucleus pulposus and annulus fibrosus of the intervertebral disks, in the inner ear and in some non-chondrogenic tissues during development (Myllyharju & Kivirikko 2001, Gelse et al. 2003). The protein is coded by a single gene, COL2A1, which is located on chromosome 12q13.11-q13.12 and consists of 54 exons (Ala-Kokko & Prockop 1990, Takahashi et al. 1990, Ala-Kokko et al. 1995). Collagen II, as well as other fibril-forming

Page 29: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

27

collagens, is synthesized in precursor form. Alternative splicing of exon 2 results in the generation of two different forms of procollagen II (IIA and IIB), of which IIB is primarily expressed by chondrocytes (Sandell et al. 1994). Following or during the secretion of procollagen into the ECM, N- and C-terminal propeptide extensions are cleaved by specific procollagen proteinases, thereby triggering the fibril formation (Hulmes 2002). The association of the collagen II bundle with other cartilage collagens and the noncollagenous components of cartilage are presented in Figure 4.

Collagen IX. Collagen IX belongs to the family of FACIT collagens, which are characterized by the presence of two highly conserved cysteine residues separated by four amino acids at the NC1-COL1 junctions, the existence of two Gly-X-Y triplet imperfections within the COL2 domain, a succession of triple-helical domains connected by short noncollagenous domains (NC), and the presence of a large N-terminal domain containing a thrombospondin subdomain next to the collagenous region (Ricard-Blum & Ruggiero 2005). FACIT collagens do not form fibrils, but associate with fibrillar collagens. Collagen IX is composed of three genetically different α chains, α1(IX), α2(IX) and α3(IX), encoded by the COL9A1, COL9A2 and COL9A3 genes, which are all mapped to different chromosomes. Each chain forms three triple-helical collagenous domains, COL1-3, flanked by noncollagenous domains NC1-4, and the mature polypeptide forms a rod-like structure (Olsen 1997). This collagen type has also been considered to be a proteoglycan, since it carries an attachment site for GAGs in the COL3 domain of the α2(IX) chain (McCormick et al. 1987). Still, no distinct function for the proteoglycan feature has been demonstrated. Collagen IX has at least seven cross-linking sites, and in cartilage it is covalently bound to the collagen II decorating the surface of the fibril, with a space between the individual molecules. It is also able to covalently bind to other collagen IX molecules. (Eyre et al. 2006.) Collagen IX has shorter and longer variants due to the use of two distinct promoters in the α1(IX) collagen gene (Nishimura et al. 1989). The longer collagen IX variant is the major form in cartilage. It contains a large NC4 domain at the tip of the COL3 domain, which projects out of the fibril surface and may allow possible interaction of the NC4 domain with matrix proteoglycans and other molecules. The shorter collagen IX, lacking the NC4 domain, is found in the vitreous of the eye, intervertebral disk (IVD) and various developing non-chondrogenic tissues (Olsen 1997, Wu & Eyre 2003).

Collagen IX is an essential part of the collagen II/IX/XI heteropolymer in AC as it gives stability to the collagenous framework and is suggested to be a

Page 30: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

28

macromolecular bridge between fibrils and other matrix constituents. It is most abundant in the areas of thinner fibrils that form the chondron and may also have a role in regulating the diameter of the collagen fibrils. (For reviews, see Eyre & Wu 1995, Olsen 1997, Hagg et al. 1998, Eyre et al. 2002, Gelse et al. 2003 and Eyre et al. 2006.)

Collagen XI. The quantitatively minor fibril-forming collagen XI is the third important component of the AC collagen network, where it is buried within the major collagen II fibril. It is a heterotrimeric molecule composed of three α chains, α1(XI), α2(XI) and α3(XI), that are products of three distinct genes, COL11A1, COL11A2 and COL2A1, all located in different chromosomes. The α3(XI) is a product of the same gene that encodes collagen II, but is more extensively modified posttranslationally (Wu & Eyre 1995). Collagens II and XI are structurally related, differing mainly in their N-propeptides. Collagen XI contains a long triple-helical part (COL1), short N-terminal triple-helical part (COL2) and three noncollagenous domains (NC1-3) in each α chain. In contrast to other fibrillar collagens, α1(XI) chains retain at least to some extent their N-terminal propeptides even as mature molecules (Wu & Eyre 1995). These retained propeptides are thought to regulate the fibrillogenesis of collagen II (Gregory et al. 2000). Collagen XI is closely related to collagen V in biological and structural features, suggesting that these two collagens constitute a single type of collagen form, V/XI collagen (Fichard et al. 1995). Crosslinked to each other in a head-to-tail manner, collagen XI molecules are likely to form a template that constrains the lateral growth of the collagen II fibrils as well as stabilize them (Blaschke et al. 2000). Even though collagen XI represents a relatively small amount of the collagen in cartilage fibrils, its presence is essential for the regulated assembly, organization and development of cartilage (Gregory et al. 2000).

Page 31: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

29

Fig. 4. Schematic presentation of cartilage ECM showing a heterotypic collagen fibril consisting of collagens II, IX and XI, and its association with some of the noncollagenous components of cartilage, including aggrecan, cartilage oligometric matrix protein (COMP), hyaluronan, matrilin-3 and link protein. The figure has been originally published in Eveliina Jakkula�s thesis, Acta Universitatis Ouluensis, D832.

Other collagens in cartilage. Other collagens are also present in cartilage, but are not considered cartilage-specific due to their main expression in non-cartilaginous tissues. Collagen III, for example, is quite abundantly present in cartilage matrix accounting for some 10% of the collagen mass (Eyre et al. 2006). It has been found to associate with collagen II in the same banded fibrils (Young et al. 2000), perhaps as part of the matrix repair or remodelling processes (Eyre et al. 2006). Collagen VI is a beaded filament that forms dimers and tetramers. It is a heteropolymer of three distinct α chains that are bundled together containing triple-helical domains as well as N- and C-terminal domains. In normal articular cartilage, it is exclusively present in the PCM, where it interacts with different ECM components including collagen II, and may also form a network that anchors the chondrocyte to the PCM. (Guilak et al. 2006.) Collagen X is expressed only by hypertrophic chondrocytes and is shown to form latticework in the AC (Bruckner & van der Rest 1994). Articular cartilage also contains a small amount of collagens XII, XIII, XIV and XVI (Kassner et al. 2003, Eyre et al. 2006). In addition, collagens XX and XXVII have been detected in some cartilaginous tissues (Koch et al. 2001, Pace et al. 2003). So far, relatively little is known about the role of these minor collagens in the functioning of AC, their organization within the collagen II meshwork as well as their influence on development, aging, and disease processes (Young et al. 2000).

Page 32: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

30

Proteoglycans of the ECM

Aggrecan. Proteoglycans constitute a family of glycoconjugates with a central core protein to which GAG side chains are covalently attached (Wight et al. 1992). Cartilage contains up to 10% of proteoglycans consisting primarily of aggrecan (90% of the total cartilage proteoglycan mass) (Kiani et al. 2002). Carrying a substantial amount of GAG chains of keratan sulfate (KS) and chondroitin sulfate (CS) on its protein core, the aggrecan molecule greatly resembles a bottle brush in shape. It interacts with HA and link protein (LP) to form large aggregates, each consisting of up to 100 aggrecan molecules radiating from the central HA string (Roughley 2006a). The multiple GAG side chains attached to aggrecan include negatively charged anionic groups resulting in an imbalance in ion concentration between the cartilage and surrounding tissue.

The osmotic imbalance causes a large amount of water (about 70% by wet weight of the tissue) to be drawn into cartilage causing swelling and expanding of matrix. Large swollen glycoprotein aggregates are entrapped within the organized collagen network, forming a space-filling gel critical to the biomechanical properties of cartilage. (Kiani et al. 2002, Dudhia 2005.) In this manner, aggrecan provides the dominant mechanism of energy dissipation in shock absorbing tissues and endows AC with strength against compression and mechanical strain (Rizkalla et al. 1992, Iozzo 1998, Papagiannopoulos 2006). The concentration of aggrecan is lowest in the ECM near the articular surface, where tensile strength is mainly provided by tight collagen bundles, and increases when moving into the deeper zones of AC (Poole 2001). A schematic presentation of the structure of aggrecan and its role in forming a large proteoglycan aggregate is presented in Figure 5.

Page 33: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

31

Fig. 5. Electron microscopic image of an aggregate extracted from bovine AC, containing 31 aggrecan molecules attached to a single hyaluronan filament (marked with arrows). B. Schematic presentation of a single aggrecan molecule, consisting of globular domains (G1-G3) and attachment sites for chondroitin sulfate (CS) and keratan sulfate (KS) side chains. Aggrecan is non-covalently bound to hyaluronan (HA) and the complex is further stabilized by link protein (LP). (Reprinted with permission of Duodecim. Electron microscopic image by Jyrki Parkkinen.)

The human aggrecan gene, AGC1, is located on chromosome 15q26 (Korenberg et al. 1993). It consists of 19 exons with an organization that strongly correlates to the specific domains of the core protein (Valhmu et al. 1995). The core protein of aggrecan consists of three disulphide-bonded globular domains (G1, G2 and G3). G1 and G2 domains are connected by an extended interglobular domain (IGD), and a large sequence characterized by KS and CS side chains is situated between the G2 and G3 domains (Kiani et al. 2002).

Beginning from the N-terminal region, the G1 domain encoded by exons 3–6 (Roughley 2006a) is involved in the interaction with HA and LP, gluing all three together thus stabilizing the matrix network. The G1 domain also mediates interactions between chondrocytes and the matrix network, regulates product processing, and maintains proteoglycan quality control (Kiani et al. 2002). The IGD, encoded by exon 7 of AGC1, contains proteolytic cleavage sites for different proteinases such as matrix metalloproteinases (MMPs), serine proteases and acid proteases (Hardingham & Fosang 1995, Mort et al. 1998). Among these proteinases is aggrecanase-1, a member of the protein family of a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS), which is able to cleave aggrecan at a specific site in the IGD region (Tortorella et al. 1999). In

Page 34: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

32

addition, cytokines target the degradation of aggrecan via cleavage at the proteolytic sites in the IGD region (Loulakis et al. 1992). Aggrecan cleavage in the IGD region results in rapid loss of the whole GAG region; thus, IGD appears to be involved in the physiological turnover of aggrecan (Kiani 2002). The G2 domain is encoded by exons 8–10 (Roughley 2006a). Its exact function is still unknown, but it seems that this domain inhibits aggrecan secretion and ensures that only fully glycosylated aggrecan monomers are secreted, thereby maintaining product quality control (Kiani et al. 2001). The CS-rich domain and much of the KS-rich domains are encoded by the large exon 12 (Roughley 2006a) and constitute the most dominant feature of aggrecan, making up some 80% of the protein´s molecular mass (Dudhia 2005). About 30 KS and 100 CS chains are attached to the protein core and contribute approximately 8000–10000 negatively charged groups to the molecule, resulting in the osmotic activity described above (Dudhia 2005). Interestingly, and unique in humans, a variable number of tandem repeat (VNTR) polymorphism in the aggrecan gene area encoding the CS domain results in a range of core protein sizes and the ability to bear different numbers of CS chains depending on the length of the tandem repeat sequence (Doege et al. 1997).

Lastly, in the C-terminal region is the G3 domain encoded by exons 13–18 (Roughley 2006a). It is a complex region produced by alternative splicing of exons during posttranslational processing and appears to be involved in the processing of the proteoglycan, while also facilitating GAG chain attachment and enhancing product secretion (Kiani et al. 2002). The G3 domain is known to be removed from aggrecan in mature cartilage, probably by proteolytic cleavage, and it also seems to decrease with aging (Dudhia et al. 1996). (For reviews, see Kiani et al. 2002, Dudhia 2005 and Roughley 2006a.)

Other proteoglycans of the extracellular matrix. Of the family of aggregating proteoglycans, versican is also expressed in cartilage, but at much lower levels than aggrecan (Roughley 2001). The large proteoglycan aggregates are stabilized by a small proteoglycan LP, which is also needed in the process of aggregate formation (Melching & Roughley 1990). In addition, LP forms a protein coat covering the surface of HA and protecting it from degradation (Roughley 2006a). HA binds to the collagen framework and is essential in the retention of aggrecan in the ECM (Poole et al. 1982). The fundamental role of both LP and HA in normal cartilage function has been demonstrated in mouse models (Watanabe & Yamada 1999, Roughley 2006a). (For reviews, see Huber et al. 2000 and Roughley 2006a.)

Page 35: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

33

The ECM also contains different leucine-rich repeat proteoglycans (SLRPs) known as decorin, biglycan, fibromodulin and lumican, which are minor components in size. All of them consist of a core protein and a different amount of attachment sites for GAGs of either CS or KS (Hocking et al. 1998, Iozzo 1999). SLRPs are most concentrated in the surface layer or AC, where they bind to the other matrix macromolecules, especially to collagen fibrils, helping to regulate fibril diameter during its formation and possibly fibril-fibril interactions as well as giving stability to the matrix (Huber et al. 2000, Geng 2006, Roughley 2006a). SLRPs are also able to interact with different growth factors, thus affecting chondrocyte metabolism (Hildebrand et al. 1994, Schönherr et al. 2005). In addition, by providing a coating on the fibril surface, SLRPs seem to be able to protect cartilage collagens from proteolytic degradation by collagenases, also known as the MMPs (Geng et al. 2006). Proline/arginine-rich end leucine-rich repeat protein (PRELP) and chondroadherin also belong to the SLRP family and are found in cartilage ECM (Roughley 2001).

Perlecan, another large cartilage matrix proteoglycan that can bear heparan sulfate (HS) and CS side chains, was originally described as a basement membrane proteoglycan (Iozzo 1998). Perlecan has proved to be an important component in the pericellular matrix and is likely to coordinate proper cartilage growth and development (Gomes et al. 2002). It can bind to growth factors, such as fibroblast growth factor-2, and seems to also have a role in a process where mechanical signals are transduced through cartilage to change chondrocyte gene expression (Vincent et al. 2007).

In addition, there is yet another important glycoprotein operating solely on the surface of AC, that has earlier been poorly understood (Chang et al. 2008). This is a mucinous glycoprotein known as lubricin (also described as superficial zone protein), secreted by surface chondrocytes and synovial cells (Jay et al. 2007). It is indispensable in maintaining frictionless articular movement as it functions as a boundary lubricant between opposing carilage surfaces (Jones et al. 2007). Lubricin interacts with HA and is able to protect the articular surface, provide synovial fluid with the ability to dissipate strain, and inhibit overgrowth of synovial cells (Rhee et al. 2005, Chang et al. 2008).

Page 36: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

34

Other components of the ECM

Among the structural proteins of ECM, cartilage oligomeric matrix protein (COMP) is suggested to have an important role in matrix assembly. As the fifth member of the thrombospondin gene family, it is a large pentameric glycoprotein expressed at high levels in the interterritorial region of the matrix (Newton et al. 1994, Shen et al. 1995). The function of COMP is still poorly understood. The absence of this protein does not affect normal growth or development (Svensson et al. 2002), but mutations in its gene cause two skeletal dysplasias, pseudoachondroplasia (PSACH) and multiple epiphyseal dysplasia (MED) (Briggs et al. 1995, Hecht et al. 1995, Briggs et al. 1998). COMP is likely to mediate the interactions of collagen fibrils with other matrix constituents (Chen et al. 2008). It binds to both collagens II and IX as well as to all of the matrilins and aggrecan (Mann et al. 2004, Pihlajamaa et al. 2004, Chen et al. 2007) and has proved to be important in the fibril formation of cartilage collagens (Halász et al. 2007). (For a review, see also Hecht et al. 2005.)

The matrilins are noncollagenous proteins made up of von Willebrand factor A (VWA) domains, epidermal growth factor-like domains and a coiled coil α-helical module (Deák et al. 1999). Matrilin-1 and -3 are expressed mainly in cartilage and are able to form protein complexes with other matrix components, which is a typical feature for molecules with VWA domains (Wagener et al. 2005). Matrilins are able to interact with at least aggrecan (Hauser et al. 1996), COMP (Mann et al. 2004), some SLRPs (Wiberg et al. 2003), fibrillar collagens (Winterbottom et al. 1992) and collagen VI (Wiberg et al. 2003), serving as adaptors in the assembly of these proteins into supramolecular structures and as modulators of collagen fibrillogenesis (Wagener et al. 2005, Nicolae et al. 2007). The importance of matrilin-1 and -3 has recently been demonstrated by their absence, which causes ultrastructural abnormalities to the AC (Nicolae et al. 2007).

Among the structural components of the ECM, there is also a monomeric glycoprotein that is expressed mainly in the intermediate zone of human cartilage. Due to this localization, it is known as the cartilage intermediate layer protein (CILP) (Lorenzo et al. 1998). Its expression is influenced by TGF-β signaling and the expression substantially increases in association with aging in human articular cartilage; still, the definite function of the protein is unclear (Lorenzo et al. 1998, Mori et al. 2006). CILP seems to be an important component for cartilage

Page 37: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

35

homeostasis and has been linked to various diseases affecting cartilage (Mori et al. 2006).

Moreover, cartilage ECM is comprised of dozens of other proteins that vary greatly in structure, distribution and function. Among the structural components worth mentioning are also fibronectin, fibrillin, C-type lectin, elastin, and tenascin-C. In addition, the matrix harbors different regulatory molecules, such as pleiotrophin, chondromodulins, various growth factors as well as proteinases and inhibitors. (Roughley 2001.)

2.3 Osteoarthritis (OA)

Osteoarthritis is the most common joint affecting disease and the leading cause of disability, particularly among the aging population (Issa & Sharma 2006). OA is characterized by the degeneration and break-down of articular cartilage, leading to the development of typical symptoms, i.e. pain and stiffness in the affected joint. The onset of OA is strongly related to aging. Other known risk factors include trauma, obesity, over-usage of joints, gender and ethnicity. In addition, the disease also has a strong genetic component. (Haq et al. 2003.) So far, there is no cure for OA as we are unable to repair or stop the degradation of AC once it has started. Therefore, as the disease progresses, the growing rate of disability ultimately leads to the need for joint replacement surgery. Other preceding therapies are mainly focused on treating the symptoms of OA, while the primary cause of this most prevalent joint disease in our society still remains unknown. A review of OA as a disease, its epidemiology, and current understanding of the processes that lead to cartilage destruction in its pathogenesis are presented in the following sections.

2.3.1 Symptoms and clinical characteristics of osteoarthritis

According to the current understanding, OA is not only a disease of AC but involves the entire joint organ, including subchondral bone, menisci if present, ligaments, periarticular muscle, joint capsule and synovium (Hunter & Felson 2006). OA can appear in any synovial joint, but most commonly it affects the joints of the hands, knees, hips and spine (Goldring & Goldring 2006). At the onset of OA, the joint space filled with articular cartilage begins to narrow. Later, shallow fibrillation occurs on the surface of AC, which turns into deeper fissures as the disease progresses. Ultimately, this onward process results in the regional

Page 38: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

36

lack of cartilage called erosions. (Konttinen et al. 2003.) The developing process of the degenerative changes takes a variable amount of time depending on the individual. A typical symptom arising from the developing OA is joint pain, which is usually exacerbated by activity of the joint and relieved by rest. In later stages of the disease, joint pain is often experienced also at rest and at night. (Hunter & Felson 2006.) Characteristic symptoms for OA include reduced joint function and short-term morning stiffness of the affected joint (Konttinen et al. 2003).

When examining a patient suffering from OA, typical findings include reduced function of the affected joint, crepitation, swelling, thickenings of bone around the joint, and tenderness. The diagnosis of OA is usually based on the radiographic imaging of a joint manifesting some or all of the preceding symptoms and findings. In a radiograph of an OA-affected joint, the breakdown of AC is seen as joint space narrowing (JSN). In the more advanced disease, the JSN is accompanied by hypertrophic bone changes, which are seen in radiographic images as a formation of shelves of new bone at the joint margins (osteophytes), thickening and sclerosis of the subchondral bone, and as the development of subchondral bone cysts. Sometimes in advanced disease, subluxation, or dislocation of the joint, is also seen. In very severe OA, images may show the deformation of the whole joint structure. (For reviews, see Konttinen et al. 2003, Martel-Pelletier 2004, Goldring & Goldring 2006 and Hunter & Felson 2006.)

Classification and grading of the radiographic features of OA have traditionally been based on a generic method described by Kellgren and Lawrence (KL) (1957). In this classification, OA is divided into five grades (0–4) based on the severity of the radiologic findings described above; grade 0 representing definite absence of OA specific changes and grade 4 indicating severe OA (Kellgren & Lawrence 1957). Although the KL scale is widely adopted, it has been criticized for placing too much emphasis on the presence of osteophytes and for causing inconsistencies in the interpretation of the grading descriptors (Menz et al. 2007). Therefore, grading of OA for research purposes is often based on different modifications of the KL scale, and alternative grading systems have also been developed for OA of the hand (Kallman et al. 1989), hip (Croft et al. 1990) and knee (Ahlbäck 1968).

OA is usually classified as primary or secondary. In primary OA, no distinct predisposing environmental factor is identified and the etiology is for the most part unclear. It has a complex genetic component (see the review in more detail in

Page 39: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

37

section 2.4). Most OA patients suffer from primary OA, which can be localized or generalized (affecting multiple joint areas), the latter being more common in postmenopausal women (Haq et al. 2003). Secondary OA has an underlying cause, most commonly trauma, heavy work, obesity or a certain cartilage affecting disorder, such as Paget’s disease, rheumatoid arthritis (RA) or other inflammatory arthritis (Haq et al. 2003).

The diagnosis of OA can be made by joint symptoms, radiographic changes, or a combination of the two (Menz et al. 2007). It is, however, very common that a patient with clear radiographic changes specific for OA is totally asymptomatic. This radiographic OA represents the majority of the population prevalence of OA and rather than being a disease, it is even suggested to be part of normal aging in humans (Dieppe 2005). It is not known why pain symptoms are initially provoked in certain individuals with radiographic OA. Nor is it known why among the individuals with symptomatic OA, the structural severity of the disease usually correlates poorly with the intensity of the symptoms, making the attempts for treatment as well as evaluation of the disease outcome in clinical trials difficult (Kean et al. 2004, Dieppe 2005). Furthermore, it is noteworthy that according to epidemiologic follow-up studies, the experiencing of symptoms, particularly pain, is likely to be a risk factor for the progression of OA (Dougados 2004).

OA of the hand

For a long time, the focus of research has mainly been on hip and knee OA and as a result, knowledge of the pathogenesis of hand osteoarthritis (HOA) is still limited (Kloppenburg et al. 2007a). Yet, the functional impact of HOA is undisputed as it causes considerable disability in society and is characterized by restricted mobility and motion pain in the hand joints (Kloppenburg 2007b). Magnetic resonance imaging (MRI) has shown that the disease process in HOA includes cartilage loss, bone edema, synovial enhancement, osteophytosis and erosions (Tan et al. 2005). In the late stage, subluxation and fibrotic ankylosis are common, but in some patients, the dominant aspect is bone erosion (Fumagalli et al. 2005). Typical features for HOA are nodules, which may also result from other conditions (Fumagalli et al. 2005). Heberden�s nodes are found in the dorsolateral sites of the distal interphalangeal (DIP) joints and are more common than the Bouchard�s nodes found in the same sites of proximal interphalangeal (PIP) joints (Alexander 1999). These hard, visible lumps are strongly heritable and are thought to arise from the osteophytes. They may grow slowly and be

Page 40: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

38

painful or painless. The familial incidence of nodes may reflect the inheritance of anatomic characteristics that determine the level of resistance to osteophyte growth. (Alexander 1999, Fumagalli et al. 2005.) Recently, the probability of hand OA was estimated to be 20% when Heberdens nodes are present without any other findings, increasing up to 88% in individuals over 40 years of age having JSN in any finger joint and a family history of nodes (Zhang et al. 2008).

Hand joints comprise 30% of all affected joints in OA (Cushnaghan & Dieppe 1991). HOA is often polyarticular and its specific radiographic changes are most prevalent in the DIP, PIP and the first carpometacarpal (CMC) joints (Wilder et al. 2006). Symptoms of HOA are most commonly experienced in DIP joints (Niu et al. 2003). Symptomatic HOA is less studied, although it is a more relevant indicator of the proportion of patients in need of treatment (Kloppenburg 2007b). Women have a higher prevalence of both radiographic and symptomatic HOA and the prevalence among women also increases with age (Jones et al. 2002, Niu et al. 2003, Oulette & Makowski 2006). The polyarticular nature of HOA is clearly seen in different patterns of joint involvement associated with the disease. The most important patterns in which the joints are affected, in descending order of importance, are symmetry (the same joint affected in opposite hands), clustering by row (the same joint affected in several fingers) and clustering by rays (multiple joints affected in the same finger), which are all seen especially in women (Egger et al. 1995, Chaisson et al. 1997, Niu et al. 2003, Poole et al. 2003, Solovieva et al. 2005, Toba et al. 2006). Unlike in other types of OA, there is a positive association between radiographic HOA and hand pain, as well as between hand pain and the severity of radiographic findings (Dahaghin et al. 2006). Furthermore, OA of the hands may be a risk factor for osteoarthritis of the weight bearing knee and hip joints (Dahagin et al. 2005). (For more detailed reviews, see Fumagalli et al. 2005, Oulette & Makowski 2006 and Kloppenburg 2007b.)

