ABSTRACT

Articular cartilage is a mechanosensitive tissue containing chondrocytes that respond to biomechanical signals by upregulating synthetic activity and/or production of inammatory mediators.1-3 This process, termed mechanotransduction, refers to the way in which chondrocytes respond to mechanical loading and convert biomechanical signals into a cellular response. The mechanotransduction process, in turn, inuences the composition and structure of the extracellular matrix, enabling chondrocytes to adapt to their physical environment. Over nearly three decades, researchers of cartilage mechanobiology have identied several key biochemical pathways that are involved in the signal transduction process.4 However, the identity of the mechanosensors and the way in which they contribute to changes in gene expression are complicated by the types of in vitro models and bioreactor systems used to study the cell signaling process.5-6 It is now widely accepted that anabolic and catabolic transcriptional activities in chondrocytes are tightly regulated by distinct mechanical signals.7-8 The general consensus, derived from previous in vitro mechanical loading studies, is that static compression which mimics excessive loading or an injurious response, inhibits some gene expression and biosynthesis of matrix proteins.9-12 In contrast, dynamic compression, which could be interpreted as a physiological environment of cartilage, causes an increase in expression of extracellular matrix components.13-15 Several research groups, including our own, have shown a strong interplay between the signal transduction pathways induced by both mechanical loading and interleU.K.in-1β (IL-1β) in chondrocytes seeded

19.1 Introduction .......................................................................................................................... 271 19.2 Methods ................................................................................................................................ 272

19.2.1 Chondrocyte/Agarose Culture and Experimental Conditions .................................. 272 19.2.2 Optimization of Quantitative Real-Time PCR Assays ............................................. 273 19.2.3 Quantitative Real-Time PCR Assays ........................................................................ 273 19.2.4 Data Analysis and Normalization ............................................................................. 274

19.3 Results ................................................................................................................................... 275 19.3.1 Optimization and Validation of qPCR Conditions ................................................... 275 19.3.2 Selection of Reference Genes ................................................................................... 275 19.3.3 Data Normalization and Statistical Analysis ............................................................ 277

19.4 Concluding Remarks ............................................................................................................ 277 Acknowledgments .......................................................................................................................... 277 References ......................................................................................................................................280

in 3D agarose gels and in monolayer culture.4,16-20 There is recent evidence that demonstrates the involvement of the mitogen-activated protein kinase (MAPK) and nuclear factor kappa B (NFκB) pathways in mediating the transcriptional response of chondrocytes to mechanical loading and/or IL-1β.21-23 Nevertheless, the molecular mechanisms underlying the specic mechanotransduction pathways are complex and will vary depending on the pathological environment of the tissue.1,7-8 Furthermore, the nature of the mechanical stimulus will additionally determine how mechanical loading controls the activation or inhibition of anabolic-and catabolic-associated genes in chondrocytes.24 For instance, frequent bursts of intermittent compression for longer time periods favored expression of matrix synthesis whereas shorter bursts of intermittent compression had the opposite effect. A similar response was described by other research groups who showed that gene expression associated with chondrocyte anabolic and catabolic activities were dependent on the duration and type of compression regime employed.11,15 Ultimately, elucidation of the intracellular pathways in response to physiologically relevant mechanical loading conditions will therefore facilitate the successful identication of mechanotherapeutic agents to treat degenerative joint disorders like osteoarthritis. These types of therapies are termed chondroprotective agents because they are focused on preventing the loss of extracellular matrix and/or regeneration of damaged tissue. However, identication of the relevant signaling events has proved difcult due to cross talk with other signaling pathways and in some instances led to contradictory results. The differences in gene-expression data are in part due to the wide range of in vitro models and bioreactors available as tools to explore mechanotransduction pathways. For instance, it is relatively easy to extract high yields of total RNA from isolated cells cultured in monolayer when compared to 3D models such as agarose. The latter is complicated by the low cell density and presence of large amounts of agarose, leading to inadequate RNA yields and less robust PCR assay performance. However, one of the key challenges encountered in cartilage mechanobiology research is the analysis of real-time quantitative PCR (qPCR) data for monitoring mRNA gene-expression proles of chondrocyte anabolic-and catabolic-associated genes. The results can be signicantly different for studies with multiple experimental test conditions involving mechanical loading, chemical and time-dependent variables since the data are dependent on the appropriate selection of normalization strategy. Thus, inadequate data normalization and errors in statistical analysis makes it difcult for the researcher to obtain meaningful results and compare or even repeat experiments from previously published studies. Additional drawbacks include poor optimization and validation of conditions for qPCR and unreliable choice of reference genes for each experimental test condition. This chapter will therefore contain a discussion of the steps required to analyze qPCR data for mechanotransduction studies utilizing a well-characterized in vitro model and bioreactor system; namely the chondrocyte/agarose model in conjunction with a compressive strain bioreactor. The time-dependent effect of IL-1β and mechancal loading on the relative expression of well-known chondrocyte anabolic (aggrecan, collagen type II) and catabolic (iNOS, COX-2) genes will be presented using a comparative quantication cycle (Cq) PCR approach, as described by others.25 In addition, data normalization and validation will be discussed using relevant statistical methods.