Automation of the in vitro 3D Pannus Model in 96-Well Format

Abstract

Tissue engineering offers the possibility of developing complex in vitro tissue models that reflect a realistic picture of the structure and function of the tissue or organ under investigation. On this basis the in vitro 3D pannus model was developed, an interactive co-culture of primary chondrocytes and rheumatoid arthritis synovial fibroblasts cultivated as pellet cultures in 96-well format. This thesis automates the cultivation process with the CyBi-Disk liquid handling system and validates that the automated process is biologically equivalent to manual handling at three independent levels: morphology by plate scanning, extracellular matrix formation by histology and immunohistology, and transcription of cartilage-specific genes by real-time RT-PCR. Automated and manually treated pellets showed no morphological or genetic differences, while manually supplied cultures were largely destroyed and automatically supplied ones were almost consistently homogeneous. Total processing time fell by 60%, with high accuracy and reproducibility of the individual process steps, establishing the basis for a high-throughput screen for drugs against rheumatoid arthritis and for reducing animal experiments.

3D

3-dimensional

200x

200-fold magnification

400x

400-fold magnification

AEC

3-amino-9-ethylcarbazole

AnTemp

melting temperature (annealing temperature)

AK

antibody

Aquadest

distilled water

BCP

1-bromo-3-chloropropane

Bp

base pair(s)

Ca²⁺

calcium ions

cDNA

reverse transcribed DNA (copy DNA)

CHO

Chinese hamster ovaries

CO₂

carbon dioxide

COMP

cartilage oligomeric matrix protein

cRNA

in vitro transcribed RNA (copy RNA)

C_t

threshold cycle

dATP

deoxyadenosine triphosphate

dCTP

deoxycytidine triphosphate

DEPC

diethyl pyrocarbonate

dGTP

deoxyguanosine triphosphate

DNA

deoxyribonucleic acid

dNTP

deoxynucleotide triphosphate

DMARD

disease modifying antirheumatic drugs

dsDNA

double strand DNA

DTT

dithiothreitol

dTTP

deoxythymidine triphosphate

dUTP

deoxyuridine triphosphate

EDTA

ethylenediaminetetraacetate

EtOH

ethanol

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

H₂O

water

HCL

hydrochloric acid

HE staining

haematoxylin-eosin staining

HTS

high throughput screening

IgG

immunoglobulin G

IL-1

interleukin-1

M

molar

Mg²⁺

magnesium ions

MgCl₂

magnesium chloride

min

minute

Mio

millions

MMPs

matrix degrading metalloproteinases

mRNA

messenger RNA

MSCs

mesenchymal stem cells

OD

optical density

Oligos

oligonucleotides

PBS

phosphate buffered saline

PCR

polymerase chain reaction

proMMPs

proenzyme matrix degrading metalloproteinases

RA

rheumatoid arthritis

RNA

ribonucleic acid

RNase

ribonuclease

RT

room temperature

RT-PCR

reverse transcriptase PCR

Sec

second

SOX

Sry-type high-mobility-group box

ssDNA

single strand DNA

ssRNA

single strand RNA

TE

tissue engineering

TE buffer

Tris-EDTA buffer

TNF-α

tumour necrosis factor alpha

U

unit

w/v

weight per volume

CHAPTER 1: Summary

Tissue engineering offers the possibility of developing complex in vitro tissue models that reflect a realistic picture of the structure and function of the tissue or organ under investigation.

On this basis the in vitro 3D pannus model was developed, which serves the investigation of destructive joint diseases such as rheumatoid arthritis (RA). It consists of an interactive co-culture of two different cell populations, primary chondrocytes and RA synovial fibroblasts. These are cultivated in 96-well format as pellet cultures in order to ensure the formation of their extracellular matrix. Through the interplay of healthy, artificially grown cartilage and the synovial cells of diseased patients, the course of RA can be simulated in vitro. Cultivation in 96-well format and automation of the system make it possible to establish a high-throughput screen for drugs or biologically active substances.

Porcine chondrocytes were isolated from articular cartilage and cultivated for two weeks in 96-well format. After 14 days in culture, artificial hyaline cartilage pellets had formed, which are then coated with synovial cells. The co-culture was cultivated for a further 14 days. Supply with culture medium was carried out manually and automatically in order to determine any differences. The liquid handling system CyBi®-Disk was used for the automation process. This automated pipettor ensures, among other things, accuracy and ease of operation. A program for medium exchange was created with the aid of the CyBio software in order to control the instrument.

The low shear stress exerted on the pellets by the automated procedure was assessed optically by means of scans. Evidence of the extracellular matrix was evaluated by histological and immunohistological methods. At the genetic level, the expression of genes specific to the cartilage matrix was determined with the aid of real-time RT-PCR. The successful automation for the creation of a test system for rheumatoid arthritis showed no morphological or genetic differences between manually and automatically treated pellets. Optically the pellets showed large differences: the manually supplied cultures were for the most part destroyed, while the automatically produced ones appeared almost consistently homogeneous. The automation process brought a reduction of the total processing time of 60%. In addition to high throughput, a high accuracy and reproducibility of the individual process steps is achieved.

On the basis of this result a high-throughput screen for pharmaceuticals can be developed and thereby the detailed effect of these on the course of the disease can be determined.

Furthermore, pathogenetically relevant intercellular processes can be captured and analytical procedures can be created depending on the question being asked. The saving or replacement of animal experiments could in this way be successfully implemented.

CHAPTER 2: Introduction

By the year 2020 the share of the population over 65 years of age in the industrialised nations will rise by up to 82%. Typical age-related diseases such as osteoarthritic joint diseases, for example rheumatoid arthritis, will continue to increase. The treatment of these arthroses will become one of the most frequent orthopaedic interventions, and the causes of these arthroses are manifold. On the one hand, malalignment, particular loading and injuries to individual joints can lead to the development of arthrosis. Studies in twins and families have shown that genetic components also play an important role. A large number of genes and molecular control circuits have meanwhile been discovered that govern the formation of joints and their structure (Kornak and Mundlos et al, 2003).

Rheumatoid arthritis (RA) is a chronic inflammatory rheumatic disease. With a share of 10% of rheumatic conditions it ranks among the most severe diseases of the locomotor system (Smolian et al., 2001). With a disease duration of 10 years, 50% of patients become unable to work, after 30 years it is 90% (Yelin et al., 1987; Makisara, 1982). Far more alarming is the mortality rate in severe cases, where the survival rate at five years is below 50% (Pincus et al., 1987). The typical symptoms of RA are joint pain at night and in the morning, morning stiffness and swelling of the joints, a general feeling of illness such as exhaustion and fatigue, the presence of rheumatoid factors, and typical radiographic findings such as erosion of the bone and destroyed cartilage. At least four of these symptoms must be present in order to diagnose RA (Arnett et al., 1988).

The pathogenesis of this disabling disease is not well understood, but activated RA synovial fibroblasts are regarded as the cause of cartilage destruction (Muller-Ladner et al., 1995). The therapeutic methods already available rest predominantly on anti-inflammatory drugs, injections of cortisone preparations into the joints, physiotherapy and physical therapy. To this day there is no cure for the disease, only a relief of the symptoms, which is entirely insufficient.

The development of an in vitro model for the study of destructive joint diseases, the so-called in vitro pannus, could provide information about the precise course of the disease (Schultz et al., 1997). For this purpose the clinical picture of RA is simulated in vitro by cultivating articular chondrocytes and RA synovial cells interactively. Because this in vitro 3D culture model corresponds to the specific cellular microenvironment, the effect of pharmaceuticals or biologically active substances with regard to the course of the disease can be tested. Research into new active substances focuses on the better investigation of the pathogenetic mechanisms of RA (Blackburn, 1996). The transfer from manual to automated high throughput screening makes it possible to investigate thousands of active substances in less than a month (Taylor et al., 2000). In order to obtain the highest possible throughput rate with suitable screening assays, the test system was cultivated in 96-well format. A low expenditure of time and the reproducibility of the screen are further parameters that a high-throughput assay should fulfil. The process of cultivating the system therefore has to be automated. This was done with the aid of the CyBi®-Disk pipetting robot.

The shear forces during the pipetting procedure could be reduced and uniform pellets thereby produced. The homogeneity of the cultures is a decisive point with regard to the distribution of the synovial fibroblasts. It had to be ensured that no difference occurs in the composition of the pellets, which was investigated at the histological, immunohistological and molecular genetic level. Through the automation of the cultivation process it is possible to carry out a screen for potential active substances against RA.

CHAPTER 3: State of knowledge

3.1 Cartilage

3.1.1 Chondrogenesis

Cartilage is a firm, pressure-stable supporting tissue consisting of water-rich cartilage cells and intercellular substance. Cartilage tissue develops from the mesenchymal cells of the embryo. The cells are rich in glycogen and lie densely next to one another; through the formation of the intercellular substance (matrix) they move further and further apart. The matrix consists of a network of collagen fibrils embedded in a hydrogel of extremely large aggregates of proteoglycans and hyaluronic acid, together with weakly basophilic ground substances (Sittinger et al., 1995). The dividing cells of the precartilage are the chondroplasts. The growth of the cartilage takes place through the continuous production of intercellular substance and the mitosis of the chondroplasts, from which the complete cartilage tissue results. Besides its structural and physical properties, the extracellular matrix plays a central role in various cell activities (Sittinger et al., 1995). At the same time the perichondrium, the cartilage skin, differentiates at the surface of the cartilage. This gives rise to the synovium, the so-called joint lubricant (Waldeyer et al., 1993).

Cartilage tissue in the adult organism shows no vascular or nerve structures (Wirth et al., 1996); nutrition takes place by diffusion from the perichondrium and the synovium (Waldeyer et al., 1993). The cartilage cells (chondrocytes) are responsible for matrix synthesis and degradation and have an active metabolism. The overall metabolic activity is nevertheless not very high, owing to the low cell density (Buckwalter et al., 1997). The ability to regenerate cartilage tissue is severely limited in the adult organism (Histologie I, 1991).

3.1.2 Types of cartilage

Cartilage tissue is divided into three different types: hyaline, elastic and fibrous cartilage (Tab. 3.1.1) (Histologie I, 1991).

Hyaline cartilage is the most frequently occurring form of cartilage tissue. The intercellular tissue consists of an amorphous ground substance in which collagen fibres are embedded. When the chondroplasts die, bone-forming cells (osteoblasts) develop in their place. In this way it enables longitudinal growth up to adult size. After growth has finished it remains at the joint-forming bone ends, which it covers as articular cartilage roughly 2 mm thick. Mechanical loading is absorbed by various components of the cartilage tissue. The pressure-elastic elements are the chondrons, held under tension by a tensile-strong wrapping, and the totality of the hydrated proteoglycan molecules. Tensile strength arises from the collagen fibre lamellae and bundles (Metz, 2001).

Elastic cartilage contains, in addition to the substances of hyaline cartilage, networks of elastic fibres. These give this cartilage a characteristic yellow appearance. The elastic cartilage cells form chondrons. The chondrons are smaller and poorer in cells. Elastic cartilage cannot regenerate, but neither does it ossify. Elastic cartilage occurs considerably less frequently than hyaline cartilage.

The third type of cartilage tissue is fibrous cartilage. In its intercellular substance fibrous cartilage is densely packed with collagen fibres. It contains only a very small proportion of amorphous ground substance. The chondrons are very small and poor in cells. Fibrous cartilage is the most resistant cartilage.

Of particular interest in medicine is hyaline cartilage, since it can show signs of wear through strong mechanical loading and the need for medical care is very high. Elastic cartilage is of great interest in plastic reconstructive medicine.

Tab. 3.1.1: Comparison of the human cartilage types in occurrence and properties (Sittinger, 1994).

3.1.3 The structure of hyaline cartilage

The cartilage tissue of hyaline cartilage consists of chondrocytes (1–10%), water (65–85%), collagens (10–25%) and proteoglycans (5–10%) (Tillmann, 1987; Mankin et al., 1994). The main constituents of the extracellular matrix are collagens and proteoglycans, which can make up as much as 98% of the cartilage tissue (Sittinger, 1994a). The collagens represent a significant extracellular component in the body (Burgeson, 1988), with at least 19 genetically distinct types existing (Ayad et al., 1994). Collagen type I accounts for approximately 90% in the organism as a whole (Molecular Biology of the Cell, 1994), while in articular cartilage collagen type II is the main constituent of the extracellular matrix and accounts for 85–95% of the total collagen (Wirth et al., 1996). Collagens of types V, VI, IX, X and XI are further constituents of cartilage (Burgeson et al., 1992).

Proteoglycans have the ability to bind large quantities of water and are responsible for the pressure elasticity (Wirth et al., 1996). They can be compressed to 20% of their volume and consist of a core protein to which glycosaminoglycans are covalently bound (Wong et al., 2001). With regard to protein content, molecular size and the number and types of glycosaminoglycans, proteoglycans can be extremely complex and heterogeneous in their structure (Sittinger, 1995). The most frequent representatives in the extracellular matrix of cartilage are chondroitin-4-sulfate, keratan sulfate and hyaluronic acid (Sittinger, 1994a), whereby hyaluronic acid is in fact only a glycosaminoglycan without a bond to a core protein (Ayad et al., 1994).

Through the interactions between the collagens, the proteoglycans and the water content, cartilage shows a viscoelastic behaviour which leads to an extremely strong biomechanical structure (Mow et al., 1994).

During the synthesis of the matrix, the chondroitin sulfate secreted by the chondrocytes accumulates around the cells and forms the cartilage capsule. The chondron denotes the group of chondrocyte, capsule and pericellular halo in the extracellular matrix.

The structure of the cartilage from the cartilage surface to the subchondral bone can be divided into four zones (Buckwalter et al., 1994; Wirth et al, 1996).

At the surface lies the tangential zone, in which the chondrocytes are flat and widely dispersed but show a high cell density. In the transitional zone the cells are rounder and arranged in small groups. In the thickest zone of hyaline cartilage, the radial zone, the chondrocytes can be recognised in columns. A basophilic tidemark separates the radial zone from the zone of mineralisation, in which calcium crystals form.

The extracellular matrix of the different zones varies in the concentrations of water, collagens and proteoglycans and in the size of the aggregates (Buckwalter et al., 1990). The cells of the different zones vary not only in size, shape and orientation, but also in metabolic activity (Aydelotte et al., 1992). It is possible that the reaction of the cartilage zones with regard to mechanical loading differs (Buckwalter et al., 1997).

3.1.4 Age-related change in cartilage

With increasing age a wearing of the articular cartilage occurs, above all when there is a general axial deviation of the bones (bow legs or knock knees), when the body mass is disproportionately high, or when constant, long-lasting loading acts on the cartilage. This degenerative and destructive joint disease is called arthrosis (Metz, 2001). The causes of arthrosis first lead to cartilage damage. Various additional factors besides the ageing process can lead to cartilage damage, for example accidents, overloading, congenital malformations of the joints, metabolic disorders and presumably also dietary errors. Thus a meniscus tear that has been present for a longer time can, because of the mechanical disturbance, cause cartilage damage at the adjacent cartilaginous joint surfaces of the upper and lower leg.

Young, healthy articular cartilage is bluish white, smooth, transparent and slightly extensible; with advancing age it becomes yellowish, cloudy and brittle.

The severity of the cartilage damage differs and is divided into four different grades.

Grade I: discolouration of the cartilage, softening, swelling of the cartilage
Grade II: fissures in the cartilage, fraying of the cartilage
Grade III: defects down to the bone, breaks in the cartilage
Grade IV: exposed bone, complete absence of cartilage

In age-related change of the cartilage, the loss of chondrocytes and a deteriorated metabolic situation of the chondrocytes are responsible for the destruction. This could be the cause of an inadequate synovial supply of the chondrocytes with nutrients. Because of the metabolic disturbance of the cell activity, insufficient proteoglycans are synthesised in altered form (for example with shorter mucopolysaccharide chains). The consequence is a reduced water storage capacity, which leads to the unmasking of the collagen fibres (fibrous cartilage, Tab. 3.1.1). The elasticity of the cartilage is greatly reduced and a loss of cartilage substance occurs (Fuchs, 2002).

3.2 Tissue engineering

The failure of an organ or the loss of tissue is one of the most frequent, most widespread and most costly problems in medicine (Langer et al., 1993).

The clinical aim of tissue engineering (TE) is the restoration, repair or replacement of damaged or degraded tissue in the body. In recent years great advances have been made in the field of TE, such as the use of carrier materials made of cells or polymers, as well as combinations of cells and polymers in the 3D model (Levenberg et al., 2004).

Particular interest attaches to the production of artificial cartilage and bone (Bizios 1994; Frangos and Hillsley 1993; Oliva et al., 1996), in order to treat articular tissue injuries or diseases (Reddi 1994; Glowacki and Mizuno 1996) or to compensate defects of the neck, nose or ear in plastic surgery (Puelacher et al., 1994; Sittinger et al., 1996a, b).

Besides skin, cartilage TE is the most highly developed field, with several clinical studies already in application (Stock et al., 2001).

Tissue engineering has the potential to provide constructs of cartilage tissue that are used for the restoration of the normal function of native articular cartilage after joint injuries or destruction (Lee et al., 2004).

In recent years the development of three-dimensional grafts has brought a new perspective to this field (Liu Tsang et al., 2004). Especially in the area of cartilage tissue engineering the creation of 3D constructs is of importance, since the autologous chondrocytes require an ideal distribution and stability for the build-up of the extracellular matrix. Highly porous non-woven fleeces or hydrogels are used above all as carrier materials. Depending on the materials used, promising results have already been achieved (Vacanti et al., 1991; Freed et al., 1993; Vacanti et al., 1994a; Vacanti et al., 1994b; Sittinger et al., 1996b; Sims et al., 1998; Ting et al., 1998; Tsang et al., 2004). These biomaterials do offer a three-dimensional structure in order to form a tissue or to guide tissue growth, but at the same time they represent a foreign body in the organism. In contrast to living structures they lack any renewal processes. While xenogeneic or allogeneic grafts suffer predominantly from immunological rejection reactions (Shinichi Terada et al., 2000; Hummer et al., 1992; Kastenbauer 1983; Westhues et al., 1970), the remaining materials are mostly endangered by corrosion and degradation (Westhues et al., 1970).

