Annals of Biomedical Engineering, Vol. 33, No. 8, August 2005 (© 2005) pp. 1071–1077 DOI: 10.1007/s10439-005-5775-y
Osmolarity Regulates Gene Expression in Intervertebral Disc Cells Determined by Gene Array and Real-Time Quantitative RT-PCR LAWRENCE M. BOYD,1 WILLIAM J. RICHARDSON,2 JUN CHEN,1 VIRGINIA B. KRAUS,3 ALOK TEWARI,1 and LORI A. SETTON1,2 1 Department of Biomedical Engineering, Duke University, Durham, North Carolina; 2 Division of Orthopaedic Surgery, Department of Surgery, Duke University Medical Center, Durham, North Carolina; and 3 Division of Rheumatology and Immunology, Department of Medicine, Duke University Medical Center, Durham, North Carolina 27708
(Received 22 November 2004; accepted 16 March 2005)
pH, and intracellular and extracellular osmolarities.36,37 In particular, changes in osmolarity are an important component of the physicochemical environment of the IVD as variations in disc loading lead to significant changes in disc hydration.36 As for other cell types, isolated cells of the IVD are known to exhibit both passive and active cell volume changes following exposure to hyper- or hypo-osmotic media in vitro.21,32,33 Osmotic stimuli have also been shown to elicit calcium transients in IVD cells that are modulated, in part, by stability of the actin cytoskeleton.32,33 These identified changes in cell volume and second messengers following altered osmotic stimuli can be expected to impact metabolic events in the cells of the intervertebral disc. Previous studies have shown that osmotic pressure is a potent regulator of biosynthesis for IVD cells in culture. Fibrochondrocytes of the disc have been shown to alter post-translational biosynthesis of proteoglycans (i.e., 35 S-incorporation) following exposure to hyper- or hypoosmotic media in both explant and isolated cell culture.5,6,21 These studies demonstrate that maximal proteoglycan synthesis is achieved at osmolarities believed to represent in situ values (−430 mOsm), with decreased proteoglycan synthesis found at osmolarities above or below these values. A recent study of IVD cells in a three-dimensional alginate culture system confirmed that the biological response to altered osmolarity is mediated, in part, by changes at the transcriptional level.10 Short periods of culture in hypo-osmotic media (4 h) resulted in increased gene expression for aggrecan and type II collagen in porcine IVD cells, whereas gene expression for the small proteoglycans, decorin and biglycan, increased under both hypo- and hyper-osmotic conditions. The responses of these cells to hyper-osmotic stimuli is considered to be of particular interest, as mechanical loading of the disc in compression will generate increased proteoglycan-associated charge densities, that are associated with increased osmotic pressures within the tissue. A wide range of genes may be transcriptionally activated in multiple cell types following exposure to
Abstract—Intervertebral disc (IVD) cells experience a broad range of physicochemical stimuli under physiologic conditions, including alterations in their osmotic environment. Cellular responses to altered osmolarity have been documented at the transcriptional and post-translational level, but mainly for extracellular matrix proteins. In this study, the gene expression profile of human IVD cells was quantified with gene array technology following exposure to increased osmolarity in order to capture the biological responses for a broad set of targets. A total of 42 genes were identified in IVD cells as significantly changed following culture under hyper-osmotic conditions. Gene expression patterns were verified using RT-PCR. Genes identified in this study include those related to cytoskeleton remodeling and stabilization (ephrinB2, muskelin), as well as membrane transport (ion transporter SLC21A12, osmolyte transporter SLC5A3, monocarboxylic acid SLC16A6). An unexpected finding was the differential regulation of the gene for the neurotrophin, brain-derived neurotrophic factor, by hyper-osmotic stimuli that suggests a capability of IVD cells to respond to physicochemical stimuli with factors that may regulate discogenic pain.
Keywords—Osmotic pressure, Osmolarity, Intervertebral disc, Cell culture, Gene array, Microarray, Gene expression, Hyperosmolarity.