OA of the hip and knee

The joints of the hip and knee are the essential weight-bearing joints in humans and their faultless functioning is essential for movement. As this functioning is disturbed by OA, it is likely to raise a more relevant clinical problem to the affected indivual than that of HOA. In general, majority of the economic burden of OA is caused by the disabilities resulting from the affected hip and knee joints (Loza et al. 2009). The symptoms and clinical findings present in hip and knee OA include all of those described earlier in section 2.3.1.

Page 41: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

39

Characteristic to both of these conditions is the reduced ability for joint movement (extension-flexion and also rotations in hip OA) and persistent pain that typically occurs during walking, running, climbing stairs and while standing up. Knee OA most commonly affects the femoro-tibial joint and the pain symptoms are mainly restricted to the areas in and around the joint. In hip OA, the localization of symptoms is often not as specific and may be experienced widely around the upper thigh area (O’Reilly & Doherty 2003). At least concerning the knee joint, there seems to be a significant discordance between the clinical outcome and radiological findings of OA (Hannan et al. 2000, Toivanen et al. 2007).

2.3.2 Prevalence of osteoarthritis

A majority of the individuals over the age of 65 have radiographic and/or clinical evidence of OA, underlining the importance of aging behind the increasing prevalence of the disease (Goldring & Goldring 2006). Due to the endemic nature of OA, diverse clinical symptoms, and variability in the radiologic classification criteria, it is difficult to compare the exact prevalence of OA between different populations and subgroups. Hence, the frequency of OA in a general population is not well established. Recently, in a survey of a health care database including 4 million people, the overall prevalence of OA was reported to be 10.8%: 8.9% in men and 12.6% in women (the diagnosis of OA was based on international classification of diseases (ICD)) (Kopec et al. 2007). When evaluating the prevalence of OA, it is important to distinguish between radiographic OA and symptomatic OA, the latter being less studied but clinically more relevant.

Joints of the hands, hips and knees are most commonly affected by OA (Goldring & Goldring 2006). According to pathological studies, OA specific changes are present in the knees of 60% of men and 70% of women who die in the seventh and eighth decades of life (Arden & Nevitt 2006). Radiographic knee OA is less frequent than hand disease. A survey of radiographic hand OA in a population of 40 years and older reported the prevalence rates for the second DIP joint, third PIP joint and first CMC joint to be 35%, 18% and 21%, respectively (Wilder et al. 2006). Among Finns aged 30 or older, the prevalence of radiographic OA in any finger joint and in at least two symmetrical pairs of DIP joints was estimated to be 44.8% and 16.0%, respectively (Haara et al. 2003). In the CMC joint of the thumb the prevalence was 7% for men and 15% for women (Haara et al. 2004). In general, population-based studies in the US and Europe

Page 42: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

40

suggest the prevalence rate of severe radiographic disease among people aged 25–34 to be 1%, rising up to 30% in those aged 75 and above. In contrast, only 3–5% of the elderly population have radiographic hip OA making it clearly less common. (See Arden & Nevitt 2006.)

The prevalence of symptomatic OA in general is lower than that of radiographic OA. In the Australian population, it occurs in fewer than 5% of people under the age of 40, increasing somewhat linearly to more than 50% of women and more than 30% of men aged 85 years (March & Bagga 2004). The estimated prevalence of symptomatic OA in different populations is presented in Table 1. A variation in the prevalence rates according to geographical origin of the population is apparent. It is supported by the finding that symptomatic knee OA was more common in Chinese women than in American women (Zhang et al. 2001) and that the prevalence of hip OA significantly varies even within Scandinavia (Ingvarsson 2000).

In summary, the prevalence of symptomatic OA varies between different epidemiological studies, mostly due to different methodological assessments, the geographical origin of the population and discrepancies in defining OA. Still, the prevalence rate in general has been shown to be substantial, demonstrating the clinical burden caused by the different forms of this chronic disease.

Table 1. Estimated prevalence of symptomatic OA of the hand, hip and knee in different populations.

Type of OA Age group

(years)

Sex Prevalence (%) Population Reference

Hand > 60 F+M 8 US Dillon et al. 2007

Hand > 65 F+M 5.6 Greek Andrianakos et al. 2006

Knee ≥ 30 M 6.1 Finnish Kaila-Kangas 2007

Knee ≥ 30 F 8.0 Finnish Kaila-Kangas 2007

Knee > 45 M 13.5 US Jordan et al. 2007

Knee > 45 F 18.7 US Jordan et al. 2007

Knee 40–75 F+M 7.6 French Roux et al. 2007

Knee > 65 F+M 20.1 Greek Andrianakos et al. 2006

Hip ≥ 30 M 5.7 Finnish Kaila-Kangas 2007

Hip ≥ 30 F 4.6 Finnish Kaila-Kangas 2007

Hip 40–75 F+M 5 French Roux et al. 2007

Hip > 65 F+M 2.5 Greek Andrianakos et al. 2006

Page 43: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

41

2.3.3 Etiopathogenesis of osteoarthritis

OA is considered to be a heterogeneous group of distinct diseases characterized by an adaptive response of synovial joints to a variety of environmental, genetic, and biomechanical stresses (Haq et al. 2003). According to the current knowledge, the risk factors of the disease relate to one of the two fundamental mechanisms in OA development: the harmful effect of abnormal loading on normal cartilage, or normal loading on abnormal cartilage. A schematic depicting the factors that are believed to be essential for the onset of OA are presented in Figure 6.

Fig. 6. Etiopathogenesis of OA.

Risk factors

Aging is the primary risk factor for OA and it represents normal loading on abnormal cartilage (Heliövaara et al. 1993, Tepper & Hochberg 1993, Hart et al. 1999, Karlson et al. 2003). Aging cartilage loses its tensile strength as the stiffness of the matrix decreases and the articular surface begins to soften. This is thought to derive from the changes in the structural composition of the ECM, particularly changes in content and organization of collagen II and proteoglycans. The capacity of chondrocytes to remodel and repair cartilage ECM also diminishes with age. (Goldring & Goldring 2007.)

Obesity is one of the best studied risk factors for OA (Heliövaara et al. 1993, Felson et al. 1997, Lievense et al. 2002, Cooper et al. 2000). Cartilage disruption is promoted by the increased dynamic stress and abnormal loading that the joints

Page 44: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

42

need to support when overweight, and it is a risk factor particularly for bilateral knee and hip OA. Obese individuals also have a higher bone mass, which may contribute to cartilage breakdown, increasing the stiffness of the subchondral bone. Interestingly, obesity is also associated with HOA. This suggests that there may be a systemic factor related to obesity, which affects the risk of OA.

Important risk factors also include a continuous physical work load caused by occupational factors or sporting activities (Heliövaara et al. 1993, Kujala et al. 1994, Sandmark et al. 2000, Lievense et al. 2001). Fractures and other injuries of the limbs are linked to hip and knee OA in cross-sectional and case-control studies and constitute an important risk factor for secondary OA (Heliövaara et al. 1993, Lau et al. 2000, Wilder et al. 2002).

In addition, a combination of multiple genetic factors constitutes an important risk factor group for OA. Heritability of OA is discussed in more detail in section 2.4.

Women are generally more frequently affected by OA than men, and gender is thus considered to be an important risk factor. Sexual hormones may be involved in the pathogenesis because the occurrence of OA and joint symptoms increase at menopause. There is also some evidence for a protective effect of estrogen in OA. Results of the cohort and case-control studies are not unanimous, but it is likely that estrogen replacement therapy (ERT) has a modest protective effect against hip and knee OA. Furthermore, as a minor risk factor, joint laxity may cause malalignment of the joint components, thus predisposing to OA. More recently, metabolic factors such as high plasma glucose level have also been suggested to affect OA pathogenesis, probably by causing changes in matrix macromolecules. (For more detailed reviews of the risk factors, see Carman et al. 1994, Cicuttini et al. 1996, Cimmino & Parodi 2005, Bierma-Zeinstra & Koes 2007, Fontana et al. 2007 and Riyazi et al. 2008).

The disease process and molecular changes in OA cartilage

Pathological cartilage breakdown in OA is a result of the imbalance between the catabolic (matrix degradation) process and the anabolic attempts at repair (matrix biosynthesis) by chondrocytes (Laadhar et al. 2007). Three stages have been identified in the process, first, the water content of AC increases and the matrix network degrades on a molecular level. Aggrecan size and content decrease and the collagen network is damaged, leading to reduced stiffness of the cartilage. Secondly, the chondrocytes try to compensate the occurred damage by increasing

Page 45: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

43

metabolic activity and matrix biosynthesis. Finally, the chondrocytes can no longer maintain the repair activity and the result is a complete loss of cartilage tissue. (Lorenz & Richter 2006.) As OA advances, the water content of AC is decreased along with the overall proteoglycan content. Chondrocytes become unable to maintain the collagen network and the catabolism and cleavage of proteoglycans and collagen II is further enhanced by the proteolytic enzymes. The changes in the proteoglycan components seem to occur before the collagen network is disrupted and are likely to result from proteolytic cleavage of the native molecules (Horton et al. 2006).

Matrix degrading enzymes such as MMPs and aggrecanases are involved in the disease process. Their synthesis is affected by proinflammatory cytokines, including different interleukins (ILs) and tumor necrosis factor (TNF) –α that are produced originally by the synovium (Martel-Pelletier 1998). Chondrocytes have receptors, including integrins that are activated from the binding of fibronectin and collagen II fragments and in turn stimulate the production of proteinases, inflammatory cytokines and chemokines (Pulai et al. 2005, Xu et al. 2007). Thus, the remains of cleaved matrix components seem to play a role in further inducing degradation of cartilage. In OA, ECM matrix synthesis is also stimulated by different anabolic factors. These include insulin-like growth factor (IGF)-1, TGF-β, fibroblast growth factors (FGFs) and bone morphogenetic proteins (BMPs) (Lorenz & Richter 2006).

In early or mild OA, biochemical analyses have shown that both collagen and proteoglycan synthesis as well as their degradation are upregulated (Lippiello et al. 1977, Sandy et al. 1984, Carney et al. 1992, Squires et al. 2003). Collagen II is downregulated in the upper regions of AC and normally minor components, collagens I and III, are upregulated (Young et al. 2005). Also expression of collagen VI is increased in the lower middle and upper deep zones of AC (Hambach et al. 1998). In addition, collagen X is expressed as a sign of hypertrophy (Aigner et al. 1993).

Regulation of noncollagenous components of the matrix is also affected in early OA. Fibronectin, tenascin, COMP, MMPs including collagenase-1 and -3, and CILP are all upregulated (Fernandes et al. 1998, Wagner et al. 2003, Lorenzo et al. 2004). Also, gene expression analyses of aggrecan and SLRPs have shown upregulation in early OA (Young et al. 2005). In healthy cartilage, there is a homeostasis between MMPs and tissue inhibitors of metalloproteinases (TIMPs), which is disturbed in OA. The activities of MMPs are controlled by TIMPs and the expression of both molecule families has been observed to change in OA.

Page 46: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

44

Particularly the increase in MMP-13 activity is thought to be involved in cartilage degeneration via effective degrading of collagen II. (Lorenz & Richter 2006.) In addition, aggrecanases (ADAMTSs), members of the MMP family, have been shown to contribute in the pathological process of OA (Glasson et al. 2005) by cleaving the important proteoglycan aggregates. It is also possible that the loss of AC in OA reflects the insufficiency of anabolic growth factors, such as IGF-1 and TGF-β, and that inhibition of the growth factors may contribute to the development of the disease (Trippel 2004).

In addition to the proteolytic enzymes, the destructive process involves nitric oxide (NO); it is produced in osteoarthritic joints, where it mediates the destructive effects of different proinflammatory cytokines (Vuolteenaho et al. 2007.) It is notable that the changes in the expression and synthesis of various ECM molecules during OA differ between the zones of AC, as they do also in healthy cartilage. Many of the changes are likely to represent attempts to repair the emerging damage, while the continuing alteration of matrix components in advanced OA demonstrates the break-down of these repair mechanisms. To date, many aspects of the molecular pathology of OA have been revealed and the extensive amount of knowledge only underlines the complexity of the disease process. The underlying trigger of cartilage degradation and the different pathways involved remain to be found in future studies. (For reviews, see Martel-Pelletier 2004, Trippel 2004, Lorenz & Richter 2006 and Goldring & Goldring 2007.)

2.3.4 Relevance of inflammation in osteoarthritis

OA has been classified as a noninflammatory arthritis, partly due to the low amount of leukocytes found in the synovial fluid, and because of the lack of systemic manifestations of inflammation. Nevertheless, the clinical outcome of OA with pain, swelling, effusions and stiffness clearly reflects inflammation (Krasnokutsky et al. 2007). Today, there is mounting evidence that inflammation of joint synovium and AC at the molecular level plays an important role in the pathogenesis of OA, even though it is not usually seen as extensive as in classic inflammatory arthritis, such as RA (Pelletier et al. 2001, Bonnet & Walsh 2005, Kristoffersen et al. 2006).

The mechanism by which production of inflammatory mediators is initiated is yet unclear (Goldring & Goldring 2007). It has been suggested that the inflammation of synovium develops due to the breakdown of matrix, which is

Page 47: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

45

catalyzed by proteolytic enzymes. As a result of the release of increased amounts of matrix fragments into the fluid, inflammatory mediators are created by the synovium. This causes a vicious cycle with more cartilage being degraded, subsequently provoking more inflammation (Martel-Pelletier 2004). From the synovial membrane, inflammatory mediators are able to diffuse through the synovial fluid into cartilage and activate chondrocytes (Martel-Pelletier 1998). Abnormal mechanical forces also seem to activate chondrocytes to produce different inflammatory mediators, such as cytokines, chemokines, leukemia inhibitory factor (LIF) and reactive oxygen species (ROS) including NO. These factors along with lipid-derived inflammatory mediators, such as prostaglandins and leukotrienes, provoke cells to produce MMPs and aggrecanases that may ultimately cause destruction of the ECM (for reviews, see Loeser 2006).

Cytokines

At present, cytokines can be regarded as diverse molecules including interferons, interleukins, mesenchymal growth factors, adipokines, and the TNF and chemokine protein families. Cytokines affect nearly every biological process, including responses to infection and antigens, embryonic development, disease pathogenesis and aging. Some of them are primarily lymphocyte growth factors, some function as proinflammatory or anti-inflammatory molecules, whereas other cytokines mediate the immune responses to antigens. (Dinarello 2007.) Cytokines bind to high affinity cell surface receptors, which mediate their actions (Martel-Pelletier 2004). To date, there are 33 known ILs among the cytokines. ILs can be classified into proinflammatory (IL-1, IL-11, IL-17, IL18), proinflammatory/ regulatory (IL-6 and IL-8) and anti-inflammatory (IL-4, IL-10 and IL-13) cytokines (Malemud 2004).

Proinflammatory cytokines are able to affect chondrocytes in various ways. They can increase enzyme synthesis and inhibit the synthesis of important matrix components, including collagens and proteoglycans, as well as reduce the synthesis of their own physiological inhibitors (Martel-Pelletier 2004). Particularly IL-1β and TNF-α seem to be the key mediators in joint destruction. In OA cartilage, chondrocytes synthesize IL-1β at concentrations that can evoke the expression of MMPs, aggrecanases and other catabolic genes. IL-1β molecules colocalize with TNF-α, MMPs and collagen II cleavage epitopes in the damaged matrix. (Goldring & Goldring 2007.) IL-1β and TNF-α also decrease collagen synthesis and stimulate the expression of other inflammatory mediators, such as

Page 48: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

46

IL-6, IL-8, IL-17, IL-18, LIF, prostaglandin E2 and NO (Goldring & Goldring 2007, Krasnokutsky 2007).

IL-1β is synthesized as an inactive precursor, which is activated through protease cleavage by IL-1β converting enzyme (ICE or caspase-1) (Fernandes et al. 2002). ICE is produced in the synovial membrane and in the AC and its production is significantly increased in OA tissues (Saha et al. 1999). Biological activities of IL-1 are mediated via binding to two cell-surface receptors, IL-1R1 and IL-1R2 (Slack et al. 1993), and the number of IL-1R1 is reported to be significantly increased in OA (Sadouk et al. 1995). Both types of IL-1R can also be shed from the cell surface, and exist extracellularly in truncated forms known as IL-1 soluble receptors (Fernandes et al. 2002). TNF-α is also synthesized as a precursor and is activated proteolytically by TNF-α converting enzyme (TACE) (Black et al. 1997). TACE has been found to be upregulated in OA (Attur et al. 2002). TNF-α binds to two specific cell membrane receptors and the one has been observed to multiply in OA chondrocytes (Malemud 2004). The actions of proinflammatory cytokines, IL-1β and TNF-α, are regulated in three different ways: (i) their activity can be blocked by a receptor antagonist, such as IL-1Ra; (ii) the soluble receptor binds the free cytokine, thus blocking its activity; or (iii) an anti-inflammatory cytokine decreases the synthesis of IL-1β and TNF-α (Martel-Pelletier 2004).

IL-6 is also believed to be one of the major factors in joint destruction, being a pleiotropic proinflammatory cytokine that is markedly upregulated at times of tissue inflammation. It is believed to be an important contributor to the pathogenesis of OA as it is able to increase the amount of inflammatory cells in synovial tissue, amplify the effects of IL-1 on the increased MMP synthesis and decreased proteoglycan production, as well as stimulate chondrocyte proliferation (Martel-Pelletier 1998). Human recombinant (hr) IL-6 has been shown to enhance hrIL-1β-induced proteoglycan degradation and to inhibit chondrocyte proliferation (Jikko et al. 1998). Furthermore, a significant rise in the level of IL-6 mRNA has been detected in OA-affected cartilage, and IL-6 levels in the serum and synovial fluid have been reported to be elevated among OA patients (Kaneko et al. 2000).

Other cytokines involved in the inflammation reaction in OA are IL-8, IL-17, IL-18 and LIF (Malemud 2004). Natural inhibitors, which reduce the production of proinflammatory cytokines and/or their activity, include TGF-β and the anti-inflammatory cytokines IL-4, IL-10 and IL-13. (For reviews, see Martel-Pelletier

Page 49: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

47

1998, Attur et al. 2002, Fernandes et al. 2002, Malemud 2004, Martel-Pelletier 2004 and Goldring & Goldring 2007.)

2.4 Inheritance of osteoarthritis

The concept of OA as a hereditary disease began to form in the 1940s, when familial clustering of Heberdens nodes of the fingers was discovered (Stecher 1941). Replication and widening of this study by others provided more understanding of the genetic component of OA, and later it became apparent that the disease is not caused by a single gene, but is rather transmitted in a nonmendelian manner as a complex, multifactorial trait. (For reviews, see Peach et al. 2005.) Comprehensive investigation on the subject was not undertaken significantly until the mid-1990s, when several twin-pair, sibling-risk and segregation studies were conducted. Studies from the past decade have estimated the influence of genetic factors on radiographic OA in women to be between 39% and 65% concerning hand joints and around 60% for the hip joint (Spector & MacGregor 2004). It has also been shown that the degree of heritability between different joint sites and between the sexes may differ, suggesting heterogeneity in the nature of the encoded susceptibility (Loughlin 2005).

Earlier, it was commonly thought that OA is a generalized systemic disease that has a major genetic component. However, this concept of the disease did not provide expected breakthroughs and it has been suggested that a joint-specific and gender-specific approach may be more rewarding (Loughlin 2004). Today, there is further evidence that multiple genes involved in OA pathogenesis may work differently in men and women, as well as affect different joint sites and cause distinct disease phenotypes in terms of severity. This notion is supported by the fact that statistical evidence in favor of chromosomal regions or independent genes often increases when the studied patient cohorts are stratified into subgroups based on sex, affected joint site and severity of the disease (Bukulmez et al. 2006). However, these stratifications often result in insufficient sample sizes with limited statistical power, which is likely to increase the possibility of false positive findings.

The field of complex genetic diseases has experienced a major revolution during the past few years. This is mostly due to the advanced technology which enables fast and cost-effective analysis of multiple SNPs simultaneously (van der Helm-van Mil et al. 2008). Traditionally, investigators have sought to identify genes that affect OA susceptibility using two different approaches, candidate gene

Page 50: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

48

association analysis and genome-wide linkage analysis. Furthermore, the completion of the human genome project has produced a third and perhaps the most promising method at the moment: genome-wide association (GWA) studies. These methods and their contributions to OA genetics are reviewed in more detail later in sections 2.4.1 and 2.4.2. Finally, a new alternatively used strategy is gene expression analysis, which is targeted to identify genes that are significantly up- or downregulated in OA compared to normal tissue, thus indicating a role in disease development or progression (Loughlin 2005). During the past few years, all of these strategies have provided extensive new information about the inheritance patterns and genome location of potential causative variants; yet, the results have been generally inconsistent at least partly due to the methodological shortcomings described below (Valdes et al. 2006).

Requirements for the quality and quantity of study populations have increased along with the growing data of complex disease genetics. In general, the major portion of the nearly one hundred previously reported different OA-associated variants have been obtained with samples that do not meet all the current standards required for studying a complex trait. If the type 1 error rate is set to a 5% level (p = 0.05) in a study of complex disease, in addition to a sufficient sample size, a careful definition of the phenotype is required and the population stratification also needs to be taken into account in order to obtain valid results. Furthermore, possible findings need correction for multiple testing if tested for independent associations (van der Helm-van Mil et al. 2008).

Limited-sized case-control studies of individual variations are laborious and require a hypothesis of the biological function or location of the candidate gene. Linkage analysis, on the other hand, is often not powerful enough to detect common alleles that have low penetrance and low frequency (Hirschhorn & Daly 2005). In contrast, during the past few years many common genetic variations with low penetrance have been successfully shown to associate with OA (Valdes et al. 2008a) as well as to various other complex diseases including cardiovascular, metabolic and autoimmune conditions, using the GWA approach with large (often thousands of individuals) study samples (Frazer et al. 2009). However, unlike in many of the candidate gene association analyses, the meaning of the results obtained by GWA studies are often unclear and, like the candidate analyses, they also require replication in independent case-control studies (Van der Helm-van Mil et al. 2008).

Page 51: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

49

2.4.1 Association analyses of candidate genes

In association analysis, the genotype and allelic distribution of a certain sequence variation is compared between healthy and OA affected individuals. Concerning OA, the genes of interest have most often included those that code for the important structural proteins of AC, such as collagens and proteoglycans (Reginato & Olsen 2002). Also the genes of different noncollagenous proteins and regulatory and signalling molecules of AC have been widely screened for association with OA (Spector & MacGregor 2004). Taken together, numerous candidate gene studies performed in different geographical populations and OA subsets worldwide have yielded dozens of statistically significant associations within a variety of different genes (Bukulmez et al. 2006). To date, only a few of the positive findings have been successfully replicated in an independent population, and most of the associated variations still lack solid evidence for causality and functional differences between susceptibility alleles. (For a review, see Ikegawa 2007.) The genetic variations that have shown the most compelling evidence of association in different studies to date are summarized in Table 2.

Despite the reported associations, the role of major structural genes, particularly COL2A1 and AGC1, in OA is still unclear. SNP haplotypes within COL2A1 have been demonstrated to associate with OA in two different ethnic groups (Meulenbelt et al. 1999, Ikeda et al.2002) and were replicated more recently (Valdes et al. 2007, Hämäläinen et al. 2008). Still, the function of the associated haplotypes is unknown and in general, the findings do not support the idea of COL2A1 as a major susceptibility gene for the common form of OA (Jakkula et al. 2005). Also, the functional VNTR polymorphism within the AGC1 gene has been shown to associate with OA in two independent samples, but the results have been inconsistent (Horton et al. 1998, Kirk et al. 2003).

The asporin gene (ASPN) and the secreted frizzled-related protein 3 gene (FRZB) are susceptibility genes for which the disease causality and functional differences in disease alleles have been convincingly shown (Loughlin et al. 2004, Kizawa et al. 2005). Asporin belongs to the SLRP family and is considered to be important in mediating cartilage metabolism through binding to TGF-β. ASPN contains a unique aspartic acid (D) repeat in its N-terminal region. The D14 allele of this polymorphic site has been shown to inhibit TGF-β activity more than other alleles, and was recently found to be overrepresented in two independent Japanese osteoarthritic populations (Kizawa et al. 2005). Thus, this variation seems to have functional significance and it provides a link between ECM proteins, TGF-β

Page 52: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

50

activity and the disease. To date, the association has been confirmed in various independent populations (Mustafa et al. 2005, Jiang et al. 2006, Shi et al. 2007). Some of the recent studies have not confirmed the association, but in contrast, have suggested the role of D13 as a protective allele (Kaliakatsos et al. 2006, Valdes et al. 2007, Song et al. 2008). Also, a total lack of association has been reported in some ethnic groups (Rodriguez-Lopez et al. 2006, Atif et al. 2008, Limer et al. 2008). In conclusion, a meta-analysis suggested that the association between the ASPN D14 allele and knee OA has global relevance, but the effect seems to have ethnic differences (Nakamura et al. 2007).