A further possibility for the formation of cartilage replacement tissue is the use of stem cells or precursor cells. Embryonic stem cells are pluripotent and can theoretically differentiate into every kind of cell, tissue or organ of the body (Grande et al., 1995; Hall et al., 1989). They therefore have an unlimited capacity for use in cell and tissue replacement therapy (Thomson et al., 1998). However, before clinical applications with embryonic stem cells the ethical and social concerns must first be clarified. For this reason great attention is paid to the mesenchymal stem cells (MSCs) from bone marrow. These so-called precursor cells (progenitor cells) are multipotent and already further developed (Johnstone et al., 1998). They supply a large reservoir and a regenerative pool for various mesenchymal tissues, such as muscle, bone, cartilage, fat, bone marrow and ligaments (Sittinger et al., 2002). They possess the ability to proliferate in culture, to replicate extensively and, under defined culture conditions, to differentiate into various cell types (Kadiyala et al., 1997), for example into chondrocytes, which form cartilage through the synthesis of the extracellular matrix (Prockop, 1997; Pittenger et al., 1999). The application of this method offers many advantages, such as the simple recovery of the cells, a low required cell number, the higher biosynthetic activity, which is of advantage especially in older organisms, and the possibility of bone and cartilage compositions for the repair of osteochondral defects (Berry et al., 1992; Butnario-Ephrat et al., 1996). However, the induction of differentiation into the desired cell type still causes difficulties. Increased attention is paid to this complex regulation of differentiation by growth factors (Johnstone et al., 1998; Machay et al., 1998). Furthermore the MSCs could be used for application in gene therapy, in order for example to support the process of tissue regeneration and repair. Genetic defects that would be accessible to MSC therapy are degenerative disorders such as osteoarthritis and osteoporosis, and diseases with inflammatory processes such as rheumatoid arthritis (Sittinger et al., 2002).

But tissue engineering also opens up new approaches for the development of in vitro models of the extracellular matrix, or of diseases in which this matrix is predominantly affected (Sittinger 1995).

3.2.1 Cartilage tissue engineering of in vitro 3D cultures

Three-dimensional tissue models offer a new possibility of investigating cell functions under physiological and pathophysiological conditions and, where appropriate, of supplementing or replacing animal models (Sittinger et al., 2002). In customary monolayer culture systems, isolated cells lack a tissue-typical extracellular matrix, so their functional properties often change: they dedifferentiate. In monolayer cultures chondrocytes take on a fibroblastoid shape (von der Mark et al., 1977) and switch their collagen synthesis from collagen type II, specific to hyaline cartilage, to collagen type I, which occurs in fibrous cartilage (von der Mark et al., 1986). Tissue engineering makes it possible to deliver culture models in which a functioning extracellular matrix is included.

In vitro models and assay systems are used to investigate cellular differentiation and the influence of biomolecules on the morphology and the behaviour of the cells. These models also find application in genetics, where an overexpression or a deletion of defined genes takes place in the induced system (Sittinger et al., 2002).

3D tissue models offer the advantage of opening up the local microenvironment of the cells with their interactions between cell populations and molecules of the extracellular matrix. There is the possibility of isolating a defined quantity of cells from a particular tissue type and of determining genetically the phenotype and the function of the cells under reproducible and specific culture conditions. Here, depending on the experimental strategy, a particular focus can be placed on structural or functional changes, cellular activation, migration, infiltration or degradation of the cellular matrix, the synthesis of specific proteins, or apoptosis (Sittinger et al., 2002). Some application examples of in vitro 3D tissue models are presented in Tab. 3.2.1. From these, in vitro test systems for pharmaceuticals or bioactive substances can be developed (Schultz et al., 1997). A substantial advance can be delivered by culture systems in which a functioning extracellular matrix is included. Especially in diseases in which the cell matrix is predominantly affected, TE opens new perspectives for the development of complex models for the investigation of pathogenetic mechanisms. In this way in vitro models for research into rheumatoid arthritis or arthrosis can be developed (Sittinger, 1995).

Tab. 3.2.1: In vitro applications for 3D tissue models (Sittinger et al., 2002).

3.2.2 Molecular characterisation of the chondrocytes

Chondrocytes in in vitro 3D culture systems are described on the basis of a few cartilage-specific genes. Of particular importance are the genes that are expressed during the formation of the extracellular matrix. During cultivation there is always the risk of a dedifferentiation, in the course of which the cartilage expresses, among other things, genes that are not formed in native cartilage, so these are of interest as well. Among them is collagen type I, formed in fibrous cartilage. Since during dedifferentiation one also speaks of fibroblastoid cells, these genes are likewise designated as fibroblast-specific.

Collagen type II is defined as the typical matrix protein in hyaline cartilage (Velikonja et al., 2001). Because of the high proportion of collagen type II in hyaline cartilage, it is used as a marker gene in the characterisation of cartilage tissue. In order to determine the level of differentiation of chondrocytes in hyaline cartilage, the synthesis of collagen type II is one of the best and most sensitive metabolic markers (Benya and Schaffer et al., 1982). It is detected both immunohistologically and at the RNA level by RT-PCR, and it has been shown that collagens are regulated at the transcriptional level (Stokes et al., 2001).

During the dedifferentiation of the chondrocytes, for example in monolayer cultures, a clear down-regulation of the synthesis of collagen type II can be seen. For in vitro 3D culture models of hyaline cartilage, the detection of this matrix protein is a decisive proof of the formation of the extracellular matrix. A further possibility for the detection of the regulation of collagen type II is indirect, via the transcription factor SOX9 (Sry-type high-mobility-group box) (Sakano et al., 1999).

Collagen type I is a constituent of the matrix typical of fibrous cartilage. Since this kind of cartilage consists predominantly of fibroblastoid cells, collagen type I also counts as a fibroblast-associated gene, as do collagen type III and V. The expression of this gene is strongly up-regulated in the course of dedifferentiation (Von der Mark et al., 1977; Elima et al., 1989; Stokes et al., 2001) and is therefore regarded as an undesirable matrix protein in 3D culture systems. When chondrocytes are cultivated in monolayer, a rise in collagen type I and a fall in collagen type II already occur after one week, that is, a switch of collagen synthesis (Saldanha et al., 2000).

Further collagens in the extracellular matrix of hyaline cartilage are, for example, collagen types IX and XI, which behave like collagen type II, while collagens that do not belong to the main constituents of the matrix are expressed like collagen type I (collagen types III and V) (Bonaventure et al., 1994). Since these collagens make up a very small proportion of the matrix compared with types I and II (Diab et al., 1996), they are only rarely used as characterisation genes.

Aggrecan is a further constituent of the extracellular cartilage matrix (Kuettner, 1992). This relatively large chondroitin sulfate proteoglycan aggregate can be detected by means of RT-PCR. During the dedifferentiation of the chondrocytes a lowering of the expression takes place considerably faster than for the collagen type II gene.

A further large proteoglycan is versican, which is synthesised in fibroblasts and behaves in the opposite way to aggrecan (Zimmermann et al., 1989; Bonaventure et al., 1994).

COMP (cartilage oligomeric matrix protein) is a non-collagenous glycoprotein of the extracellular matrix with cartilage-specific spatial and temporal expression patterns (Hedbom et al., 1992; Garnero et al., 2000; Issack et al., 2000; Fuchs, 2002). The molecule is able to bind calcium via specific domains, which implies that calcium plays a role in the interaction with other matrix constituents (Fuchs, 2002). This assumption finds support in the occurrence of dominant bone malformations in mutations in this region (Briggs et al., 1995). A further enzyme that is used as a cartilage-specific marker gene is hyaluronan synthase-2 (Nishida et al., 2000). It is involved in the synthesis of hyaluronic acid, which is a large constituent of the extracellular matrix.

Further constituents of the cartilage-specific matrix could certainly function as marker genes, but their significance as evidence is low.

3.3 Rheumatoid arthritis

In the Federal Republic of Germany, rheumatic diseases occupy the first 10 places in the disease statistics. Almost half of the population complains of at least one rheumatic symptom, in about 20% a rheumatic disorder can be diagnosed, and 20–30% of the cases of incapacity for work are attributable to diseases of the locomotor apparatus. Analgesics and antirheumatic drugs have for many years been the most frequently prescribed medicines (Smolian et al., 2001).

Rheumatoid arthritis is the most severe rheumatic disease and affects 0.5% of the population, with the disease frequency of women to men standing in a ratio of 3:1 (Seidel et al, 2004). RA can occur at any age, but most frequently between the ages of 35 and 45 and above the age of 60. With a disease duration of 10–20 years, up to 70% of patients become unable to work (Medicine-Worldwide, 2003).

Rheumatoid arthritis, which is also called polyarthritis, is a chronic inflammatory disease which predominantly affects the joints, but can also attack internal organs such as the eyes and skin. The cause of its development is still not fully clarified, but connections exist between genetic factors and autoimmune processes. Two hypotheses exist concerning its origin: on the one hand autoimmune mechanisms mediated by T cells, on the other hand there is the theory of rheumatoid arthritis (RA) as a bacterial or viral infection (Smolian et al., 2001).

At the beginning of the disease process, cells of the immune system are activated which, in an autoimmune reaction, turn against the body's own tissue. The so-called T lymphocytes, a particular form of white blood cell, are activated. The genetic predisposition is supported by the increased occurrence of certain surface molecules in patients with severe RA (Kornak and Mundlos et al, 2003).

In the course of the disease an inflammation of the joint lining, the so-called synovium, occurs, involving a large proportion of the cells of the immune system. The inflammatory reaction leads to the formation of the so-called pannus tissue, which invades cartilage and bone tissue and finally destroys it (Fig. 3.3.1). The cellular interactions are controlled by the cytokines, and in RA there is a release of pro-inflammatory cytokines and growth factors. Under physiological conditions these maintain the balance between matrix formation and matrix degradation.

[Figure reproduced from a third-party publication in the original thesis; not republished here. See the PDF.]

Fig. 3.3.1: Schematic representation of RA in the joint (Smolian et al, 2001).

In RA this balance is disturbed by the excessive production of pro-inflammatory cytokines and shifts in favour of matrix degradation. Tumour necrosis factor alpha (TNF-α) and interleukin-1 (IL-1) are regarded as important cytokines in the development of rheumatoid arthritis. TNF-α appears to be responsible for the destructive inflammatory process, while IL-1 promotes the destruction of cartilage tissue and activates the bone-degrading cells. This effect is normally regulated by the receptor antagonist, in that it blocks the receptors on the cell membrane and the destructive cell response is prevented. This makes it clear that the balance between protective and destructive cytokines is more important for the resulting destruction than the absolute level of destructive mediators (van den Berg et al, 1999).

Through the excessive production of these cytokines, the rheumatoid synovial membrane is activated to release matrix-degrading enzymes from the synovial cells. The process of matrix degradation is mediated mainly by the enzymes also designated matrix metalloproteinases (MMPs). These are proteolytic enzymes which degrade various matrix proteins with differing specificity. They are divided into the subfamilies collagenases, stromelysins and gelatinases. Most MMPs are secreted by a multitude of cells as inactive proenzymes (proMMPs) and subsequently activated by various proteases such as plasmin, or by MMPs, via a stepwise mechanism. The regulation of the expression of most proMMPs is subject to various stimulators such as cytokines, growth factors and hormones (Smolian et al, 2001).

RA is a chronic functional disorder in the system which induces an inflammation of the joints (Gay et al, 2001; Feldmann et al, 1996). Cells of the synovial membrane begin to grow on the cartilage, and macrophages and fibroblasts organise themselves into the aggressive pannus tissue (Fassbender et al, 1983; Pap et al, 2000; Zvaifler et al, 1994). Activated endothelial cells are important for the neovascularisation of the pannus tissue, while the accumulation of T cells can be interpreted as a local reaction to a hitherto unknown antigen (Bombara et al, 1993). Chondrocytes, as producers of the cartilage matrix, are of importance for the regenerative processes. The significance of the cartilage matrix for the invasive behaviour of the synovial membrane lies, among other things, in the presence of matrix constituents that enable a cell adhesion to the articular cartilage. Inflammatory cells can attach to matrix constituents and trigger cell activation processes that lead to the release of matrix-degrading enzymes.

The typical symptoms of RA are joint pain at night and in the morning, morning stiffness and swelling of the joints, mostly in the metacarpophalangeal joints (MCP joints) and the proximal interphalangeal joints (PIP joints), and a general feeling of illness such as exhaustion and fatigue. In the further course of the disease the following changes of the joints can develop:

Ulnar deviation = slipping of the fingers to the outside
Swan-neck deformity = downward bending of the last finger joint
Boutonniere deformity = upward protrusion of the finger knuckle
Rheumatoid nodules = growth of rubbery nodules on the extensor sides of the joints

The organs can also be affected, as for example connective tissue proliferation of the lung (pulmonary fibrosis), inflammation of the pericardium (pericarditis), inflammation of various layers of the eye wall (scleritis) and perforation of the eyeball, small vessel inflammation (vasculitis) with in some cases large skin defects, and chronic inflammation with dryness of mouth and eyes (sicca syndrome).

The diagnosis is made on the basis of several findings: symptoms, physical examination, laboratory data such as blood values, for example a rise in the inflammatory values such as the erythrocyte sedimentation rate and C-reactive protein. The detection of the so-called rheumatoid factor (various auto- antibodies) and inflammatory anaemia with a fall in the haemoglobin value, as well as radiographic examinations, also play an important role in the diagnosis.

The therapeutic methods already available rest predominantly on anti-inflammatory drugs, injections of cortisone preparations into the joints, physiotherapy and physical therapy, occupational therapy and rehabilitation, and surgical therapy (synovectomies and reconstructive surgery). To this day there is no cure for the disease, only a relief of the symptoms.

In order to halt the joint-destroying course of RA, a consistent and sufficient drug therapy must be carried out, with the aim of controlling the inflammation of the joints as far as possible. An indispensable part is played here by the so-called basic medicines, also called disease modifying antirheumatic drugs (DMARD), which can positively influence the course of the disease. Every active rheumatoid arthritis should be treated with at least one or several basic medicines (Tab. 3.3.1) (Blackburn, 1996).

Frequently used active substances that have an inhibitory effect on the formation of the T lymphocytes are methotrexate and leflunomide; sulfasalazine acts anti-inflammatorily, hydroxychloroquine leads to the lysis of the cells, and auranofin presumably prevents the immigration of white blood cells into the joint lining.

The basic medicine methotrexate is an effective DMARD with a relatively safe efficacy profile compared with other therapies (Borchers et al, 2004). There are promising developments in the use of inhibitors of TNF-α, the cytokine which is above all responsible for inflammatory processes in RA. The basic medicine etanercept is a genetically engineered variant of the TNF receptor and is produced with the aid of CHO cells. Tumour necrosis factor binds to the physiologically occurring TNF receptors, whereby reactions are triggered which play a large role in, among other things, the development of inflammation. Etanercept is, so to speak, an artificial TNF receptor which binds two molecules each of TNF-α, a subtype of tumour necrosis factor, in the blood. This leads to a clear lowering of the concentration of TNF-α in the blood and thereby to an inhibition of the inflammatory process. A further promising treatment is the genetic engineering of recombinant receptor antagonists of IL-1, anakinra. It binds to the subunit of the interleukin receptor and blocks it against the body's own IL-1. In this way the disturbed balance between IL-1 and the IL-1 receptor antagonist can be stabilised.

Tab. 3.3.1: List of the common basic medicines for RA and their trade names (Medicine-worldwide, 2003).

Although benign courses of RA also occur, the disease usually runs a gradually progressive course and almost always leads to joint destruction. With today's medicines the course can in part be slowed and the inflammation and the pain can often be controlled well over a longer period, yet a high probability of permanent disability remains. Patients with severe courses also have a shortened life expectancy. In order to make a lasting and successful treatment possible, the causes of the development and of the course of the disease must be researched more precisely.

3.4 The in vitro pannus model

The three-dimensional (3D) in vitro model for rheumatoid arthritis (RA) was established with the aim of developing a standardised in vitro test system for the analysis of the effect of pharmaceuticals and various biological active substances. There is still no ideal model for the pathogenesis of RA. Several approaches to the investigation of the interactions of pannus and cartilage tissue are known, such as the SCID mouse system with implanted synovial tissue (for example Sack et al., 1993; Kaul et al., 1995; Pap et al., 2000) and the use of HLA-DR4-CD4 transgenic mice (Sonderstrup et al., 1999), and of course established experimental arthritides. The models named cannot, however, dispense with laboratory animals and are transferable to human pathology only to a limited extent. The established cell culture systems are indeed standardised and easily reproducible, but they have the disadvantage of the dedifferentiation of the cells, that is, the loss of their typical properties. Research into the cell-cell and cell-matrix interactions of the in vivo situation is therefore hardly possible. In addition, the data obtained from human surgical material or two-dimensional cell cultures represent only the snapshot of a long-term process (Smolian et al., 2001).

[Figure reproduced from a third-party publication in the original thesis; not republished here. See the PDF.]

Fig. 3.4.1: Schematic representation of the 3D in vitro pannus model (Smolian et al., 2001).

The in vitro pannus consists of chondrocyte pellet cultures which interact with human RA synovial cells. The co-culture of different cell populations is cultivated in a three-dimensional matrix which corresponds to the specific microenvironment. The interactive co-cultures consist of defined rheumatoid synovial cell populations which are brought into direct contact with the chondrocyte pellet cultures by coating (Fig. 3.4.1).

The composition of the matrix of porcine and human articular cartilage is very similar (Frye et al., 1996), which is why this was used for the isolation of the chondrocytes. In addition the availability of porcine cartilage is considerably higher than that of human material. A high cell number is needed in order to form a cartilage-specific extracellular matrix.

