Profiling microRNA expression in bovine articular cartilage and implications for mechanotransduction.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 8, August 2009, pp 2333–2339 DOI 10.1002/art.24678 © 2009, American College of Rheumatology Profiling MicroRNA Expression in Bovine Articular Cartilage and Implications for Mechanotransduction Walter Dunn, Grayson DuRaine, and A. Hari Reddi Objective. Articular cartilage is an avascular tissue with precise polarity and organization comprising 3 distinct functional zones: the surface, middle, and deep zones. Each zone has a different gene expression pattern that plays a specific role in articular cartilage development and maintenance. MicroRNA (miRNA) are small noncoding gene products that play an important regulatory role in determining cell differentiation and function. The purpose of this study was to test our hypothesis that miRNA expression profiles in the different articular cartilage zones as well as between regions subjected to different levels of weight-bearing stresses are unique. Methods. Using an miRNA microarray approach in conjunction with quantitative reverse transcription– polymerase chain reaction, we identified miRNA in bovine articular cartilage that were differentially expressed in the different functional zones and in the anterior weight-bearing and posterior non–weightbearing regions of the medial femoral condyle (M1 and M4, respectively). Results. We identified miRNA-221 and miR-222 as part of a subset of differentially expressed miRNA that were up-regulated in articular cartilage in the anterior, M1, greater weight-bearing location. Additionally, miR-126, miR-145, and miR-335 were down- regulated in monolayers of tissue-cultured chondrocytes as compared with levels determined directly from intact native cartilage. Conclusion. In conclusion, miR-222 expression patterns in articular cartilage are higher in the weightbearing anterior medial condyle as compared with the posterior non–weight-bearing medial condyle. Thus, miR-222 might be a potential regulator of an articular cartilage mechanotransduction pathway. These data implicate miRNA in the maintenance of articular cartilage homeostasis and are therefore targets for articular cartilage tissue engineering and regenerative medicine. MicroRNA (miRNA) are a class of ⬃22-nucleotide noncoding RNAs that play an important role in modulating gene expression during embryonic development and cell differentiation (1). These RNAs are encoded within the genome as part of a predicted RNA hairpin precursor. Generation of miRNA begins with a long RNA transcript, termed primary miRNA, which is initially processed into an ⬃60-nucleotide hairpin RNA precursor (pre-miRNA) by the nuclear ribonuclease Drosha (1). The resultant pre-miRNA is then processed by Dicer to generate a ⬃22-bp duplex. Subsequently, one strand of the duplex is incorporated into the RNAinduced silencing complex (RISC) of the RNA interference (RNAi) pathway and guides the RISC to the target transcript that is complementary to the miRNA sequence (1). RISC binding results in either cleavage or translation inhibition of the messenger RNA, depending on the degree of complementarity between the miRNA and the target transcript (1). Although the function of most miRNA have yet to be determined, the importance of miRNA-mediated gene regulation in humans is underscored by the fact that miRNA have been found in every human tissue type examined thus far. Interestingly, subsets of these miRNA are expressed in a tissue-specific manner and Supported in part by NIH grant 1R01-AR-47345 and by the Lawrence J. Ellison Endowed Chair in Musculoskeletal Molecular Biology held by Dr. Reddi. Dr. Dunn’s work was supported by a Medical Student Research Fellowship from the University of California, Davis School of Medicine. Walter Dunn, PhD, Grayson DuRaine, BS, A. Hari Reddi, PhD: University of California, Davis at Sacramento. Address correspondence and reprint requests to A. Hari Reddi, PhD, Lawrence Ellison Center for Tissue Regeneration and Repair, Department of Orthopaedic Surgery, University of California, Davis, 4635 Second Avenue, Research Building I, Room 2000, Sacramento, CA 95817. E-mail: email@example.com. Submitted for publication December 4, 2008; accepted in revised form April 15, 2009. 