There are two single nucleotide polymorphisms (SNPs) in the area of the FRZB gene that encode for a substitution of highly conserved arginine residues to either tryptophan (Arg200Trp) or glycine (Arg324Gly) and result in reduction in the ability of the mature protein to antagonize Wnt signalling, which is crucial for chondrocyte metabolism (Loughlin 2005). These SNPs have been repeatedly shown to associate or suggested to associate with OA in different ethnic populations (Loughlin et al. 2004, Min et al. 2005, Lane et al. 2006, Valdes et al. 2007 and Rodriguez-Lopez et al. 2007). Together with the functional causality, the association studies provide evidence of the contribution of FRZB in the pathogenesis of OA, although they could not be confirmed in a recent meta-analysis (Kerkhof et al. 2008).

Another promising susceptibility gene for OA is growth and differentiation factor 5 (GDF5), which belongs to the BMP gene family. An individual SNP within the gene may reduce the transcriptional activity of chondrogenic cells, and to date it has been shown to associate with OA in three distinct Asian populations (Miyamoto et al. 2007). The association has also been confirmed in European populations, and the SNP was reported to be functional, causing decreased expression of the gene (Southam et al. 2007, Vaes et al. 2008, and Valdes et al. 2008).

The matrilin 3 gene (MATN3) was originally identified as an OA susceptibility gene through a genome wide linkage scan, and a missense Thr303Met mutation in the gene has repeatedly been reported to increase the risk and severity of HOA (Stefánsson et al. 2003, Eliasson et al. 2006, Pullig et al. 2007). Furthermore, a disintegrin and metalloproteinase gene, ADAM12, important in mediating cellular interactions and responses, has been seen to strongly associate with knee OA in two separate populations (Valdes et al. 2004, 2006).

Page 53: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

51

At present, the role of cytokines in the genetics of OA is unclear. Markers within the IL-1 gene cluster have earlier shown moderate evidence of association with knee OA (Loughlin et al. 2002a). Further evidence of this association was more recently gained by haplotype analyses in two small independent cohorts (Meulenbelt et al. 2004, Smith et al. 2004). Confusingly, neither one of the findings could be confirmed in a population that was more than six times larger and comparable in terms of ethnicity and disease classification (Chapman & Loughlin 2006). Yet, the SNPs and different haplotypes in the IL-1 gene cluster have repeteadly been positively linked to knee and hand OA also in other populations (Stern et al. 2003, Moxley et al. 2007 and Kanoh et al. 2008), further implying that there are likely one or more genetic features within the area that contribute in the etiology of OA at some level. Also, contradictory results have been reported (Ni et al. 2009). Furthermore, association of variants within the interleukin 4 receptor (IL-4R), IL-6 and IL-10 genes have also been reported (Forster et al. 2004a, Pola et al. 2005, Fytili et al. 2005, Riyazi et al. 2005), but the findings still lack confirmation in independent samples. In summary, the results concerning the IL-1 gene cluster in particular are diverse, consisting of multiple susceptible alleles and haplotypes. More research is needed in order to accurately define the genetic effect of the common inflammatory mediators on OA susceptibility.

There is also evidence of the role of the vitamin D receptor (VDR) gene in OA. Polymorphisms in the gene that are detectable with TaqI and ApaI restriction enzymes have been shown to associate with knee and hand OA (Keen et al. 1997, Utterlinden et al. 1997 and 2000, Solovieva et al. 2006). Nevertheless, many attempts to reproduce these associations have failed (Huang et al. 2000, Baldwin et al. 2002, Valdes et al. 2006, Limer et al. 2008). Furthermore, no association was confirmed in a recent meta-analysis between the VDR TaqI or ApaI polymorphisms and OA susceptibility (Lee et al. 2008).

Page 54: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

Tabl

e 2.

A s

umm

ary

of g

enes

and

und

erly

ing

vari

ants

tha

t ha

ve s

how

n co

nvin

cing

ass

ocia

tion

with

OA

in

mul

tiple

stu

dies

, co

nduc

ted

amon

g po

pula

tions

with

diff

eren

t phe

noty

pes

and

ethn

ic b

ackg

roun

ds.

Gen

e

(Loc

us)

Var

iatio

n

OA

* S

ex

Cas

es (n

) /

Con

trols

(n)

Min

or a

llele

freq

uenc

y

% c

ases

/ co

ntro

ls

OR

p-

valu

e P

opul

atio

n R

efer

ence

AS

PN

A

llele

D14

K

F>

M

137

/ 234

10

.9 /

4.7

2.63

0.

0008

4 Ja

pane

se

Kiz

awa

et a

l. 20

05

(9q2

2.31

) A

llele

D14

H

F>

M

593

/ 374

7.

9 / 4

.8

1.70

0.

0078

Ja

pane

se

Kiz

awa

et a

l. 20

05

A

llele

D14

H

M

36

4 / 3

56

17.4

/ 13

.1

1.48

0.

016

Brit

ish

Mus

tafa

et a

l. 20

05

A

llele

D14

K

F>

M

218

/ 454

9.

4 / 4

.9

2.04

0.

0013

C

hine

se

Jian

g et

al.

2006

A

llele

D14

K

F>

M

354

/ –

8.8

/ –

na

0.00

4 C

hine

se

Shi

et a

l. 20

07a

FRZB

A

rg32

4Gly

H

F

378

/ 760

11

.0 /

7.0

1.5

0.

04

Brit

ish

Loug

hlin

et a

l. 20

04

(2q3

2.1)

A

rg20

0Trp

and

Arg

324

Gly

H

F

558

/ 760

2.

6 / 0

.6

4.1

0.

007

Brit

ish

Loug

hlin

et a

l. 20

04

A

rg32

4Gly

G

F+

M

545

/ 136

2 11

.0 /

8.0

1.6

0.02

D

utch

M

in e

t al.

2005

A

rg20

0Trp

and

Arg

324

Gly

H

F+

M

570

/ 131

7 5.

0 / 2

.0

1.90

0.

01

US

La

ne e

t al.

2006

A

rg20

0Trp

and

Arg

324

Gly

K

F

603

/ 599

2.

0 / 0

.7

2.87

0.

04

Cau

casi

an

Val

des

et a

l. 20

07

GD

F5

C+1

04T

K

F+M

71

8 / 8

61

21.2

/ 25

.9

1.30

0.

0021

Ja

pane

se

Miy

amot

o et

al.

2007

(20q

11.2

2)

C+1

04T

H

F+M

10

00 /

981

16.0

/ 26

.0

1.79

1.

8x10

−13

Japa

nese

M

iyam

oto

et a

l. 20

07

C

+104

T K

F+

M

313

/ 485

21

.6 /

29.8

1.

54

0.00

028

Chi

nese

M

iyam

oto

et a

l. 20

07

C

+104

T K

/H

F+M

24

87 /

2018

12

.8 /

15.9

1.

28

0.00

4 S

pani

sh U

KS

outh

am e

t al.

2007

C

+104

T K

/H

F+M

18

42 /

1166

32

.4 /

40.5

1.

29

0.8x

10−4

U

K

Val

des

et a

l. 20

08

C+1

04T

C+1

04T

Ha K

F F

604

/ 110

2

494

/ 117

4

na

na

0.68

0.68

0.8x

10−5

0.00

3

Cau

casi

an

Cau

casi

an

Vae

s et

al.

2008

Vae

s et

al.

2008

MA

TN3

Th

r303

Met

H

a F+

M

2162

/ 87

3 2.

1 / 1

.0

2.12

na

Ic

elan

d S

tefá

nsso

n et

al.

2003

(2p2

4.1)

Th

r303

Met

H

a F>

M

226

/ 356

10

/ 2.

5 4.

28

0.00

7 G

erm

an

Pul

lig e

t al.

2007

aA

ssoc

iatio

n fo

r ear

ly o

nset

OA

; *G

= g

ener

aliz

ed; H

a =

hand

; K =

kne

e; H

= h

ip; n

a =

not a

vaila

ble.

52

Page 55: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

53

2.4.2 Genome-wide studies

Another approach to OA genetics has been genome wide linkage analyses that aim to uncover chromosomal regions predicted to contain variants that contribute to OA pathogenesis. Traditionally, these studies have been conducted using cohorts of affected sibling pairs or families where OA is exceptionally prevalent and seems to be inherited in a Mendelian manner.

Two different genomic loci are considered to be in linkage with each other if they are transmitted together from parent to offspring more frequently than would be probable with independent inheritance (Schork 1997). A large amount of polymorphic microsatellite markers consisting of repeating units typically 2–4 base pairs in length have been identified in the genome and their inheritance can be accurately monitored (Goldstein et al. 2005). These microsatellite repeats are usually not associated with any clinical trait and are called genetic markers. If a genetic marker is situated close enough to a locus that harbors a disease-associated variant, it is likely that they are inherited without a recombination occurring between them during meiosis and are thus linked. Analyses aiming to identify new susceptibility loci for OA can be conducted by genotyping polymorphic microsatellite markers and following their segregation through pedigrees with multiple OA-affected family members.

Since OA is a multifactorial disease with no clear mode of inheritance, non-parametric (model-free) linkage analysis would be expected to yield more accurate results. Parametric analysis requires specification of inheritance model and disease allele frequency. It is more powerful than non-parametric linkage analysis if the penetrance model and disease allele frequencies are estimated correctly (Kruglyak et al. 1996). Parametric and non-parametric linkage analyses can be performed in two ways: two-point and multipoint. In two-point analysis, the probability of linkage (the marker and disease are inherited together more often than expected under Mendelian law) is calculated with different recombination fractions separately between each marker and the disease locus (Aston & Wilson 1986). In multipoint analysis, all markers on the same chromosome are tested jointly for linkage and the position that is in linkage with the disease is the one that maximizes the LOD score. The LOD score is proportional to the strength of the evidence of linkage (traditionally ≥3, p = 0.001 is considered the limit for significance). The LOD score is calculated from the best estimation of the recombination fraction (maximum likelihood ratio) and is

Page 56: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

54

presented in the form of log to the base of 10. (For a more detailed review concerning linkage methodology and concepts, see Dawn Teare & Barrett 2005.)

Genome-wide linkage studies have concentrated on HOA (Leppävuori et al. 1999, Demissie et al. 2002, Stefánsson et al. 2003, Hunter et al. 2004) and hip OA (Ingvarsson et al. 2001, Chapman et al. 2002, Loughlin et al. 2002b, Forster et al. 2004b, Southam et al. 2004). These studies have uncovered loci with at least suggestive linkage in most chromosomes. Following the evidence that genes may contribute differently in men and women and vary also according to the site of the joint, recent linkage studies have yielded more convincing results using the preceding stratifications in the analyses (Creig et al. 2006, Livshits et al. 2007). A summary of positive linkage results to date is presented in Table 3, in which the limit for the positive linkage in terms of logarithm of odds (LOD) score is set to 2.0; however, lower suggestive results (LOD > 1.5) are also presented if the locus of linkage is close or overlaps with another suggestive region.

At present, Chr 2 is perhaps the strongest candidate for carrying important OA-associated variants. Regions in its short arm have earlier been linked to HOA (Demissie et al. 2002, Stefánsson et al. 2003) and have led to the identification of MATN3 as an important candidate gene (Stefánsson et al. 2003). Furthermore, one of the largest OA linkage studies performed to date recently provided strong evidence of linkage (LOD 4.0) at a locus on Chr 2p to HOA (Livshits et al. 2007). There are also loci with significant linkage in the long arm of Chr 2 (Leppävuori et al. 1999, Loughlin et al. 2002b, Hunter et al. 2004), and FRZB has been identified as an OA-associating candidate gene in the area (Loughlin et al. 2004). In an adjacent locus in the telomeric direction from the preceding areas, a strong linkage (LOD 6.1) was obtained within families manifesting early-onset generalized OA (Meulenbelt et al. 2006). The following fine mapping of the area has led to the identification of two potential variants in the NRP2 and IDH1 genes (Min et al. 2007). Overall, these results suggest that genomic variants particularly within Chr 2 are important in the development of OA.

Taken together, there are numerous loci within 1q, 3p, 4q, 11q and 12q with suggestive linkage to different phenotypes of HOA and to hip OA (see Table 3). The reported linked regions do not tend to overlap, but mostly lie in relatively close proximity to each other. A strong linkage to HOA in a locus 15q was reported by Hunter et al. (2004) with a LOD-score of 6.3, but the locus has not shown positive linkage in other studies. Furthermore, Chr 16 has multiple loci next to each other that have been suggested to be in linkage with hip or HOA (Ingvarsson et al. 2001, Forster et al. 2004b, Greig et al. 2006). IL4R has been

Page 57: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

55

identified as a candidate gene within the region showing linkage to OA (Forster et al. 2004a). There are new linkage results concerning the long arm of Chr 19 that are of particular interest. Evidence for linkage in 19q was earlier reported by Demissie et al. (2002) and the same locus was recently confirmed by Livshits et al. (2007) with a LOD-score of 4.3. In the most recent genome wide analysis, 14q32.11 was linked to generalized OA with a LOD of 3.03 with a potential susceptibility gene (DIO2) identified within the locus (Meulenbelt et al. 2008).

So far, numerous chromosomal loci among different populations have been suggested to harbor OA susceptibility variants. Fine mapping of areas detected by genome-wide linkage studies have resulted in the identification of many candidate genes, but the following association studies have identified a susceptible variation only in a few cases (Stefánsson et al. 2003, Forster et al. 2004a, Loughlin et al. 2004, see also Min et al. 2007). The inconsistency seen in results gained from the linkage analyses to date suggests that the genetic etiology of OA may be heterogenic among geographical populations. Furthermore, it implies that genetic factors may result in different disease outcomes in terms of joint specificity and severity. As ongoing and future linkage studies are performed in even more carefully selected populations, it is possible that they will provide information that will lead to the identification of new important candidate genes for OA.

Table 3. A summary of chromosomal areas that have been reported to be in linked (LOD-score > 2.0) to different OA outcomes. Areas with suggestive linkage (LOD > 1.5) are also presented if they overlap or are in close proximity to another suggestive locus.

Chr locus cM LOD Phenotype Country Reference 1p32.2 102 3.0 JSN in hand USA Demissie et al. 2002 1q31.1 202 3,0 DIP (women) USA Hunter et al. 2004 1q32.1 218 2.1 DIP USA Hunter et al. 2004 1q 42.1 250 3.0 Hand Tot-KL UK Livshits et al. 2007 2p23.3–24.1 42–47 4.4 DIP/CMC1 Iceland Stefánsson et al. 2003 1 2p23.3 48 2.2 JSN in hand USA Demissie et al. 2002 2p13.2–2p14 90 2.9 DIP UK Livshits et al. 2007 2p12–13.3 95 4.0 Hand Tot-KL UK Livshits et al. 2007 2q12–21 116–155 2.3 DIP Finland Leppävuori et al. 1999 2q31.1 181 1.6 Thumb IP USA Hunter et al. 2004 2q24.3–31.1 175–184 1.6 Hip UK Loughlin et al. 2002b

Loughlin et al. 2004 2 2q33.3 202 6.1 GOA Netherlands Meulenbelt et al. 2006

Min et al. 20073 3p25.1–25.2 30 2.8 DIP UK Livshits et al. 2007

Page 58: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

56

Chr locus cM LOD Phenotype Country Reference 3p21.2 71 2.7 PIP USA Hunter et al. 2004 3p14.1 91 2.5 Hand JSN & OST UK Greig et al. 2006 3p14.1 96 1.8 DIP Iceland Stefánsson et al. 2003 4p11 57 3.0 JSN in hand UK Greig et al. 2006 4q12–21.2 57–73 3.1 Hip (women) UK Loughlin et al. 1999 4q13.3 79 3.1 Hip UK Forster et al. 2004b 4q 26–27 128 2.3 DIP Finland Leppävuori et al. 1999 4q31.3 154 3.3 DIP Iceland Stefánsson et al. 2003 4q32.3 170 3.8 Hand Tot-KL UK Livshits et al. 2007 6p11.1 78 4.8 Hip (women) UK Southam et al. 2004 6q11.2–12 80 3.1 Hand Tot-KL UK Livshits et al. 2007 6p21.1–q22.1 56–120 2.1 Hip UK Loughlin et al. 1999 7p11.2 50 2.3 JSN in hand USA Demissie et al. 2002 7q35 155 3.1 DIP USA Hunter et al. 2004 8p12 61 2.6 JSN in hand UK Greig et al. 2006 8q12–21 105 2.1 DIP USA Hunter et al. 2004 9q21.2 76 2.3 JSN in hand USA Demissie et al. 2002 9q34.2–34.3 155 4.5 DIP UK Livshits et al. 2007 10p11.1 63 2.7 First CMC USA Hunter et al. 2004 11q13.4 76 1.6 JSN in hand USA Demissie et al. 2002 11q12.1–23.2 72–96 2.4 Hip (women) UK Chapman et al. 2002 12q21.3–22 100 3.9 DIP UK Livshits et al. 2007 12q24.3 166 1.7 JSN in hand USA Demissie et al. 2002 12q24.3 166 1.8 DIP USA Hunter et al. 2004 13q 33.1 87 2.3 First CMC USA Hunter et al. 2004 13q22.1 56 3.57 Hipa Japan Mabuchi et al. 2006 14q32.11 119–123 3.0 GOA multiple Meulenbelt et al. 20084 15q25.3 81 6.3 First CMC USA Hunter et al. 2004 16p12.1–13.1 28–49 2.6 Hipa Iceland Ingvarsson et al. 2001 16p 12.2 46 1.7 Hip (women) UK Forster et al. 2004b 16p12.1 52 2.6 JSN in hand UK Greig et al. 2006 16q22.1 89 1.9 Hip (women) UK Forster et al. 2004b 17p13.1 22 2.2 First CMC USA Hunter et al. 2004 17q 25.1 100 2.3 DIP USA Hunter et al. 2004 19p11 52 1.8 JSN in hand USA Demissie et al. 2002 19q13.2 65 4.3 Hand Tot-KL UK Livshits et al. 2007 19q13.3 68 1.8 Hand Tot-KL USA Demissie et al. 2002 19q13.4 90 4.0 DIP UK Livshits et al. 2007 20p13 4 3.7 DIP (women) USA Hunter et al. 2004 20q11.1 51 2.2 OST in hand UK Greig et al. 2006 aLinkage study of one family; Associating candidate genes found: 1MATN3; 2FRZB; 3NRP2 and IDH1; 4DIO2; cM = cent morgan; DIP = distal interphalangeal; GOA = generalized OA; OST = osteophyte; PIP = proximal interphalangeal; JSN = joint space narrowing; Tot-KL = Kellgren-Lawrence score sum for both hands.

Page 59: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

57

The GWA studies enable the comparison of the allele frequencies of a large number of SNPs between cases and controls. The International HapMap Project has resulted in the identification of over 3.1 million SNPs in the human genome with a density of approximately one per kilobase, including approximately 25–35% of all the 9–10 million common SNPs (The International HapMap Consortium 2007). With optimized analysis of these SNPs, most of the common alleles in the entire genome can be covered. The GWA is a more efficient method than the traditional linkage scan and is likely to result in the identification of variations that are very close to or even in the gene that is associated with the disease. Currently, the most up-to-date microarrays, or chips, have a capability of assessing more than 1 million SNPs in a single sample. Nevertheless, this method also has certain weaknesses as pointed out by Hunter & Kraft (2007). Even the GWA method can not cover the variations of the entire genome (false negative) and can also easily lead to false positive results. In order to reduce the possibility of a type 1 error, the study population in GWA analysis needs to be substantially large (preferably thousands of individuals), which is also essential in order to correct for multiple testing and still maintain the statistical significance of the finding (Valdes & Spector 2009).

A case-control study by Mototani et al. (2005) with 72,000 markers found the calmodulin-1 gene (CALM1) to be strongly associated with hip OA in a Japanese population (428 cases / 1008 controls, OR 2.40, p = 0.00065). However, the replication attempts in two European populations (920 cases / 752 controls and 158 cases / 193 controls) failed (Loughlin et al. 2006, Poulou et al. 2008, respectively). In a study of more than 25,000 examined SNPs, the leucine-rich repeats and calponin homology domain containing protein 1 gene (LRCH1) with an unknown function was reported to associate with radiographic knee OA in a sample of 335 cases and 335 controls (OR 1.44, p = 0.0078) (Spector et al. 2006). Again, a replication attempt in a sufficiently powered independent sample set (1145 cases / 1266 controls) failed (Jiang et al. 2008). One of the newest potential candidate genes is also the previously unknown gene DVWA at 3p24.3. It was identified in a recent genome-wide association study and the variations within the gene were shown to increase the risk of knee OA in Asian populations with a combined p-value of 7.3 × 10−11 (Miyamoto et al. 2008). One of the reported SNPs in the gene associated with the knee OA also in a large replication study with a population (2602 cases and 2147 controls) of European origin (a meta-analysis with the previous asian populations, OR 1.29, p = 2.70 × 10−5) (Meulenbelt et al. 2009).

Page 60: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

58

Perhaps the largest GWA study to date with 500,000 markers, 1564 knee OA cases and 2627 control individuals identified variants within the 2q33 linkage region that associated with knee OA (OR 1.55, p < 6.9 × 10−7 Valdes et al. 2008a). This region is currently being tested in a large (n = 8000) consortium and other large cohorts including Rotterdam, Framingham and TwinsUK are also being utilized for GWA analyses (Valdes & Spector 2009).

Page 61: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

59

3 Outlines of the present study

The genetic foundation of OA has proven to be more complex than previously assumed. Prior to the beginning of this doctoral study, numerous chromosomal regions had been linked to OA in different populations and many genes important for cartilage homeostasis had been shown to harbor OA associated variants. To date, the general view of OA genetics with various linkage results and genetic associations still has not taken shape. It is, however, probable that once a sufficient amount of information on OA is provided through various independent genetic studies, we will be able to obtain a better understanding of the cause of OA as a whole and perhaps in the future develop new therapeutic interventions. This study was thus undertaken in order to provide new information on the genetic background of this common disease. The specific aims were:

1. to study the role of the aggrecan gene (AGC1) VNTR polymorphism in hand osteoarthritis in a Finnish population,

2. to study the role of common promoter variations within the gene of interleukin 6 (IL6) in different forms of DIP OA,

3. to investigate the association of known SNPs and the VNTR polymorphism within the interleukin 1 gene cluster with DIP OA;

4. to perform a genome-wide linkage analysis in a sample of Finnish families manifesting primary osteoarthritis of the hip and/or knee in order to reveal chromosomal susceptibility loci and possible new candidate gene(s) for the disease.

Page 62: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

60

Page 63: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

61

4 Materials and methods

The materials and study populations used in this thesis are described in detail in original articles I–IV. Also, detailed descriptions of the common human genetic and statistical methods used in this study together with their references are presented in original articles I–IV and summarized in Table 4 below.

Table 4. Methods used in the present study.

Method Original publication

Laboratory procedures

DNA extraction I, II, III, IV

Polymerase Chain Reaction (PCR) I, II, III, IV

Gel electrophoresis I,II, III, IV

Restriction enzyme analysis I, III

Sequencing II, III, IV

Southern hybridization I

Statistical Methods

Fisher�s test I, II, III

Chi-square test I, II, III

Student�s t test II

Haploview II, III

PHASE II, III

Pedcheck IV

MLINK/LINKAGE IV

Simwalk IV

GeneMapper IV

4.1 Study populations

Two independent populations were used in this study. Five hundred forty-three women aged 45 to 63 years with similar academic backgrounds (dentists and teachers) were recruited for studies I–III through the registers of the Finnish Dental Association and the Finnish Teachers� Trade Union. Participating in a clinical examination were 295 (67.7%) dentists and 248 (56.9%) teachers. In addition to the clinical examination, all the participants were radiographed and evaluated for the presence and grade of OA in different hand joints. Disease outcomes for the AGC1 VNTR analysis (I) were based on the common disease patterns frequently observed among females with HOA (Solovieva et al. 2005). Because the OA of the DIP joints is a homogenous disease form and has yielded

Page 64: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

62

positive results in genome scans compared to other OA phenotypes (Leppävuori et al. 1999, Hunter et al. 2004), it was chosen for a phenotype in the subsequent studies (II and III). The DIP joints were also the most commonly affected joints in this sample. Due to the assumption that inflammatory mediators are likely to associate with the symptoms of OA, the presence of symptoms was also taken into account in studies II and III. Bilateral OA was required to reduce the possibility of OA-specific changes occurring in a single joint by chance. Furthermore, due to the limited number of OA cases, only one pair of affected joints was required for the symmetrical OA definition in study II.

Another study sample (IV) consisted of 15 independent families with multiple subjects affected with primary OA of the hip and/or knee joint(s). All families originated from Central Finland and family members were evaluated for the presence of primary OA clinically and by radiologic imaging. Ten families with a total of 225 subjects were used for the genome-wide linkage analysis and an additional five families including 54 members were recruited for the following fine mapping and association analyses. A total of 279 individuals in 15 families were analyzed, consisting of 58 subjects with OA diagnosis of the hip and/or knee and 34 subjects that were considered healthy. The OA status of 185 subjects was considered unknown due to the strict exclusion criteria, including possible secondary causes and inflammatory joint diseases resembling OA.