In this way the essential cellular interactions in the process of matrix destruction in rheumatoid arthritis can be simulated and investigated under in vitro conditions. The co-cultures make possible the investigation of individual cell populations and their interactions in the system.

3.5 Description of the CyBi®-Disk automated pipettor

The CyBi®-Disk (Fig. 3.5.1) is based on the screening performance of the simultaneous pipettor CyBi®-Well. Here, however, the plate transport takes place by means of a rotating disk instead of a track, carriage and stacker.

The multiple dispenser CyBi®-Disk is a simultaneously operating 96-fold or 384-fold dispensing system for the automatic processing of microplates in chemical, biological and physical research laboratories. The instrument has six basic functions which offer further application possibilities (Tab. 3.5): dosing, pipetting, dispensing, diluting, rinsing or mixing, and tip exchange.

[Figure not reproduced in the web edition. See page 28 of the PDF.]

Fig. 3.5.1: CyBi®-Disk with 96-well plates under the sterile bench.

Through its high precision, even in the sub-microlitre volume range, and its high flexibility, the CyBi®-Disk covers a large field of application. The process control takes place with the aid of the CyBio Control software. The program is an instrument control software which makes it possible, by means of a personal computer, to program and control various instruments through a uniform user interface.

The instrument control software simplifies the programming through the use of Windows-typical functions and elements. Instrument commands are displayed as a symbol with their corresponding parameters in a working window (see Sect. 6.1.3) (CyBio Manual, 2001–2003).

Tab. 3.5: Short description of the basic functions of the CyBi®-Disk.

The advantage of the automation of 3D tissue models lies in the low shear loading and the saving of time. For the creation of a high throughput screening system, reproducibility in particular is important in order to carry out meaningful tests. Furthermore the possible miniaturisation of the assay is at present an interesting field in research, and this is of particular importance for pharmaceutical companies in drug discovery (Berg et al., 2000), since it makes it possible to carry out a test series, depending on its objective, within a period of weeks or months (Beggs et al., 1999). The quality of a liquid handling system depends in general on the transfer rate of the liquids during the pipetting procedure, the mixing behaviour in the wells, the development of foam and bubbles, and evaporation (Berg et al., 2001). In general a high accuracy, simple handling and a short processing time should be ensured.

When using the CyBi®-Disk in the field of cell culture, particular attention had to be paid to a sensitive pipetting procedure. The suction during the aspiration of the old medium must not be felt directly at the surface of the cells. Bubble or foam formation should also be avoided and no excessive shear stress should arise on the impact of the droplets. Furthermore the time factor plays a large role here.

CHAPTER 4: Problem statement

The in vitro 3D pannus model serves for the testing of antirheumatic drugs and other biologically active substances on artificial cartilage. The system offers many possible uses (Fig. 4.1) for investigating disease-relevant cellular interactions and gives the possibility of researching the efficacy profile of substances under in vitro conditions.

This model is to be developed further into an in vitro assay in order to research the effect of pharmaceuticals, cytokines, growth factors and inhibitors of proteolytic enzymes. With the technical implementation of a high throughput system it would be possible to supplement or reduce costly animal experiments already at early stages of development. This would on the one hand lead in the long term to the replacement of animal experiments, and on the other hand to cost savings through a shortened test duration and the saving of animal husbandry.

[Figure reproduced from a third-party publication in the original thesis; not republished here. See the PDF.]

Fig. 4.1: Schematic representation of possibilities for the investigation of disease-relevant cellular interactions and fields of application of the developed in vitro pannus model (Smolian et al., 2001).

In order to be able to carry out an effective screen or an extensive patient testing, it is necessary to automate the process. The CyBi®-Disk, a 96-well automated pipettor, is available for this.

Within the framework of this thesis, programs for culture management are to be developed and optimised. A high throughput screening system is to be implemented technically (Fig. 4.1). The subsequent analytical procedures are to ensure that the automation has no negative influence on the cells. For this, established methods of histology and immunohistology and molecular biological working methods such as real-time RT-PCR are used.

For the implementation, a sequence program for the daily medium exchange is to be created. For this the software of the CyBio company is to be used. The problem here lies in ensuring the gentlest possible pipetting procedure, in order to reduce the shear loading compared with the manual procedure. The cultures are created from primary chondrocytes, which are transferred into 96-well plates directly after the isolation. Supply is carried out in each case manually and automatically in order to compare the processes with one another.

A successful automation is the cornerstone for the further possible applications of the in vitro 3D pannus model (Fig. 4.1).

CHAPTER 5: Materials and methods

5.1 Cell isolation and cultivation

Instruments for cell isolation and cultivation

CO₂ incubator

(Heraeus)

Neubauer counting chamber

(Neubauer Feinoptik)

Spinner flask

(Wheaton)

Sterile bench

(Heraeus)

Water bath

(MLW)

Microscope

(Hund Wetzlar)

Centrifuge (GS-6R)

(Beckman)

CyBi®-Disk pipetting robot

(CyBio AG)

Material for cell isolation and cultivation

Porcine femoral knee condyles

(Fleischerei Staske, Schöneweide, Berlin)

Media and supplements

Culture medium: RPMI 1640 medium (FG1385, Biochrom), 10% FBS inactivated (Biochrom), 100 U/ml penicillin, 100 µg/ml streptomycin (Biochrom)

Digestion medium: RPMI 1640 medium (FG1385, Biochrom), 10% FBS inactivated (Biochrom), 100 U/ml penicillin, 100 µg/ml streptomycin (Biochrom), 1.5 U/ml collagenase P (Boehringer-Mannheim), 500 U/ml collagenase II (Biochrom), 50 U/ml hyaluronidase (Sigma)

HANKS solution (Dulbecco)

Consumables (sterile)

Falcon centrifuge tubes (50 ml)

(Falcon)

Gloves

(Semperit)

Cannulas

(Braun Melsungen)

Compresses

(Bard Parker)

Petri dishes (10 mm)

(Greiner Labortechnik)

Forceps

(Inox)

Scalpels

(Ansell)

(Falcon)

Serological polystyrene pipettes (1, 2, 5, 10 and 25 ml)

(Braun Melsungen)

Syringes (20 ml)

(Schleicher & Schuell)

Sterile filter FP30/0.2CA-S

(Falcon)

96-well plates

(Falcon)

Cell sieve (100 µm nylon)

Working solutions

0.4% trypan blue

(Sigma Aldrich)

70% ethanol

(J.T. Baker)

HANKS salt solution w/o Ca²⁺, Mg²⁺

(Biochrom)

Phosphate buffered saline (PBS) w/o Ca²⁺, Mg²⁺

(Biochrom)

Trypsin/EDTA (0.5%/0.2% w/v)

(Biochrom)

Gap in the source document: sections 5.1.1 – 5.1.4

Pages 29–31 of the source document (5.1.1 Isolation of porcine articular cartilage, 5.1.2 Preparation of the pellet cultures, 5.1.3 Cultivation of the pellet cultures, 5.1.4 Creation of a program for medium exchange with the CyBi®-Disk pipetting robot) are not contained in the scan provided. As soon as the pages are supplied, the text can be inserted here.

5.1.5 Cultivation of the synovial cell lines

The synovial cell lines HSE and K4IM were thawed and seeded in culture flasks. Culture medium with 10% inactivated FCS was used. After 2–3 days a medium change took place and after 6–7 days the cells were trypsinised. With the aid of the trypan blue method (see Section 5.1.2) the cell number and the viability were determined.

5.1.6 Creation of an interactive co-culture (in vitro pannus)

After 14 days of growth of the cartilage pellet cultures the co-cultures are set up. For the co-cultivation the synovial cell lines HSE and K4IM were used.

The synovial cells are trypsinised and the viable cell number is determined by trypan blue staining. The cells were taken up in serum-specific culture medium. The old culture medium was removed completely from the pellet cultures and replaced by the synovial cell suspension at a concentration of 0.06 million per well. 200 µl of cell suspension per well were seeded. The medium change takes place daily and completely for a further 14 days. It was exchanged manually and automatically in parallel. The multiwell plates are incubated in the incubator at 37 °C, 5% CO₂ and 90% humidity.

5.2 Optical assessment of the pellet cultures

The optical assessment of the pellet cultures was carried out both for the manual samples and for the automatically treated ones by scanning the 96-well plates. The plates were scanned at different times of the cultivation. The scanning procedure was carried out in colour and at a resolution of 600 dpi.

Principle: by scanning the plates, the shape of the pellet cultures in the wells can be recognised precisely, that is, whether they are damaged or have detached. This permits conclusions about the shear loading during the pipetting procedure.

5.3 Histochemistry

Instruments for histochemistry

Microtec cut 4060

(Medim)

Consumables

Tissue Tek

(Sakura)

Super Frost Plus slides

(Menzel + Gläser)

Solutions

Acetone

(Merck)

Aniline alcohol 1%

(Merck)

Antibody dilution buffer (Antibody Diluent S3022)

(DAKO)

Distilled water

(J.T. Baker)

Carnoy

(Merck)

Acetic acid 3%

(J.T. Baker)

Acetic alcohol

(J.T. Baker)

EtOH 80%, 96%, 100%

(J.T. Baker)

Phosphotungstic acid 5%

(Merck)

Polyclonal antibody for collagen type I (1:100)

(Invitrogen)

Polyclonal antibody for collagen type II (1:100)

(Invitrogen)

Rabbit IgG immunoglobulin (negative control; 1:200)

(Invitrogen)

(Merck)

Xylene

Staining solutions

Alcian blue 8GX (1% alcian blue in 3% acetic acid, pH 2.5)

Aniline blue / orange G / acetic acid mixture (dissolve 0.5 g methylene blue (Sigma), 2 g gold orange G (Aldrich) in 100 ml distilled water, add 8 ml glacial acetic acid, boil, filter and then dilute 1:4 with distilled water)

Azocarmine G solution (dissolve 0.1 g azocarmine G in 100 ml distilled water, then boil and filter and add 1% glacial acetic acid)

Eosin staining solution (3 g eosin Y in 300 ml 70% EtOH)

Haematoxylin staining solution (according to Mayer)

Nuclear fast red aluminium sulfate (dissolve 5 g aluminium sulfate in 100 ml distilled water, heat and add 0.1 g nuclear fast red)

5.3.1 Histochemical staining techniques

Sampling for histochemistry. The pellets cultivated for 2–4 weeks were carefully removed from the wells with a pipette and embedded in Tissue Tek. The pellets were placed horizontally and vertically into the Tissue Tek; for this they were pressed in flat, divided in the middle with the aid of a scalpel and embedded with the cut side facing downwards.

The labelled dishes with the samples were slowly frozen in liquid nitrogen and then stored at -80 °C. The blocks were then cut 10 µm thick with the Microtec and applied to slides.

HE staining (haematoxylin/eosin staining)

Principle: the HE staining is an overview stain in which the tissue structures in the organ section can be displayed by means of the nuclear dye Mayer's haematoxylin and, for contrast, the plasma dye eosin. The dye haemalum is obtained from logwood by ether extraction. It is initially colourless and must first mature into the dye. Through oxidation with air and NaJO₃ it becomes the actual dye haematein, which has to be heated for the nuclear staining. Through the addition of alums the different lakes arise, haematoxylin alums = haemalums. The positively charged haemalum stains, in an acidic environment, only the negatively charged nucleic acids (DNA, RNA). The staining with haemalum is a progressive endpoint stain, that is, the longer the staining time, the more dye will bind.

Eosin is an acidic plasma dye. The counterstain with eosin is regressive, that is, the section is overstained with eosin and the excess is then differentiated with water. The cell nuclei stain blue, cytoplasm and collagen fibres red.

Staining procedure: the cryostat sections are fixed for 5 min in acetone/methanol (1:1). Haematoxylin staining: staining of the nuclei. First 2–5 min in distilled water in order to make the sections receptive, then the staining follows for 7–10 min in haemalum (according to Mayer); in order to remove the excess dye, rinse briefly in distilled water, then in 1% HCl solution for differentiation.

After microscopic checking of the nuclear staining one proceeds to the blueing. This takes place for 10–30 min in tap water; the dye is stabilised by the pH value of the water and the calcium salts cause a lake to form.

Eosin staining: staining of the tissue and cytoplasm. Brief rinsing in distilled water, then 4 min in eosin staining solution. Brief immersion in distilled water, after which the dehydration could be started. This was carried out using the ascending alcohol series with 96% and 100%. The sections were then placed in xylene in order to wash out the alcohol and mounted water-free.

Azan staining according to Heidenhain

Principle: the azan staining (azocarmine and aniline blue) is a trichrome stain. It serves for the display of the collagenous connective tissue in the tissue section. It is an indirect stain, in which the nuclei are counterstained red by azocarmine.

Staining procedure: the cryosections were fixed for 10 min in Carnoy, which could be removed with distilled water. There followed 3–5 min in 1% aniline alcohol and brief rinsing in order to prepare the sections for the staining. The preparations were incubated at 60 °C for 50 min in pre-warmed azocarmine solution and then cooled for 15 min at room temperature. They were subsequently rinsed briefly with water so as not to wash out the water-soluble dye. With 1% aniline alcohol the differentiation could be carried out and the nuclear staining checked under the microscope. The differentiation was then interrupted by rinsing in acetic alcohol, since the aniline is washed out and the pH value changes. Brief rinsing with distilled water in order to prepare the sections for the following mordanting. This makes the tissue receptive for the next dye, whereby a co-differentiation can take place, which decolourises the tissue further. The mordanting took place by 90–180 min incubation in 5% phosphotungstic acid and the sections were then rinsed briefly with distilled water. The second staining step stains the collagenous connective tissue blue, with the aniline blue / orange G / acetic acid mixture for 10–40 min, depending on how large the tissue to be stained is. The remaining dye was washed out with distilled water and the sections were dehydrated with 96% and 100% alcohol. They were then immersed in xylene and mounted. The evaluation takes place under the light microscope and with the photographic camera.

Alcian blue staining

Principle: alcian blue 8GX is a water-soluble phthalocyanine dye. The dye stains the cartilage ground substance specifically; under the microscope the stained proteoglycan appears blue and the other cell constituents lighter. Cell nuclei are not stained by alcian blue; these can be counterstained with nuclear fast red aluminium sulfate.

Staining procedure: the cryosections were first placed for 3 min in 3% acetic acid, by which the pH value of 2.5 necessary for the staining was ensured. The staining took place for 30 min in 1% alcian blue 8GX in 3% acetic acid and was then rinsed off with 3% acetic acid. The preparations were rinsed with distilled water and counterstained with nuclear fast red for 3.5 min. The excess dye was removed with distilled water and the sections dehydrated with 96% and 100% alcohol; xylene removed the alcohol. The sections were mounted with balsam. The evaluation took place after drying under the light microscope and with the photographic camera.

5.3.2 Immunohistological staining

Principle: in order to demonstrate the presence of the matrix constituent collagen type II, on the one hand the expression of the genes is investigated at the genetic level and on the other hand the collagen type II produced in the cartilage matrix is stained. There is the possibility that the cells express collagen II but that nothing is found in the immunohistology, because of the regulation of the translation or because collagen type II is not anchored in the matrix. For this reason both detection methods are always applied. This immunohistological staining is based on the EnVision HRP Rabbit (DAKO K4008) method, in which a specific primary antibody binds to collagen type II. The fundamental aim of immunohistology is to identify cells or tissues by means of antigen-antibody reactions via defined epitopes. Since antigen-antibody complexes alone are difficult to identify or to make visible in cell suspensions or tissue assemblies, it is necessary to make the antibodies used visible by coupling with an enzyme and a subsequent substrate reaction, or by direct fluorochrome labelling. A first, unlabelled antibody (= primary antibody; unlabelled = unconjugated) binds to the antigen. Subsequently this antibody is detected via a second, labelled antibody (anti-antibody, which binds to the Fc fragment of the primary antibody), which brings about the actual detection reaction (through fluorescence or a chromogen-substrate reaction, depending on the kind of labelling of the antibody) (Fig. 5.3.1). It is therefore not the antigen that is detected directly, but the first antibody, and thus an indirect detection of the antigen takes place. A disadvantage is the danger of unspecific binding of the secondary antibodies to cell antigens.

Collagen type I and type II were bound with polyclonal antibodies, and at the same time a negative control was carried out with rabbit IgG (immunoglobulin G). The second antibody binds to the primary one and is detected by the substrate AEC (3-amino-9-ethylcarbazole), a chromogenic substrate system.

[Figure not reproduced in the web edition. See page 39 of the PDF.]

Fig. 5.3.1: Principle of immunohistology in the antigen-antibody system, with subsequent detection by the secondary fluorescent or chromatophoric antibody.

Method: the staining takes place in a moist chamber at room temperature. The cryosections taken from -20 °C were first dried on a paper towel (2–3 min.) and then fixed for about 5 min. in methanol/acetone mix 1:1 (+4 °C). Before the blocking of the endogenous peroxidase with blocking reagent, the sections were dried again for 30 min. They were washed 2× in TRIS buffer and incubated for 30 min with the primary antibody. The second antibody EnVision HRP-labelled rabbit (polymer conjugate) was added for 30 min after washing 3× with TRIS buffer. The samples were washed 3× again and then placed for 8–10 min in the chromogenic substrate AEC. The excess dye residues were rinsed off with distilled water and a counterstain with haematoxylin followed in order to stain the nuclei. By rinsing with tap water the blueing of the nuclei was achieved as in the HE staining (see HE staining). Mounting was done with Aquatex. The stained sections were evaluated under the light microscope and with the photographic camera.