2333 2334 sometimes only during certain developmental stages (1–3). The large number of miRNA genes found in the human and animal genomes, their unique distribution and expression patterns, along with the findings of functional studies in Caenorhabditis elegans and Drosophila melanogaster, suggest that miRNA represent a general mechanism for the regulation of gene expression in specific cell types and tissues (4). An intriguing observation regarding miRNA expression and function is that their function depends on the particular tissue type in which they are found and the cellular environment in which they are expressed. For example, although a specific miRNA species may be expressed similarly in 2 differing tissue types, their downstream function may be radically different. In one tissue type, the miRNA may target a cellular replication inhibitor, and thus, the expression of the miRNA may lead to up-regulation of cellular replication. In another tissue type, an miRNA may target an inducer of cell proliferation, and thus, the expression of the miRNA may suppress that inducer, resulting in inhibition of cell proliferation. Thus, the role of miRNA is not only a function of their own expression patterns, but also those of their potential targets and other associated factors (1,4). The profiling of miRNA expression patterns continues to be an active area of research because there are a multitude of tissue types and developmental stages yet to be examined. Additionally, as more novel miRNA sequences are identified, previously screened samples need to be reexamined for the presence or absence of the new miRNA. The investigation of miRNA expression in articular cartilage may provide opportunities for the identification of targets in development and in diseases such as osteoarthritis. Articular cartilage in diarthrodial joints provides a low-friction interface for skeletal movement during locomotion. This unique tissue, although avascular and recalcitrant to regenerative capacity, has the unique capacity to resist significant mechanical loading stresses during the lifetime of the organism. The surface, middle, and deep zones within the articular cartilage are distinct domains, and they exhibit differential gene expression patterns and attendant functional roles. The tissue architecture and secretion of superficial zone protein is unique to the surface of articular cartilage and functions as a lubricant in friction-free movement (5). To date, there have been no published reports on the miRNA expression patterns in articular cartilage. Using a bovine stifle joint articular cartilage model (5), we examined the miRNA expression patterns DUNN ET AL across the regions of the femoral condyle in articular cartilage with precise geometry and differing mechanical loading conditions. Our results demonstrated a differential pattern of miRNA expression, suggesting that some miRNA may be up-regulated in response to mechanical stress. Furthermore, we examined the miRNA expression profiles of the different zones of articular cartilage, interrogating the microRNA expression patterns in the superficial zone versus the middle zone of the articular cartilage. MATERIALS AND METHODS Cartilage procurement and RNA extraction. Cartilage procurement was performed as described previously (5). Briefly, stifle joints from 2–3-month-old calves were harvested within 18 hours of slaughter. A 5-mm coring reamer (Fisher Scientific, Fair Lawn, NJ) was used to extract cores of articular cartilage down to the depth of the subchondral bone. The medial condyle was divided into 4 regions, designated M1 through M4, with M1 representing the most anterior (weightbearing) region and with M4 representing the most posterior (non–weight-bearing) region. Plugs of articular cartilage were taken from each region and further sectioned into superficial, middle, and deep zones. Each plug was placed in a custom cutting jig and sectioned into a superficial zone section measuring 0.5 mm in thickness and a middle zone measuring 1.25 mm in thickness, with the remainder representing the deep zone. Cartilage sections were then frozen in liquid nitrogen and homogenized with a custom tissue homogenizer. A modified RNA extraction procedure was used to purify total RNA from the cartilage sections, as previously described (5). After resuspension and centrifugation of homogenized cartilage in TRIzol reagent (Invitrogen, Carlsbad, CA), the supernatant was removed, and total RNA was extracted using a mirVana miRNA isolation kit (Ambion, Austin, TX) according to the manufacturer’s total RNA extraction protocol. RNA species ⬍200 bp were preserved with this protocol. RNA samples from 7 different animals were pooled, the purity and quantity of the isolated total RNAs were assessed using a spectrophotometer measurement, and then the isolated total RNAs were submitted to LC Sciences (Houston, TX) for miRNA microarray hybridization and analysis. MicroRNA microarray analysis. Each customized microarray chip contained a probe set based on the miRBase version 10.1 release of human/mouse/rat and bovine sequences (online at http://microrna.sanger.ac.uk). The