4.2 Southern hybridization

The Southern hybridization technique used in original article I was performed as follows: five micrograms of genomic DNA from each sample was digested with HaeIII and fractionated on a 1.3% agarose gel for 20 h at 46V using a 100 bp DNA Step Ladder as a molecular weight marker. DNA on the gel was denaturated (0.5 M sodium hydroxide, sodium chloride 1.5 M), neutralized (1.5 M sodium chloride, 0.5 M Tris - hydrochloric acid, 1 mM EDTA) and transferred to a nylon membrane using capillary action. The membrane was then exposed to ultraviolet radiation to permanently attach the transferred DNA and pre-washed for 1 h at 65 ºC in 0.1 X SSC (SSC, 0.15 M sodium chloride, 0.015 M sodium citrate, pH 7.0), 0.5% SDS. The probe used in the analysis was a PCR amplified fragment from the VNTR area. The primer sequences used were provided previously by Doege et al. (1997). Labelling of the probe was performed with [α-32P]dCTP using the Redprime II DNA Labelling system (Amersham Corp.). Prehybridization and hybridization were executed in hybridization buffer (SSC, 50% formamide, 10%

Page 65: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

63

50- fold Denhardt�s solution, 1 M sodium phosphate, 0.5% SDS, 10mg/ml ssDNA) at a temperature of 42 ºC for 2 h and 16 h, respectively. After hybridization, excess probe was washed from the membrane (2 x SSC, 0.1%SDS) in 37 ºC and then 65 ºC. The pattern of hybridization was visualized on X-ray film by autoradiography and the number of 57 bp repeats was then estimated from the film.

4.3 Linkage analysis

Genotyping for genome-wide screening was performed at the Finnish Genome Center (Helsinki, Finland) using the Applied Biosystems Linkage Mapping Set (MD 10) and an automated instrument (Megabase 1000; Molecular Dynamics). In the first fine mapping, an additional 10 markers and 7 markers (http://www.ncbi.nlm.nih.gov) spaced 2–5 cM apart were selected from the regions of interest on chromosomes 2 and 11, respectively. In the second fine mapping, 15 additional markers from chromosome 2 and 10 markers from chromosome 11 were genotyped to narrow down the space between markers to ~1 cM. Genetic distances (cM) and microsatellite marker locations were taken from the MAP-O-MAT sex-average linkage map (available at http://compgen.rutgers.edu/mapomat) and the database of the National Center for Biotechnology Information (NCBI). Standard PCR amplifications of the marker areas were performed according to the conditions described in detail in the article IV. The genotyping was performed using standard fluorescence-based genotyping methodologies (ABI PRISM 3100 Genetic Analyser, Applied Biosystems) and analyzed using GeneMapper Software (version 4.0, Applied Biosystems).

Both parametric and non-parametric methods were used in the study. Penetrance models were age-dependent (see Table 2 in article IV) and the disease allele frequency was set to 0.001. Parametric and non-parametric linkage analyses were performed in two-point and multipoint ways.

Calculation of the likelihood is complex due to the multiple markers and family members included in the study. Also, parental genotype data is often inadequate, with the result that all possible genotype and haplotype configurations must be taken into account. Furthermore, all possible disease locus genotypes need to be considered due to the incomplete penetrance. Thus, specific computer programs were used to estimate the linkage in different loci. In this study, both parametric and non-parametric multipoint LOD scores were calculated with Simwalk2 version 2.9 (Sobel & Lange 1996). Two-point LOD scores were

Page 66: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

64

obtained using the MLINK of the LINKAGE program package version 5.2 (Lathrop et al. 1984).

Page 67: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

65

5 Results

5.1 Aggrecan VNTR polymorphism and hand osteoarthritis (I)

According to the radiographic measurements, three different definitions of outcomes were chosen for hand OA in this study: (i) the subject was defined to have OA if at least two of the finger joints were affected by OA of grade 2 (modified Kellgren and Lawrence system) or more, (ii) OA was moderately severe if at least one finger joint was affected by OA of grade 3 or more, and (iii) symmetrical OA if at least two pairs of joints were symmetrically affected by OA of grade 2 or more. The number of 57 bp repeats in the aggrecan VNTR locus was successfully detected by the Southern hybridization method from 530 subjects (dentists, n = 288 and teachers, n = 242). Among these subjects, the prevalence of OA, moderately severe OA and symmetrical OA were 42.6%, 14.7% and 19.9%, respectively.

5.1.1 VNTR allele frequency

Fourteen different alleles with 18, 21–30 and 32–34 repeats were identified (alleles A18, A21–A30 and A32–A34, respectively). Alleles A25–A29 represented 96.3% of the total allele count (n = 1060). The most common allele was A27 (38.6%) followed by A28 (30.7%) and A26 (16.8%). The total frequencies of the observed alleles are presented in Table 5 below. There were no differences between the studied occupational groups. Furthermore, individuals with and without radiographic findings showed no significant differences in terms of allele frequencies.

Table 5. Allele frequencies of the aggrecan gene VNTR polymorphism

Alleles n % Alleles n %

A18 1 0.1 A27 409 38.6

A21 6 0.6 A28 325 30.7

A22 8 0.7 A29 57 5.4

A23 2 0.2 A30 9 0.8

A24 2 0.2 A32 2 0.2

A25 51 4.8 A33 9 0.8

A26 178 16.8 A34 1 0.1

Page 68: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

66

5.1.2 The most common aggrecan VNTR allele, A27, is protective against hand osteoarthritis

Ninety-five of the studied 530 women were homozygous for the most common A27 allele (the A27A27 genotype) and had a clearly reduced risk of OA compared to 216 non-carriers of the same allele with odds ratios (ORs) of 0.46 for OA, 0.40 for symmetrical OA and 0.42 for moderately severe OA. The corresponding 95% confidence intervals (CIs) were 0.27–0.78, 0.20–0.83 and 0.18–0.96, respectively.

Due to the low frequency of independent shorter alleles A18–A25 and the longer alleles A30–A34 (< 1%), these were combined in the statistical analysis. Women that carried two copies of the longer alleles (A28–A34, n = 93) had an increased risk of OA with an OR of 1.73 (95% CI 1.03–2.89). Furthermore, the risk of moderately severe OA was elevated among women carrying two copies of the shorter alleles (A18–A26, n = 50), but this result was not formally significant after adjustment for multiple testing. The ORs with corresponding CIs for the different disease outcomes are presented in Table 6.

Table 6. Association between the aggrecan VNTR alleles and radiographic findings.

OA Moderately severe OA Symmetrical OA Allele Genotype

N OR 95%CI N OR 95%CI N OR 95%CI

A18–

A26

Homozygous 25 1.30 0.69–2.42 13 2.45a 1.17–5.12 13 1.45 0.71–2.96

Heterozygous 60 0.93 0.62–1.41 24 1.37 0.78–2.40 30 1.12 0.68–1.86

Absence 139 1.00 41 1.00 62 1.00

A27 Homozygous 28 0.46b 0.27–0.78 8 0.42c 0.18–0.96 11 0.40d 0.20–0.83

Heterozygous 83 0.85 0.57–1.29 28 0.82 0.48–1.43 37 0.78 0.47–1.25

Absence 102 1.00 38 1.00 52 1.00

A28–

A34

Homozygous 47 1.73e 1.03–2.89 15 1.00 0.51–1.98 23 1.57 0.85–2.87

Heterozygous 94 1.36 0.91–2.04 28 0.82 0.47–1.42 44 1.28 0.78–2.12

Absence 83 1.00 35 1.00 38 1.00

Odds ratios (OR) and their 95% confidence intervals (CI) are adjusted for age and occupation. Subjects

were defined as having OA if at least two of the finger joints were affected by OA of grade 2 or more,

moderately severe OA if at least one finger joint was affected by OA of grade 3 or more, and symmetrical

OA if at least two pairs of joints were symmetrically affected by OA of grade 2 or more. apcorr = 0.11; bpcorr = 0.012; cpcorr = 0.12; dpcorr = 0.039; epcorr = 0.036.

Page 69: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

67

5.2 Interleukin 6 promoter polymorphisms and symptomatic distal interphalangeal osteoarthritis (II)

In this study, DIP OA was defined by the presence of radiographic findings of grade 2 (modified KL system) in at least two DIP joints. Symmetrical DIP OA was defined by the presence of radiographic findings of grade 2 or more in at least one symmetrical pair of the DIP joints, and symptomatic DIP OA was defined by the same radiographic findings as DIP OA, accompanied by common symptoms, such as pain and tenderness in the affected joints. Three different sites with SNPs at promoter positions -597, -572 and -174 (G-597A, G-572C and G-174C) were investigated successfully from 535 participants, of whom 224 subjects were diagnosed as having DIP OA, 205 as having symmetrical DIP OA and 48 as having symptomatic OA. None of these conditions could be identified in 309 subjects.

All genotype frequencies were in Hardy-Weinberg equilibrium, and the two observed occupational groups showed no statistically significant differences in genotype frequencies or carriage rates. No significant association was observed between any of the studied SNP genotypes and DIP OA (radiographic OA).

5.2.1 G-alleles of the promoter polymorphisms associate with symptomatic distal interphalangeal osteoarthritis

A significant association was observed between symptomatic DIP OA and the G alleles at promoter positions -174 and -597, with the G allele being more common in subjects with symptomatic DIP OA (at -174, 57.3% compared to 43.4%, pcorrected = 0.01, and at -597 58.5% and 44.5%, pcorrected = 0.02). Furthermore, 67.6% (n = 328) of the subjects without symptomatic DIP OA had at least one G allele at -174 versus 87.5% (n = 42) of the subjects with the disease. The corresponding difference for G allele carriage rates at -597 were 68.2% (n = 331) and 87.2% (n = 41), respectively (p = 0.007). At G-174C the combined GG and GC genotypes increased the risk of the disease in comparison with the CC genotype (p = 0.008). The same pattern was also observed with the G-597A polymorphism (p = 0.006). None of the genotypes at the G-572C polymorphic location had any effect on the risk of symptomatic DIP OA. No significant differences were seen in the genotypes or G allele frequencies between subjects with asymptomatic DIP OA or symmetrical DIP OA and subjects without these conditions.

Page 70: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

68

5.2.2 Haplotype analysis of the promoter polymorphisms

A total of five haplotypes were observed of which the two most common ones were CGA (0.53) and GGG (0.40). The combined frequency of the other three haplotypes was only 0.07. No significant association was observed between a haplotype containing G allele at each locus and radiographic DIP OA or symmetrical DIP OA. Yet, the G-G-G haplotype was overrepresented in the subjects with symptomatic DIP OA in comparison with those without the disease (0.51 versus 0.39, p = 0.023). Furthermore, analysis of the haplotype pairs (diplotypes) showed that the G-G-G/other diplotype was also overrepresented among the subjects with symptomatic DIP OA (0.71 versus 0.46, p = 0.001). The risk of symptomatic DIP OA was also increased among the carriers of the G-G-G haplotype. Results of the haplotype analysis with ORs and corresponding confidence intervals (CI) are presented in Table 7 below.

Page 71: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

69

Table 7. Frequency of the IL-6 G-G-G diplotypes according the DIP OA status (n = 533 to 535).

No Yes OR (95% CI) Diplotypes

N %

N % Crude Adjusted

DIP OA 309 224

other/other 117 37.9 74 33.0 1.00 1.00

G-G-G/other 148 47.9 107 47.8 1.14 (0.78–1.68) 1.17 (0.77–1.76)

G-G-G/G-G-G 44 14.2 43 19.2 1.54 (0.93–2.58) 1.48 (0.86–2.55)

G-G-G carriage 192 62.1 150 67.0 1.23 (0.86–1.77) 1.24 (0.85–1.83)

G-G-G frequency 236 38.2 193 43.1 1.22 (0.96–1.57) 1.20 (0.92–1.56)

Symmetrical DIP OA 329 205

other/other 127 38.6 64 31.2 1.00 1.00

G-G-G/other 155 47.1 101 49.3 1.29 (0.87–1.91) 1.35 (0.89–2.04)

G-G-G/G-G-G 47 14.3 40 19.5 1.69 (1.01–2.83) 1.65 (0.95–2.84)

G-G-G carriage 202 61.4 141 68.8 1.38 (0.96–2.00) 1.42 (0.96–2.10)

G-G-G frequency 229 35.9 181 42.1 1.30 (1.01–1.67)1 1.29 (0.99–1.68)

Symptomatic DIP OA 487 48

other/other 185 38.0 6 35.7 1.00 1.00

G-G-G/other 223 45.8 34 70.8 4.70 (1.93–11.44)a 5.03 (2.02-12.51)a

G-G-G/G-G-G 79 16.2 8 16.7 3.12 (1.05–9.29) 3.01 (0.98-9.22)

G-G-G carriage 302 62.0 42 87.5 4.29 (1.79–10.28)a 4.45 (1.82-10.88)a

G-G-G frequency 381 39.1 50 51.0 1.83 (1.19–2.81)a 1.62 (1.07-2.45)a

DIP OA was defined by the presence of radiographic findings of grade 2 or more in at least two DIP joints.

Symmetrical DIP OA was defined by the presence of radiographic findings of grade 2 or more in at least

one symmetrical pair of the DIP joints. Symptomatic DIP OA was defined by the presence of both

radiographic findings of grade 2 or more and symptoms in at least two DIP joints. ap-value remained statistically significant after correcting for multiple testing.

5.3 Interleukin 1 gene cluster polymorphisms and bilateral distal interphalangeal osteoarthritis (III)

In this study, the chosen disease outcome was bilateral DIP OA. If the subject had radiographic findings (grade ≥ 2) in at least one symmetrical pair of the DIP joints (if one DIP joint of the hand is affected, the same joint of the opposite hand is also affected), she was classified as having bilateral DIP OA. Among the 295 dentists and 248 teachers, the prevalence of bilateral DIP OA was 38% (46% in teachers and 31% in dentists). Within the interleukin 1 gene cluster, a total of nine SNPs were genotyped within the genes coding for the interleukin-1 type I receptor (IL1R1 on 2q12), interleukin-1 receptor-like 2 protein (IL1RL2 on 2q12),

Page 72: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

70

interleukin-1α (IL1A on 2q13) and interleukin-1β (IL1B on 2q13). Also, a VNTR polymorphism in intron 2 of the IL1RN gene (IL1RN on 2q14.2) was genotyped resulting in the identification of four different alleles: allele 1 (four repeated units) had a frequency of 0.73, allele 2 (two repeats) a frequency of 0.25, allele 3 (three repeats) a frequency of 0.002, and allele 4 (five repeats) a frequency of 0.001.

To ensure laboratory quality control in genotyping tests, two independent readers interpreted the results. In every genotyping analysis, internal control samples with known genotypes were included. Any sample with ambiguous results was re-tested, and a random selection of 10% of all samples was repeated. No discrepancies were discovered upon replicate testing with the method used. For technical reasons, seven of the 543 samples could not be genotyped for the IL1R1, IL1RL2 and IL1RN polymorphisms.

5.3.1 Frequency of the alleles, genotypes and carriage rates of the studied polymorphisms

No statistically significant differences between the two studied occupational groups were seen in the frequency of alleles, genotypes or carriage rates of the studied polymorphisms. Nevertheless, a statistically significant association between one SNP in IL1B (rs1143633) and bilateral DIP OA was obtained with OR of 3.03 (95% CI 1.35-6.83, p = 0.006) under the recessive model of inheritance. A significant association between the SNPs in IL1R1, IL1RL2, IL1A and IL1B (rs16944) and between the IL1RN VNTR polymorphism and bilateral DIP OA could not be seen. Results of the association analysis between all studied polymorphisms and bilateral DIP OA are presented in Table 8. A considerable linkage equilibrium (LD) occurred between the studied SNPs in the IL1R1 and IL1RL2 genes and between the IL1RN VNTR polymorphism and either IL1A or IL1B SNPs. (See the LD plot in original article III).

Page 73: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

71

Table 8. Association of the IL1 gene cluster polymorphisms with bilateral DIP OA.

p-value1 SNPs ID

Gene

Geno-type Counts

No/Yes

OR2 95% CI

General D A R

rs1465325 T/T 204/124 1 0.42 0.39 0.25 0.24

IL1R1 T/C 115/72 1.13 0.76–1.66

C/C 9/10 1.86 0.70–4.93

rs956730 G/G 164/94 1 0.51 0.30 0.47 0.82

IL1R1 G/A 138/97 1.25 0.85–1.84

A/A 26/15 1.02 0.50–2.12

rs2287047 C/C 168/103 1 0.79 0.49 0.51 0.78

IL1R1 C/T 135/87 1.13 0.77–1.67

T/T 25/16 1.17 0.57–2.40

rs1922290 G/G 138/82 1 0.42 0.53 0.99 0.40

IL1RL2 G/C 128/90 1.24 0.82–1.86

C/C 62/34 0.90 0.53–1.53

rs1922295 G/G 137/82 1 0.45 0.56 0.99 0.39

IL1RL2 G/A 145/98 1.20 0.81–1.79

A/A 46/26 0.87 0.48–1.56

rs1800587 C/C 149/91 1 0.63 0.59 0.90 0.56

IL1A C/T 145/95 1.16 0.79–1.71

T/T 41/21 0.90 0.48–1.68

rs1143634 C/C 189/100 1 0.06 0.02 0.02 0.27

IL1B C/T 123/88 1.53 1.03–2.25

T/T 23/19 1.76 0.87–3.55

rs1143633 G/G 170/118 1 0.001 0.13 0.88 0.006

IL1B G/A 154/69 0.62 0.42–0.92

A/A 11/20 2.48 1.08–5.68

rs16944 C/C 121/85 1 0.48 0.24 0.37 0.91

IL1B C/T 167/92 0.78 0.52–1.16

T/T 47/30 0.85 0.48–1.50

IL1RN 1/1 185/114 1.00 0.14 0.86 0.54 0.08

VNTR 1/2 110/77 1.20 0.80–1.79

2/2 35/14 0.58 0.28–1.18 1Uncorrected p-value of the association analysis; D = Dominant; A = Additive; R = Recessive. 2ORs and

95% CIs are adjusted for age, occupation, BMI and smoking history. For the IL1RN variable-number

tandem repeat (VNTR) polymorphism, allele 1 consists of the alleles with 4 repeat units, 3 repeat units

and 5 repeat units; allele 2 has 2 repeat units.

Page 74: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

72

5.3.2 Analysis of the extended haplotypes

Separate haplotype analyses for the IL1R1-IL1RL2 and IL1B-IL1RN SNPs were conducted due to the weak LD between the two regions. Seven different haplotypes were analyzed and no difference was observed between subjects with and without bilateral DIP OA in the IL1R1-IL1RL2 haplotype distribution. Yet, the IL1R1-IL1RL2 112 haplotype was associated with an increased risk of OA among the dentists (OR = 1.75, 95% CI 1.15–2.68, p = 0.009) and a reduced risk of OA among the teachers (OR = 0.66, 95% CI 0.43–1.01, p = 0.05).

It was hypothesized that among the dentists who are exposed to continuous repetitive joint usage, the haplotype might modify this environmental factor. Thus, the possible interaction between the haplotype and occupational hand load among the dentists was also examined. It seemed that the haplotype exacerbates the effect of low variation in dental work tasks on the risk of OA, but there was no effect of the haplotype in dentists with high variation in work tasks (dentists with low variation in work tasks and the carriers of the haplotype: OR = 2.25, 95% CI 0.91–5.54, p = 0.08, and dentists with high variation in work tasks and non-carriers of the haplotype: OR = 0.73, 95% CI 0.28–2.01, p = 0.54).

A total of 16 different IL1B-IL1RN VNTR haplotypes were reconstructed and in general, none of the haplotype allele frequencies differed significantly between subjects with and without DIP OA. Due to the low LD between the rs180057 (IL1A) and other investigated polymorphisms, four polymorphisms within the IL1B-IL1RN region were used to group the haplotypes for further analysis. The distribution of the 211-1 diplotypes fitted an additive mode of inheritance, while the distribution of the 121-1 diplotypes fitted a recessive mode (OR = 1.42, 95% CI 1.05–1.94, p = 0.025 and OR = 4.36, 95% CI 1.48–12.83, p = 0.007, respectively). A difference between the two occupational groups was also seen in extended haplotypes constructed from IL1B-IL1RN. The 211-1 haplotype was associated with an increased risk of OA among the dentists only with OR 2.39, 95% CI 1.30–4.41, p = 0.005. In addition, the 111-1 haplotype was associated with a reduced risk of OA among the teachers (OR = 0.43, 95% CI 0.20–0.93, p = 0.03) and an increased risk among the dentists (OR = 2.23, 95% CI 1.04–4.80, p = 0.04).

Concerning the amount of dental tasks, both low variation in work tasks and the carriage of the 211-1 haplotype had an effect on the risk of OA (OR = 3.45, 95% CI 1.22–9.74, p = 0.02 and OR = 5.69, 95% 1.86–17.42, p = 0.002, respectively). However, the combined effect of these two factors on the risk of

Page 75: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

73

OA was similar to the effect of the haplotype on the disease risk (OR = 5.63, 95% 1.93–16.46, p = 0.002).

5.4 Linkage analysis of primary hip and knee osteoarthritis (IV)

In this study, the diagnosis of hip and knee OA was based on the following criteria: Subjects were required to have suffered chronically from common OA symptoms in the affected hip or knee joint(s) and typical radiological findings associated with OA were to be seen in the standard x-ray joint projections. Subjects that were known to have any secondary cause of OA or symptoms of any other confounding rheumatoid joint disease were excluded from the study.

In this sample of a total of fifteen families, the average age of the disease onset was 50 years, although in many cases the first symptoms of joint disease had been noticed before the age of 40. Among the 279 studied individuals, OA of the hip joint was present in 23 subjects, OA of the knee joint in 20 subjects and 15 subjects were affected with both conditions. Characteristics of the studied families by the OA status are presented in Table 9.

Table 9. Characteristics of the studied families with the definitions of OA outcomes used for the linkage analysis.

Gender OA status Family n

Male Female Hip OA Knee OA Knee & Hip OA

Healthy Unknown

11 22 11 11 0 1 0 4 17 2 11 4 7 0 3 0 2 6 3 20 10 10 0 3 1 5 11 4 29 14 15 2 0 1 3 23 5 14 8 6 1 1 2 2 8 6 31 13 18 1 2 2 2 24 7 14 6 8 0 3 0 1 10 8 14 8 6 3 0 1 1 9 9 37 16 21 0 2 2 3 30 10 33 17 16 5 1 4 6 17 11 14 8 6 3 2 0 1 8 12 13 9 4 5 0 0 1 7 13 10 6 4 2 0 1 0 7 14 8 4 4 2 0 1 2 3 15 9 3 6 1 2 0 1 5 All 279 137 142 23 20 15 34 185 1 Three affected individuals were identified in family 1, but DNA was available only from one.

Page 76: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

74

5.4.1 Genome-wide scan

The whole genome scan of 225 individuals in ten different families resulted in the identification of four interesting loci on chromosomes 2, 11, 13 and 20 with logarithm of odds (LOD) scores above 1.5 using two-point linkage analysis. The same region on chromosome 2 (between markers D2S112 and D2S142) resulted in a LOD score of 2.14 in multipoint analysis, which was the highest peak detected in the genome wide scan. The second highest peak was detected on chromosome 11 with a maximum multipoint LOD score of 1.82.

5.4.2 Fine mapping

The two loci with the highest LOD scores at Chr 2 and 11 were further analyzed by fine mapping. Five additional families were recruited with 54 individuals. The linkage signal for Chr 11 disappeared in the fine mapping. Nevertheless, the susceptibility of Chr 2 was confirmed with a multipoint LOD score of 3.91 under the recessive model of inheritance. The same, approximately 7 cM region at 2q21 was also positive in two-point linkage analysis with a LOD score of 3.63 (see Figure 7 below). Surprisingly, mainly one family (family 10) contributed to the positive LOD scores in all analyses. This family consisted of a total of 33 studied members. The gender distribution was 17 males and 16 females from three generations, of whom primary OA of the hip joint was identified in 5 family members, OA of the knee joint in 1 family member and both conditions in 4 family members. Only 6 family members could be certainly defined as healthy concerning OA. Among 17 family members, the presence/absence of OA could not be confirmed and the status of these individuals was therefore considered unknown in the analyses. The pedigree of this family is presented in Figure 8.

Page 77: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

75

Fig. 7. Markers showing positive linkage in the fine mapping of Chr 2.

5.4.3 Association analyses of potential candidate genes

Several variations within the candidate genes at the positive linkage loci were analyzed. Coding regions and intron boundaries of TSG6, NURR1, CXCR4, FRZB, ACVR1 and CFC1 were analyzed and several single nucleotide polymorphisms in the IL1A, IL1B, IL1R1 and IL1RL2 genes were genotyped. There was no statistically significant difference in the genotype frequencies between the affected and healthy family members in any of the analyzed variations.