INTRODUCTION The intervertebral discs contain a sparse population of cells distributed in a large volume of extracellular matrix composed primarily of water (60–99% by weight), collagens and negatively charged proteoglycans.31 Loading of the intervertebral disc (IVD) under static or dynamic conditions gives rise to complex physicochemical stimuli, including interstitial hydrostatic pressures, matrix stresses and strains, as well as associated changes in tissue hydration that may modify extracellular cation concentrations,
Address correspondence to L.A. Setton, PhD, Department of Biomedical Engineering, Duke University, Room 136, Hudson Hall, Box 90281, Durham, North Carolina 27708. Electronic mail:
[email protected]
1071 C 2005 Biomedical Engineering Society 0090-6964/05/0800-1071/1
1072
BOYD et al.
hypo- or hyper-osmotic conditions.23,29,39 The objective of this study was to examine the transcriptional response of IVD cells following culture under hyper-osmotic conditions. The gene expression profile of human IVD cells was quantified with high-density oligonucleotide gene array technology (GeneChip, Affymetrix, Santa Clara, CA) following exposure to hyper-osmolarity in a three-dimensional matrix. Cells were embedded in alginate gel beads, a welldefined system for characterizing disc cell phenotype and environmental conditions.11,25,40 Results of the array were verified via quantitative RT-PCR. The results of the study demonstrate that genes encoding proteins related to cytoskeletal organization, osmolyte and ion transport, cytokines and growth factors, are regulated by osmotic conditions in cells of the IVD. MATERIALS AND METHODS Cell Isolation and Culture Human IVD tissue was obtained from patients (average age 51 years) undergoing discectomy prior to surgery for lumbar interbody fusion (n = 3) or lumbar disc herniation (n = 1). All tissue was harvested from the central nucleus pulposus region and was of a uniformly fibrous appearance. Any tissues containing endplate, bone or cartilage were discarded prior to cell isolation. Cells were isolated by sequential pronase–collagenase digestion.2 Cells were passaged twice in monolayer and suspended at a density of 2 × 106 cells/ml in 1.2% alginate (low viscosity, Sigma Chemical, St. Louis, MO) dissolved in 150 mM NaCl. Multiple alginate beads were formed by drop-wise addition of the alginate from a 22 gauge needle into 102 mM CaCl2 , followed by 10 min of curing, as described previously.11,25 Cell-gel beads were incubated in cell culture media consisting of Ham’s F-12 medium (Gibco BRL, Grand Island, NY), supplemented with 10% FBS (HyClone, Logan, UT), 25 µg/ml ascorbic acid (Sigma, St. Louis, MO), 100 U/ml penicillin, 100 µg/ml streptomycin, and 1 µg/ml Fungizone at 5% CO2 and 37◦ C. After 24 h, the cell culture medium was removed via pipette and exchanged for either iso-osmotic or hyperosmotic media.10,21,32 The iso-osmotic media consisted of a defined cell culture medium (Ham’s F-12 with supplements as described above; 293 mOsm/kg H2 O). The hyper-osmotic media consisted of the same cell culture media supplemented with sucrose to a final osmolarity of 450 mOsm/kg H2 0. Previous studies using both NaCl and sucrose to alter osmolarity have demonstrated that the biological response to changes in osmolarity are not mediated by ionic concentration changes, but rather depend directly on extracellular osmotic pressure alone.21 The osmolarity of all solutions was determined using a freezing-point osmometer (Advanced Laboratory Wide Range 3W2, Advanced Instrument, Needham Heights, MA) as described
previously.10 Cell-alginate beads were cultured for a 4 h period under these two conditions, after which the cells were released from alginate in a dissolving buffer (55 mM Na-citrate and 150 mM NaCl), lysed and stored at −80◦ C. Gene Array Experiments The Human Genome U133A array (Affymetrix, Santa Clara, CA) was used to study the expression of 22,283 gene sequences, transcript variants and expressed sequence tags. Total RNA was isolated from each cell sample with the RNeasy mini kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. 