5.4 Gene expression analysis

Instruments for gene expression analysis

10 mm quartz cuvette

(Stibo Group)

Heating block

(Eppendorf)

iCycler

(BioRad)

Imager with CCD camera

(Appligene)

Refrigerated centrifuge, Biofuge fresco

(Heraeus)

Laptop cooler

(Merck)

Photometer

(Stibo Group)

Thermocycler

(Biozym)

Vortex Genie 2

(Merck)

Centrifuge, Biofuge pico

(Heraeus)

Zero Cooler

(Biozym)

Material for gene expression analysis

1-bromo-3-chloropropane (BCP)

(Fluka)

DEPC water

(Sigma)

Diethyl pyrocarbonate (DEPC); 0.2% in H₂O

(Sigma)

dNTP set (100 mM), used as dNTP mix: ATP, CTP, GTP and TTP

(Invitrogen)

Eppendorf tubes (1.5 and 2 ml)

(Eppendorf)

Ethanol (100% undenatured)

(J.T. Baker)

Isopropanol (absolute)

(J.T. Baker)

Reaction tubes

(Applied Biosystems)

Reaction caps

(Applied Biosystems)

Oligo (dT) (12-18) primer 0.5 µg/µl

(Invitrogen)

Plexi Cooler, acrylic glass

(Biozym)

Rnasin RNase inhibitor 40 U/µl

(Promega)

Superscript RNase H reverse transcriptase 200 U/µl, contains: 5× first strand buffer and 0.1 M DTT

(Invitrogen)

SYBR Green® PCR core reagents, contains: AmpliTaqGold (5 U/µl), 10× SYBR buffer, MgCl₂ (25 mM), dNTP (12.5 mM)

(PE-Biosystems)

Taq polymerase (Thermus aquaticus)

(Sigma)

TriReagent

(Sigma)

TRIS-EDTA buffer (TE buffer) (100× 1 M TRIS HCL, 0.1 M EDTA, pH 8.0)

(Merck)

Vortex Genie 2

(Merck)

Zero Cooler

(Biozym)

Primers for the real-time PCR.

5.4.1 Cartilage processing for RNA isolation

Sampling. At different times of the cultivation, samples were taken for the gene expression analysis. In each case 5 pellets were transferred with the aid of forceps into a 2 ml Eppendorf tube and taken up in 1 ml TriReagent. The sample was mixed several times in order to achieve a homogeneous suspension, then it was stored at -80 °C until further processing.

TriReagent contains phenol for the disruption of the cells. Guanidine thiocyanate serves to inhibit released cell-derived RNases, so that the RNA remains intact during storage.

RNA isolation and quantification. During the in vitro cultivation of chondrocytes a dedifferentiation or redifferentiation takes place; in order to investigate this more precisely, one can consider the actively transcribed genes of the cells. A common method for this is the isolation of the total RNA and, via the mRNA, the drawing of conclusions about the transcribed genes. This can take place through the synthesis of cDNA and subsequent amplification of particular genes by means of PCR, or directly through hybridisation of the mRNA with known oligos on an array.

For the isolation of the RNA the samples were first thawed for 5 min at RT and incubated for 15 min. The cartilage cell pellets were homogenised by simple pipetting up and down; in addition the Ultra Turrax was used. This had to be made RNase-free before use, by placing the dispersing attachment for some time first in PBS and then in DEPC water (RNase-free). The cell pellets were each comminuted for about 15 seconds with the aid of the Ultra Turrax in order to obtain a homogeneous suspension. Longer times should be avoided because of the high shear forces, which could destroy the RNA. The samples were mixed with 133 µl BCP and, after vigorous shaking, incubated for 15 min at RT. During this incubation a phase separation occurred, which was formed by a centrifugation (16,000×g) for 45 min at 4 °C. Three phases form: in the lower, reddish one is the DNA, the white interphase contains proteins, and in the upper clear phase is the RNA. This aqueous phase was transferred very carefully, without damaging the other phases, into a 1.5 ml Eppendorf tube. The clear phase was then mixed 1:1 with ice-cold isopropanol and shaken vigorously in order to guarantee mixing. In order to precipitate the RNA this mixture was stored at -20 °C for 30 min and subsequently centrifuged at 4 °C (16,000×g) for 30 min in order to pellet the RNA. The supernatant was then removed and the resulting pellet washed 3 times with 70% ethanol and dried briefly, in order then to be taken up in 10–20 µl DEPC water, depending on the pellet size. For better dissolution of the sample, it was shaken at 60 °C for 12 min on the thermomixer.

The concentration of the RNA was measured photometrically in a 1:50 dilution. The quantity was determined via the OD (optical density) at 260 nm, on the assumption that an absorbance of 1 corresponds to an amount of 40 µg/µl ssRNA. The absorbance of the dilution should lie in the linear range (0.1–1) of the instrument, with the value of the solvent DEPC water used as the zero value. At the same time the protein contamination was determined from the ratio of the OD values at 260 nm to 280 nm; highly pure RNA solutions have an OD ratio of 2. The samples should not fall below the value 1.4. In the case of solvent contamination the sample must be processed once again. The dissolved RNA was stored at -80 °C.

5.4.2 cDNA synthesis

The further investigation of the RNA by means of PCR was made possible by the preparation of cDNA. The reaction was carried out by the reverse transcriptase, an RNA-dependent DNA polymerase which recognises RNA molecules as template strand.

For a targeted transcription of the total RNA an oligo-(dT) primer was used, which is complementary to the 3′polyA+ end of the mRNA, so that a double-stranded starting point for the cDNA synthesis was found by the enzyme. Single-stranded cDNA complementary to the mRNA was synthesised.

First, 3 µg of RNA each were diluted in 10 µl total volume of DEPC water in a PCR tube. To this mixture 1 µl of the primer (0.5 µg dT primer) is then pipetted and incubated at 72 °C for 10 min. in the thermocycler, whereby the oligo-dT primer anneals to the mRNA. The samples, placed on ice, are each mixed with 9 µl of the master mix, consisting of:

and incubated for 1 hour 2 minutes at 37 °C in the thermocycler. In this step the synthesis of the cDNA takes place. The reaction was stopped by the addition of 60 µl DEPC water and a further incubation for 10 min. at 94 °C in the thermocycler, since at this temperature the enzyme is inactivated and the cDNA strand was separated from the mRNA strand. Between the individual incubation times the batch was to be placed on ice. After completion of the cDNA synthesis the samples were stored at -20 °C.

5.4.3 Real-time RT-PCR

In the PCR, specific DNA or cDNA fragments are amplified, the specificity being given by short oligonucleotides (primers) complementary to the target DNA. These primers flank the 5′ and 3′ end of the DNA region to be amplified by hybridising complementarily to one of the two ssDNA strands. The dsDNA regions are the starting points of the thermostable Taq polymerase, which extends the primers according to the template strand, in each case in opposite directions.

These three steps – denaturation of the DNA, hybridisation of the primers and synthesis of the DNA – constitute one PCR cycle. Through the repetition of the cycles an exponential amplification of the DNA section flanked by the primers is achieved.

A quantification of the amount of DNA formed is possible in the normal PCR only if the PCR reaction is in the linear range of the amplification efficiency. This is different for every PCR reaction and can be determined only with great difficulty. This check is unnecessary if the amount of DNA formed can be measured during the PCR. The real-time PCR makes it possible to follow the product amplification in all cycles. The quantity of the amount of DNA is determined by measuring the released fluorescence of the SYBR Green® dye. The dye intercalates with the dsDNA strands and emits a detectable fluorescence, the strength of which correlates with the amount of dsDNA. Since in every cycle an exponential increase of the amplification products takes place, an increase in fluorescence can also be observed during the reaction. The increase in fluorescence is directly proportional to the amount of dsDNA, so the intercalating dye does not permit a distinction between the desired product and possible by-products. For this reason the amplification product is examined for possible by-products in every run with the aid of a melting curve analysis.

For the reaction batch of the real-time PCR, 2 µl SYBR PCR buffer, 25 mM MgCl₂, 1.25 mM dNTPs, 5 pmol primer and 0.5 U AmpliTaqGold polymerase were added to the cDNA (about 2 µl) in a PCR reaction vessel. The batch was made up with PCR water to a total volume of 20 µl. Each reaction cycle consists of -1- denaturation at 95 °C for 30 sec, -2- primer hybridisation (annealing) and synthesis for 30 sec at a temperature specific to the primers. This touch-down cycle was repeated 40× in total. Before the amplification cycles were started, the AmpliTaqGold polymerase had to be activated and the starting DNA completely denatured, by incubating them for 10 min at 95 °C. The data of the PCR were recorded in every cycle at the annealing temperature.

In conclusion a melting curve analysis was carried out. An example of a melting curve can be seen in Figure 5.4.1. Each denaturation of the DNA for 30 sec at 95 °C was followed by a reannealing of the PCR products for 30 sec at 55 °C. The melting curve consisted of 80 cycles of temperature increase of 0.5 °C each for 7 sec, from 55 °C to 95 °C. In the process a fall in the fluorescence occurred through the denaturation of the PCR products at their specific melting temperature. All data of the melting curve were recorded in every cycle and every 0.5 °C.

[Figure not reproduced in the web edition. See page 45 of the PDF.]

Fig. 5.4.1: Representation of an arbitrary melting curve, fluorescence intensity plotted against temperature.

Evaluation of the real-time PCR. The evaluation of the PCR delivers digital data which were recorded during the PCR. The evaluation takes place via the threshold cycle (CT), which indicates the intersection of the fluorescence curve with the threshold value (Fig. 5.4.2). The exponential rise of the cDNA passes into a plateau after a variable number of cycles (Fig. 5.4.2 c). The CT value is a measure of the amount of DNA synthesised in the sample, whereby the more target gene is present in the starting sample, the smaller the CT value. The minimum detectable amount is thus reached at an early PCR cycle.

The CT value depends on the starting amount of the target sequence and on the effectiveness of the PCR reaction, which is strongly influenced by the primers. The optimal PCR reaction would lead in every cycle to a doubling of the DNA molecules, which is only rarely the case.

[Figure not reproduced in the web edition. See page 46 of the PDF.]

Fig. 5.4.2: Schematic evaluation of a real-time PCR experiment: a) signal measured from the residual fluorescence of the probes and the noise signal of the detector, from which the baseline is determined; b) quantification of the sample via the threshold; c) plateau formation.

The reaction effectiveness must therefore be taken into account in the calculation of the amount of target molecules (Eq. 1-5 to 5-5).

Eq. 1-5. X_T: number of target molecules at the threshold cycle; X₀: number of target molecules at time zero; E_X: effectiveness of the PCR target gene amplification; C_T,X: threshold cycle; S: threshold value of the instrument.

The fluorescence of different amplicons is regarded as equal at their respective CT values and refers to the set threshold value (S) of the instrument.

It is assumed that the number of target molecules (X_T) at the same fluorescence intensity (S) corresponds to the number of reference molecules (R_T) (Eq. 2-5).

X_T: number of target molecules at the threshold cycle; X₀: number of target molecules at time zero; E_X: effectiveness of the PCR target gene amplification; C_T,X: threshold cycle; S: threshold value of the instrument; R: corresponding reference values of GAPDH. (Eq. 2-5)

From this the following relationship results for the semi-quantification of the PCR samples (Eq. 3-5):

Eq. 3-5. X₀: number of target molecules at time zero; E_X: effectiveness of the PCR target gene amplification; C_T,X: threshold cycle; R: corresponding reference values of GAPDH.

The effectiveness of the PCR can be determined once via a defined dilution series; the CT value can be read directly on the instrument. The CT values are logarithmic and are plotted against the dilution series, which is therefore also used logarithmically. The slope of the resulting straight line can determine the effectiveness of the amplification.

This is to be shown by way of example for the GAPDH gene (Fig. 5.4.3). The GAPDH gene is a common control gene, since its expression shows no demonstrable transcriptional regulation and it is therefore always expressed at the same level: it is a constitutive gene.

[Figure not reproduced in the web edition. See page 48 of the PDF.]

Fig. 5.4.3: Example of the determination of the PCR effectiveness of GAPDH.

The effectiveness was determined by a PCR with the aid of a dilution series (undiluted, 1:10, 1:100 and 1:1000). For the PCR effectiveness E_R of GAPDH the following applies (Eq. 4-5):

Eq. 4-5. PCR effectiveness E_R of GAPDH.

The individual samples contained different amounts of cDNA and had therefore to be matched to one another via the internal standard GAPDH. With the aid of dilutions a particular CT value was set. These dilutions were determined by the dilution series of the PCR effectiveness of GAPDH.

In addition a fine adjustment was carried out according to (Eq. 5-5):

V_new = new volume of cDNA stock solution to be used; E = effectiveness of the PCR GAPDH amplification; C_T,old = old CT value from the first adjustment; C_T,new = new CT value to be set; V_old = volume of cDNA stock solution used. (Eq. 5-5)

The samples matched in this way were run in the PCR with primers for the following genes:

1. Collagen type I – this matrix protein is not a specific constituent of articular cartilage, but its expression is strongly up-regulated during the dedifferentiation of the cells.
2. Collagen type II – is a main constituent of the extracellular matrix of hyaline cartilage; because of its high content compared with other tissue it counts as a marker gene.
3. COMP (cartilage oligomeric matrix protein) – non-collagenous glycoprotein of the extracellular matrix with a cartilage-specific expression pattern.
4. Aggrecan – is a proteoglycan which makes up the largest share of these in the extracellular matrix. The degradation products count as markers for cartilage destruction.

The evaluation took place via the route described, which represents a modification of the ΔΔCT method (Heid et al., 1996; Winer et al., 1999; User Bulletin #2, 2001).

CHAPTER 6: Results

6.1 Cell isolation and cultivation

6.1.1 Determination of cell number and viability

After the isolation of the chondrocytes, the cell number and the viability of the cells were determined by means of trypan blue and the Neubauer chamber. Trypan blue is a negatively charged chromophore which reacts unspecifically with the proteins. In dead cells it can diffuse through the defective cell membrane into the interior of the cell and react there with the proteins. The dead cells then appear permanently blue. Cell membranes of living cells are impermeable to trypan blue. The results presented in Tab. 6.1 show an average viability of 76%, which is a satisfactory result for primary cells. Cells with a viability below 60% were discarded.

Tab. 6.1: Cell numbers and viability after the isolation of primary chondrocytes.

For the production of the 3D pellet cultures from the primary chondrocytes, 0.6 million per well (200 µl) were seeded. In order to avoid the dedifferentiation of the chondrocytes, these were transferred into the wells immediately after the isolation or after cryopreservation. From the first day the medium was supplemented with vitamin C in order to stimulate matrix formation.

6.1.2 Optical assessment of the pellet cultures

From the scans of the plates it can be clearly seen that the manually treated cultures were exposed to greater shear forces than the automatically supplied ones. With the manual medium change using the multiwell pipette, no constant and reproducible sequence can be ensured, which is an essential precondition for a meaningful test system for pharmaceuticals. As can be seen in Figure 6.1.1, the manually treated pellets show clear defects, such as flipping over or detaching. The cultures in which the medium change was carried out with the aid of the CyBi®-Disk show, as can be seen in Figure 6.1.1, no damage from shear forces: they give a uniform picture.

Fig. 6.1.1: Scans of porcine cartilage pellets cultivated for 9 days: A. manual, B. automatically treated.

6.1.3 Creation of a program for medium exchange

The automated medium change was carried out with the aid of the CyBi®-Disk (see Sect. 3.5). The plate transport takes place by means of a rotating disk, so that access to the plates is ensured at any time. The instrument stood under a sterile bench in order to keep the contamination risk as low as possible. In the creation of the program, attention was paid to the requirements of the pellet cultures. In the first three days only half, that is 100 µl, of the total volume was exchanged, in order not to damage the still unstable pellet. In order to keep the shear stress low, the aspiration and the addition of the medium took place in three steps in the slowest speed mode. In Figure 6.1.2 a section of the sequence program from the 4th day in culture is presented. Here the complete volume of the medium (200 µl) was exchanged. This takes place in 3 steps: 1. 50 µl; 2. 100 µl; 3. 50 µl. The new medium is added at equal intervals. The external washing program (Sect. 10 Appendix) fills the tips completely with 70% ethanol and afterwards washes with distilled water; this takes place before and after every contact with the culture.

These parameters also apply to the manually performed process, only in this case they are extremely difficult to implement. The manual pipetting procedure is not completely reproducible and a consistent treatment of every pellet on every day is impossible to realise.

[Figure not reproduced in the web edition. See page 53 of the PDF.]

Fig. 6.1.2: Section of the sequence program for the medium exchange with the liquid handling system CyBi®-Disk from the 4th day in culture. The total volume of 200 µl is exchanged.

The complete supply of one 96-well plate takes 7 min 45 sec including the washing steps; per well 3 min 23 sec are needed. In comparison, the manually performed medium change takes 19 min, but only 12 sec per well. The simultaneous processing of the CyBi®-Disk reduces the processing time by 60%.

6.2 Histochemistry

The histochemical stainings of the frozen tissue sections of the pellet cultures at various time points of the cultivation can make clear the detection of components that are also constituents of the extracellular matrix.

Pellet cultures after 14 days in culture were stained. The 3D co-cultures were examined after 16, 19 and 28 days in culture. The RA synovial cells, that is cells of the lines HSE (invasive-aggressive) and K4IM (healthy synovial tissue), were used for the co-cultivation.

6.2.1 Overview staining

HE staining. The haematoxylin-eosin staining is an overview stain with the aid of which conclusions about the cell distribution in the pellet culture could be drawn. In Figure 6.2.1 a-b, sections of pellet cultures cultivated for 14 days without synovial cells can be seen, whereby a was treated automatically and b manually. It can be clearly seen that the surface of the automatically supplied pellet is considerably more even and uniform. The distribution of the cells is good in both samples, whereby in this section the manually treated sample shows a high cell density. Clear distances between the cells can be seen.

[Figure not reproduced in the web edition. See page 54 of the PDF.]

Fig. 6.2.1: HE staining, 14 days, magnification 400-fold; a) automated, b) manually supplied. The arrows show the pellet surface.

Figure 6.2.2 shows stained sections after 16 days of cultivation. After 14 days these were wetted with synovial cells of the cell line HSE. c-d were supplied by the CyBi®-Disk and e-f manually. In every sample a uniform cell distribution can be seen, whereby the surface of the pellets in the automatically treated ones (a-b) appears smoother and more even. In the marginal areas a higher cell number can also be seen here. The use of the synovial cell line K4IM likewise shows the same results after 16 days of cultivation (Fig. 6.2.3 g-j).