Page 78: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

76

Fig. 8. Pedigree of family 10. OA affected individuals are indicated by black circles and unaffected by white. Circles represent females and squares, males. Individuals marked with a diagonal line across the symbol are deceased. Status of OA could not be determined from individuals marked with a question mark.

Page 79: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

77

6 Discussion

6.1 The complex genetics of osteoarthritis

OA is a very common disease resulting in the destruction of synovial joints. Many of the mechanisms involved in the disease cascade of OA have been identified, and important molecular and signaling mechanisms behind its pathology have been described. Furthermore, it is known that there are a number of environmental factors predisposing to the disease and that it is more prevalent among women. It can also be stated that the contribution of genetic factors to this condition is complex, yet undisputable. Despite all this knowledge, the treatment of OA is still primarily based on a desperate fight against the progressing symptoms that arise from the continuous slow destruction of articular cartilage. The onset of the disease cannot be prevented because we have not been able to find or recognize the underlying cause that initiates this process. Nevertheless, scientists studying the etiology of OA are unanimous that the answer to this problem, at least to some extent, lies in the genome.

Since the discovery of the familial accumulation of nodes, we have come a long way in learning the genetic etiology of OA. An important basis for the current knowledge has been the identification of multiple Mendelian disorders and syndromes characterized by the juvenile arthritic phenotype. It has also been shown that the genetics of these chondrodysplasias often involve genes of important structural proteins of cartilage (for a review, see Kannu et al. 2009). Nearly all of the 22 autosomal chromosomes have been reported to putatively harbor one or more loci that are transmitted from parent to offspring together with OA more often than expected under independent inheritance. Only a few of these regions showing linkage to OA overlap or are in close proximity with another linked region. Also, only a few of the reported loci have shown significant evidence for linkage. In addition to all of the susceptible loci, there are dozens of individual genes that are considered to harbor OA-related variants. Again, only a handful of the reported susceptibility variants have provided significant evidence for association and have been successfully replicated in independent populations/study cohorts. This raises a question: Are some of the reported linkage and association findings false positives, or is it possible that the genetic etiology of OA would be this complicated? It is likely that both of these alternatives are at least partly true.

Page 80: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

78

The search for susceptibility genes has progressed during the last few years. At least to some extent, it might be because OA is now assumed to be not a single entity, but more likely a group of distinct diseases. This means that the contribution of genes may differ between sexes and according to the localization and pattern of the affected joints. This doctoral study suggests that even the development of symptoms in this disease may, at least partly, arise from a genetic variation. Applying some of the preceding stratifications to carefully selected patient material has generally led to better results in OA genetics. It is noteworthy that even though some of the candidate genes, for example ASPN, are repeatedly shown to associate with OA, many times the results have not been successfully confirmed in other ethnic groups (Rodriguez-Lopez et al. 2006). Generally, this leads to the assumption that the genetic etiology of OA may be significantly different among populations with varying ethnic backgrounds. This hypothesis is supported by the fact that there seems to be also small groups, even independent families that demonstrate the typical OA phenotype with unique predisposing genetic factors that cannot be identified in larger populations of the same origin, as discussed later in section 6.4. Another likely reason for the discrepancies seen in the results between different populations lies in the low statistical power of the study material, which is still common in OA studies.

The preceding points make it difficult to give a comprehensive description of the genetic etiology of OA because the results obtained to date are divergent. However, the recent success in GWA studies has begun to reveal common OA-associated variations that seem to be modest but universal. Despite the wealth of genetic data obtained from OA, a lot of the needed information is yet to be found. The significance and contribution of the work presented in this thesis on uncovering of the mystery of OA genetics is discussed in detail in the following sections.

6.2 Characteristics of the study populations

The study sample consisting of dentists and teachers was well characterized and large enough to show that the reported associations are plausible. Possible population stratification was avoided by random selection of subjects from a common geographical origin. However, defining the phenotype of OA in multiple ways significantly reduced the statistical power in these analyses and increased the possibility of a type 1 error. Thus, a larger sample size would have been beneficial. The study setting was cross-sectional, meaning that the relationship

Page 81: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

79

between symptoms of OA, radiological findings and different genotypes were evaluated at a certain point in time. Therefore, one must take into account the possibility that the information collected concerning the symptoms of OA might have been different if the study setting was longitudinal. Utilizing two different occupational groups enabled the investigation of the relationship between different patterns of joint usage and analyzed genotypes. This was a clear strength of the study. Analyzed genetic variations were chosen based on their previously reported associations or known functionality. This increases the value of the reported results because, if valid, they may be useful in the estimation of the proportional risk of OA, as suggested previously (Valdes et al. 2009).

Families used in the linkage analysis had a common geographical origin. Only symptomatic OA was studied, which may have affected the results. Sizes of the families were generally adequate, but the number of affected individuals was relatively low, which is likely the reason why only one family was shown to contribute significantly to the resulting linkage. This matter is discussed in more detail in section 6.5. In light of recent studies, combining the hip and knee OA cases in the analysis could also be criticized. Unfortunately, the statistical power of this sample would have been inadequate for separate analysis.

6.3 Is the effect of the aggrecan VNTR polymorphism significant in the development of osteoarthritis? (I)

Aggrecan is the most abundant structural proteoglycan in articular cartilage and its degradation is among the first identified molecular events that take place in the beginning of OA (Magdalou et al. 2008). The common VNTR polymorphism in AGC1 has a clear biological function as the number of repeat sequences in the VNTR area correlate to the amount of GAG side chain binding sites in the mature protein. Prior to this study, only two previous association studies concerning the VNTR polymorphism and OA had been conducted with substantially smaller sample sizes and somewhat ambiguous results. Horton et al. (1998) reported that in a sample of 93 Caucasian men, allele A27 with 27 repeats increased risk of hand and knee OA (OR 3.23, CI 1.24–8.41), whereas Kirk et al. (2003) reported the deletorius effect of A27 and protective effect of alleles A25 and A28 among 134 twins for OA of the hand, hip and knee. In contrast to these previous findings, the results reported in the current study suggest that A27 is protective against HOA if the subject is homozygous for it. With the allele frequencies reported in this study (0.4 for OA cases and 0.53 for controls), the sufficient amount of cases

Page 82: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

80

and controls in order to detect the association with 80% certainty were 210 and 294, respectively. Thus, the sample size in this study (224 affected and 306 controls with A27) was sufficient in terms of statistical power. A27 was also the most common allele in our sample. Furthermore, it was observed that carrying two copies of any of the shorter numbers of repeat sequences than 27 predisposes to more serious forms of HOA. Also, subjects with two copies of any longer alleles than 27 repeats had a slightly increased risk for severe HOA, but the low number of the alleles reduced the statistical power concerning this finding. The first two observations are supported by the biological function of the polymorphism. As A27 was the most common allele in our sample, it is logical to think that the amount of GAG attachment sites it encodes is sufficient for the functioning of the mature protein, and thus associates with lower disease frequency. Furthermore, the HOA predisposing effect of two shorter alleles could arise from the resulting inadequate amount of GAG attachment sites in the core protein that leads to reduced osmotic swelling pressure of the proteoglycan network during physical strain of the tissue. Another explanation of this finding could be the easier diffusion of partially cleaved aggrecan molecules, encoded by shorter alleles, into the synovial fluid as suggested in a recent study describing a similar association of the shorter alleles with RA (de Souza et al. 2008).

Despite the findings presented in this study and the preceding conclusions that can be drawn based on the known effect of the VNTR polymorphism on the molecular structure of aggrecan, it still remains unclear whether or not there is a direct link between aggrecan VNTR and OA. The number of reported associations so far is not convincing and needs to be confirmed in other populations that are large enough to demonstrate the possible predisposing effect of the rare alleles with either a very low or high repeat number. The low frequency of certain alleles may be another reason why the significance of the AGC1 gene in cartilage breakdown has been difficult to confirm. As discussed previously, the number of subjects with two rare alleles may be only one or two in a sample of hundreds of individuals. Yet, if at risk for OA, these subjects would represent a substantial proportion in the entire population (Roughley et al. 2006b). It is also a possibility that the AGC1 VNTR polymorphism is not directly linked to OA susceptibility, but is in linkage equilibrium with some other nearby susceptibility locus that has not yet been identified.

Even if the role of this VNTR polymorphism in the etiology of OA can be confirmed in the genetic studies undertaken in the future, a set of functional

Page 83: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

81

studies are still needed in order to test the effect of different polymorphic alleles on the actual homeostasis of the proteoglycan network in the ECM.

6.4 Genetic factors affect the inflammation reaction observed in osteoarthritis cartilage (II, III)

IL-6 is an essential cytokine of the inflammation cascade. In OA cartilage, the level of IL-6 mRNA has proven to be elevated along with the amount of this cytokine in serum and synovial fluid in OA patients (Kaneko et al. 2000). Recently it was also reported that the expression of IL-6 is increased in the subchondral bone osteoblasts of OA tissue (Sakao et al. 2008) suggesting it is affecting cartilage metabolism during the disease. Furthermore, mechanical compression seems to increase the IL-6 production by osteoblasts (Sanchez et al. 2008). Among the various polymorphisms recognized in the IL-6 gene, those in the promoter area in particular have been reported to affect the transcriptional activity of the gene (Osiri et al. 1999, Jordanies et al. 2000, Ota et al. 2001). In the present study, it was observed that the G alleles at promoter positions -174 and -597 individually and also together were more common among subjects with symptomatic DIP OA. The predisposing effect of the G alleles was also evident when haplotypes constructed from variants G-597A, G-572C and G-174C were compared between subjects with or without symptomatic DIP OA. The haplotype G-G-G was overrepresented in the women with symptomatic DIP OA in comparison with those without the disease. Several previous and recent findings support these results and their possible contribution to the inflammation reaction present in OA: G alleles at -597 and -174 are related to increased ex vivo production of IL-6 (Rivera-Chavez et al. 2003), and at -174 individually, the G allele is related to increased expression and plasma levels of the molecule (Bennermo et al. 2004). The CC genotype at -174 was significantly higher among the control subjects in a hip OA population (Pola et al. 2005), and the G alleles in these three loci generally associate with increased transcription of IL-6 (Terry et al. 2000). The number of affected individuals in this study was rather low (n = 48, 9%). Furthermore, the only other reported association study by Pola et al. (2005) that showed OR 0.4 (CI 0.1–0.9, p = 0.04) for the carriage of the CC genotype had an even smaller sample size (75 patients and 107 controls). Thus, the reported associations need to be replicated in a larger OA sample. No associations between the studied genotypes or haplotypes and DIP OA without symptoms (ROA) were seen, which suggests that the studied cytokine may affect the development of the

Page 84: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

82

symptoms through inflammatory pathways. Pain is a common feature of the inflammation reaction; its connection to IL6 is evident and has been observed concerning genotype GG at -174 also in patients with juvenile RA (Oen et al. 2005).

Taken together, these observations provoke the idea that the influence of IL-6 on the synovial joint could be therapeutically modified in order to treat OA. One such intervention could be a specific receptor antagonist that binds to the IL-6 receptor, preventing the response in target cells. However, designing a molecule with the desired properties could be difficult due to the fact that many cytokines share the same signaling pathway as IL-6, with some of the same structural receptor elements (Heinrich et al. 1998). Concerning IL-1, such an antagonist, named anakinra, is already in use to treat RA (Bingham 2008).

When studying the IL-1 gene cluster, it was observed in this sample that carrying of the minor alleles of two IL1B polymorphisms (rs1143633 and rs1143634) and the IL1B-IL1RN extended haplotypes (211-1 and 121-1) were associated with an increased risk of bilateral DIP OA. Also, extended haplotypes of IL1R1-IL1RL2 and IL1B-IL1RN associated with the disease, although, differently between the two occupational groups. It is possible that this kind of controversy within the cohort is caused by chance due to the multiple haplotype alleles analyzed in a study sample of limited size.

Numerous association studies have suggested the role of different SNPs and haplotypes contributing to the etiology of OA as stated in section 2.4.1. Functional studies have suggested that the combination of alleles may be an important aspect in the regulation of IL1 gene expression. The inflammatory response is regulated by a balance between pro- and anti-inflammatory cytokines. A risk haplotype reported here includes the alleles that can affect the production of both IL-1Ra and IL-1β (Hurme & Santila 1998, Vamvakopoulos et al. 2002). A certain combination of haplotypes could, therefore, result in the imbalance in cytokine production and contribute to the destructive process of cartilage (Martel-Pelletier et al. 1999). Subjects representing two different academic groups in terms of repetitive joint usage were analyzed in this study. It is known that the mechanical environment can have a substantial effect on cartilage homeostasis and on the production of inflammatory mediators. Stronger associations between the IL-1 gene cluster polymorphisms and OA observed among the dentists in this sample supports the idea that environmental factors can modulate the effect of modest genetic predisposing alleles on OA risk. Yet, due to the high LD rates observed in the area, the possibility remains that none of the reported IL-1 gene

Page 85: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

83

cluster alleles are independently contributing to OA risk, but are in linkage equilibrium with unidentified risk alleles situated nearby. In summary, it seems rather certain that this group of genes is somehow contributing to the etiology of OA, but further investigation is needed in order to clarify the role of the numerous polymorphisms and their possible interaction in different forms of OA.

Despite the fact that the IL-1 gene cluster and IL-6 can be shown to play a role in OA, one must acknowledge that there are also multiple other cytokines present in OA cartilage and that the mechanisms of their interrelated regulation as well as hierarchy are complex. Therefore, more information is needed before the role of these molecules can be accurately defined in the disease process. It may also be speculated that the recognized inflammation process in OA is actually a secondary phenomenon, the consequence of the progressing degradation of cartilage that has been originally initiated by totally different molecular events. Nevertheless, if the role of the cytokines will prove to be significant in the etiology of OA, modifying their effects in AC is likely to result in the development of new therapeutic interventions.

6.5 Linkage analysis, a way to identify new osteoarthritis-associated chromosomal loci (IV)

In this study, a new chromosomal locus that is in linkage with the primary OA of hip and knee was identified in Chr 2q21 with a LOD score of 3.91. Chr 2 has previously been shown to contain multiple loci that are in linkage with nodal OA (Wright et al. 1996), HOA (Leppävuori et al. 1999, Demissie et al. 2002, Stefánsson et al. 2003, Hunter et al. 2004, Livshits et al. 2007), hip OA (Loughlin et al. 2002b) and generalized OA (Meulenbelt et al. 2006). Nevertheless, the locus reported here does not overlap with any of the previously reported loci. In the current study, the identification of a specific predisposing variation within the linked region has been unsuccessful and thus, these results at this point mainly strengthen the hypothesis that Chr 2 harbors multiple genetic variants that are likely to be important in the disease etiology. Mainly, only one of the studied families contributed significantly to the observed LOD. It may therefore be that the underlying genetic variation is restricted to this family and does not necessarily contribute to disease prevalence in others. These kinds of independent families with unique disease-associated variations have been reported previously, and the clinical outcome of the disease in these cases can be very similar to or even indistinguishable from OA (Meulenbelt et al. 1997, Ingvarsson et al. 2001,

Page 86: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

84

Jakkula et al. 2005, Mabuchi et al. 2006). It may be due to the limited number of affected individuals, however, which resulted in none of the other families contributing strongly to the observed LOD. It is also noteworthy that no significant linkage was observed using the dominant inheritance model in the analyses. Concerning complex diseases, it has been thought that the disease is more likely inherited according to the dominant Mendelian model. Nevertheless, this study suggests it is possible that some of the genetic features predisposing to the disease are transmitted from parent to offspring following the recessive model of inheritance. This assumption is also supported by the results obtained in the IL1 gene cluster association analysis (III).

Page 87: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

85

7 Conclusions

Several genome-wide scans and an even greater number of association reports of individual genes conducted to date represent important milestones on the way towards better understanding of OA. So far, none of these milestones alone can be considered major enough to offer a convincing explanation of the disease etiology. If anything, individual results are pieces in a puzzle that is proving to be much larger than anticipated by former investigators of OA genetics. Many promising individual discoveries have been made since the beginning of this doctoral study; nevertheless, we are still unable to see what the puzzle illustrates.

A key to the mystery of OA genetics has traditionally been sought from the genes that are expected to be important for cartilage homeostasis. Mostly, this has not been as rewarding as one would expect. Variants within some of the genes of structural proteins clearly harbor OA susceptibility, but cannot be considered as major disease-causing factors. Recently, the focus of investigation has been more on the genes of different signaling molecules found in cartilage. Convincing results obtained with ASPN, FRZB, GDF5 and MATN3 prove that it is rather the signal transduction pathway, which is likely to be a major component in OA susceptibility. This is an important discovery, since these pathways are potentially modifiable and can therefore be suitable targets for therapeutic interventions. The question is: Will any of the known susceptibility loci prove to be compelling enough to provoke the development of a new treatment?

There is already a wealth of genetic data on OA that keeps building up as new susceptibility loci are reported more frequently than before. The knowledge of the molecular genetics is developing quickly, and we are already taking advantage of new powerful genotyping techniques that are likely to revolutionize the understanding of OA genetics in the near future. It is possible that as the genetics of OA will sharpen, the genes considered important in this study will be replaced by others that prove to harbor variations that are more essential for the disease etiology as a whole. Still, every locus that has proven to affect the susceptibility of OA may be important. The information from different association analyses could be gathered together in order to design a gene test to evaluate the individual risk of OA. Thus, a person carrying multiple genetic risk factors could avoid the known environmental factors that further predispose to the disease. This information could also be used to genotype chondrocytes that are used for autologous transplantation treatment in order to choose those cells with minimal OA predisposing loci.

Page 88: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

86

This doctoral study underlines the importance of the careful selection of populations for future studies. Ethnically homogenous subjects with a common clinical outcome in terms of joint location and symptoms as well as strict exclusion of other possible confounding factors will help to ensure that the investigated genetic contribution, which in most cases has proven to be modest, will not be overwhelmed. It would also be beneficial if the same definitions and classifications of OA could be used among different populations. It is likely that the standards concerning these definitions of disease outcomes will develop in the future as our understanding of OA grows.

It remains to be seen whether we will continue to collect the small pieces of the puzzle or eventually find a major OA locus. The growing evidence of the complexity of this disease makes the latter option seem rather unlikely. Nevertheless, by gaining a little amount of new information step by step, we are on the way towards understanding why cartilage ultimately breaks down. It is a fundamental matter to solve if we are determined to overcome the burden of osteoarthritis.

Page 89: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

87

References Abu-Hijleh G, Reid O & Scothorne RJ (1997) Cell death in the developing chick knee

joint: I. Spatial and temporal patterns. Clin Anat 10(3): 183–200. Ahlbäck S (1968) Osteoarthrosis of the knee. A radiographic investigation. Acta Radiol

Diagn (Stockh) 277(Suppl): 7–72. Aigner T, Reichenberger E, Bertling W, Kirsch T, Stöss H & von der Mark K (1993) Type

X collagen expression in osteoarthritic and rheumatoid articular cartilage. Virchows Arch B Cell Pathol Incl Mol Pathol 63(4): 205–211.

Aigner T & Stöve J (2003) Collagens–major component of the physiological cartilage matrix, major target of cartilage degeneration, major tool in cartilage repair. Adv Drug Deliv Rev 55(12): 1569–1593.

Ala-Kokko L & Prockop DJ (1990) Completion of the intron-exon structure of the gene for human type II procollagen (COL2A1): variations in the nucleotide sequences of the alleles from three chromosomes. Genomics 8(3): 454–460.

Ala-Kokko L, Kvist AP, Metsäranta M, Kivirikko KI, de Crombrugghe B, Prockop DJ & Vuorio E (1995) Conservation of the sizes of 53 introns and over 100 intronic sequences for the binding of common transcription factors in the human and mouse genes for type II procollagen (COL2A1). Biochem J 308(3): 923–929.

Alexander CJ (1999) Heberden´s and Bouchard´s nodes. Ann Rheum Dis 58: 675–678. Andrianakos AA, Kontelis LK, Karamitsos DG, Aslanidis SI, Georgountzos AI, Kaziolas

GO, Pantelidou KV, Vafiadou EV & Dantis PC (2006) Prevalence of symptomatic knee, hand, and hip osteoarthritis in Greece. The ESORDIG study. J Rheumatol 33(12): 2507–2513.

Archer CW, Morrison H & Pitsillides AA (1994) Cellular aspects of the development of diarthrodial joints and articular cartilage. J Anat 184(3): 447–56.

Arden N & Nevitt MC (2006) Osteoarthritis: epidemiology. Best Pract Res Clin Rheumatol 20(1): 3–25.

Aston CE & Wilson SR (1986) Two-point versus multipoint linkage analysis: a statistical view. Genet Epidemiol Suppl 1: 113–116.

Atif U, Philip A, Aponte J, Woldu EM, Brady S, Kraus VB, Jordan JM, Doherty M, Wilson AG, Moskowitz RW, Hochberg M, Loeser R, Renner JB & Chiano M (2008) Absence of association of asporin polymorphisms and osteoarthritis susceptibility in US Caucasians. Osteoarthritis Cartilage 16(10): 1174–1177.

Attur MG, Dave M, Akamatsu M, Katoh M & Amin AR (2002) Osteoarthritis or osteoarthrosis: the definition of inflammation becomes a semantic issue in the genomic era of molecular medicine. Osteoarthritis Cartilage 10(1): 1–4.

Baldwin CT, Cupples LA, Joost O, Demissie S, Chaisson C, Mcalindon T, Myers RH & Felson D (2002) Absence of linkage or association for osteoarthritis with the vitamin D receptor/type II collagen locus: the Framingham Osteoarthritis Study. J Rheumatol 29(1): 161–165.

Benedek TG (2006) A history of the understanding of cartilage. Osteoarthritis Cartilage 14(3): 203–9.

Page 90: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

88

Bennermo M, Held C, Stemme S, Ericsson CG, Silveira A, Green F, Tornvall P (2004) Genetic predisposition of the interleukin-6 response to inflammation: implications for a variety of major diseases? Clin Chem 50: 2136–2140.

Bierma-Zeinstra SM & Koes BW (2007) Risk factors and prognostic factors of hip and knee osteoarthritis. Nat Clin Pract Rheumatol 3(2): 78–85.

Bingham CO 3rd (2008) Emerging therapeutics for rheumatoid arthritis. Bull NYU Hosp Jt Dis 66(3): 210–215.

Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ & Cerretti DP (1997) A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385(6618): 729–733.

Blaschke UK, Eikenberry EF, Hulmes DJ, Galla HJ & Bruckner P (2000) Collagen XI nucleates self-assembly and limits lateral growth of cartilage fibrils. J Biol Chem 275(14): 10370–10378.

Bonnet CS & Walsh DA (2005) Osteoarthritis, angiogenesis and inflammation. Rheumatology (Oxford) 44(1): 7–16.

Briggs MD, Hoffman SM, King LM, Olsen AS, Mohrenweiser H, Leroy JG, Mortier GR, Rimoin DL, Lachman RS, Gaines ES, Cekleniak JA, Knowlton RG & Cohn DH (1995) Pseudoachondroplasia and multiple epiphyseal dysplasia due to mutations in the cartilage oligomeric matrix protein gene. Nat Genet 10(3): 330–336.

Briggs MD, Mortier GR, Cole WG, King LM, Golik SS, Bonaventure J, Nuytinck L, De Paepe A, Leroy JG, Biesecker L, Lipson M, Wilcox WR, Lachman RS, Rimoin DL, Knowlton RG & Cohn DH (1998) Diverse mutations in the gene for cartilage oligomeric matrix protein in the pseudoachondroplasia-multiple epiphyseal dysplasia disease spectrum. Am J Hum Genet 62(2): 311–319.

Bruckner P & van der Rest M (1994) Structure and function of cartilage collagens. Microsc Res Tech 28(5): 378–384.

Bukulmez H, Matthews AL, Sullivan CM, Chen C, Kraay MJ, Elston RC, Moskowitz RW, Goldberg VM & Warman ML (2006) Hip joint replacement surgery for idiopathic osteoarthritis aggregates in families. Arthritis Res Ther 8(1): R25.

Carman WJ, Sowers M, Hawthorne VM & Weissfeld LA (1994) Obesity as a risk factor for osteoarthritis of the hand and wrist: a prospective study. Am J Epidemiol 139(2): 119–129.

Carney SL, Billingham ME, Caterson B, Ratcliffe A, Bayliss MT, Hardingham TE & Muir H (1992) Changes in proteoglycan turnover in experimental canine osteoarthritic cartilage. Matrix 12(2): 137–147.

Chaisson CE, Zhang Y, McAlindon TE, Hannan MT, Aliabadi P, Naimark A, Levy D & Felson DT (1997) Radiographic hand osteoarthritis: incidence, patterns, and influence of pre-existing disease in a population based sample. J Rheumatol 24(7): 1337–1343.

Chang DP O, Abu-Lail NI O, Guilak F, Jay GD & Zauscher S (2008) Conformational Mechanics, Adsorption, and Normal Force Interactions of Lubricin and Hyaluronic Acid on Model Surfaces. Langmuir 24(4): 1183–1193.