10 µg of total RNA was used for synthesis of cDNA via reverse-transcription using SuperScriptTM reverse transcriptase (Gibco BRL, Rockville, MD) and a T7 promotor (Genset, San Diego, CA). Biotin-labelling of the cRNA target was carried out by in vitro transcription using the BioArray High Yield RNA Transcription Labeling Kit (Enzo Diagnostics, Farmingdale, NY). The cRNA was fragmented and hybridized to the U133A array. After hybridization, arrays were washed, stained with streptavidin phycoerythrin and scanned to obtain images for analysis with proprietary software (Microarray Suite v5.0, Affymetrix). Images were analyzed for segmentation, background correction, scaling for arrayto-array comparison, as well as statistical analyses to determine the presence/absence of a transcript or a differential expression profile following protocols established by the manufacturer (Statistical Algorithms Reference Guide, Affymetrix Part No. 701110). Each transcript is represented by 10–20 sets of oligonucleotide probes, each set containing perfect match (PM) and mismatch (MM) probe cells, together representing a probe pair for each transcript. The statistical algorithm (One-sided Wilcoxon’s signed rank test) used probe pair intensities to generate a detection p-value to make a detection call that the transcript for a particular target was present, marginal or absent from the hybridization profile for the complete set of probes. The complete datasets, in MIAME format, are available under GEO accession numbers as series GSE1648 at the following link www.ncbi.nlm.nih.gov/geo. The iso-osmotic condition for each patient sample (H1 to H4) was used as a control for assessment of genes differentially expressed under hyper- osmotic (n = 4 paired comparisons) conditions. All arrays for a given patient were processed at the same time in order to eliminate any potential for batch/day sources of variability (due to unintended day-to-day processing variability).17 The data for each sample represents the results of a single hybridization run, as prior studies have found chip-to-chip variability with Affymetrix arrays to be very small (R 2 = 0.981).3 All array data were normalized by scaling to a target signal of 500 (Statistical Algorithms Description Document, Affymetrix, 2002). Pair-wise comparisons were made between control and experimental arrays for each target gene
Osmolarity Regulates Gene Expression in Intervertebral Disc Cells
based on intensity information from all oligo sets. First, differences in the values for hybridization to the perfect match and mismatch probe pairs (PM-MM) were compared between the baseline array (iso-osmotic) and experimental array (hyper-osmotic) using a Wilcoxon’s signed-rank test. In order to focus on the most significantly regulated genes, a p-value for the Wilcoxon’s signed rank test of 1.3-fold change) and found 16 of 17 (94%) were correctly identified by the MicroArray Suite 4.0 software (Affymetrix). Our findings of an 80% correlation between array and RT-PCR are consistent with these prior studies. As mentioned, our prior work using RT-PCR to study gene expression in porcine disc chondrocytes under osmotic loading found that the small proteoglycans, decorin and biglycan, increased under both hypo- and hyper-osmotic conditions. In the current study, gene expression levels for these small proteoglycans were not found to be significantly altered in human IVD cells. This may reflect differences in the behavior of the degenerate IVD cells compared to the younger porcine cells or it may relate to the reduced sensitivity of the gene array to detect smaller fold differences compared to RT-PCR, as discussed above. In summary, changes in the osmotic environment of IVD cells was shown to modify gene expression for proteins involved in cell cycle regulation, cell–cell interactions and adhesion, signal transduction mechanisms, growth factors and cytokines, and proteases. Although the majority of the genes regulated by osmotic stimuli have poorly understood functions in the intervertebral disc, a significant number of genes identified in this study were related to cytoskeleton remodeling and stabilization, as well as osmolyte transport.
BOYD et al.