[Figure not reproduced in the web edition. See page 55 of the PDF.]

[Figure not reproduced in the web edition. See page 55 of the PDF.]

Fig. 6.2.2: HE staining after 16 days of cultivation, after 14 days synovial cells (HSE) were seeded: a- (200-fold) b (400-fold) automated; c- (200-fold) d (400-fold) manual.

[Figure not reproduced in the web edition. See page 56 of the PDF.]

[Figure not reproduced in the web edition. See page 56 of the PDF.]

Fig. 6.2.3: HE staining after 16 days of cultivation, after 14 days synovial cells (K4IM) were seeded: a- (200-fold) b (400-fold) automated; c- (200-fold) d (400-fold) manual.

In both cases it can be clearly seen that in the automatically supplied pellets the surface is coated with synovial cells, whereas in the manually treated ones these are hardly recognisable.

In Figure 6.2.4 a-f, co-cultures after 19 days of cultivation can be seen, stained with haematoxylin/eosin and automatically supplied. Figure 6.2.4 b/e shows two magnifications of pellets with HSE synovial cells; these can be seen on the surface of the culture. The cell distribution is uniform, whereby in the lower part of the pellets a slight condensation of the cells can be seen (Fig. 6.2.4 e).

[Figure not reproduced in the web edition. See page 57 of the PDF.]

[Figure not reproduced in the web edition. See page 57 of the PDF.]

Fig. 6.2.4: HE staining after 19 days of cultivation, after 14 days the synovial cells were seeded: a (200x)/b (400x) synovial cells K4IM; b (200x)/e (400x) synovial cells HSE; c (200x)/f (400x) without synovial cells; a-f automatically supplied.

In Figure 6.2.4 a/b the pellets can be seen which were overlaid after 14 days with cells of the synovial cell line K4IM. The cells are uniformly distributed and the synovial cells are visible on the surface. A pellet culture after 19 days without synovial cells is presented in Figure 6.2.4 e/f.

Even after 28 days of cultivation of the co-culture, a homogeneous cell distribution of the chondrocytes can be clearly shown by the overview staining (Fig. 6.2.5: a-b). The layer of the respective synovial cell line is visible.

[Figure not reproduced in the web edition. See page 57 of the PDF.]

Fig. 6.2.5: HE staining after 28 days of cultivation of the pellets, after 14 days the synovial cells were seeded: a with HSE synovial cells; b with K4IM synovial cells, a/b automatically supplied.

All HE stainings demonstrate a homogeneous cell distribution and a typically round cell morphology in the pellet cultures and the co-cultures. Distances between the chondrocytes can be seen.

6.2.2 Stainings for the detection of the extracellular matrix

Azan staining. With the aid of the azan staining the total collagen formed is stained by aniline blue. The cell nuclei are counterstained red. In this way it can be shown that extracellular matrix was formed.

In Figure 6.2.6, pellets without synovial cells after 14 days of cultivation can be seen. Figure 6.2.6 a shows the manually supplied culture, Figure 6.2.6 b the automatically treated one. In both cases a staining can be seen. The collagens in the cartilage tissue are stained blue with differing intensity.

[Figure not reproduced in the web edition. See page 58 of the PDF.]

Fig. 6.2.6: Azan staining after 14 days of cultivation; magnification: 400x: a manual; b automated.

In Figures 6.2.7 a-d and 6.2.8 a-d, co-cultures after 16 days of cultivation are presented, 6.2.7 a-d with the synovial cell line HSE and 6.2.8 a-d with K4IM.

[Figure not reproduced in the web edition. See page 59 of the PDF.]

[Figure not reproduced in the web edition. See page 59 of the PDF.]

Fig. 6.2.7: Azan staining after 16 days of cultivation, after 14 days HSE synovial cells were seeded: a- (200x) b (400x) manually supplied; c- (200x) d (400x) automatically supplied.

[Figure not reproduced in the web edition. See page 59 of the PDF.]

[Figure not reproduced in the web edition. See page 59 of the PDF.]

Fig. 6.2.8: Azan staining after 16 days of cultivation, after 14 days K4IM synovial cells were seeded: a- (200x) b (400x) manually supplied; c- (200x) d (400x) automatically supplied.

The collagens are stained blue; they can be clearly seen between the homogeneously distributed cartilage cells. The synovial cells are not visible, since they would be seen stained reddish on the pellet surface (Fig. 6.2.7 and Fig. 6.2.8).

[Figure not reproduced in the web edition. See page 60 of the PDF.]

[Figure not reproduced in the web edition. See page 60 of the PDF.]

Fig. 6.2.9: Azan staining after 19 days of cultivation, after 14 days the synovial cells were seeded: a (200x)/d (400x) synovial cells K4IM; b (200x)/e (400x) synovial cells HSE; c (200x)/f (400x) without synovial cells; a-f automatically supplied.

After 19 days of cultivation of the co-culture, supplied only automatically, the stained collagen appears somewhat irregular (Fig. 6.2.9: a/d; b/e), but not in the pellet culture without synovial cells (Fig. 6.2.9: c/f), where a higher cell number can also be seen. In Figure 6.2.9: k/n and l/o the respective synovial cell layer can also be detected, albeit very irregularly.

[Figure not reproduced in the web edition. See page 60 of the PDF.]

Fig. 6.2.10: Azan staining after 28 days of cultivation of the pellets, after 14 days the synovial cells were seeded: a with HSE synovial cells; b with K4IM synovial cells, a/b automatically supplied.

The azan staining of the co-cultures after 28 days (Fig. 6.2.10: a-b) shows a good distribution of the cells and collagen synthesis. The blue staining is stained very intensely, especially around the chondrocytes. Both pellets are supplied automatically; Figure 6.2.10: a was coated after 14 days with synovial cells of the cell line HSE and Figure 6.2.10: b shows the co-culture with K4IM cells. The synovial cell layer can be recognised on the surface of the pellets through the counterstaining of the nuclei in red.

Alcian blue staining. Alcian blue stains the proteoglycans in the extracellular matrix. These appear in light blue. The cell nuclei can be recognised in pink through the counterstain, so the synovial cells of the co-cultures can be identified very well on the surface of the cell pellets. The presence of proteoglycans can be seen at every time point in culture.

In Figure 6.2.11: a-b, pellet cultures cultivated for 14 days without synovial cells can be seen. Figure 6.2.11: a was supplied manually and Figure 6.2.11 b automatically. The automatically treated pellets (Fig. 6.2.11: b) show a more intense staining and clearly more cells than the manually supplied ones in Figure 6.2.11: a.

[Figure not reproduced in the web edition. See page 61 of the PDF.]

Fig. 6.2.11: Alcian blue staining, 14 days, magnification: 400x: a manual; b automated.

In Figure 6.2.12: a-d the pellets are presented which were coated after 14 days with HSE synovial cells and cultivated for a further 2 days before sampling, a-b treated manually and d-c automatically. The proteoglycans can be clearly recognised by the blue staining, the cell nuclei appear pink and show a homogeneous distribution. The synovial cell layer cannot be seen; it would be stained pink and seen as a layer on the pellet surface.

[Figure not reproduced in the web edition. See page 62 of the PDF.]

[Figure not reproduced in the web edition. See page 62 of the PDF.]

Fig. 6.2.12: Alcian staining after 16 days of cultivation, after 14 days HSE synovial cells were seeded: a- (200x) b (400x) manually supplied; c- (200x) d (400x) automatically supplied.

[Figure not reproduced in the web edition. See page 62 of the PDF.]

[Figure not reproduced in the web edition. See page 62 of the PDF.]

Fig. 6.2.13: Alcian staining after 16 days of cultivation, after 14 days K4IM synovial cells were seeded: a- (200x) b (400x) manually supplied; c- (200x) d (400x) automatically supplied.

The same observation could be made with co-cultures cultivated for 16 days with K4IM synovial cells (Fig. 6.2.13: a-d). Here individual synovial cells can be seen in pink at the surface of the pellets.

In Figure 6.2.14: a-f the pellets have already been cultivated for 19 days, a/d with HSE cells, b/e with K4IM cells and c/f without synovial cells. In every case the presence of proteoglycans can be seen. The synovial cells can be recognised in Figure 6.2.14 a/b/d/e as a long, pink layer, although slightly detached from the pellet surface. The pellets without co-culture show a more homogeneous cell distribution and a strongly pronounced extracellular matrix (Fig. 6.2.14: c/f).

The co-culture after 28 days of cultivation and automated supply is presented in Figure 6.2.15 a-b. The synovial cell layer can be seen very well, in Figure 6.2.15 a HSE cells and in Figure 6.2.15 b K4IM cells. Here too a homogeneous distribution of the cartilage cells and proteoglycans can be detected.

[Figure not reproduced in the web edition. See page 63 of the PDF.]

[Figure not reproduced in the web edition. See page 63 of the PDF.]

Fig. 6.2.14: Alcian staining after 19 days of cultivation, after 14 days the synovial cells were seeded: a (200x)/c (400x) synovial cells K4IM; b (200x)/e (400x) synovial cells HSE; c (200x)/f (400x) without synovial cells; a-f automatically supplied.

[Figure not reproduced in the web edition. See page 64 of the PDF.]

Fig. 6.2.15: HE staining after 28 days of cultivation of the pellets, after 14 days the synovial cells were seeded: a with HSE synovial cells; b with K4IM synovial cells, a/b automatically supplied.

In all histological stainings it could be clearly seen that the chondrocytes retained their typically round phenotype during the entire cultivation phase.

The size of the 3D pellet cultures corresponded in diameter to the well of a 96-well plate and corresponded to a thickness of about 1–2 mm.

6.3 Immunohistochemistry

In immunohistochemistry the cartilage-specific constituents of the extracellular matrix are stained by an antigen-antibody reaction. Collagen type II is the main constituent of the hyaline cartilage tissue; in general the collagen fibres are responsible for the tensile strength in cartilage. Collagen type I is hardly a natural constituent of the hyaline cartilage tissue, it is found in fibrous cartilage. Through a dedifferentiation of the cartilage cells in vitro, however, collagen type I synthesis can also occur. The respective collagens are stained orange and for every staining a negative control was prepared, in that the first antibody, which binds specifically to the collagen, is not added. The dye-associated antibody can then not bind and a staining does not occur.

Figure 6.3.1: a-f shows pellet cultures after 14 days in culture, a-c were supplied with medium manually, d-f with the aid of the CyBi®-Disk liquid handling system. Both controls show no staining and are thus negative (Fig. 6.3.1: a/d). In the immunohistological staining of collagen type I (Fig. 6.3.1: b/e) a staining in the marginal area can be seen in the manually treated pellets. The remaining matrix shows only a weak presence of collagen type I. The automatically supplied cultures (Fig. 6.3.1: e), by contrast, show no orange staining in the marginal area. Collagen type II was detected in every case (Fig. 6.3.1: c/f). In Figure 6.3.1 f a clear orange staining can be seen in the marginal area of the pellet and much collagen type II can also be seen in the remaining matrix.

[Figure not reproduced in the web edition. See page 65 of the PDF.]

[Figure not reproduced in the web edition. See page 65 of the PDF.]

Fig. 6.3.1: Immunohistochemistry of pellets cultivated for 14 days. Magnification 400x; a-c manually supplied, d-f automatically treated. a/d negative control; b/e collagen type I; c/f collagen type II.

After 16 days in culture with K4IM cells, the staining of collagen type II can be seen very strongly in the automatically supplied pellets (Fig. 6.3.2: b), whereas collagen type I was hardly formed here.

The manually treated co-cultures show in the marginal areas a staining of collagen type I as well as II (Fig. 6.3.2: e-f), but compared with Fig. 6.3.2 c the collagen type II is not stained so intensely. Both controls are unstained and thus negative (Fig. 6.3.2: a/d).

[Figure not reproduced in the web edition. See page 66 of the PDF.]

[Figure not reproduced in the web edition. See page 66 of the PDF.]

Fig. 6.3.2: Immunohistochemistry of pellets cultivated for 16 days, after 14 days synovial cells K4IM were applied. Magnification 400x; a-c automatically supplied, d-f manually treated. a/d negative control; b/e collagen type I; c/f collagen type II.

[Figure not reproduced in the web edition. See page 66 of the PDF.]

[Figure not reproduced in the web edition. See page 66 of the PDF.]

Fig. 6.3.3: Immunohistochemistry of pellets cultivated for 16 days, after 14 days synovial cells HSE were applied. Magnification 400x; a-c automatically supplied, d-f manually treated. a/d negative control; b/e collagen type I; c/f collagen type II.

In the co-cultures complemented with HSE cells too, hardly any detection of collagen type I can be seen in the automatically supplied pellets, but collagen type II was detected, especially at the margins (Fig. 6.3.3: b-c). Figure 6.3.3 e shows the staining of collagen type I in the manually treated pellets, only little is present. In Figure 6.3.3 f the presence of collagen type II can be seen. The controls are unstained in both cases, that is, negative (Fig. 6.3.3 a/d).

After 19 days of cultivation too, hardly any collagen type I is detected (Fig. 6.3.4: b/e/h) in all the cultures, HSE, K4IM and without synovial cells. All pellets were supplied automatically. Collagen type II is detected in the co-cultures by orange staining; in the pure pellet the staining appears somewhat more intense.

The marginal areas show in all three cases an increased occurrence of collagen type II (Fig. 6.3.4: c/f/i). In Figure 6.3.4 a/d/g no staining can be seen, from which it follows that the controls are negative.

[Figure not reproduced in the web edition. See page 67 of the PDF.]

[Figure not reproduced in the web edition. See page 67 of the PDF.]

[Figure not reproduced in the web edition. See page 67 of the PDF.]

Fig. 6.3.4: Immunohistochemistry of pellets cultivated for 19 days, after 14 days synovial cells were applied. a-i automatically supplied; magnification 400x; a-c HSE; d-e K4IM; g-i without synovial cells. a/d/g negative control; b/e/h collagen type I; c/f/i collagen type II.

Even after 28 days in culture the co-cultures show similar stainings, but the detection of collagen type I is clearer here than in younger pellets (Fig. 6.3.5: b/e). Much collagen type II is formed, especially in the marginal areas (Fig. 6.3.5: c/f). The controls are unstained (Fig. 6.3.5: a/d).

In all immunohistological stainings the chondrocytes retain their round shape and the typical matrix protein collagen type II shows a strong synthesis. Collagen type I, which is a protein formed by fibroblasts, is produced more strongly only in the co-culture after 28 days in culture. In the pellet cultures without synovial cells hardly any synthesis takes place.

[Figure not reproduced in the web edition. See page 68 of the PDF.]

[Figure not reproduced in the web edition. See page 68 of the PDF.]

Fig. 6.3.5: Immunohistochemistry of the co-cultures after 28 days in culture, after 14 days synovial cells were seeded. Magnification: 400x; a-c HSE, d-f K4IM; a/d negative control; b/e collagen type I; c/f collagen type II. a-f automatically supplied.

6.4 Gene expression analysis

In order to analyse the gene expression of important matrix constituents in the in vitro pannus model, some characteristic genes involved in the build-up of the extracellular matrix were selected. Since the model consists of porcine chondrocytes and human synovial cells, it was possible, through the use of species-specific primers, to consider the gene expression of the chondrocyte genes separately. Because of differing sample quality and quantity the results obtained can fluctuate, so besides a quantification of the target sequence a relativising factor is indispensable. For this purpose reference genes, so-called housekeeping genes, are used, which are constitutively expressed in every cell. By relating the genes under investigation to the amplification of such a transcript, an assessment of the sample quality is possible and at the same time one obtains a measure of the gene expression in the tissue concerned (Foss, 1998). cDNA from co-cultures of porcine chondrocyte pellet cultures and human synovial cells was used as template. GAPDH was chosen as reference gene; by relating to this constitutively expressed gene the samples could be quantified indirectly. The CT value is recorded in the cycle at which the first significant rise of the fluorescence signal appears. It is directly proportional to the starting template number (see Sect. 5.4). At an effectiveness of 100% of the reaction, 3.5–4.0 cycles are needed in order to increase the number of copies by one exponent (Kreuzer et al., 1999).

The CT values of the genes COMP, collagen type I and II and aggrecan were each placed in ratio to GAPDH and the values plotted over the cultivation period (Fig. 6.4.1–11). Figures 6.4.1–5 showed the pellet cultures without synovial cells after 3 and 10 days in culture, the first day being the starting value of the native chondrocytes. In the pellet cultures, manual and automated supply were compared (Fig. 6.4.1–5).

Collagen type II showed no down-regulation of the expression between the 3rd and the 10th day of the cultivation. The manually supplied cultures are only slightly less expressed (Fig. 6.4.1).

[Figure not reproduced in the web edition. See page 69 of the PDF.]

Fig. 6.4.1: Representation of the mean values of the expression of collagen type II in relation to GAPDH. Pellet culture on the 1st, 3rd and 10th day in culture, manual and automated in comparison.

The expression of the non-cartilage-specific matrix protein collagen I remained unchanged low up to the 10th day in culture, both manually and automatically supplied. After 10 days of cultivation the expression level rose, especially in the automatically supplied pellets (Fig. 6.4.2).

[Figure not reproduced in the web edition. See page 70 of the PDF.]

Fig. 6.4.2: Representation of the mean values of the expression of collagen type I in relation to GAPDH. Pellet culture on the 1st, 3rd and 10th day in culture, manual and automated in comparison.

The aggrecan gene falls in its expression in the robot pellets from the starting value and remains largely the same up to day 10 in culture (Fig. 6.4.3). The manually treated cultures show a stronger fall of the expression level but likewise remain unchanged between the 3rd and the 10th day in culture (Fig. 6.4.3).

[Figure not reproduced in the web edition. See page 71 of the PDF.]

Fig. 6.4.3: Representation of the mean values of the expression of aggrecan in relation to GAPDH. Pellet culture on the 1st, 3rd and 10th day in culture, manual and automated in comparison.