Page 91: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

89

Chapman K, Mustafa Z, Dowling B, Southam L, Carr A & Loughlin J (2002) Finer linkage mapping of primary hip osteoarthritis susceptibility on chromosome 11q in a cohort of affected female sibling pairs. Arthritis Rheum 46(7): 1780–1783.

Chapman K & Loughlin J (2006) Association of the interleukin-1 gene cluster with osteoarthritis of the hip: comment on the article by Meulenbelt et al and the letter by Smith et al. Arthritis Rheum 54(11): 3722–3723.

Chen FH, Herndon ME, Patel N, Hecht JT, Tuan RS & Lawler J (2007) Interaction of cartilage oligomeric matrix protein/thrombospondin 5 with aggrecan. J Biol Chem 282(34): 24591–24598.

Chen TL, Posey KL, Hecht JT & Vertel BM (2008) COMP mutations: Domain-dependent relationship between abnormal chondrocyte trafficking and clinical PSACH and MED phenotypes. J Cell Biochem 103(3): 778–787.

Choi JB, Youn I, Cao L, Leddy HA, Gilchrist CL, Setton LA & Guilak F (2007) Zonal changes in the three-dimensional morphology of the chondron under compression: the relationship among cellular, pericellular, and extracellular deformation in articular cartilage. J Biomech 40(12): 2596–2603.

Cicuttini FM, Baker JR & Spector TD (1996) The association of obesity with osteoarthritis of the hand and knee in women: a twin study. J Rheumatol 23(7): 1221–1226.

Cimmino MA & Parodi M (2005) Risk factors for osteoarthritis. Semin Arthritis Rheum 34(6 Suppl 2): 29–34.

Cooper C, Snow S, McAlindon TE Kellingray S, Stuart B, Coggon D & Dieppe PA (2000) Risk factors for the incidence and progression of radiographic knee osteoarthritis. Arthritis Rheum 43: 995–1000.

Corvol MT (2000) The chondrocyte: from cell aging to osteoarthritis. Joint Bone Spine 67(6): 557–560.

Craig FM, Bentley G & Archer CW (1987) The spatial and temporal pattern of collagens I and II and keratan sulphate in the developing chick metatarsophalangeal joint. Development 99(3): 383–91.

Croft P, Cooper C, Wickham C & Coggon D (1990) Defining osteoarthritis of the hip for epidemiologic studies. Am J Epidemiol 132(3): 514–522.

Cushnaghan J & Dieppe P (1991) Study of 500 patients with limb joint osteoarthritis. I. Analysis by age, sex, and distribution of symptomatic joint sites. Ann Rheum Dis 50(1): 8–13.

Dahaghin S, Bierma-Zeinstra SM, Reijman M, Pols HA, Hazes JM & Koes BW (2005) Does hand osteoarthritis predict future hip or knee osteoarthritis? Arthritis Rheum 52(11): 3520–3527.

Dahaghin S, Bierma-Zeinstra SM, Hazes JM & Koes BW (2006) Clinical burden of radiographic hand osteoarthritis: a systematic appraisal. Arthritis Rheum 55(4): 636–647.

Dawn Teare M & Barrett JH (2005) Genetic linkage studies. Lancet 366(9490): 1036–1044.

Deák F, Wagener R, Kiss I & Paulsson M (1999) The matrilins: a novel family of oligomeric extracellular matrix proteins. Matrix Biol 18(1): 55–64.

Page 92: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

90

Demissie S, Cupples LA, Myers R, Aliabadi P, Levy D & Felson DT (2002) Genome scan for quantity of hand osteoarthritis: the Framingham Study. Arthritis Rheum 46(4): 946–952.

Dieppe PA (2005) Relationship between symptoms and structural change in osteoarthritis: what are the important targets for therapy? J Rheumatol 32(6): 1147–1149.

Dillon CF, Hirsch R, Rasch EK & Gu Q (2007) Symptomatic hand osteoarthritis in the United States: prevalence and functional impairment estimates from the third U.S. National Health and Nutrition Examination Survey, 1991–1994. Am J Phys Med Rehabil 86(1): 12–21.

Dinarello CA (2007) Historical insights into cytokines. Eur J Immunol 37(Suppl 1): S34–S45.

Doege KJ, Coulter SN, Meek LM, Maslen K & Wood JG (1997) A human-specific polymorphism in the coding region of the aggrecan gene. Variable number of tandem repeats produce a range of core protein sizes in the general population. J Biol Chem 272(21): 13974–13979.

Dougados M (2004) Structural progression is also driven by clinical symptoms in patients with osteoarthritis. Arthritis Rheum 50(5): 1360–1365.

Dudhia J, Davidson CM, Wells TM, Vynios DH, Hardingham TE & Bayliss MT (1996) Age-related changes in the content of the C-terminal region of aggrecan in human articular cartilage. Biochem J 313(3): 933–940.

Dudhia J (2005) Aggrecan, aging and assembly in articular cartilage. Cell Mol Life Sci 62(19–20): 2241–2256.

Edwards JC (1994) The nature and origins of synovium: experimental approaches to the study of synoviocyte differentiation. J Anat 184( 3): 493–501.

Egger P, Cooper C, Hart DJ, Doyle DV, Coggon D & Spector TD (1995) Patterns of joint involvement in osteoarthritis of the hand: the Chingford Study. J Rheumatol 22(8): 1509–1513.

Eliasson GJ, Verbruggen G, Stefansson SE, Ingvarsson T & Jonsson H (2006) Hand radiology characteristics of patients carrying the T(303)M mutation in the gene for matrilin-3. Scand J Rheumatol 35(2): 138–142.

Eyre DR (1991) The collagens of articular cartilage. Semin Arthritis Rheum 21(3 Suppl 2): 2–11.

Eyre DR & Wu JJ (1995) Collagen structure and cartilage matrix integrity. J Rheumatol Suppl 43: 82–85.

Eyre DR, Wu JJ, Fernandes RJ, Pietka TA & Weis MA (2002) Recent developments in cartilage research: matrix biology of the collagen II/IX/XI heterofibril network. Biochem Soc Trans 30(6): 893–899.

Eyre DR, Weis MA & Wu JJ (2006) Articular cartilage collagen: an irreplaceable framework? Eur Cell Mater 12: 57–63.

Felson DTR, Zhang Y, Hannan MT, Naimark A, Weissman B, Aliabadi P & Levy D (1997) Risk factors for incident radiographic knee osteoarthritis in the elderly: The Framingham Study. Arthritis Rheum 40: 728–733.

Page 93: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

91

Fernandes JC, Martel-Pelletier J, Lascau-Coman V, Moldovan F, Jovanovic D, Raynauld JP & Pelletier JP (1998) Collagenase-1 and collagenase-3 synthesis in normal and early experimental osteoarthritic canine cartilage: an immunohistochemical study. J Rheumatol 25(8): 1585–2594.

Fernandes JC, Martel-Pelletier J & Pelletier JP (2002) The role of cytokines in osteoarthritis pathophysiology. Biorheology 39(1–2): 237–246.

Fichard A, Kleman JP & Ruggiero F (1995) Another look at collagen V and XI molecules. Matrix Biol 14(7): 515–531.

Fontana L, Neel S, Claise JM, Ughetto S & Catilina P (2007) Osteoarthritis of the thumb carpometacarpal joint in women and occupational risk factors: a case-control study. J Hand Surg 32(4): 459–465.

Forster T, Chapman K & Loughlin J (2004a) Common variants within the interleukin 4 receptor alpha gene (IL4R) are associated with susceptibility to osteoarthritis. Hum Genet 114(4): 391–5.

Forster T, Chapman K, Marcelline L, Mustafa Z, Southam L & Loughlin J (2004b) Finer linkage mapping of primary osteoarthritis susceptibility loci on chromosomes 4 and 16 in families with affected women. Arthritis Rheum 50(1): 98–102.

Frazer KA, Murray SS, Schork NJ & Topol EJ (2009) Human genetic variation and its contribution to complex traits. Nat Rev Genet 10(4): 241–251.

Fumagalli M, Sarzi-Puttini P & Atzeni F (2005) Hand osteoarthritis. Semin Arthritis Rheum. 34(6 Suppl 2): 47–52.

Fytili P, Giannatou E, Karachalios T, Malizos K & Tsezou A (2005) Interleukin-10G and interleukin-10R microsatellite polymorphisms and osteoarthritis of the knee. Clin Exp Rheumatol 23(5): 621–627.

Ge Z, Hu Y, Heng BC, Yang Z, Ouyang H, Lee EH & Cao T (2006) Osteoarthritis and therapy. Arthritis Rheum 55(3): 493–500.

Gelse K, Pöschl E & Aigner T (2003) Collagens--structure, function, and biosynthesis. Adv Drug Deliv Rev 55(12): 1531–1546.

Geng Y, McQuillan D & Roughley PJ (2006) SLRP interaction can protect collagen fibrils from cleavage by collagenases. Matrix Biol 25(8): 484–491.

Glasson SS, Askew R, Sheppard B, Carito B, Blanchet T, Ma HL, Flannery CR, Peluso D, Kanki K, Yang Z, Majumdar MK & Morris EA (2005) Deletion of active ADAMTS5 prevents cartilage degradation in a murine model of osteoarthritis. Nature 434(7033): 644–648.

Goldring MB (2006) Update on the biology of the chondrocyte and new approaches to treating cartilage diseases. Best Pract Res Clin Rheumatol 20(5): 1003–1025.

Goldring SR & Goldring MB (2006) Clinical aspects, pathology and pathophysiology of osteoarthritis. J Musculoskelet Neuronal Interact 6(4): 376–378.

Goldring MB & Goldring SR (2007) Osteoarthritis. J Cell Physiol 213(3): 626–634. Goldstein DB, Ruiz Linares A, Cavalli-Sforza LL & Feldman MW (2005) An evaluation

of genetic distances for use with microsatellite loci. Genetics 139(1): 463–471.

Page 94: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

92

Gomes R, Kirn-Safran C, Farach-Carson MC & Carson DD (2002) Perlecan: an important component of the cartilage pericellular matrix. J Musculoskelet Neuronal Interact 2(6): 511–516.

Gregory KE, Oxford JT, Chen Y, Gambee JE, Gygi SP, Aebersold R, Neame PJ, Mechling DE, Bächinger HP & Morris NP (2000) Structural organization of distinct domains within the non-collagenous N-terminal region of collagen type XI. J Biol Chem 275(15): 11498–11506.

Greig C, Spreckley K, Aspinwall R, Gillaspy E, Grant M, Ollier W, John S, Doherty M, & Wallis G (2006) Linkage to nodal osteoarthritis: quantitative and qualitative analyses of data from a whole-genome screen identify trait-dependent susceptibility loci. Ann Rheum Dis 65(9): 1131–1138.

Guilak F, Alexopoulos LG, Upton ML, Youn I, Choi JB, Cao L, Setton LA & Haider MA (2006) The pericellular matrix as a transducer of biomechanical and biochemical signals in articular cartilage. Ann N Y Acad Sci 1068: 498–512.

Haara MM, Manninen P, Kröger H, Arokoski JP, Kärkkäinen A, Knekt P, Aromaa A & Heliövaara M (2003) Osteoarthritis of finger joints in Finns aged 30 or over: prevalence, determinants, and association with mortality. Ann Rheum Dis 62(2): 151–158.

Haara MM, Heliövaara M, Kröger H, Arokoski JP, Manninen P, Kärkkäinen A, Knekt P, Impivaara O & Aromaa A (2004) Osteoarthritis in the carpometacarpal joint of the thumb. Prevalence and associations with disability and mortality. J Bone Joint Surg Am 86-A(7): 1452–1457.

Hagg R, Bruckner P & Hedbom E (1998) Cartilage fibrils of mammals are biochemically heterogeneous: differential distribution of decorin and collagen IX. J Cell Biol 142(1): 285–294.

Haines RW (1941) The Tetrapod knee joint. J Anat 76: 270–301. Haines RW (1942) The evolution of epiphyses and of endochondral bone. Biol Rev 17:

267–292. Halász K, Kassner A, Mörgelin M & Heinegård D (2007) COMP acts as a catalyst in

collagen fibrillogenesis. J Biol Chem 282(43): 31166–31173. Hämäläinen SH, Solovieva S, Hirvonen A, Vehmas T, Takala EP, Riihimäki H, Leino-

Arjas P (in press) COL2A1 gene polymorphisms and susceptibility to hand osteoarthritis in Finnish women. Ann Rheum Dis.

Hambach L, Neureiter D, Zeiler G, Kirchner T & Aigner T (1998) Severe disturbance of the distribution and expression of type VI collagen chains in osteoarthritic articular cartilage. Arthritis Rheum 41(6): 986–996.

Hannan MT, Felson DT & Pincus T (2000) Analysis of the discordance between radiographic changes and knee pain in osteoarthritis of the knee. J Rheumatol 27: 1513–1517.

Haq I, Murphy E & Dacre J (2003) Osteoarthritis. Postgrad Med J 79(933): 377–383. Hardingham TE & Fosang AJ (1995) The structure of aggrecan and its turnover in

cartilage. J Rheumatol Suppl 43: 86–90.

Page 95: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

93

Hart DJ, Doyle DV & Spector TD (1999) Incidence and risk factors for radiographic knee osteoarthritis in middle-aged women: The Chingford Study. Arthritis Rheum 42: 17–24.

Hartmann C & Tabin CJ (2001) Wnt-14 plays a pivotal role in inducing synovial joint formation in the developing appendicular skeleton. Cell 104: 341–351.

Hauser N, Paulsson M, Heinegârd D & Mörgelin M (1996) Interaction of cartilage matrix protein with aggrecan. Increased covalent cross-linking with tissue maturation. J Biol Chem 271(50): 32247–32252.

Hecht JT, Nelson LD, Crowder E, Wang Y, Elder FF, Harrison WR, Francomano CA, Prange CK, Lennon GG, Deere M & Lawler J (1995) Mutations in exon 17B of cartilage oligomeric matrix protein (COMP) cause pseudoachondroplasia. Nat Genet 10(3): 325–329.

Hecht JT, Hayes E, Haynes R & Cole WG (2005) COMP mutations, chondrocyte function and cartilage matrix. Matrix Biol 23(8): 525–533.

Heino J (2007) The collagen family members as cell adhesion proteins. Bioessays 29(10): 1001–1010.

Heinrich PC, Behrmann I, Müller-Newen G, Schaper F & Graeve L (1998) Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem J 1;334: 297–314.

Heliövaara M, Mäkelä M, Impivaara O, Knekt P, Aromaa A & Sievers K (1993) Association of overweight, trauma and workload with coxarthrosis. A health survey of 7,217 persons. Acta Orthop Scand 64: 513–518.

van der Helm-van Mil AH, Padyukov L, Toes RE, Klareskog L & Huizinga TW (2008) Genome-wide single-nucleotide polymorphism studies in rheumatology: Hype or hope? Arthritis Rheum 58(9): 2591–2597.

Hildebrand A, Romarís M, Rasmussen LM, Heinegård D, Twardzik DR, Border WA & Ruoslahti E (1994) Interaction of the small interstitial proteoglycans biglycan, decorin and fibromodulin with transforming growth factor beta. Biochem J 302(2): 527–534.

Hirschhorn JN & Daly MJ (2005) Genome-wide association studies for common diseases and complex traits. Nat Rev Genet 6(2): 95–108.

Hocking AM, Shinomura T & McQuillan DJ (1998) Leucine-rich repeat glycoproteins of the extracellular matrix. Matrix Biol 17(1): 1–19.

Holder N (1977) An experimental investigation into the early development of the chick elbow joint. J Embryol Exp Morphol 39: 115–27.

Horton WE Jr, Lethbridge-Cejku M, Hochberg MC, Balakir R, Precht P, Plato CC, Tobin JD, Meek L & Doege K (1998) An association between an aggrecan polymorphic allele and bilateral hand osteoarthritis in elderly white men: data from the Baltimore Longitudinal Study of Aging (BLSA). Osteoarthritis Cartilage 6(4): 245–251.

Horton WE Jr, Bennion P & Yang L (2006) Cellular, molecular, and matrix changes in cartilage during aging and osteoarthritis. J Musculoskelet Neuronal Interact 6(4): 379–381.

Page 96: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

94

Huang J, Ushiyama T, Inoue K, Kawasaki T & Hukuda S (2000) Vitamin D receptor gene polymorphisms and osteoarthritis of the hand, hip, and knee: acase-control study in Japan. Rheumatology (Oxford) 39(1): 79–84.

Huber M, Trattnig S & Lintner F (2000) Anatomy, biochemistry, and physiology of articular cartilage. Invest Radiol 35(10): 573–580.

Hulmes DJ (2002) Building collagen molecules, fibrils, and suprafibrillar structures. J Struct Biol 137(1–2): 2–10.

Hunter DJ, Demissie S, Cupples LA, Aliabadi P & Felson DT (2004) A genome scan for joint-specific hand osteoarthritis susceptibility: The Framingham Study. Arthritis Rheum 50(8): 2489–2496.

Hunter DJ & Felson DT (2006) Osteoarthritis. BMJ 332(7542): 639–642. Hunter DJ & Kraft P (2007) Drinking from the fire hose–statistical issues in genomewide

association studies. N Engl J Med 357(5): 436–439. Hunter W (1743) Of the structure and disease of articulating cartilages. Philos Trans R Soc

Lond 42: 514–521. Hurme M & Santtila S (1998) IL-1 receptor antagonist (IL-1Ra) plasma levels are co-

ordinately regulated by both IL-1Ra and IL-1beta genes. Eur J Immunol 28: 2598–2602.

Ikeda T, Mabuchi A, Fukuda A, Kawakami A, Ryo Y, Yamamoto S, Miyoshi K, Haga N, Hiraoka H, Takatori Y, Kawaguchi H, Nakamura K & Ikegawa S (2002) Association analysis of single nucleotide polymorphisms in cartilage-specific collagen genes with knee and hip osteoarthritis in the Japanese population. J Bone Miner Res 17(7): 1290–1296.

Ikegawa S (2007) New gene associations in osteoarthritis: what do they provide, and where are we going? Curr Opin Rheumatol 19(5): 429–434.

Ingvarsson (2000) Prevalence and inheritance of hip osteoarthritis in Iceland. Acta Orthop Scand Suppl 298: 1–46.

Ingvarsson T, Stefánsson SE, Gulcher JR, Jónsson HH, Jónsson H, Frigge ML, Pálsdóttir E, Olafsdóttir G, Jónsdóttir T, Walters GB, Lohmander LS & Stefánsson K (2001) A large Icelandic family with early osteoarthritis of the hip associated with a susceptibility locus on chromosome 16p. Arthritis Rheum 44(11): 2548–2555.

Iozzo RV (1998) Matrix proteoglycans: from molecular design to cellular function. Annu Rev Biochem 67: 609–652.

Iozzo RV (1999) The biology of the small leucine-rich proteoglycans. Functional network of interactive proteins. J Biol Chem 274(27): 18843–18846.

Issa SN & Sharma L (2006) Epidemiology of osteoarthritis: an update. Curr Rheumatol Rep 8(1): 7–15.

Ito MM, Kida MY (2000) Morphological and biochemical re-evaluation of the process of cavitation in the rat knee joint: cellular and cell strata alterations in the interzone. J Anat 197(4): 659–79.

Page 97: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

95

Jakkula E, Melkoniemi M, Kiviranta I, Lohiniva J, Räinä SS, Perälä M, Warman ML, Ahonen K, Kröger H, Göring HH & Ala-Kokko L (2005) The role of sequence variations within the genes encoding collagen II, IX and XI in non-syndromic, early-onset osteoarthritis. Osteoarthritis Cartilage 13(6): 497–507.

Jay GD, Torres JR, Rhee DK, Helminen HJ, Hytinnen MM, Cha CJ, Elsaid K, Kim KS, Cui Y & Warman ML (2007) Association between friction and wear in diarthrodial joints lacking lubricin. Arthritis Rheum 56(11): 3662–3669.

Jiang Q, Shi D, Yi L, Ikegawa S, Wang Y, Nakamura T, Qiao D, Liu C & Dai J (2006) Replication of the association of the aspartic acid repeat polymorphism in the asporin gene with knee-osteoarthritis susceptibility in Han Chinese. J Hum Genet 51(12): 1068–1072.

Jiang Q, Shi D, Nakajima M, Dai J, Wei J, Malizos KN, Qin J, Miyamoto Y, Kamatani N, Liu B, Tsezou A, Nakamura T & Ikegawa S (2008) Lack of association of single nucleotide polymorphism in LRCH1 with knee osteoarthritis susceptibility. J Hum Genet 53(1): 42–7.

Jikko A, Wakisaka T, Iwamoto M, Hiranuma H, Kato Y, Maeda T, Fujishita M & Fuchihata H (1998) Effects of interleukin-6 on proliferation and proteoglycan metabolism in articular chondrocyte cultures. Cell Biol Int 22: 615–621.

Jones AR, Gleghorn JP, Hughes CE, Fitz LJ, Zollner R, Wainwright SD, Caterson B, Morris EA, Bonassar LJ & Flannery CR (2007) Binding and localization of recombinant lubricin to articular cartilage surfaces. J Orthop Res 25(3): 283–292.

Jones G, Cooley HM & Stankovich JM (2002) A cross sectional study of the association between sex, smoking, and other lifestyle factors and osteoarthritis of the hand. J Rheumatol 29(8): 1719–1724.

Jordan JM, Helmick CG, Renner JB, Luta G, Dragomir AD, Woodard J, Fang F, Schwartz TA, Abbate LM, Callahan LF, Kalsbeek WD & Hochberg MC (2007) Prevalence of knee symptoms and radiographic and symptomatic knee osteoarthritis in African Americans and Caucasians: the Johnston County Osteoarthritis Project. J Rheumatol 34(1): 172–180.

Jordanies N, Eskdale J, Stuart R & Gallagher G (2000) Allele associations reveal four prominent haplotypes at the human interleukin-6 (IL-6) locus. Genes Immun 1: 451–455.

Kaila-Kangas L (2007) Musculoskeletal disorders and diseases in Finland, Helsinki. Arokoski J, Manninen P, Kröger H, Heliövaara M, Nykyri E & Impivaara O (2007) Hip and knee pain and osteoarthritis. Publications of the National Public Health Institute B25

Kaliakatsos M, Tzetis M, Kanavakis E, Fytili P, Chouliaras G, Karachalios T, Malizos K & Tsezou A (2006) Asporin and knee osteoarthritis in patients of Greek origin. Osteoarthritis Cartilage 14(6): 609–611.

Kallman DA, Wigley FM, Scott WW Jr, Hochberg MC & Tobin JD (1989) New radiographic grading scales for osteoarthritis of the hand. Reliability for determining prevalence and progression. Arthritis Rheum 32(12): 1584–1591.

Page 98: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

96

Kaneko S, Satoh T, Chiba J, Ju C, Inoue K & Kagawa J (2000) Interleukin-6 and interleukin-8 levels in serum and synovial fluid of patients with osteoarthritis. Cytokines Cell Mol Ther 6: 71–79.

Kannu P, Bateman JF, Belluoccio D, Fosang AJ & Savarirayan R (2009) Employing molecular genetics of chondrodysplasias to inform the study of osteoarthritis. Arthritis Rheum 60(2): 325–334.

Kanoh T, Hasegawa Y, Masui T, Yamaguchi J, Ishiguro N & Hamajima N (2008) Interleukin-1beta gene polymorphism associated with radiographic signs of osteoarthritis of the knee. J Orthop Sci 13(2): 97–100.

Karlson EW, Mandl LA, Aweh GN, Sangha O, Liang MH & Grodstein F (2003) Total hip replacement due to osteoarthritis: the importance of age, obesity, and other modifiable risk factors. Am J Med 114: 93–98.

Kassner A, Hansen U, Miosge N, Reinhardt DP, Aigner T, Bruckner-Tuderman L, Bruckner P & Grässel S (2003) Discrete integration of collagen XVI into tissue-specific collagen fibrils or beaded microfibrils. Matrix Biol 22(2): 131–143.

Kean WF, Kean R & Buchanan WW (2004) Osteoarthritis: symptoms, signs and source of pain. Inflammopharmacology 12(1): 3–31.

Keen RW, Hart DJ, Lanchbury JS & Spector TD (1997) Association of early osteoarthritis of the knee with a Taq I polymorphism of the vitamin D receptor gene. Arthritis Rheum 40(8): 1444–1449.

Kellgren JH & Lawrence JS (1957) Radiological assessment of osteo-arthrosis. Ann Rheum Dis 16: 494–502.

Kerkhof JM, Uitterlinden AG, Valdes AM, Hart DJ, Rivadeneira F, Jhamai M, Hofman A, Pols HA, Bierma-Zeinstra SM, Spector TD & van Meurs JB (2008) Radiographic osteoarthritis at three joint sites and FRZB, LRP5, and LRP6 polymorphisms in two population-based cohorts. Osteoarthritis Cartilage 16(10): 1141–1149.

Khan IM, Redman SN, Williams R, Dowthwaite GP, Oldfield SF & Archer CW (2007) The development of synovial joints. Curr Top Dev Biol 79: 1–36.