Additional insight into the importance of gene expression changes in IVD cell biology regulated by osmotic stimuli would require studies of multiple timepoints. With this additional information, techniques such as cluster analysis, principle components analysis and other data mining techniques would provide value in determining co-regulatory pathways and kinetics of gene regulation. This work was beyond the scope of the current study. One finding of particular interest related to the differential regulation of the gene for the neurotrophin, brainderived neurotrophic factor, in IVD cells cultured under hyper-osmotic conditions. The expression of RNA for this protein in disc cells is novel, and may further be relevant to modulation of pain subsequent to altered loading. In this manner, the gene array technology has been useful for identifying otherwise unanticipated responses of IVD cells to osmotic stimuli and has yielded potentially novel insights into physical regulation of transcriptional behaviors. ACKNOWLEDGMENTS We acknowledge Dr Holly Dressman of the Duke Gene Array Core Facility for helpful discussions and technical assistance. This study was supported with funds from the NIH (R01AR47442, T32GM08555, R29AG15108) and NSF (BES 9703299). REFERENCES 1
Apfel, S. C. Neurotrophic factors and pain. Clin. J. Pain 16:S7– S11, 2000. 2 Baer, A. E., J. Y. Wang, V. B. Kraus, and L. A. Setton. Collagen gene expression and mechanical properties of intervertebral disc cell-alginate cultures. J. Orthop. Res. 19:2–10, 2001. 3 Bakay, M., Y.-W. Chen, R. Borup, P. Zhao, K. Nagaraju, and E. P. Hoffman. Sources of variability and effect of experimental approach on expression profiling data interpretation. BMC Bioinform. 3:2002. 4 Battaglia, A. A., K. Sehayek, J. Grist, S. B. McMahon, and I. Gavazzi. Eph-b receptors and ephrin-b ligands regulate spinal sensory connectivity and modulate pain processing. Nat. Neurosci. 6:339–340, 2003. 5 Bayliss, M. T., B. Johnstone, and J. P. O’Brien. Proteoglycan synthesis in the human intervertebral disc. Variation with age, region and pathology. Spine 13:972–981, 1988. 6 Bayliss, M. T., J. P. Urban, B. Johnstone, and S. Holm. In vitro method for measuring synthesis rates in intervertebral disc. J. Orthop. Res. 4:10–17, 1986. 7 Bennett, D. L. H. Neurotrophic factors: Important regulators of nociceptive function. Neurosci. Update 7:13–17, 2001. 8 Bush, P. G., and A. C. Hall. The osmotic sensitivity of isolated and in situ bovine articular chondrocytes. J. Orthop. Res. 19:768–778, 2001. 9 Caterson, B., C. R. Flannery, C. E. Hughes, and C. B. Little. Mechanisms involved in cartilage proteoglycan catabolism. Matrix Biol. 19:333–244, 2000. 10 Chen, J., A. E. Baer, P. Y. Paik, W. Yan, and L. A. Setton. Matrix protein gene expression in intervertebral disc cells subjected to altered osmolarity. Biochem. Biophys. Res. Commun. 293:932– 938, 2002.
Osmolarity Regulates Gene Expression in Intervertebral Disc Cells 11
Chiba, K., G. B. J. Andersson, K. Masuda, and E. J.-M. A. Thonar. Metabolism of the extracellular matrix formed by intervertebral disc cells cultured in alginate. Spine 22:2885–2893, 1997. 12 Cutforth, T., and C. J. Harrison. Ephs and ephrins close ranks. Trends Neurosci. 25:332–334, 2002. 13 Dodelet, V. C., and E. B. Pasquale. Eph receptors and ephrin ligands: Embryogenesis to tumorigenesis. Oncogene 19:5614– 5619, 2000. 14 Erickson, G. R., L. G. Alexopoulos, and F. Guilak. Hyperosmotic stress induces volume change and calcium transients in chondrocytes by transmembrane, phospholipid, and G-protein pathways. J. Biomech. 34:1527–1535, 2001. 15 Gradient, R. A., and U. H. Otten. Interleukin-6 (IL-6) - a molecule with both beneficial and destructive potentials. Prog. Neurobiol. 52:379–390, 1997. 16 Ha, S. O., J. K. Kim, H. S. Hong, D. S. Kim, and H. J. Cho. Expression of brain-derived neurotrophic factor in rat dorsal root ganglia, spinal cord and gracile nuclei in experimental models of neuropathic pain. Neuroscience 107:301–309, 2001. 17 Han, E.-S., Y. Wu, R. McCarter, J. F. Nelson, A. Richardson, and S. G. Hilsenbeck. Reproducibility, sources of variability, pooling, and sample size: Important considerations for the design of high-density oligonucleotide array experiments. J. Gerontol. Biol. Sci. 59A:306–315, 2004. 