The expression of COMP falls in both cases, but remains almost the same in the automatically cultivated pellets (Fig. 6.4.4).

[Figure not reproduced in the web edition. See page 72 of the PDF.]

Fig. 6.4.4: Representation of the mean values of the expression of COMP in relation to GAPDH. Pellet culture on the 1st, 3rd and 10th day in culture, manual and automated in comparison.

The pellet cultures were completed after 14 days by the synovial fibroblasts. Here only the automatically supplied co-cultures with the synovial cell line K4IM are presented. In the co-cultures with K4IM cells a steady increase of the expression of collagen type II can be seen (Fig. 6.4.5). Relative to the constitutively expressed gene GAPDH, the matrix protein typical of hyaline cartilage shows a strong expression up to the 28th day in culture.

[Figure not reproduced in the web edition. See page 73 of the PDF.]

Fig. 6.4.5: Representation of the mean values of the expression of collagen type II in relation to GAPDH. Co-culture K4IM after 16, 19 and 28 days in culture, automatically supplied.

The cartilage-atypical collagen type I likewise shows an expression pattern similar to that of collagen type II, but overall the expression level is very low compared with collagen type II (Fig. 6.4.6).

[Figure not reproduced in the web edition. See page 74 of the PDF.]

Fig. 6.4.6: Representation of the mean values of the expression of collagen type I in relation to GAPDH. Co-culture K4IM after 16, 19 and 28 days in culture, automatically supplied.

The transcription of aggrecan shows a sudden rise after 19 days of the co-cultivation with K4IM cells, but falls again after 28 days (Fig. 6.4.7). In general the expression level of aggrecan is very low.

[Figure not reproduced in the web edition. See page 75 of the PDF.]

Fig. 6.4.7: Representation of the mean values of the expression of aggrecan in relation to GAPDH. Co-culture K4IM after 16, 19 and 28 days in culture, automatically supplied.

The expression of COMP shows, as presented in Fig. 6.4.8, a fall after 19 days of cultivation of the co-culture. After 28 days in culture, however, the transcription rises to a maximum.

[Figure not reproduced in the web edition. See page 76 of the PDF.]

Fig. 6.4.8: Representation of the mean values of the expression of COMP in relation to GAPDH. Co-culture K4IM after 16, 19 and 28 days in culture, automatically supplied.

In order to be able to analyse the course of the expression over the entire cultivation period better, these are presented in Figures 6.4.9–11. Only the automatically supplied cultures were taken into consideration. From the 14th day the co-culture was completed with K4IM synovial cells.

Fig. 6.4.9: Representation of the mean values of the expression of collagen type I and II in relation to GAPDH over the entire course of the cultivation. After 14 days K4IM synovial cells were seeded.

Collagen type I and II were presented in one figure in order to compare them better (Fig. 6.4.9). Collagen type II initially showed a clearly higher expression than collagen type I, which was only expressed at a low level after the 3rd day. From the 16th day in culture both proteins show the same low expression level.

The course of the expression level of aggrecan is comparable with that of collagen type II, a fall to a low level from the 16th day (Fig. 6.4.10).

[Figure not reproduced in the web edition. See page 78 of the PDF.]

Fig. 6.4.10: Representation of the mean values of the expression of aggrecan in relation to GAPDH over the entire course of the cultivation. After 14 days K4IM synovial cells were seeded.

A relatively constant level can be observed in the expression of COMP; only on the 19th day in culture were slight fluctuations apparent (Fig. 6.4.11).

[Figure not reproduced in the web edition. See page 79 of the PDF.]

Fig. 6.4.11: Representation of the mean values of the expression of COMP in relation to GAPDH over the entire course of the cultivation. After 14 days K4IM synovial cells were seeded.

CHAPTER 7: Discussion

The in vitro test system for the investigation of pathogenetic mechanisms in destructive joint diseases such as rheumatoid arthritis has already been examined for its functionality (Smolian et al., 2001). It was shown that the destructive course of this disease can be simulated in vitro with this model (Schultz et al., 1997; Smolian et al., 2001). For this, articular chondrocytes and human RA synovial cells were cultivated interactively. Activated RA synovial fibroblasts (RASFs) appear to play the main role in the destruction of the joint (Muller-Ladner et al., 2000). This could already be shown in the SCID mouse model, where the invasive behaviour of the RASFs destroyed the human cartilage, independently of T cells, macrophages and the inflammatory process (Muller-Ladner et al., 1996). The invasion of the synovial cells into the cartilage goes hand in hand with a strong expression of adhesion molecules, matrix-degrading enzymes and the inhibition of apoptosis (Schedel et al., 2002; Franz et al., 2000). It is, however, independent of the proliferation of the cells, which was connected with the invasion in the pathogenetic process (Seemayer et al., 2003).

In order to be able to apply this test system to the desired extent, and thereby to enable its use for drug testing in the field of destructive joint diseases, the cultivation process must be automated.

Particular interest attaches here to the build-up of the extracellular matrix and the homogeneous distribution of the synovial cells. The automation procedure must ensure that an intact cartilage pellet is formed and that the demonstrated functionalities of the manually cultivated system are preserved. The unique biological and mechanical properties of cartilage depend on the tissue structure and the interactions between chondrocytes and the extracellular matrix. The cartilage cells form the macromolecular scaffold of the tissue from three kinds of molecule: collagen, proteoglycans and non-collagenous proteins (Buckwalter, 1998).

The analysis of the constituents of this typical hyaline cartilage matrix was carried out histologically and immunohistologically, which is a standardised method for the investigation of cartilage tissue samples (Hyllsted, 2002).

The isolation of primary chondrocytes was carried out with porcine cartilage because of the increased demand for cells. Furthermore a consistent quality had to be ensured, which is difficult with human material. Since the structure of the matrix of porcine and human articular cartilage is very similar (Frye et al., 1996), this presents no problem. With an average viability of the cells of 76%, the cell isolation always met the requirement of at least 60%. The pellet culture offers the possibility of producing 3D cultures simply, without the use of carrier materials such as hydrogel (Liu Tsang, 2004). A choice of the culture vessels was made with regard to the growth conditions of the cells and the requirements of automation, whereby on the one hand the supply of high cell densities is prevented by culture vessels that are too small, while for the automation process there were initially few restrictions on miniaturisation. A format smaller than 96-well leads to difficulties in the cell supply and can only with difficulty ensure the required growth conditions. A successful cultivation in 48-well format had already been shown (Smolian et al., 2001), which is why the next smaller step, the 96-well format, was selected.

High throughput screening assays are a key component in the identification of new pharmaceuticals, which makes it possible to carry out a test series, depending on its objective, within a period of weeks or months (Beggs et al., 1999). The technical quality of a liquid handling process depends on the transfer rate of the residual liquid during the pipetting procedure, the mixing behaviour in the wells, the formation of foam and bubbles and the evaporation of the medium. Studies with the CyBi®-Well have already been carried out on this (Berg et al., 2001). Within the framework of this thesis the use of this system was also to be tested for application in the field of cell culture.

The creation of a program for culture management was realised with regard to the sterility of the process, the saving of time and the reduction of the shear forces. Through a double washing step of the pipette tips before and after every contact with the culture, the contamination risk is reduced. The tips of the CyBi®-Well were washed with 70% ethanol and distilled water. The rinsing with distilled water is necessary for the removal of the ethanol, since alcohol residues can bring about the destruction of the culture (Me'lnikova, 1989; Chen et al., 2002). In general the automation process brought a 60% saving of time per plate compared with the manual supply of the pellet cultures. The pipetting in each individual well was increased, compared with the manual process, from about 12 seconds to 3 minutes 23 seconds. The simultaneous supply of the entire plate in the automated procedure brought, despite slow pipetting, a clear reduction of the processing time.

In order to keep the shear loading during the pipetting procedure as low as possible, both the aspiration and the pipetting of the fresh medium were carried out in 3 stages. In this, the first 50 µl were aspirated at the surface, at the next movement of the pipette tips into the middle of the well 100 µl of medium were removed and the last 50 µl were aspirated relatively close above the pellet surface. These three steps were also carried out in reverse when adding the medium. The shear loading influences the formation of the extracellular matrix in animal cells, whereby on the one hand it can be reduced by this (Lane Smith, 2000; Raimondi et al., 2002), and on the other hand an induced shear stress can trigger an increased synthesis (Waldmann, 2003). With regard to this thesis it is a matter of avoiding the mechanical destruction of the pellets; for this it is necessary to ensure a minimal shear loading. The pellets are very sensitive in the first three days, so the medium change should not be complete during this time. The low shear forces of the sequence program could be shown with the optical assessment of the pellets. In the scans of the plates supplied by the CyBi®-Disk, homogeneous pellets can be seen which extend over the entire floor area of the wells. No destruction of the pellets is observed. The manual medium change led in some cases to damage of the pellets, which is attributable to pipetting that is too fast or uneven and to the associated high shear forces. A flipping over of the pellets leads, in the completion of the 3D pannus model after coating with the synovial cells, to an irregular distribution of these on the surface of the chondrocyte culture and to an insufficient reproducibility. In order to analyse the matrix degradation during the pathogenesis of rheumatoid arthritis and, where appropriate, the effect of biologically active molecules, a homogeneous distribution of the synovial cells and a uniform thickness of the cartilage layer are necessary. The program created was optimised with regard to this and, evaluated optically, brought the desired result.

The composition of the healthy articular cartilage is attributable to the balance between synthesis and degradation of the extracellular matrix, which comes about through the anabolic and catabolic activities of the chondrocytes (Dingle, 1984). The production of artificial cartilage brings with it the problem that there are differences from articular hyaline cartilage. Thus cartilage tissue grown in vitro often has the typical content of proteoglycans compared with native cartilage, but reduced collagen proportions, which ensures only a part of the mechanical stability (Waldmann et al., 2003). In vivo, human tissue is three-dimensional and multicellular. These criteria are important for the achievement of tissue-specific functions. Here the extracellular matrix plays a critical role, it is very specific and therefore its formation is of decisive importance in the binding of the cells in the tissue (Goessler et al., 2004). The most widespread method for the detection of matrix constituents is histochemistry and immunohistochemistry, but these stainings are not stoichiometric (Hyllested, 2002). In order to obtain an overall impression of the cultured cartilage pellet, an HE overview staining was carried out. The results show a homogeneous cell distribution at every time point of the cultivation. All cells were evidently involved in the formation of the extracellular matrix. This uniform cell density was shown by the histological results. The cells display their chondrocyte-typical round shape, which suggests that no change of the morphology took place. In earlier work it was shown that the cultivation of chondrocytes in monolayer leads to a dedifferentiation, in which the cells show a fibroblastoid morphology and lose the gene expression specific to chondrocytes (Benya et al., 1978; Benya and Schaffer, 1982; Watt, 1988; Holtzer et al., 1960; von der Mark, 1977; Horton, 1989). Chondrocyte proliferation, differentiation and homeostasis are influenced not only by the interaction between mediators and the genome of the cells; the extracellular matrix also plays an important role for the behaviour of the chondrocytes (Goessler et al., 2004). The extracellular matrix of articular cartilage contains a large number of non-collagenous components, of which the proteoglycans constitute a large group. They differ in structure, distribution and function. Some occur only in specific kinds of cartilage or in particular zones of the matrix, some vary with the age or the development of the individual and others are universally expressed (Roughley, 2001). The highest proportion of the proteoglycans in hyaline cartilage tissue is held by aggrecan; it belongs to the family of the aggregating proteoglycans, which form large multimolecular complexes with hyaluronan. They fulfil various biological functions, act in tissue organisation, which influences cell growth, modulate the activities of growth factors and regulate collagen fibrillogenesis. The family further includes versican, neurocan and brevican, of which only versican has a small share in cartilage tissue (Iozzo, 1998).

The histochemical staining with alcian blue displays the proteoglycans of the extracellular matrix in blue. After 14 days in culture the manually supplied pellets show a weaker blue staining than the automatically treated ones. This could be connected with the higher shear loading, which lowers the formation of matrix proteins (Roughley, 2001), but conversely can also up-regulate it. The shear forces can also be responsible for the detachment of the synovial cell layer in some cultures. The presence of proteoglycans after 28 days in culture suggests the formation of a specific extracellular matrix. The uniform cell distribution of the chondrocytes is a proof of a well-organised tissue. The cells show their cartilage-typical round morphology during the entire cultivation time. Chondrocytes are responsible for the synthesis of the proteoglycans, so a good distribution of these and the retention of their morphology play a decisive role in the matrix build-up (Handley et al., 1985), since a change of the morphology can lead to a change of synthesis.

Collagens are responsible, among other things, for the elasticity, a low-friction surface and the absorption of high physical forces (Cremer et al., 1998). 19 different collagen types have been identified in tissue, 5 of them occur in isolable proportions in articular cartilage (types II, VI, IX–XI) (Seyer and Klang, 1996; Eyre and Wu, 1995). With the aid of the azan staining the collagenous connective tissue is stained blue and the cell nuclei are counterstained with azocarmine. At every time point of the pellets in culture a clear blue staining was apparent, which demonstrates the presence of collagens.

The results of the stainings with alcian blue and azan clearly showed the synthesis of proteoglycans and total collagen as the ground substance of the cartilage (Lohmander, 1988; Toole, 1991). Since the histological stainings are not stoichiometric, no conclusion about the quantity of the proteoglycans and collagens formed can be drawn from the colour intensity. It can, however, be shown in this way that a formation of extracellular matrix tissue took place at every investigated time point of the cultivation. A quantitative analysis of the expression levels can take place through the gene expression analysis. The extracellular matrix scaffold and 66% of the dry mass of adult cartilage tissue consist of polymeric collagen (Eyre, 2002). Collagen type II makes up 80–85% of the collagens in articular cartilage tissue. It forms the backbone of the heteropolymeric fibrils in cartilage, exactly as collagen type I does in fibrous cartilage (Cremer et al., 1998). Through the immunohistological stainings, collagen type I and type II can be detected specifically. It is of particular importance to detect these collagens, since their synthesis can change through ageing and other functional disorders. In the course of a dedifferentiation the synthesis of collagen type II can switch to type I (Alsalameh et al., 1992). This is of particular importance in the destructive process of rheumatoid arthritis, since the products of the metabolic degradation of these matrix molecules are continuously released into the synovial fluid and further into the bloodstream. They thus serve as markers for the metabolic changes of the cartilage (Kuettner, 1992).

Within the framework of this thesis the formation of collagen type II in particular was investigated, in order to demonstrate the specific extracellular matrix of the hyaline cartilage. Three-dimensional cell culture models have the advantage over monolayer models of reflecting the three-dimensional interactions of the cells (Bouhadir and Mooney, 1998). The conventional culture systems often lack the complex interplay of cell-cell and cell-matrix interactions (Adams, 1993; Horwitz, 1994; Gumbiner, 1996). The immunohistological stainings show collagen type II synthesised to differing degrees, whereby only limited conclusions about the quantity can be drawn from the antibody-antigen reaction. In the automatically treated pellets almost no synthesis of collagen type I was apparent, which is not a natural constituent of the hyaline cartilage tissue but of fibrous cartilage (Kühn, 1987). Biochemical analyses show, however, that the collagen synthesis can switch from the cartilage-specific collagen type II to the atypical collagen type I (Alsalameh et al., 1992). Compared with the automatically supplied cultures, the manually treated ones showed, especially in the marginal areas, an increased production of collagen type I, which was demonstrated by the immunohistological reaction. This result could be explained by the increased shear stress through the manual, irregular supply and thus a higher degree of dedifferentiation of the chondrocytes into fibroblastoid cells. The substrate supply plays an important role in the synthesis of collagen type I and II. With a good supply of nutrients, increased collagen type I is produced as a result of the dedifferentiation, and with poor supply, as in native cartilage, collagen type II (Alsalameh et al., 1992). Because of the small medium volume in the wells compared with the high cell density, a good supply is present precisely in the marginal areas, which led there to an increased synthesis of collagen type I. The co-cultures show after 16 and 19 days a weakening of the collagen type II synthesis, whereby no difference between the invasive and the non-invasive synovial fibroblasts was apparent. The co-culture after 19 days, automatically supplied, shows a higher proportion of collagen type II than the co-cultures. After four weeks of cultivation of the co-culture, clearly more collagen type I is formed and less collagen type II, which suggests a switch in the expression from type II to type I (Stokes et al., 2001). In vivo, RA synovial fibroblasts have the poten- tial to destroy the cartilage tissue, which was also observed in vitro in the SCID mouse model (Muller-Ladner et al., 1996). In the interaction between chondrocytes and RA synovial fibroblasts in the in vitro pannus model, the invasive behaviour of the cells was demonstrated (Smolian et al, 2001).

A detection at the genetic level can give precise conclusions about the expression of specific matrix proteins. For this, however, it must be ensured that the regulation of the genes takes place for the most part at the transcriptional level, which was shown by Maatta et al., 1993. The up- and down-regulation of the genes is connected with the change of the phenotype, whereby for example the expression level of collagen type II expression is influenced (Stokes et al., 2001). The RNA isolation from the pellet cultures was the basis for the gene expression analysis. Compared with the isolation from monolayer this is problematic, since in native cartilage a lower cell density is present and the RNA can be contaminated by constituents of the extracellular matrix (Gehrsitz et al., 2001). On average 16 µg of RNA were isolated and 3 µg of each transcribed into cDNA. For the investigation the genes relevant to the matrix build-up were selected, collagen type II, aggrecan and COMP. In order to check a possible dedifferentiation of the chondrocytes, the expression of collagen type I, typical of fibrous cartilage, was also analysed.

In the method of the real-time RT-PCR for the determination of the gene expression, as in every method, some assumptions must be made in order to be able to carry out an evaluation of the results. Thus the result is only indirectly quantitative; since every gene is expressed in different proportions, they cannot be compared directly with one another. The expression of every gene is related to a constitutively expressed reference gene. The housekeeping gene GAPDH was used, with the assumption of a constant expression level (Winer et al., 1999). Furthermore, in the calculation the fluorescence threshold value of the instrument is always related to an equal amount of fluorescing dsDNA. This assumption cannot, however, be regarded as correct, since the fluorescence intensity depends not only on the number of molecules but also on the length of the amplicon. In addition the SYBR Green® dye intercalates to differing degrees depending on the base pairing of the dsDNA (Users Bulletin #2, 2001). The assumptions must be taken into account, but they are not decisive for the evaluation of the regulation of expression.