Kiani C, Lee V, Cao L, Chen L, Wu Y, Zhang Y, Adams ME & Yang BB (2001) Roles of aggrecan domains in biosynthesis, modification by glycosaminoglycans and product secretion. Biochem J 354(1): 199–207.

Kiani C, Chen L, Wu YJ, Yee AJ & Yang BB (2002) Structure and function of aggrecan. Cell Res 12(1): 19–32.

Kierszenbaum AL (2007) Histology and cell biology: An introduction to pathology. , Philadelphia, Mosby Elsevier: 124–129.

Kimura T & Shiota K (1996) Sequential changes of programmed cell death in developing fetal mouse limbs and its possible roles in limb morphogenesis. J Morphol 229(3): 337–46.

Kirk KM, Doege KJ, Hecht J, Bellamy N & Martin NG (2003) Osteoarthritis of the hands, hips and knees in an Australian twin sample–evidence of association with the aggrecan VNTR polymorphism. Twin Res 6(1): 62–66.

Page 99: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

97

Kizawa H, Kou I, Iida A, Sudo A, Miyamoto Y, Fukuda A, Mabuchi A, Kotani A, Kawakami A, Yamamoto S, Uchida A, Nakamura K, Notoya K, Nakamura Y & Ikegawa S (2005) An aspartic acid repeat polymorphism in asporin inhibits chondrogenesis and increases susceptibility to osteoarthritis. Nat Genet 37(2): 138–44.

Kloppenburg M, Stamm T, Watt I, Kainberger F, Cawston TE, Birrell FN, Petersson IF, Saxne T, Kvien TK, Slatkowsky-Christensen B, Dougados M, Gossec L, Breedveld FC & Smolen JS (2007a) Research in hand osteoarthritis: time for reappraisal and demand for new strategies. An opinion paper. Ann Rheum Dis 66(9): 1157–1161.

Kloppenburg M (2007b) Hand osteoarthritis–an increasing need for treatment and rehabilitation. Curr Opin Rheumatol 19(2): 179–183.

Koch M, Foley JE, Hahn R, Zhou P, Burgeson RE, Gerecke DR & Gordon MK (2001) alpha 1(Xx) collagen, a new member of the collagen subfamily, fibril-associated collagens with interrupted triple helices. J Biol Chem 276(25): 23120–23126.

Konttinen YT, Lindroos L, Ruuttila P, Lähdeoja T, Lassus J, Nordström DC & Santavirta S (2003) Clinical picture and treatment of osteoarthritis. Duodecim 119(16): 1537–1544.

Kopec JA, Rahman MM, Berthelot JM, Le Petit C, Aghajanian J, Sayre EC, Cibere J, Anis AH & Badley EM (2007) Descriptive epidemiology of osteoarthritis in British Columbia, Canada. J Rheumatol 34(2): 386–393.

Korenberg JR, Chen XN, Doege K, Grover J & Roughley PJ (1993) Assignment of the human aggrecan gene (AGC1) to 15q26 using fluorescence in situ hybridization analysis. Genomics 16(2): 546–548.

Koyama E, Ochiai T, Rountree RB, Kingsley DM, Enomoto-Iwamoto M, Iwamoto M & Pacifici M (2007) Synovial joint formation during mouse limb skeletogenesis: roles of Indian hedgehog signaling. Ann N Y Acad Sci 1116: 100–112.

van der Kraan PM, Buma P, van Kuppevelt T & van den Berg WB (2002) Interaction of chondrocytes, extracellular matrix and growth factors: relevance for articular cartilage tissue engineering. Osteoarthritis Cartilage 10(8): 631–637.

Krasnokutsky S, Samuels J & Abramson SB (2007) Osteoarthritis in 2007. Bull NYU Hosp Jt Dis 65(3): 222–228.

Kristoffersen H, Torp-Pedersen S, Terslev L, Qvistgaard E, Holm CC, Ellegaard K & Bliddal H (2006) Indications of inflammation visualized by ultrasound in osteoarthritis of the knee. Acta Radiol 47(3): 281–286.

Kruglyak L, Daly MJ, Reeve-Daly MP & Lander ES (1996) Parametric and nonparametric linkage analysis: a unified multipoint approach. Am J Hum Genet 58(6): 1347–1363.

Kujala UM, Kaprio J & Sarna S (1994) Osteoarthritis of weight bearing joints of lower limbs in former elite male athletes. Br Med J 308: 231–234.

Laadhar L, Zitouni M, Kalle-Sellami M, Mahjoub M, Sellami S & Makni S (2007) Physiopathology of osteoarthritis. From normal cartilage to osteoarthritic cartilage: risk factors and inflammatory mechanisms. Rev Med Interne 28(8): 531–536.

Lamb KJ, Lewthwaite JC, Bastow ER & Pitsillides AA (2003) Defining boundaries during joint cavity formation: going out on a limb. Int J Exp Pathol 84: 55–67.

Page 100: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

98

Lane NE, Lian K, Nevitt MC, Zmuda JM, Lui L, Li J, Wang J, Fontecha M, Umblas N, Rosenbach M, de Leon P & Corr M (2006) Frizzled-related protein variants are risk factors for hip osteoarthritis. Arthritis Rheum 54(4): 1246–1254.

Lathrop GM, Lalouel JM, Julier C & Ott J (1984) Strategies for Multilocus Analysis in Humans. PNAS 81: 3443–3446.

Lau EC, Cooper C, Lam D, Chan VN, Tsang KK & Sham A (2000) Factors associated with osteoarthritis of the hip and knee in Hong Kong Chinese: obesity, joint injury, and occupational activities. Am J Epidemiol 152: 855–862.

Lee YH, Woo JH, Choi SJ, Ji JD & Song GG (2008) Vitamin D receptor TaqI, BsmI and ApaI polymorphisms and osteoarthritis susceptibility: A meta-analysis. Joint Bone Spine 76(2): 156–161.

Leppävuori J, Kujala U, Kinnunen J, Kaprio J, Nissilä M, Heliövaara M, Klinger N, Partanen J, Terwilliger JD & Peltonen L (1999) Genome scan for predisposing loci for distal interphalangeal joint osteoarthritis: evidence for a locus on 2q. Am J Hum Genet 65(4): 1060–1067.

Lievense A, Bierma-Zeinstra S, Verhagen A, Verhaar J & Koes B (2001) Influence of work on the development of osteoarthritis of the hip: a systematic review. J Rheumatol 28: 2520–2528.

Lievense A, Bierma-Zeinstra S, Verhagen AP, van Baar ME, Verhaar JA & Koes BW (2002) Influence of obesity on the development of osteoarthritis of the hip: a systematic review. Rheumatology (Oxford) 41: 1155–1162.

Limer KL, Tosh K, Bujac SR, McConnell R, Doherty S, Nyberg F, Zhang W, Doherty M, Muir KR & Maciewicz RA (in press) Attempt to replicate published genetic associations in a large, well-defined osteoarthritis case-control population (the GOAL study). Osteoarthritis Cartilage.

Lippiello L, Hall D & Mankin HJ (1977) Collagen synthesis in normal and osteoarthritic human cartilage. J Clin Invest 59(4): 593–600.

Livshits G, Kato BS, Zhai G, Hart DJ, Hunter D, MacGregor AJ, Williams FM & Spector TD (2007) Genomewide linkage scan of hand osteoarthritis in female twin pairs showing replication of quantitative trait loci on chromosomes 2 and 19. Ann Rheum Dis 66(5): 623–627.

Loeser RF (2006) Molecular mechanisms of cartilage destruction: mechanics, inflammatory mediators, and aging collide. Arthritis Rheum 54(5): 1357–1360.

Lorenz H & Richter W (2006) Osteoarthritis: cellular and molecular changes in degenerating cartilage. Prog Histochem Cytochem 40(3): 135–163.

Lorenzo P, Neame P, Sommarin Y & Heinegård D (1998) Cloning and deduced amino acid sequence of a novel cartilage protein (CILP) identifies a proform including a nucleotide pyrophosphohydrolase. J Biol Chem 273(36): 23469–23475.

Lorenzo P, Bayliss MT & Heinegård D (2004) Altered patterns and synthesis of extracellular matrix macromolecules in early osteoarthritis. Matrix Biol 23(6): 381–391.

Page 101: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

99

Loughlin J, Mustafa Z, Irven C, Smith A, Carr AJ, Sykes B & Chapman K (1999) Stratification analysis of an osteoarthritis genome screen-suggestive linkage to chromosomes 4, 6, and 16. Am J Hum Genet 65(6): 1795–1798.

Loughlin J, Dowling B, Mustafa Z & Chapman K (2002a) Association of the interleukin-1 gene cluster on chromosome 2q13 with knee osteoarthritis. Arthritis Rheum 46(6): 1519–1527.

Loughlin J, Dowling B, Mustafa Z, Southam L & Chapman K (2002b) Refined linkage mapping of a hip osteoarthritis susceptibility locus on chromosome 2q. Rheumatology (Oxford) 41(8): 955–956.

Loughlin J, Dowling B, Chapman K, Marcelline L, Mustafa Z, Southam L, Ferreira A, Ciesielski C, Carson DA & Corr M (2004) Functional variants within the secreted frizzled-related protein 3 gene are associated with hip osteoarthritis in females. Proc Natl Acad Sci U S A 101(26): 9757–9762.

Loughlin J (2004) Familial inheritance of osteoarthritis: documented family subsets. Clin Orthop Relat Res (427 Suppl): S22–S25.

Loughlin J (2005) The genetic epidemiology of human primary osteoarthritis: current status. Expert Rev Mol Med 7(9): 1–12.

Loughlin J, Sinsheimer JS, Carr A & Chapman K (2006) The CALM1 core promoter polymorphism is not associated with hip osteoarthritis in a United Kingdom Caucasian population. Osteoarthritis Cartilage 14(3): 295–298.

Loulakis P, Shrikhande A, Davis G & Maniglia CA (1992) N-terminal sequence of proteoglycan fragments isolated from medium of interleukin-1-treated articular-cartilage cultures. Putative site(s) of enzymic cleavage. Biochem J 284(2): 589–593.

Loza E, Lopez-Gomez JM, Abasolo L, Maese J, Carmona L & Batlle-Gualda E (2009) Economic burden of knee and hip osteoarthritis in Spain. Arthritis Rheum 61(2): 158–65.

Mabuchi A, Nakamura S, Takatori Y & Ikegawa S (2006) Familial osteoarthritis of the hip joint associated with acetabular dysplasia maps to chromosome 13q. Am J Hum Genet 79(1): 163–168.

Magdalou J, Netter P, Fournel-Gigleux S & Ouzzine M (200) Agrecan and articular cartilage: assessment of glycosyltransferases for the restoration of cartilage matrix in osteoarthritis. J Soc Biol 202(4): 281–288.

Malemud CJ (2004) Cytokines as therapeutic targets for osteoarthritis. BioDrugs 18(1): 23–35.

Mann HH, Ozbek S, Engel J, Paulsson M & Wagener R (2004) Interactions between the cartilage oligomeric matrix protein and matrilins. Implications for matrix assembly and the pathogenesis of chondrodysplasias. J Biol Chem 279(24): 25294–25298.

March LM & Bagga H (2004) Epidemiology of osteoarthritis in Australia. Med J Aust 180(5 Suppl): S6–10.

Martel-Pelletier J (1998) Pathophysiology of osteoarthritis. Osteoarthritis Cartilage 6(6): 374–376.

Martel-Pelletier J, Alaaeddine N & Pelletier JP (1999) Cytokines and their role in the pathophysiology of osteoarthritis. Front Biosci 4: D694–D703.

Page 102: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

100

Martel-Pelletier J (2004) Pathophysiology of osteoarthritis. Osteoarthritis Cartilage 12 Suppl A: S31–33.

McCormick D, van der Rest M, Goodship J, Lozano G, Ninomiya Y & Olsen BR (1987) Structure of the glycosaminoglycan domain in the type IX collagen-proteoglycan. Proc Natl Acad Sci USA 84(12): 4044–4048.

Melching LI & Roughley PJ (1990) Studies on the interaction of newly secreted proteoglycan subunits with hyaluronate in human articular cartilage. Biochim Biophys Acta 1035(1): 20–28.

Menz HB, Munteanu SE, Landorf KB, Zammit GV & Cicuttini FM (2007) Radiographic classification of osteoarthritis in commonly affected joints of the foot. Osteoarthritis Cartilage 15(11): 1333–1338.

Meulenbelt I, Bijkerk C, Breedveld FC & Slagboom PE (1997) Genetic linkage analysis of 14 candidate gene loci in a family with autosomal dominant osteoarthritis without dysplasia. J Med Genet 34: 1024–1027.Meulenbelt I, Bijkerk C, De Wildt SC, Miedema HS, Breedveld FC, Pols HA, Hofman A, Van Duijn CM & Slagboom PE (1999) Haplotype analysis of three polymorphisms of the COL2A1 gene and associations with generalised radiological osteoarthritis. Ann Hum Genet 63(5): 393–400.

Meulenbelt I, Seymour AB, Nieuwland M, Huizinga TW, van Duijn CM & Slagboom PE (2004) Association of the interleukin-1 gene cluster with radiographic signs of osteoarthritis of the hip. Arthritis Rheum 50(4): 1179–1186.

Meulenbelt I, Min JL, van Duijn CM, Kloppenburg M, Breedveld FC & Slagboom PE (2006) Strong linkage of 2q33.3 to familial early-onset generalized osteoarthritis and a consideration of two positional candidate genes. Eur J Hum Genet 14: 1280–1287.

Meulenbelt I, Min JL, Bos S, Riyazi N, Houwing-Duistermaat JJ, van der Wijk HJ, Kroon HM, Nakajima M, Ikegawa S, Uitterlinden AG, van Meurs JB, van der Deure WM, Visser TJ, Seymour AB, Lakenberg N, van der Breggen R, Kremer D, van Duijn CM, Kloppenburg M, Loughlin J & Slagboom PE (2008) Identification of DIO2 as a new susceptibility locus for symptomatic osteoarthritis. Hum Mol Genet 17(12): 1867–1875.

Meulenbelt I, Chapman K, Dieguez-Gonzalez R, Shi D, Tsezou A, Dai J, Malizos KN, Kloppenburg M, Carr A, Nakajima M, van der Breggen R, Lakenberg N, Gomez-Reino JJ, Jiang Q, Ikegawa S, Gonzalez A, Loughlin J & Slagboom EP (2009) Large replication study and meta-analyses of DVWA as an osteoarthritis susceptibility locus in European and Asian populations. Hum Mol Genet 18(8): 1518–1523.

Min JL, Meulenbelt I, Riyazi N, Kloppenburg M, Houwing-Duistermaat JJ, Seymour AB, Pols HA, van Duijn CM & Slagboom PE (2005) Association of the Frizzled-related protein gene with symptomatic osteoarthritis at multiple sites. Arthritis Rheum 52(4): 1077–1080.

Min JL, Meulenbelt I, Kloppenburg M, van Duijn CM & Slagboom PE (2007) Mutation analysis of candidate genes within the 2q33.3 linkage area for familial early-onset generalised osteoarthritis. Eur J Hum Genet 15(7): 791–799.

Page 103: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

101

Mitrovic D (1978) Development of the diarthrodial joints in the rat embryo. Am J Anat 151(4): 475–85.

Miyamoto Y, Mabuchi A, Shi D, Kubo T, Takatori Y, Saito S, Fujioka M, Sudo A, Uchida A, Yamamoto S, Ozaki K, Takigawa M, Tanaka T, Nakamura Y, Jiang Q & Ikegawa S (2007) A functional polymorphism in the 5' UTR of GDF5 is associated with susceptibility to osteoarthritis. Nat Genet 39(4): 529–533.

Miyamoto Y, Shi D, Nakajima M, Ozaki K, Sudo A, Kotani A, Uchida A, Tanaka T, Fukui N, Tsunoda T, Takahashi A, Nakamura Y, Jiang Q & Ikegawa S (2008) Common variants in DVWA on chromosome 3p24.3 are associated with susceptibility to knee osteoarthritis. Nat Genet 40(8): 994–998.

Mori M, Nakajima M, Mikami Y, Seki S, Takigawa M, Kubo T & Ikegawa S (2006) Transcriptional regulation of the cartilage intermediate layer protein (CILP) gene. Biochem Biophys Res Commun 341(1): 121–127.

Mort JS, Magny MC & Lee ER (1998) Cathepsin B: an alternative protease for the generation of an aggrecan ‘metalloproteinase’ cleavage neoepitope. Biochem J 335(3): 491–494.

Mototani H, Mabuchi A, Saito S, Fujioka M, Iida A, Takatori Y, Kotani A, Kubo T, Nakamura K, Sekine A, Murakami Y, Tsunoda T, Notoya K, Nakamura Y & Ikegawa S (2005) A functional single nucleotide polymorphism in the core promoter region of CALM1 is associated with hip osteoarthritis in Japanese. Hum Mol Genet 14(8): 1009–1017.

Mow VC, Wang CC & Hung CT (1999) The extracellular matrix, interstitial fluid and ions as a mechanical signal transducer in articular cartilage. Osteoarthritis Cartilage 7(1): 41–58.

Moxley G, Han J, Stern AG & Riley BP (2007) Potential influence of IL1B haplotype and IL1A-IL1B-IL1RN extended haplotype on hand osteoarthritis risk. Osteoarthritis Cartilage 15: 1106–1112.

Mustafa Z, Dowling B, Chapman K, Sinsheimer JS, Carr A & Loughlin J (2005) Investigating the aspartic acid (D) repeat of asporin as a risk factor for osteoarthritis in a UK Caucasian population. Arthritis Rheum 52(11): 3502–3506.

Myllyharju J & Kivirikko KI (2001) Collagens and collagen-related diseases. Ann Med 33(1): 7–21.

Myllyharju J & Kivirikko KI (2004) Collagens, modifying enzymes and their mutations in humans, flies and worms. Trends Genet 20(1): 33–43.

Nakamura T, Shi D, Tzetis M, Rodriguez-Lopez J, Miyamoto Y, Tsezou A, Gonzalez A, Jiang Q, Kamatani N, Loughlin J & Ikegawa S (2007) Meta-analysis of association between the ASPN D-repeat and osteoarthritis. Hum Mol Genet 16(14): 1676–1681.

Nalin AM, Greenlee TK & Sandell LJ (1995) Collagen gene expression during development of avian synovial joints: transient expression of types II and XI collagen genes in the joint capsule. Dev Dyn 203(3): 352–62.

Newton G, Weremowicz S, Morton CC, Copeland NG, Gilbert DJ, Jenkins NA & Lawler J (1994) Characterization of human and mouse cartilage oligomeric matrix protein. Genomics 24(3): 435–439.

Page 104: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

102

Ni H, Shi D, Dai J, Qin J, Xu Y, Zhu L, Yao C, Shao Z, Chen D, Xu Z, Yi L, Ikegawa S & Jiang Q (in press) Genetic polymorphisms of interleukin-1beta (-511C/T) and interleukin-1 receptor antagonist (86-bpVNTR) in susceptibility to knee osteoarthritis in a Chinese Han population. Rheumatol Int.

Nicolae C, Ko YP, Miosge N, Niehoff A, Studer D, Enggist L, Hunziker EB, Paulsson M, Wagener R & Aszodi A (2007) Abnormal collagen fibrils in cartilage of matrilin-1/matrilin-3-deficient mice. J Biol Chem 282(30): 22163–22175.

Nishimura I, Muragaki Y & Olsen BR (1989) Tissue-specific forms of type IX collagen-proteoglycan arise from the use of two widely separated promoters. Biol Chem 264(33): 20033–20041.

Niu J, Zhang Y, LaValley M, Chaisson CE, Aliabadi P & Felson DT (2003) Symmetry and clustering of symptomatic hand osteoarthritis in elderly men and women: the Framingham Study. Rheumatology 42(2): 343–348.

Oen K, Malleson PN, Cabral DA, Rosenberg AM, Petty RF, Nickerson P & Reed M (2005) Cytokine genotypes correlate with pain and radiologically defined joint damage in patients with juvenile rheumatoid arthritis. Rheumatology 44: 1115–1121.

Olsen BR (1997) Collagen IX. Int J Biochem Cell Biol 29(4): 555–558. O’Reilly S & Doherty M (2003) Osteoarthritis. Brandt K, Doherty M & Lohmander S

(2003) Signs, symptoms and laboratory tests. Oxford University Press, Oxford. Osiri M, McNicholl J, Moreland LW & Bridges SL Jr. (1999) A novel single nucleotide

polymorphism and five probable haplotypes in the 5' flanking region of the IL-6 gene in African-Americans. Genes Immun 1: 166–167.

Ota N, Nakajima T, Nakazawa I, Suzuki T, Hosoi T, Orimo H, Inoue S, Shirai Y & Emi M (2001) A nucleotide variant in the promoter region of the interleukin-6 gene associated with decreased bone mineral density. J Hum Genet 46: 267–272.

Ouellette EA & Makowski AL (2006) How men and women are affected by osteoarthritis of the hand. Orthop Clin North Am 37(4): 541–548.

Pace JM, Corrado M, Missero C & Byers PH (2003) Identification, characterization and expression analysis of a new fibrillar collagen gene, COL27A1. Matrix Biol 22(1): 3–14.

Pacifici M, Koyama E & Iwamoto M (2005) Mechanisms of synovial joint and articular cartilage formation: recent advances, but many lingering mysteries. Birth Defects Res C Embryo Today 75(3): 237–48.

Pacifici M, Koyama E, Shibukawa Y, Wu C, Tamamura Y, Enomoto-Iwamoto M & Iwamoto M (2006) Cellular and molecular mechanisms of synovial joint and articular cartilage formation. Ann N Y Acad Sci 1068: 74–86.

Papagiannopoulos A, Waigh TA, Hardingham T & Heinrich M (2006) Solution structure and dynamics of cartilage aggrecan. Biomacromolecules 7(7): 2162–2172.

Peach CA, Carr, AJ & Loughlin J (2005) Recent advances in the genetic investigation of osteoarthritis. Trends Mol Med 11(4): 186–191.

Pelletier JP, Martel-Pelletier J & Abramson SB (2001) Osteoarthritis, an inflammatory disease: potential implication for the selection of new therapeutic targets. Arthritis Rheum 44(6): 1237–1247.

Page 105: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

103

Pihlajamaa T, Lankinen H, Ylöstalo J, Valmu L, Jäälinoja J, Zaucke F, Spitznagel L, Gösling S, Puustinen A, Mörgelin M, Peränen J, Maurer P, Ala-Kokko L & Kilpelaïnen I (2004) Characterization of recombinant amino-terminal NC4 domain of human collagen IX: interaction with glycosaminoglycans and cartilage oligomeric matrix protein. J Biol Chem 279(23): 24265–24273.

Pola E, Papaleo P, Pola R, Gaetani E, Tamburelli FC, Aulisa L & Logroscino CA (2005) Interleukin-6 gene polymorphism and risk of osteoarthritis of the hip: a case-control study. Osteoarthritis Cartilage 13(11): 1025–1028.

Poulou M, Kaliakatsos M, Tsezou A, Kanavakis E, Malizos KN & Tzetis M (2008) Association of the CALM1 core promoter polymorphism with knee osteoarthritis in patients of Greek origin. Genet Test 12(2): 263–265.

Poole AR, Pidoux I, Reiner A & Rosenberg L (1982) An immunoelectron microscope study of the organization of proteoglycan monomer, link protein, and collagen in the matrix of articular cartilage. J Cell Biol 93(3): 921–937.

Poole AR, Kojima T, Yasuda T, Mwale F, Kobayashi M & Laverty S (2001) Composition and structure of articular cartilage: a template for tissue repair. Clin Orthop Relat Res 391 Suppl: S26–S33.

Poole J, Sayer AA, Hardy R, Wadsworth M, Kuh D & Cooper C (2003) Patterns of interphalangeal hand joint involvement of osteoarthritis among men and women: a British cohort study. Arthritis Rheum 48(12): 3371–3376.

Pulai JI, Chen H, Im HJ, Kumar S, Hanning C, Hegde PS & Loeser RF (2005) NF-kappa B mediates the stimulation of cytokine and chemokine expression by human articular chondrocytes in response to fibronectin fragments. J Immunol 174(9): 5781–5788.

Pullig O, Tagariello A, Schweizer A, Swoboda B, Schaller P & Winterpacht A (2007) MATN3 (matrilin-3) sequence variation (pT303M) is a risk factor for osteoarthritis of the CMC1 joint of the hand, but not for knee osteoarthritis. Ann Rheum Dis 66(2): 279–280.

Rhee DK, Marcelino J, Baker M, Gong Y, Smits P, Lefebvre V, Jay GD, Stewart M, Wang H, Warman ML & Carpten JD (2005) The secreted glycoprotein lubricin protects cartilage surfaces and inhibits synovial cell overgrowth. J Clin Invest 115(3): 622–631.

Reginato AM & Olsen BR (2002) The role of structural genes in the pathogenesis of osteoarthritic disorders. Arthritis Res 4(6): 337–345.

van der Rest M & Mayne R (1987) Type IX collagen proteoglycan from cartilage is covalently cross-linked to type II collagen. J Biol Chem 263(4): 1615–1618.