18 Hirano, T., and T. Kishimoto. “Interleukin-6.” In: Peptide Growth Factors and Their Receptors, edited by Sporn, M. B. and A. B. Roberts. New York: Springer-Verlag, 1991, pp. 633– 665. 19 Holder, N., and R. Klein. Eph receptors and ephrins: Effectors of morphogenesis. Development 126:2033–2044, 1999. 20 Hopewell, B., and J. P. G. Urban. Adaptation of articular chondrocytes to changes in osmolality. Biorheology 20:73–77, 2003. 21 Ishihara, H., K. Warensjo, S. Roberts and J. P. G. Urban. Proteoglycan synthesis in the intervertebral disk nucleus: The role of extracellular osmolality. Am. J. Physiol. 272:C1499–C1506, 1997. 22 Kuno, K., N. Kanada, E. Nakashima, F. Fujiki, F. Ichimura, and K. Matsushima. Molecular cloning of a gene encoding a new type of metalloproteinase-disintegrin family protein with thrombospondin motiffs as an inflammation associated gene. J. Biol. Chem. 272:556–562, 1997. 23 Lang, F., G. L. Busch, M. Ritter, H. Volkl, S. Waldegger, E. Gulbins, and D. Haussinger. Functional significance of cell volume regulatory mechanisms. Physiol. Rev. 78:247–306, 1998. 24 Livak, K. J., and T. D. Schmittgen. Analysis of relative gene Ct expression data using real-time quantitative PCR and the 2− method. Methods 25:402–408, 2001. 25 Maldonado, B. A., and J. Theodore R. Oegema. Initial characterization of the metabolism of intervertebral disc cells encapsulated in microspheres. J. Orthop. Res. 10:677–690, 1992. 26 Mannion, R. J., M. Costigan, I. Decosterd, F. Amaya, Q. P. Ma, J. C. Holstege, R. R. Ji, A. Acheson, R. M. Lindsay,
1077
A. A. Wilkinson, and C. J. Woolf. Neurotrophins: Peripherirally and centrally acting modulators of tactile stimulus-induced inflammatory pain hypersensitivity. Proc. Natl. Acad. Sci. U.S.A. 96:9385–9390, 1999. 27 Miletic, G., and V. Miletic. Increase in the concentration of brain derived neurotrophic factor in the lumbar spinal dorsal horn are associated with pain behavior following chronic constriction injury in rats. Neurosci. Lett. 319:137–140, 2002. 28 Millan, M. J. The induction of pain: An integrative review. Prog. Neurobiol. 57:1–164, 1999. 29 O’Neill, W. C. Physiological significance of volume-regulatory transporters. Am. J. Physiol. Cell Physiol. 45:C995–C1011, 1999. 30 Obata, K., H. Tsujino, H. Yamanaka, D. Yi, T. Fukuoka, N. Hashimoto, K. Yonenobu, H. Yoshikawa, and K. Noguchi. Expression of neurotrophic factors in the dorsal root ganglion in a rat model of lumbar disc herniation. Pain 99:121–132, 2002. 31 Oegema, T. R. Biochemistry of the intervertebral disc. Clin. Sports Med. 12:419–439, 1993. 32 Pritchard, S., G. R. Erickson, and F. Guilak. Hyperosmotically induced volume change and calcium signalling in intervertebral disc cells: The role of the actin cytoskeleton. Biophys. J. 83:2502–2510, 2002. 33 Pritchard, S., and F. Guilak. The role of F-actin in hypoosmotically induced cell volume change and calcium signaling in anulus fibrosus cells. Ann. Biomed. Eng. 32:103–111, 2004. 34 Rajeevan, M. S., S. D. Vernon, N. Taysavang, and E. R. Unger. Validation of array-based gene expression profiles by real-time (kinetic) RT-PCR. J. Mol. Diagnostics 3:26–31, 2001. 35 Specchia, N., A. Pagnotta, A. Toesca, and F. Greco. Cytokines and growth factors in the protruded intervertebral disc of the lumbar spine. Eur. Spine J. 11:145–151, 2002. 36 Urban, J. P. G. “The effect of physical factors on disk cell metabolism.” In: Musculoskeletal Soft-Tissue Aging: Impact on Mobility, edited by Buckwalter, J. A., Goldberg, V. M., and Woo, S. L.-Y. Rosemont. Illinois: AAOS, 1993, pp. 391– 412. 37 Urban, J. P. G. The role of the physicochemical environment in determining disc cell behavior. Biochem. Soc. Trans. 30:858– 864, 2002. 38 Urban, J. P. G., S. Roberts, and J. R. Ralphs. The nucleus of the intervertebral disc from development to degeneration. Am. Zool. 40:53–61, 2000. 39 Waldegger, S., and F. Lang. Cell volume and gene expression. J. Membr. Biol. 162:95–100, 1998. 40 Wang, J. Y., A. E. Baer, V. B. Kraus, and L. A. Setton. Intervertebral disc cells exhibit differences in gene expression in alginate and monolayer culture. Spine 26:1747–1751, 2001. 41 Yuen, T., E. Wurmbach, R. L. Pfeffer, B. J. Ebersole, and S. C. Sealfon. Accuracy and calibration of commercial oligonucleotide and custom cdna microarrays. Nucl. Acids Res. 30:e48, 2002.