The main constituent of the hyaline cartilage tissue is collagen type II, whose synthesis was already shown in the immunohistological results. Collagen type II shows a high level of expression, which however decreases relatively quickly and from the 16th day in culture is down-regulated to the same level as collagen type I. On the one hand the dedifferentiation of the in vitro cultures can be a reason for the fall of the expression in the course of the cultivation (Stokes et al., 2001), on the other hand no increased synthesis is necessary any more after the completion of the matrix build-up. The dedifferentiation leads to a change of the phenotype, which has a change at the transcriptional level as a consequence (Stokes et al., 2001). The slight rise of the expression level of collagen type I confirms this assumption. In arthritic cartilage, changes in the activity of the chondrocytes and a loss of the collagen matrix take place (Gibson et al., 2001). Through arthrosis a decrease of the proteoglycans and a degradation of collagen occur, which is demonstrated by immunodetection of the collagen degradation products as markers for the cartilage decomposition (Hollander et al., 1995). For the gene expression analysis, exclusively co-cultures with non-invasive synovial fibroblasts were taken into consideration, so the down-regulation of the expression level of collagen type II cannot be connected with interactions with the co-culture.

One of the most important proteoglycans in the extracellular matrix is aggrecan. Of all proteoglycans, aggrecan shows the greatest structural variations, caused by degenerative changes and ageing processes. Degenerative processes take place extracellularly and are triggered by proteases. The cleavage of hyaluronic acid results in a loss of aggrecan, since they are connected (Roughley, 2001). The expression level of all proteoglycans varies with increasing patient age and influences the level of synthesis (Bolton et al., 1996; Melching et al., 1997). The RT-PCR results show a clear fall of the expression level up to the end of the cultivation after 28 days, whereby after the 10th day in culture the expression in relation to the expression of GAPDH falls almost to zero. The decomposition by proteases is only possible if these reach the cartilage matrix; a degradation of aggrecan is mostly to be attributed to MMPs and aggrecanases (Caterson et al., 2000). These processes are not relevant at the transcriptional level. The down-regulation of the expression level towards the end of the cultivation time could be connected with the completed matrix build-up. Through the histological results, proteoglycans were demonstrated up to the end of the cultivation, but the expression could already have been down-regulated. In addition there is always the possibility in vitro of a dedifferentiation, which alters the genetic expression of the cells (Stokes et al., 2001). The precise reasons for the genetic changes and the restrictions in the expression of some matrix proteins are still unclear; in vivo they could be triggered by diseases or destruction (Roughley, 2001).

The cartilage oligomeric matrix protein (COMP) is a relatively large protein with a molecular weight of 435 kD and consists of five identical subunits (Hedbom et al., 1992; Mörgelin et al., 1992; Oldberg et al., 1992). It is synthesised by chondrocytes and is localised in the extracellular matrix, whereby in young tissue it is found rather in the pericellular or territorial section of the matrix and in adult cartilage in the interterritorial space at a distance from the chondrocytes (Shen et al., 1995). COMP is also expressed by synovial fibroblasts and reacts strongly to the inflammation-mediating cytokines interleukin-1 and TNF-β. During the cultivation of isolated chondrocytes it is synthesised at a low level (Recklies et al., 1998). The expression level of the pellet cultures falls somewhat after 3 days in vitro, but then remains constant over the entire cultivation period, also with the inclusion of the synovial fibroblasts. It was observed that COMP is detected in increasing amounts in the synovial fluid of patients with RA, which goes hand in hand with the cartilage degeneration (Lohmander et al., 1994; Mansson et al., 1995; Sharif et al., 1995; Lorenzo, 1998; Bleasel et al., 1999; Petersson et al., 1998). It is, however, not yet clarified whether the COMP level reflects the stage of the inflammatory process, but it could be an indication of the first changes in the cartilage in RA and thus a marker for joint diseases (Garnero et al., 2000). Within the framework of this thesis the expression of COMP as a constituent of the extracellular matrix was to be demonstrated. The initial fall of the expression level and the relatively constant course over the further cultivation period corresponded to our expectations.

The results of this thesis show the build-up of the extracellular cartilage matrix, whereby the gene expression of collagen type II, aggrecan and COMP was steadily down-regulated. The comparison of manual with automated brings clear advantages for the automated process. The scans, the histological and the immunohistological stainings of the automatically supplied pellets show that these are reproducibly homogeneous. The objective of this project consisted in the transfer of the pannus model into a procedure for routine investigations. Through the successful automation in 96-well format a first step has been taken and it can now be continued in connection with drug testing. The liquid handling system CyBi®-Disk was suitable for use in cell culture. A transfer of the in vitro pannus into a high throughput screen leads not only to a minimisation of animal experiments, but in particular to a saving of time and of the costs associated with it.

CHAPTER 8: Outlook

The advantage of this culture model lies in the rapid analysis of the efficacy profile of the substances under in vitro conditions which reflect important aspects of the pathology in destructive joint diseases. The successful automation of the pannus model led to the development of a high throughput screening system, which can now be applied further to the testing of active substances against rheumatoid arthritis (RA). For this, technical modifications with regard to the screening sequence can be carried out.

A further field of application would be the development of analytical procedures depending on the question being asked. Degradation products of matrix proteins which arise during the destruction of the cartilage could be investigated, in order to use these as characteristic markers for the diagnosis of destructive joint diseases. For this it is necessary to investigate the fundamental mechanisms of the pathogenetically relevant cellular interactions during the destruction process. For these applications the in vitro 3D pannus model is available as a suitable system.

The saving or supplementing of animal experiments is connected with these application examples.

CHAPTER 9: Bibliography

• Adams JC, Watt FM; Regulation of development and differentiation by the extracellular matrix. Development 117: 1183-1198, (1993).
• Alsalameh S, Jahn B, Kalden JR, Burmester GR; Antigenicity and accessory cell function of human articular chondrocytes. J Rheumatol. 18: 414-421, (1991).
• Arnett FC, Edworthy SM, Bloch DA et al.; The american rheumatism association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 31: 315-324, (1988).
• Ayad S, Boot-Handford RP, Humphries MJ, Kadler KE, Shuttleworth CA; In: The extracellular matrix facts Book. Academic Press Harcourt Brace & Company, Publishers, London, pp 11-20, (1994).
• Aydelotte MB, Schumacher BL, Kuettner KE; Heterogeneity of articular chondrocytes. In: Articular cartilage and osteoarthritis. Pp 237-249, Kuettner KE, Schleyerbach R, Peyron JG, Hascall VC (eds), NY, Raven Press, (1992).
• Beggs M, Blok H, Diels A; The high throughput screening infrastructure: The right tools for the task. J Biomol Screen. 4(3): 143-149, (1999).
• Benya PD, Padilla SR, Nimni ME; Independent regulation of collagen types by chondrocytes during the loss of differentiated function in culture. Cell 15: 1313-1321, (1978).
• Benya PD, Schaffer JD; Dedifferantiated chondrocytes reexpress the differentiated collagen phenotype. Cell 30: 215224, (1982).
• Berg M, Undisz K, Thiericke R, Moore T, Posten C; Miniaturization of a functional transcription assay in yeast (human progesterone receptor) in the 384and 1536Well plate format. J Biomol Screen. 5 (2): 71-6, (2000).
• Berg M, Undisz K, Thiericke R, Zimmermann P, Moore T, Posten C; Evaluation of liquid handling conditions in microplates. J Biomol Screen. 6 (1): 47-56, (2001).
• Berry L, Grant ME, McClure J, Rooney P; Bone-marrowderived chondrogenesis in vitro. J Cell Sci 101: 333342, (1992).
• Bizios R.; Mini-Review: osteoblasts: an in vitro model of bone-implant interactions. Biotech Bioeng 43, 58285, (1994).
• Blackburn WD; Management of osteoarthritis and rheumatoid Arthritis: Prospects and possibilities. American Journal of Medicine. 100, 2A-30S, (1996).
• Bleasel JF, Poole AR, Heinegard D, Saxne T, Holderbaum D, Ionescu M; Change in serum cartilage marker levels indicate altered cartilage metabolism in families with the osteoarthritis-related type II collagen gene COL2A1 mutation. Arthritis Rheum 42: 39-45, (1999).
• Bolton MC, Dudhia J, Bayliss MT; Quantification of aggrecan and link-protein mRNA in human articular cartilage of different ages by competitive reverse transcriptase-PCR. Biochem J 319: 489-498, (1996).
• Bombara MP, Webb DL, Conrad P, Marlor CW, Sarr T, Ranges GE, Greve JM, Blue ML; Cell contact between T cells and synovial fibroblasts causes induction of adhesion molecules and cytokines. J Leukoc Biol 54(5): 399-406, (1993).
• Bonaventure J, Kadhom N, Cohen Solal L, Ng KH, Bourguignon J, Lasselin C, Freisinger P; Reexpression of cartilage-specific genes by dedifferentiated human articular chondrcytes cultured in alginate beads. Exp Cell Res 212: 97-104, (1994).
• Borchers AT, Keen CL, Cheema GS, Gershwin ME; The use of methotrexate in rheumatoid arthritis. Semin Arthritis Rheum 34(1): 465-83, (2004).
• Bouhadir KH, Mooney DJ; In vivo and in vitro models for the reconstruction of intercellular signalling. Ann NY Acad Sci. 842: 188-94, (1998).
• Briggs MD, Hoffmann SM, King LM, Olsen AS, Mohrenweiser H, Leroy JG, et al.; Pseudoachondroplasia and multiple epiphyseal dysplasia due to mutations in the cartilage oligometric matrix protein gene. Nat Genet 10: 330336, (1995).
• Buckwalter JA; Building on our strengths. J Orthop Res. 8(6): 17-20, (1990).

• Buckwalter JA, Mow VC, Ratcliff A; Restoration of injured or degenerated articular cartilage. J Am Acad Orthop Surg. 2(4): 192-201, (1994).
• Buckwalter JA, Mankin HJ; Articular Cartilage. Part 1: Tissue design and chondrocyte matrix interactions. J Bone Joint Surg [Am] 79-A: 600-611, (1997).
• Buckwalter JA, Rosenberg LC, Hunziker EB; Articular cartilage: composition, structure, response to injury, and methods of facilitation repair. In: Articular Cartilage and Knee Joint Function. Basic Science and Arthroscopy, Bristol-Myers/Zimmer Orthopaedic Symposium, pp 1956, Ewing JW (ed), Raven Press, NY, (1999).
• Burgeson RE; New collagens, new concepts. Annu Rev Cell Biol 4: 551-577, (1988).
• Burgeson RE, Nimni ME; Collagen types. Molecular structure and tissue distribution. Clin Orthop 282: 250-272, (1992).
• Butnariu-Ephrat M, Robinson D, Mendes DG, Halperin N, Nevo Z; Resurfacing of goat articular cartilage from chondrocytes derived from bone marrow. Clin Orthop 330: 234-243, (1996).
• Caterson B, Flannery CR, Hughes CE & Little, CB; Mechanism involved in cartilage proteoglycan catabolism. Matrix Biol 19: 333-344, (2000).
• CyBio Manual; Research & Technology for Life, (2001-2003).
• Chen CC, Chang JH, Lee JB, Javier J, Azar DT; Human corneal epithelial cell viability and morphology after dilute alcohol exposure. Invest Ophthalmol Vis Sci. 43, 2593602, (2002).
• Cremer MA, Rosloniec EF, Kang AH; The cartilage collagens: a review of their structure, organization, and role in the pathogenesis of experimental arthritis in animals and in human rheumatic disease. J Mol Med. 76: 275-288, (1998).
• Diab M, Wu JJ, Eyre DR; Collagen type IX from human cartilage: a structural profile of intermolecular cross-linking sites. Biochem J 314: 327-332, (1996).
• Elima K, Vuorio E; Expression of mRNA for collagens and other matrix components in dedifferentiating and redifferentiating human chondrocytes in culture. FEBS Lett 258: 195-198, (1989).
• Eyre D, Wu JJ; Collagen structure and cartilage matrix integrity. J Rheumatol. 22: 82-85, (1995).
• Eyre D; Collagen of articular cartilage. Arthritis Res. 4: 3035, (2002).
• Fassbender HG; Histomorphical basis of articular cartilage destruction in rheumatoid arthritis. Coll Relat Res 3: 141155, (1983).
• Feldmann M; What is the mechanism of action of antitumour necrosis factor alpha antibody in rheumatoid arthritis. Int Allergy Immunol. 111(4): 362-5, (1996).
• Foss D.L., Baarsch M.J., Murtaugh M.P.; Regulation of HPRT, GAPDH and b-Actin m-RNA expression in porcine immune cells ans tissues. Anim Biotechnol 9 (1), 67-78, (1998).
• Frangos JA., Hillsley MV.; Review: bone tissue engineering: the role of interstial fluid flow. Biotech Bioeng 43, 57381, (1993).
• Franz JK., Hummel KM, Nawrath M, Aicher WK, Shigeyama Y, Muller-Ladner U, Gay RE, Gay S; Expression of sentrin, a novel antiapoptotic molecule, at sites of synovial invasion in rheumatoid arthritis. Arthritis Rheum. 43: 599607, (2000).
• Freed LE, Marqui JC, Nohria A, Emmanual J, Mikos AG, Langer R; Neocartilage formation in vitro and in vivo using cells cultured on synthetic biodegradable polymers. J Biomed Mater Res 27: 11-23, (1993).
• Frye C., Yocum D., Tuan R. et al.; An in vitro model for studying mechanisms underlying synoviocyte-mediated cartilage invasion in rheumatoid arthritis. Pathol Oncol Res 2 (3), 157-166, (1996).
• Fuchs S; Molekulare Charakterisierung künstlicher humaner Knorpeltransplantate mittels knorpelspezifischer Genexpressionsprofile. Diplomarbeit, Medizinische Klinik mit Schwerpunkt Rheumatologie, Charité Berlin, (2002).
• Garnero P, Rousseau JC, Delmas PD; Molecular basis and clinical use of biochemical markers of bone, cartilage, and synovium in joint diseases. Arthritis and Rheumatism 43(5): 953-968, (2000).
• Gay S; Rheumatoid Arthritis. Curr Opin Rheumatol. 13(3): 191-2, (2001).
• Gehrsitz A, McKenna LA, Söder S, Kirchner T, Aigner T; Isolation of RNA from small human articular cartilage specimens allows quantification of mRNA expression levels in local articular cartilage defects. J Orthop Res 19: 478-481, (2001).

• Glowacki J., Mizuno S.; Chondroinduction of human dermal fibroblasts by demineralized bone in three-dimensional culture. Exp Cell Res 227, 89-97, (1996).
• Goessler UR., Hermann K., Riedel F.; Tissue engineering with chondrocytes and function of the extracellular matrix. Int J Mol Med. 13, 505-13, (2004).
• Grande DA, Southerland SS, Manji R, Pate DW, Schwartz RE, Lucas PA; Repair of articular cartilage defects using mesenchymal stem cells. Tissue Engineering 1: 345353, (1995).
• Gumbiner BM; Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell 84: 345-357, (1996).
• Hall PA, Watt FM; Stem cells: the generation and maintenance of cellular diversity. Development 106(4): 619-633, (1989).
• Handley CJ, Lowther DA, Mcquillan DJ; The structure and synthesis of proteoglycans of articular cartilage. Cell Biol Int Rep. 9 (9): 753-82, (1985).
• Hummer C, Bujia J; Immunologie vitaler und konservierter Transplantate. Eur. Arch. Oto-Rhino-Laryngol. (Suppl) 1: 3-26, (1992).
• Hedbom E, Antonsson P, Hjerpe A, Aeschlimann D, Paulsson M, Rosa-Pimental E; Cartilage matrix proteins: an acidic oligomeric protein (COMP) detected only in cartilage. J Biol Chem 267: 6132-6, (1992).
• Heid CA, Stevens J, Livak KJ, Williams PM; Real time quantitative PCR. Genome Res 6: 986-994, (1996).
• Hyllested J.L., Veje K., Ostergaard K.; Histochemical studies of the extracellular matrix of human articular cartilage – a review. Osteoarthritis Cartilage 10, 333-43, (2002).
• Holtzer H, Abbott J, Lash J, Holtzer S; The loss of phenotypic traits by differentiated cells in vitro: differentiation of cartilage cells. Proc. Natl. Acad. Sci. U.S.A. 46: 15331542, (1960).
• Horton WA; Expression of the human chondrocyte phenotype in vitro. In Vitro Cell Dev. Biol. 25: 659-668, (1989).
• Horwitz AF, Thiery JP; Cell-to-cell contact and extracellular matrix. Curr Opin Cell Biol 6: 645-647, (1994).
• Histologie I-Auxilium Repetitorium. Robert Tachmann Verlag, Berlin pp 76-79, (1991).
• Iozzo RV; Matrix proteoglycans: from molecular design to cellular function. Annu Rev Biochem 67: 609-52, (1998).
• Issack PS, Fang C, Leslie MP, Di Cesare PE; Chondrocytespecific enhancer regions in the COMP gene. J Orthop Res 18: 345-350, (2000).
• Johnstone B, Hering TM, Caplan AI, Goldberg VM, Yoo JU; In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells. Exp Cell Res 238: 265272, (1998).
• Kadiyala S, et al.; Culture expanded canine mesenchymal stem cells posses osteochondrogenic potential in vitro and in vivo. Cell Transplant 6: 125-134, (1997).
• Kastenbauer ER; Konservierung und Anwendungsmöglichkeiten allogener (homologer) Transplantate im HalsNasen-Ohren-bereich. Hals-Nasen-Klinik 31: 371380, (1983).
• Kaul R, Sharma A, Lisse JR, Christadoss P; Human recombinant IL-2 augments immunoglobin and induces rheumatoid factor production by rheumatoid arthritis lymphocytes engrafted into severe combined immunodeficient mice. Clin Immunol Immunophatol 74(3): 271-282, (1995).
• Kornak U. and Mundlos S.; Genetik disorders of the skeleton: a developmental approach. Am J Hum Genet. 73, 44774, (2003).
• Kreuzer K.A., Lass U., Landt O. et al.; Highly sensitive and specific fluorescec reverse transcription-PCR assay for pseudogene-free detection of b-actin transcripts as quantitativ reference. Clinical Chemistry 2, 297300, (1999).
• Kuettner KE; Biochemestry of articular cartilage in health and disease. Clin Biochem. 25 (3): 155-63, (1992).
• Lane Smith R, Trindade MC, Ikenoue T, Mohtai M, Das P, Carter D, Goodman SB, Schurman DJ; Effects of shear stress on articular chondrocyte metabolism. Biorheology 37(1-2): 95-107, (2000).
• Langer R. and Vacanti J.P.; Tissue Engineering. Science 260, 920-926, (1993).
• Lee JW., Kim YH., Kim SH., Han SH., Hahn SB.; Chondrogenic differentiation of mesenchymal stem cells and its clinical applications. Yonsei Med J., 45, 41-7, (2004).
• Levensberg S., Langer R.; Advances in tissue engineering. Curr Top Dev. Biol. 61, 113-34, (2004).