Ricard-Blum S & Ruggiero F (2005) The collagen superfamily: from the extracellular matrix to the cell membrane. Pathol. Biol (Paris) 53(7): 430–442.

Rivera-Chavez FA, Peters-Hybki DL, Barber RC, O’Keefe GE (2003) Interleukin-6 promoter haplotypes and interleukin-6 cytokine responses. Shock 20: 218–223.

Riyazi N, Kurreeman FA, Huizinga TW, Dekker FW, Stoeken-Rijsbergen G & Kloppenburg M (2005) The role of interleukin 10 promoter polymorphisms in the susceptibility of distal interphalangeal osteoarthritis. J Rheumatol 32(8): 1571–1575.

Page 106: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

104

Riyazi N, Rosendaal FR, Slagboom E, Kroon HM, Breedveld FC & Kloppenburg M (2008) Risk factors in familial osteoarthritis: the GARP sibling study. Osteoarthritis Cartilage 16(6): 654–549.

Rizkalla G, Reiner A, Bogoch E & Poole AR (1992) Studies of the articular cartilage proteoglycan aggrecan in health and osteoarthritis. Evidence for molecular heterogeneity and extensive molecular changes in disease. J Clin Invest 90(6): 2268–2277.

Rodriguez-Lopez J, Pombo-Suarez M, Liz M, Gomez-Reino JJ & Gonzalez A (2006) Lack of association of a variable number of aspartic acid residues in the asporin gene with osteoarthritis susceptibility: case-control studies in Spanish Caucasians. Arthritis Res Ther 8(3): R55.

Rodriguez-Lopez J, Pombo-Suarez M, Liz M, Gomez-Reino JJ & Gonzalez A (2007) Further evidence of the role of frizzled-related protein gene polymorphisms in osteoarthritis. Ann Rheum Dis 66(8): 1052–1055.

Ross JM, Sherwin AF & Poole CA (2006) In vitro culture of enzymatically isolated chondrons: a possible model for the initiation of osteoarthritis. J Anat 209(6): 793–806.

Roughley PJ (2001) Articular cartilage and changes in arthritis: noncollagenous proteins and proteoglycans in the extracellular matrix of cartilage. Arthritis Res 3(6): 342–347.

Roughley PJ (2006a) The structure and function of cartilage proteoglycans. Eur Cell Mater 12: 92–101.

Roughley P, Martens D, Rantakokko J, Alini M, Mwale F & Antoniou J (2006b) The involvement of aggrecan polymorphism in degeneration of human intervertebral disc and articular cartilage. Eur Cell Mater 11: 1–7

Roux CH, Saraux A, Mazieres B, Pouchot J, Morvan J, Fautrel B, Testa J, Fardellone P, Rat AC, Coste J, Guillemin F & Euller-Ziegler L (2007) Screening for hip and knee osteoarthritis in the general population: predictive value of a questionnaire and prevalence estimates. Ann Rheum Dis 67(10): 1406–1411.

Sadouk MB, Pelletier JP, Tardif G, Kiansa K, Cloutier JM & Martel-Pelletier J (1995) Human synovial fibroblasts coexpress IL-1 receptor type I and type II mRNA. The increased level of the IL-1 receptor in osteoarthritic cells is related to an increased level of the type I receptor. Lab Invest 73(3): 347–55.

Saha N, Moldovan F, Tardif G, Pelletier JP, Cloutier JM & Martel-Pelletier J (1999) Interleukin-1beta-converting enzyme/caspase-1 in human osteoarthritic tissues: localization and role in the maturation of interleukin-1beta and interleukin-18. Arthritis Rheum 42(8): 1577–1587.

Sakao K, Takahashi KA, Mazda O, Arai Y, Tonomura H, Inoue A, Saito M, Fujioka M, Takamiya H, Imanishi J & Kubo T (2008) Enhanced expression of interleukin-6, matrix metalloproteinase-13, and receptor activator of NF-kappaB ligand in cells derived from osteoarthritic subchondral bone. J Orthop Sci 13(3): 202–210.

Sanchez C, Gabay O, Salvat C, Henrotin YE & Berenbaum F (2008) Mechanical loading highly increases IL-6 production and decreases OPG expression by osteoblasts. Osteoarthritis Cartilage 17(4): 473–481.

Page 107: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

105

Sandell LJ, Nalin AM & Reife RA (1994) Alternative splice form of type II procollagen mRNA (IIA) is predominant in skeletal precursors and non-cartilaginous tissues during early mouse development. Dev Dyn 199(2): 129–140.

Sandmark H, Hogstedt C & Vingard E (2000) Primary osteoarthrosis of the knee in men and women as a result of lifelong physical load from work. Scand J Work Environ Health 26: 20–25.

Sandy JD, Adams ME, Billingham ME, Plaas A & Muir H (1984) In vivo and in vitro stimulation of chondrocyte biosynthetic activity in early experimental osteoarthritis. Arthritis Rheum 27(4): 388–397.

Schork NJ (1997) Genetics of complex disease: approaches, problems, and solutions. Am J Respir Crit Care Med 156(4 Pt 2): 103–109.

Schönherr E, Sunderkötter C, Iozzo RV & Schaefer L (2005) Decorin, a novel player in the insulin-like growth factor system. J Biol Chem 280(16): 15767–15772.

Shen Z, Heinegård D & Sommarin Y (1995) Distribution and expression of cartilage oligomeric matrix protein and bone sialoprotein show marked changes during rat femoral head development. Matrix Biol 14(9): 773–781.

Shi D, Nakamura T, Dai J, Yi L, Qin J, Chen D, Xu Z, Wang Y, Ikegawa S & Jiang Q (2007) Association of the aspartic acid-repeat polymorphism in the asporin gene with age at onset of knee osteoarthritis in Han Chinese population. J Hum Genet 52(8): 664–667.

Shum L & Nuckolls G (2002) The life cycle of chondrocytes in the developing skeleton. Arthritis Res 4: 94–106.

Slack J, McMahan CJ, Waugh S, Schooley K, Spriggs MK, Sims JE & Dower SK (1993) Independent binding of interleukin-1 alpha and interleukin-1 beta to type I and type II interleukin-1 receptors. J Biol Chem 268(4): 2513–2524.

Smith AJ, Keen LJ, Billingham MJ, Perry MJ, Elson CJ, Kirwan JR, Sims JE, Doherty M, Spector TD & Bidwell JL (2004) Extended haplotypes and linkage disequilibrium in the IL1R1-IL1A-IL1B-IL1RN gene cluster: association with knee osteoarthritis. Genes Immun 5(6): 451–460.

Snelling S, Sinsheimer JS, Carr A & Loughlin J (2007) Genetic association analysis of LRCH1 as an osteoarthritis susceptibility locus. Rheumatology (Oxford) 46(2): 250–252.

Sobel E & Lange K (1996) Descent graphs in pedigree analysis: applications to haplotyping, location scores, and marker sharing statistics. Am J Hum Genet 58: 1323–1337.

Solovieva S, Vehmas T, Riihimäki H, Luoma K & Leino-Arjas P (2005) Hand use and patterns of joint involvement in osteoarthritis. A comparison of female dentists and teachers. Rheumatology 44(4): 521–528.

Solovieva S, Hirvonen A, Siivola P, Vehmas T, Luoma K, Riihimäki H & Leino-Arjas P (2006) Vitamin D receptor gene polymorphisms and susceptibility of hand osteoarthritis in Finnish women. Arthritis Res Ther 8(1): R20.

Page 108: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

106

Song JH, Lee HS, Kim CJ, Cho YG, Park YG, Nam SW, Lee JY & Park WS (2008) Aspartic acid repeat polymorphism of the Asporin gene with susceptibility to osteoarthritis of the knee in a Korean population. Knee 15(3): 191–195.

Southam L, Dowling B, Ferreira A, Marcelline L, Mustafa Z, Chapman K, Bentham G, Carr A & Loughlin J (2004) Microsatellite association mapping of a primary osteoarthritis susceptibility locus on chromosome 6p12.3-q13. Arthritis Rheum 50(12): 3910–3914.

Southam L, Rodriguez-Lopez J, Wilkins JM, Pombo-Suarez M, Snelling S, Gomez-Reino JJ, Chapman K, Gonzalez A & Loughlin J (2007) An SNP in the 5'-UTR of GDF5 is associated with osteoarthritis susceptibility in Europeans and with in vivo differences in allelic expression in articular cartilage. Hum Mol Genet 16(18): 2226–2232.

de Souza TB, Mentz EF, Brenol CV, Xavier RM, Brenol JC, Chies JA, Simon D. Association between the aggrecan gene and rheumatoid arthritis (2008). J Rheumatol 35(12): 2325–2328.

Spector TD & MacGregor AJ (2004) Risk factors for osteoarthritis: genetics. Osteoarthritis Cartilage 12 Suppl A: S39–S44.

Spector TD, Reneland RH, Mah S, Valdes AM, Hart DJ, Kammerer S, Langdown M, Hoyal CR, Atienza J, Doherty M, Rahman P, Nelson MR & Braun A (2006) Association between a variation in LRCH1 and knee osteoarthritis: a genome-wide single-nucleotide polymorphism association study using DNA pooling. Arthritis Rheum 54(2): 524–532.

Squires GR, Okouneff S, Ionescu M & Poole AR (2003) The pathobiology of focal lesion development in aging human articular cartilage and molecular matrix changes characteristic of osteoarthritis. Arthritis Rheum 48(5): 1261–1270.

Stecher RM (1941) Heberden’s nodes. Heredity in hypertrophic arthritis of the finger joints. Am J Med Sci 201: 801–809.

Stefánsson SE, Jónsson H, Ingvarsson T, Manolescu I, Jónsson HH, Olafsdóttir G, Pálsdóttir E, Stefánsdóttir G, Sveinbjörnsdóttir G, Frigge ML, Kong A, Gulcher JR & Stefánsson K (2003) Genomewide scan for hand osteoarthritis: a novel mutation in matrilin-3. Am J Hum Genet 72(6): 1448–1459.

Stern AG, de Carvalho MR, Buck GA, Adler RA, Rao TP, Disler D & Moxley G (2003) Association of erosive hand osteoarthritis with a single nucleotide polymorphism on the gene encoding interleukin-1 beta. Osteoarthritis Cartilage 11: 394–402.

Storm EE & Kingsley DM (1999) GDF5 coordinates bone and joint formation during digit development. Dev Biol 209(1): 11–27.

Svensson L, Aszódi A, Heinegård D, Hunziker EB, Reinholt FP, Fässler R & Oldberg A (2002) Cartilage oligomeric matrix protein-deficient mice have normal skeletal development. Mol Cell Biol 22(12): 4366–4371.

Takahashi E, Hori T, O´Connel P, Leppert M & White R (1990) R-banding and non-isotopic in situ hybridization: precise localisation of the human type II collagen gene (COL2A1). Hum Genet 86: 14–16.

Page 109: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

107

Tan AL, Grainger AJ, Tanner SF, Shelley DM, Pease C, Emery P & McGonagle D (2005) High-resolution magnetic resonance imaging for the assessment of hand osteoarthritis. Arthritis Rheum 52(8): 2355–2365.

The International HapMap Consortium (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449: 851–861.

Tepper S & Hochberg MC (1993) Factors associated with hip osteoarthritis: data from the First National Health and Nutrition Examination Survey (NHANES-I). Am J Epidemiol 137: 1081–1088.

Terry CF, Loukaci V, Green FR (2000) Cooperative influence of genetic polymorphisms on interleukin 6 transcriptional regulation. J Biol Chem 275: 18138–18144.

Toba N, Sakai A, Aoyagi K, Yoshida S, Honda S & Nakamura T (2006) Prevalence and involvement patterns of radiographic hand osteoarthritis in Japanese women: the Hizen-Oshima Study. J Bone Miner Metab 24(4): 344–348.

Toivanen AT, Arokoski JP, Manninen PS, Heliövaara M, Haara MM, Tyrväinen E, Niemitukia L & Kröger H (2007) Agreement between clinical and radiological methods of diagnosing knee osteoarthritis. Scand J Rheumatol 36(1): 58–63.

Tortorella MD, Burn TC, Pratta MA, Abbaszade I, Hollis JM, Liu R, Rosenfeld SA, Copeland RA, Decicco CP, Wynn R, Rockwell A, Yang F, Duke JL, Solomon K, George H, Bruckner R, Nagase H, Itoh Y, Ellis DM, Ross H, Wiswall BH, Murphy K, Hillman MC Jr, Hollis GF, Newton RC, Magolda RL, Trzaskos JM & Arner EC (1999) Purification and cloning of aggrecanase-1: a member of the ADAMTS family of proteins. Science 284(5420): 1664–1666.

Trippel SB (2004) Growth factor inhibition: potential role in the etiopathogenesis of osteoarthritis. Clin Orthop Relat Res 427 Suppl: S47–S52.

Uitterlinden AG, Burger H, Huang Q, Odding E, Duijn CM, Hofman A, Birkenhäger JC, van Leeuwen JP & Pols HA (1997) Vitamin D receptor genotype is associated with radiographic osteoarthritis at the knee. J Clin Invest 100(2): 259–263.

Uitterlinden AG, Burger H, van Duijn CM, Huang Q, Hofman A, Birkenhäger JC, van Leeuwen JP & Pols HA (2000) Adjacent genes, for COL2A1 and the vitamin D receptor, are associated with separate features of radiographic osteoarthritis of the knee. Arthritis Rheum 43(7): 1456–1464.

Vaes RB, Rivadeneira F, Kerkhof JM, Hofman A, Pols HA, Uitterlinden AG & van Meurs JB (in press) Genetic variation in the GDF5 region is associated with osteoarthritis, height, hip axis length and fracture risk: the Rotterdam study. Ann Rheum Dis.

Valdes AM, Hart DJ, Jones KA, Surdulescu G, Swarbrick P, Doyle DV, Schafer AJ & Spector TD (2004) Association study of candidate genes for the prevalence and progression of knee osteoarthritis. Arthritis Rheum 50(8): 2497–2507.

Valdes AM, Van Oene M, Hart DJ, Surdulescu GL, Loughlin J, Doherty M & Spector TD (2006) Reproducible genetic associations between candidate genes and clinical knee osteoarthritis in men and women. Arthritis Rheum 54(2): 533–539.

Page 110: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

108

Valdes AM, Loughlin J, Oene MV, Chapman K, Surdulescu GL, Doherty M & Spector TD (2007) Sex and ethnic differences in the association of ASPN, CALM1, COL2A1, COMP, and FRZB with genetic susceptibility to osteoarthritis of the knee. Arthritis Rheum 56(1): 137–146.

Valdes AM, Loughlin J, Timms KM, van Meurs JJ, Southam L, Wilson SG, Doherty S, Lories RJ, Luyten FP, Gutin A, Abkevich V, Ge D, Hofman A, Uitterlinden AG, Hart DJ, Zhang F, Zhai G, Egli RJ, Doherty M, Lanchbury J & Spector TD (2008a) Genome-wide association scan identifies a prostaglandin-endoperoxide synthase 2 variant involved in risk of knee osteoarthritis. Am J Hum Genet 82(6): 1231–1240.

Valdes AM, Spector TD, Doherty S, Wheeler M, Hart DJ & Doherty M (in press) Association of the DVWA and GDF5 polymorphisms with osteoarthritis in UK populations. Ann Rheum Dis.

Valdes AM & Spector TD (2009) The contribution of genes to osteoarthritis. Med Clin North Am 93(1): 45–66.

Valhmu WB, Palmer GD, Rivers PA, Ebara S, Cheng JF, Fischer S & Ratcliffe A (1995) Structure of the human aggrecan gene: exon-intron organization and association with the protein domains. Biochem J 309(2): 535–542.

Vamvakopoulos J, Green C & Metcalfe S (2002) Genetic control of IL-1beta bioactivity through differential regulation of the IL-1 receptor antagonist. Eur J Immunol 32: 2988–2996.

Venn MF (1978) Variation of chemical composition with age in human femoral head cartilage. Ann Rheum Dis 37(2): 168–174.

Vincent TL, McLean CJ, Full LE, Peston D & Saklatvala J (2007) FGF-2 is bound to perlecan in the pericellular matrix of articular cartilage, where it acts as a chondrocyte mechanotransducer. Osteoarthritis Cartilage 15(7): 752–763.

Vuolteenaho K, Moilanen T, Knowles RG & Moilanen E (2007) The role of nitric oxide in osteoarthritis. Scand J Rheumatol 36(4): 247–258.

Wagener R, Ehlen HW, Ko YP, Kobbe B, Mann HH, Sengle G & Paulsson M (2005) The matrilins–adaptor proteins in the extracellular matrix. FEBS Lett 579(15): 3323–3329.

Wagner S, Hofstetter W, Chiquet M, Mainil-Varlet P, Stauffer E, Ganz R & Siebenrock KA (2003) Early osteoarthritic changes of human femoral head cartilage subsequent to femoro-acetabular impingement. Osteoarthritis Cartilage 11(7): 508–518.

Watanabe H & Yamada Y (1999) Mice lacking link protein develop dwarfism and craniofacial abnormalities. Nat Genet 21(2): 150–151.

Wiberg C, Klatt AR, Wagener R, Paulsson M, Bateman JF, Heinegård D & Mörgelin M (2003) Complexes of matrilin-1 and biglycan or decorin connect collagen VI microfibrils to both collagen II and aggrecan. J Biol Chem 278(39): 37698–37704.

Wight TN, Kinsella MG & Qwarnström EE (1992) The role of proteoglycans in cell adhesion, migration and proliferation. Curr Opin Cell Biol 4(5): 793–801.

Wilder FV, Hall BJ, Barrett JP Jr & Lemrow NB (2002) History of acute knee injury and osteoarthritis of the knee: a prospective epidemiological assessment. The Cleawater Osteoarthritis Study. Osteoarthritis Cartilage 10: 611–616.

Page 111: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

109

Wilder FV, Barrett JP & Farina EJ (2006) Joint-specific prevalence of osteoarthritis of the hand. Osteoarthritis Cartilage 14(9): 953–957.

Winterbottom N, Tondravi MM, Harrington TL, Klier FG, Vertel BM & Goetinck PF (1992) Cartilage matrix protein is a component of the collagen fibril of cartilage. Dev Dyn 193(3): 266–276.

Wright GD, Hughes AE, Regan M & Doherty M (1996) Association of two loci on chromosome 2q with nodal osteoarthritis. Ann Rheum Dis 55: 317–319

Wu JJ & Eyre DR (1995) Structural analysis of cross-linking domains in cartilage type XI collagen. Insights on polymeric assembly. J Biol Chem 270(32): 18865–18870.

Wu JJ & Eyre DR (2003) Intervertebral disc collagen. Usage of the short form of the alpha1(IX) chain in bovine nucleus pulposus. J Biol Chem 278(27): 24521–24525.

Xu L, Peng H, Glasson S, Lee PL, Hu K, Ijiri K, Olsen BR, Goldring MB & Li Y (2007) Increased expression of the collagen receptor discoidin domain receptor 2 in articular cartilage as a key event in the pathogenesis of osteoarthritis. Arthritis Rheum 56(8): 2663–2673.

Young RD, Lawrence PA, Duance VC, Aigner T & Monaghan P (2000) Immunolocalization of collagen types II and III in single fibrils of human articular cartilage. J Histochem Cytochem 48(3): 423–432.

Young AA, Smith MM, Smith SM, Cake MA, Ghosh P, Read RA, Melrose J, Sonnabend DH, Roughley PJ & Little CB (2005) Regional assessment of articular cartilage gene expression and small proteoglycan metabolism in an animal model of osteoarthritis. Arthritis Res Ther 7(4): 852–861.

Zhang W, Doherty M, Leeb BF, Alekseeva L, Arden NK, Bijlsma JW, Dincer F, Dziedzic K, Hauselmann HJ, Kaklamanis P, Kloppenburg M, Lohmander LS, Maheu E, Martín-Mola M, Pavelka K, Punzi L, Reiter S, Smolen J, Verbruggen G, Watt I & Zimmermann-Górska I (2009) EULAR evidence based recommendations for the diagnosis of hand osteoarthritis – report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis 68(1): 8–17.

Zhang Y, Xu L, Nevitt MC, Aliabadi P, Yu W, Qin M, Lui LY & Felson DT (2001) Comparison of the prevalence of knee osteoarthritis between the elderly Chinese population in Beijing and whites in the United States: The Beijing Osteoarthritis Study. Arthritis Rheum 44(9): 2065–2071.

Page 112: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

110

Page 113: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

111

Original articles

This thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Kämäräinen OP, Solovieva S, Vehmas T, Luoma K, Leino-Arjas P, Riihimäki H, Ala-Kokko L & Männikkö M (2006) Aggrecan core protein of a certain length is protective against hand osteoarthritis. Osteoarthritis Cartilage 14(10): 1075–1080.

II Kämäräinen OP, Solovieva S, Vehmas T, Luoma K, Riihimäki H, Ala-Kokko L, Männikkö M & Leino-Arjas P (2008) Common interleukin-6 promoter variants associate with the more severe forms of distal interphalangeal osteoarthritis. Arthritis Res Ther 10(1): R21.

III Solovieva S, Kämäräinen OP, Hirvonen A, Hämäläinen S, Laitala M, Vehmas T, Luoma K, Näkki A, Riihimäki H, Ala-Kokko L, Männikkö M & Leino-Arjas P (2009) Association between interleukin 1 gene cluster polymorphisms and bilateral dip osteoarthritis. J Rheumatol, in press.

IV Jakkula E, Taipale M, Kämäräinen OP, Barral S, Kiviranta I, Kröger H, Ott J, Ala-Kokko L & Männikkö M (2009) Primary osteoarthritis of the hip and knee shows linkage to Chromosome 2q21 in Finnish families. Manuscript.

Reprinted with permission from Elsevier (I) and The Journal of Rheumatology Publishing Company Limited (III).

Original publications are not included in the electronic version of the dissertation.

Page 114: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

112

Page 115: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

A C T A U N I V E R S I T A T I S O U L U E N S I S

Distributed byOULU UNIVERSITY LIBRARY

P.O. Box 7500, FI-90014University of Oulu, Finland

Book orders:OULU UNIVERSITY PRESSP.O. Box 8200, FI-90014University of Oulu, Finland

S E R I E S D M E D I C A

1003. Anttonen, Olli (2009) Prevalence, prognosis and characteristics of subjects withshort QT interval in an electrocardiogram

1004. Pulkkinen, Pasi (2009) Radiographical assessment of hip fragility

1005. Iinattiniemi, Sari (2009) Fall accidents and exercise among a very old home-dwelling population

1006. Westerlund, Tarja (2009) Thermal, circulatory, and neuromuscular responses towhole-body cryotherapy

1007. Kaikkonen, Kari (2009) Risk factors for sudden cardiac death from an acuteischemic event in general population. A case-control study

1008. Kuisma, Mari (2009) Magnetic resonance imaging of lumbar degenerative bonemarrow (Modic) changes. Determinants, natural course and association with lowback pain

1009. Vartiainen, Johanna (2009) Ghrelin, obesity and type 2 diabetes. Genetic,metabolic and epidemiological studies

1010. Löfgren, Johan (2009) Genetic polymorphisms in collectins and Toll-like receptor4 as factors influencing susceptibility to severe RSV infections and otitis media

1011. Korhonen, Topi (2009) Mathematical modeling of the regulation, developmentand genetically engineered experimental models of cardiac excitation-contractioncoupling

1012. Pajala, Ari (2009) Achilles tendon rupture. Comparison of two surgicaltechniques, evaluation of outcomes after complications and biochemical andhistological analyses of collagen type I and III and tenascin-C expression in theAchilles tendon

1013. Tetri, Sami (2009) Factors affecting outcome after primary intracerebralhemorrhage

1014. Utriainen, Kati (2009) Arvostava vastavuoroisuus ikääntyvien sairaanhoitajientyöhyvinvoinnin ytimenä hoitotyössä

1015. Girsén, Anna (2009) Preeclampsia and maternal type-1 diabetes: new insights intomaternal and fetal pathophysiology

1016. Hietaniemi, Mirella (2009) Studies on novel and traditional risk factors ofatherosclerosis

1017. Kyllönen, Heli (2009) Gelatinases, their tissue inhibitors and p53 in lymphomas

Page 116: OULU 2009 D 1018 UNIVERSITY OF OULU P.O.B. 7500 FI …

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Professor Mikko Siponen

University Lecturer Elise Kärkkäinen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Information officer Tiina Pistokoski

University Lecturer Seppo Eriksson

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9143-2 (Paperback)ISBN 978-951-42-9144-9 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1018

ACTA

Olli-Pekka K

ämäräinen

OULU 2009

D 1018

Olli-Pekka Kämäräinen

THE SEARCH FOR SUSCEPTIBILITY GENESIN OSTEOARTHRITIS

FACULTY OF MEDICINEINSTITUTE OF BIOMEDICINE,DEPARTMENT OF MEDICAL BIOCHEMISTRY AND MOLECULAR BIOLOGY,UNIVERSITY OF OULU;BIOCENTER OULU,CENTER FOR CELL MATRIX RESEARCH,UNIVERSITY OF OULU