• Liu Tsang V, Bhatia SN; Three-dimensional tissue fabrication. Adv Drug Deliv Rev. 56, 1635-47, (2004).
• Lohmander S; Proteoglycans of joint cartilage. Structure, function, turnover and role as markers of joint disease. Baillieres Clin Rheumatol. 2(1): 37-62, (1988).
• Lohmander SL, Saxne T, Heinegard D; Release of cartilage oligometric matrix protein (COMP) into joint fluid after knee injury and in osteoarthritis. Ann Rheum Dis 53: 813, (1994).
• Lorenzo P; Identifaction and characterization of a novel cartilage gene product CILP, which is an early indicator of osteoarthritis [thesis]. Lund (Sweden): Lund Univ.; (1998).
• Maatta A, Glumoff V, Paakkonen P, Liska D, Pellinen RP, Elima K; Nuclear factor binding to an AP-1 site is associated with the activation of pro-alpha 1(I)-collagen gene in dedifferentiating chondrocytes. Biochem J 294: 365371, (1993).
• Mackay AM, Beck SC, Murphy JM, Barry FP, Chichester CO, Pittinger MF; Chondrogenic differentiation of cultured human mesenchymal stem cells from marrow. Tissue Engineering 4: 415-428, (1998).
• Makisara GL, Makisara P; Prognosis of functional capacity and work capacity in RA. Clin Rheumatol. 1, 117125, (1982).
• Mankin HJ, van Mow C, Buckwalter JA, Lanotti JP, Ratcliff A; Form and function of articular cartilage. In: Simon SR (ed): Orthopaedic Basic Science. Rosemont, AAOS, pp 144, (1994).
• Mansson B, Carey D, Alini M, Ionescu M, Rosenberg LC, Poole AR; Cartilage and bone metabolism in rheumatoid arthritis: differences between rapid and slow progression of disease identified by serum markers of cartilage metabolism. J Clin Invest 95: 1071-7, (1995).
• Medicine-Worldwide; www.medicine-worldwide.de, (2003).
• Melching LI, Cs-Sabo G, Roughley PJ; Analysis of proteoglycan messages in human articular cartilage by a competitive PCR technique. Matrix Biol 16: 1-11, (1997).
• Me'lnikova T.N., Konovalov G.V.; Effect of ethanol on the morphological characteristics of the cerebral cortex in tissue culture. ZH Nevropatol Psikhiatr Im S S Korsakova 89, 49-52, (1989).
• Metz J; Makroskopie, Histologie und Zellbiologie des Gelenkknorpels. In: Erggelet C, Steinwachs M (eds)
• Gelenkknorpeldefekte. Steinkopff Verlag, Darmstadt, 1529, (2001).
• Molecular Biology of the cell. Alberts B, Bray D, Lewis J, Ratt M, Roberts K, Watson JD (eds), Garland Publishing Inc., NY, 3rd ed pp 980, (1994).
• Morgelin M, Heinegard D, Engel J, Paulson M; Electron microscopy of native cartilage oligomeric matrix protein purified from the swarm rat chondrosarcoma reveals a five-armed structure. J Biol Chem 267: 6137-41, (1992).
• Mow VC, Newton PM, Greisamer RP; Biomechanics of articular cartilage and meniscus. In: Fu FH, Harner CD, Vince KG (eds) Knee surgery. Williams & Wilkins, Baltimore, pp 101-130, (1994).
• Muller-Ladner U, Gay RE, Gay S; Activation of synoviocytes. Curr Opin Rheumatol. 12: 186-194, (2000).
• Muller-Ladner U, Kriegsmann J, Franklin BN, Matsumoto S, Geiler T, Gay RE, Gay S; Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice. Am J Pathol. 149: 1607-1615, (1996).
• Muller-Ladner U, Kriegsmann J, Gay RE, Gay S; Oncogenes in rheumatoid arthritis. Rheum Dis Clin North Am. 21(3): 675-90, (1995).
• Nishida Y, Knudson CB, Kuettner KE, Knudson W; Osteogenic protein-1 promotes the synthesis and retention of extracellular matrix within bovine articular cartilage and chondrocyte cultures. Osteoarthritis Cartilage 8: 127136, (2000).
• Oldberg A, Antonsson P, Lindblom K, Heinegard D; COMP (cartilage oligomeric matrix protein) is structurally related to the thrombospondins. J Biol Chem 267: 2234650, (1992).
• Oliva A., Ragione PD., Salerno A., Riccio V., Tartaro G., Cozzolino A., Amato SO., Pontoni G., Zappia V.; Biocompatibility studies on glas ionomer cements by primary cultures of human osteoblasts. Biomaterials 17, 1351-1356, (1996).
• Pap T, van der Laan WH, Auperle KR; Modulation of fibroblast-mediated cartilage degradation by articular chondrocytes in rheumatoid arthritis. Arthritis Rheum 43: 2531-2536, (2000).

• Pincus T, Callahan LF, Vaughn WK; Questionaire, walking time and button test measures of functional capacity as predictive markers for mortality in rheumatoid arthritis. J Rheumatol. 14: 240-251, (1987).
• Pittinger MF, Mackay AM, Beck SC, Jaiswal PK, Douglas R, Mosca JD; Multilineage potential of adult human mesenchymal stem cells. Science 284: 143-147, (1999).
• Prockop DJ; Marrow stromal cells as stem cells for nonhematopoietic tissues. Science 276: 71-74, (1997).
• Puelacher WC, Kim SW, Vacanti JP, Schloo B, Mooney D, Vacanti CA; Tissue-engineered growth of cartilage: the effect of varying the concentration of chondrocytes seeded onto synthetic polymer matrices. Int J Oral Maxillofac Surg. 23(1): 49-53, (1994).
• Raimondi M.T., Boschetti F., Falcone L., Fiore G.B., Remuzzi A., Marinoni E., Marazzi M., Pietrabissa R.; Mechanobiology of engineered cartilage cultured under a quantified fluid-dynamic environment. Biomech Model Mechanobiol. 1, 69-82, (2002).
• Reddi AH.; Symbiosis of biotechnology and biomaterials: applications in tissue engineering of bone and cartilage. J Cell Biochem 56, 192-5, (1994).
• Roughley PJ; Articular cartilage and changes in arthritis: Noncollagenous proteins and proteoglycans in the extracellular matrix of cartilage. Arthritis Res. 3 (6): 342347, (2001).
• Sack U, Kuhn H, Ermann J, Emmrich F; Cartilage destruction in the human/murine SCID mouse arthritis. Arthritis Rheum 36(9): 34 (Abstract), (1993).
• Sakano S, Zhu Y, Sandell LJ; Cartilage-derived retinoic acidsensitive protein and type II collagen expression during fracture healing are potential targets for SOX 9 regulation. J Bone Miner Res 14: 1891-1901, (1999).
• Saldanha V, Grande DA; Extracellular matrix protein gene expression of bovine chondrocytes cultured on resorbable scaffolds. Biomaterials 21: 2427-2431, (2000).
• Schedel J, Gay RE, Kuenzler P, Seemayer C, Simmen B, Michel BA, Gay S; FLICE-inhibitory protein expression in synovial fibroblasts and at sites of cartilage and bone erosion in rheumatoid arthritis. Arthritis Rheum. 46: 1512-1518, (2002).
• Schultz O., Keyszer G., Zacher J. et al.; Development of in vitro model system for destructive joint deseases. Arthritis Rheum 40, 1420-1428, (1997).
• Seemayer CA, Kuchen S, Kuenzler P, Rihoskova V, Rethage J, Aicher WK, Michel BA, Gay RE, Kyburz D, Neidhart M, Gay S; Cartilage destruction mediated by synovial fibroplasts does not depend on proliferation in rheumatoid arthritis. Am J Pathol. 162: 1549-1557, (2003).
• Seidel W., Pierer M., Hantzschel H.; Internist's therapie of rheumatoid Arthitis. Internist (Berl.) 45, 693-705, (2004).
• Seyer J, Klang A.; Connective tissues of the subendothelium. Creager M, Dzag V (ed) Vascular medicine. Little Brown, Boston, 39-68, (1996).
• Sharif M, Srome I, Shepstone L, Kirwan JR, Elson CJ, Heinegard D; Relationship between serum cartilage oligometric matrix protein levels and disease progression in osteoarthritis of the knee joint. Br J rheumatol 34: 30610, (1995).
• Sims CD, Butler PEM, Cao YL, Casanova R, Randolph MA, Black A; Tissue engineered neocartilage using plasma derived polymer substrates and chondrocytes. Plast Reconstr Surg 101: 1580-1585, (1998).
• Sittinger M, Bujia J, Hammer C, Minuth WW, Burmester GR; Engineering of cartilage tissue using bioresorbable polymer carriers in perfusion culture. Biomaterials 15: 451456, (1994b).
• Sittinger M, Bujia J, Rotter N, Reitzel D, Minuth WW, Burmester GR; Tissue engineering and autologous transplant formation: practical approaches with resorbable biomaterials and new cell culture techniques. Biomaterials 17: 237-242, (1996a).
• Sittinger M; In: in vitro Herstellung von vitalem Knorpelgewebe mit Hilfe resorbierender Polymere. Dissertation, Fakultät III Universität Regensburg, (1994a).
• Sittinger M; Tissue engineering: artificial tissue replacement containing vital components. Laryngorhinootologie 74(11): 695-9, (1995).
• Sittinger M, Perka C, Schultz O, Häupl T, Burmester GR; Joint cartilage regeneration by tissue engineering. Z Rheumatol 58: 130-135, (1999).
• Sittinger M, Reitzel D, Dauner M, Hierlemann H, Hammer C, Kastenbauer E, Planck H, Burmester GR, Bujia J; Resorbable polyester in cartilage engineering: Affinity and

• biocompatibility of polymer fiber structures to chondrocytes. J Biomed Mat Res 33: 57-63, (1996b).
• Sittinger M and Risbud MV; Tissue engineering: advances in vitro cartilage generation. Trends in Biotechnology 20(8): 351-357, (2002).
• Smolian H., Thiele S., Kolkenbrock H., Zacher J., Aicher W., Schultz O., Burmester G. R., Sittinger M.; Etablierung eines in vitro Testsystems für die rheumatoide Arthritis zur Testung therapeutisch relevanter Wirkstoffe. ALTEX 18, (2001).
• Sonderstrup G, Cope AP, Patel S et al.; HLA class II transgenic mice: models of the human CD4+ T-cell immune response. Immunol Rev 172: 335-43, (1999).
• Stock UA, Vacanti JP; Tissue Engineering: Current state and prospects. Annu Rev Med 52: 443-451, (2001).
• Stokes DG, Liu G, Dharmavaram R, Hawkins D, Piera-Velazquez S, Jimenez A; Regulation of type-II collagen gene expression during human chondrocyte de-differentiation and recovery of chondrocyte-spezific phenotype in culture involves Sry-type high-mobility-group box (SOX)
• transcription factors. Biocem J 360: 461-470, (2001).
• User Bulletin #2. ABI Prism 7700 Sequence Detection System, Applied Biosystems (2001).
• Taylor PB, Stewart FP, Dunnington DJ, Quinn ST, Schulz CK, Vaidya KS, Kurali E, Lane TR, Xiong WC, Sherrill TP, Snider JS, Terpstra ND, Hertzberg RP; Automated assay optimization with integrated statistics and smart robotics. J Biomol Screen. 4: 213-26, (2000).
• Terada S., Sato M., Sevy A., Vacanti J. P.; Tissue Engineering in the Twenty-First Century. Yonsei Med J. 41, 68591, (2000).
• Thomson JA, et al.; Embryonic stem cell lines derived from human blastocycts. Science 282: 1145-1147, (1998).
• Tillmann B; Skelettsystem; In: Leonhardt H, Tillmann B, Töndury G, Zilles K (eds): Anatomie des Menschen. Stuttgart, Thieme, pp 52-127, (1987).
• Ting V, Sims CD, Brecht LE, McCarthy JG, Kasabian AK, Connelly PR, Elisseeff J, Gittes GK, Longaker MT; In vitro prefabrication of human cartilage shapes using fibrin glue and human chondrocytes. Ann Plast Surg 40: 413421, (1998).
• Toole BP; Proteoglycans and hyaluronam in morphogenesis and differentiation. In Hay, E.D. (Hrsg.), Cell Biology of Extracellular Matrix (305), New York: Plenum Press, (1991).
• Vacanti CA, Kim W, Schloo B, Upton J, Vacanti JP; Joint resurfacing with cartilage grown in situ from cell-polymer structures. Am J Sports Med 22: 485-488, (1994a).
• Vacanti CA, Langer R, Schloo B, Vacanti JP; Synthetic biodegradable polymers seeded with chondrocytes provide a template for new cartilage formation in vivo. Plast Reconstr Surg 88: 753-759, (1991).
• Vacanti CA, Upton J; Tissue-engineered morphogenesis of cartilage and bone by means of cell transplantation using synthetic biodegradable polymer matrices. Clin Plast Surg 21: 445-462, (1994b).
• Velikonja NK, Wozniak G, Malicev E, Knezevic M, Jeras M; Protein synthesis of human articular chondrocytes cultured in vitro for autologous transplantation. Pflügers Arch 442: 169-170, (2001).
• Von der Mark K, Gauss V, von der Mark H, Muller P; Relationship between cell shape and type of collagen synthesised as chondrocytes lose their cartilage phenotype in culture. Nature 267: 531-532, (1977).
• Waldeyer A, Mayet A; In: Anatomie des Menschen I. de Gruyter, Berlin, NY, pp 31-34, (1993).
• Waldmann SD, Spiteri CG, Grynpas MD, Pilliar RM, Kandel RA; Long-term intermittent shear deformation improves the quality of cartilaginous tissue formed in vitro. J Orthop Res. 21, 590-6, (2003).
• Watt F; Effect of seeding density on stability of the differentiated phenotype of pig articular chondrocytes in culture. J. Cell Sci. 89: 373-378, (1988).
• Westhues M; Die antigene Wirkung des Knorpels. I. Nachweis der antigenen Wirkung des transplantierten Knorpels durch histologische Untersuchungen. Laryngol. Rhinol. Otol. 49: 150-161, (1970).
• Winer J, Jung CKS, Shackel I, Williama PM; Development and Validation of real time quantitive reverse transcriptasepolymerase chain reaction for monitoring gene expression in cardiac mycocytes in vitro. Analytical Biochemistry 270: 41-49, (1999).

• Wirth CJ, Rudert M; Techniques of cartilage growth enhancement: a review of the literature. Arthroscopy 12(3): 300-8, (1996).
• Wong M, Hunziker EB; Articular cartilage biology and biomechanics. In: Erggelet C, Steinwachs M (eds)
• Gelenkknorpeldefekte. Steinkopff Verlag, Darmstadt, 1529, (2001).
• Yelin E, Henke C, Epstein W; The work dynamics of the person with rheumatoid arthritis. Arthritis Rheum. 30: 501512, (1987).
• Yelin E, Henke C, Epstein W; The work dynamics of the person with rheumatoid arthritis. Arthritis Rheum. 30: 348357, (1988).
• Zimmermann DR, Ruoslahti E; Multiple domains of the large fibroblast proteoglykan, versican. EMBO J 8: 29752981, (1989).
• Zvaifler NJ, Firestein GS; Pannus und pannocytes: alternative models of joint destruction in rheumatoid arthritis. Arthritis Rheum 37: 783-789, (1994).

CHAPTER 10: Appendix

[Figure not reproduced in the web edition. See page 96 of the PDF.]

Fig. 10.1: Section from the sequence program for the medium exchange with the CyBi®-Disk. Supply of the pellet culture from the 4th day in culture.

[Figure not reproduced in the web edition. See page 97 of the PDF.]

Fig. 10.2: Section from the sequence program for the medium exchange with the CyBi®-Disk. Supply of the pellet culture from the 4th day in culture.

[Figure not reproduced in the web edition. See page 98 of the PDF.]

Fig. 10.3: Section from the sequence program for the washing of the tips with the CyBi®-Disk. Applied before and after every contact with the culture.

Tab. 10.1: The efficiencies of each tested gene.

Tab. 10.2: The expression level in relation to the reference gene GAPDH.

Tab. 10.3: The mean values of the threshold values (CT value) of each tested gene and of the reference gene GAPDH.

Tab. 10.4: Standard deviations of each tested gene and of the reference gene GAPDH.

Fig. 10.4: Scan of a 96-well plate with porcine pellet cultures after 12 days in culture, manually supplied.

Fig. 10.5: Scan of a 96-well plate with porcine pellet cultures after 12 days in culture, automatically supplied.