probe sets contained 1,193 unique miRNA sequences that were deposited in triplicate on the array. The array also included perfectly matched and single-basepair–mismatched control probes for a 20-mer RNA oligo spiked into the RNA samples before labeling. The microarray experiments analyzing the M1 superficial zone versus the M1 middle zone and the M1 superficial zone versus the M4 superficial zone consisted of 2-color competition hybridization with a pair of samples labeled with either Cy3 or Cy5. Subtraction of background values and normalization were then performed. For each dual-sample MicroRNA EXPRESSION PROFILES IN BOVINE ARTICULAR CARTILAGE 2335 Figure 1. Regions and zones of articular cartilage from the medial femoral condyle. A, Locations on the articular surface of the bovine femoral condyle where the plugs of articular cartilage were extracted. The medial femoral condyle was divided into 4 regions, with M1 representing the most anterior (weightbearing) region, which experiences the highest contact stresses across all the regions of the joint surface (5), and M4 representing the most posterior (non–weight-bearing) region, which experiences significantly less contact stress than the M1 site (5). B, Sectioning of the articular cartilage plugs into superficial, middle, and deep zones. experiment, a P value calculation was performed, and a list of miRNA candidates was generated based on their differential expression between the 2 samples. Quantitative reverse transcription–polymerase chain reaction (RT-PCR). Total RNA was isolated as described for the microarray hybridization. RNA used for the RT-PCR samples was pooled from 8 different animals collected on 2 separate harvest dates and did not include any RNA used for the microarray hybridization. MicroRNA levels were measured in the M1 superficial, the M1 middle, and the M4 superficial cartilage zones and in chondrocytes from tissue culture monolayers. Real-time quantitative PCR was performed on selected miRNA using TaqMan MicroRNA Assays (Applied Biosystems, Foster City, CA) according to the manufacturer’s protocol and as previously described (6). TaqMan miRNA primer sets designed for human miRNA were used to quantify bovine miRNA after verifying that the human and bovine miRNA sequences were identical. Each sample was run in triplicate. GAPDH transcript levels were determined using SYBR Green, and miRNA expression was normalized based on GAPDH expression. Tissue culture. Juvenile bovine femoral condyles were obtained within 6 hours after slaughter, and the superficial zone cartilage was manually sectioned off with a surgical dermatome. Cartilage across the entire medial and lateral face of the femoral condyle was removed in this manner. The superficial zone cartilage was then placed in a collagenase solution in a rotating incubator at 37°C for 5 hours. Chondrocytes from the surface zone were isolated and plated as a monolayer in Dulbecco’s modified Eagle’s medium/F-12 medium (Invitrogen, Carlsbad, CA) containing 1% antibiotic solution, 1% bovine serum albumin, and 50 g/ml of ascorbic acid–phosphate for 2 days before harvesting total RNA using a mirVana miRNA isolation kit. RESULTS Findings of miRNA microarray studies. MicroRNA for the microarray studies were generated by directly extracting total RNA from specific regions of articular cartilage from freshly harvested bovine stifle joint femoral condyles obtained from several animals. The regions of interest were obtained by taking cores of articular cartilage from the anterior and posterior regions of the medial femoral condyle and further sectioning them into superficial, middle, and deep zone slices (Figure 1B). Previous studies have shown that cartilage from the anterior regions of the femoral condyle exhibit higher contact pressures than does cartilage from the posterior regions (5). Thus, the medial condyle was divided into 4 regions, designated M1–M4, with the M1 cartilage plug taken from the most anterior region and the M4 plug taken from the most posterior region (5) (Figure 1A). Two-color competitive microarray experiments were then performed to determine the differential expression pattern of known miRNA based on locations within and across the articular cartilage obtained from the femoral condyles. The first array experiment examined miRNA expression in cartilage from the anterior weight-bearing M1 superficial zone, where contact pressures are maximal, compared with cartilage from the posterior non–weight bearing M4 superficial zone, 2336 DUNN ET AL Table 1. Hybridization values and ratios identified by microarray analyses of bovine articular cartilage* Normalized hybridization value miRNA analysis M1 superficial zone vs. M1 middle zone Cartilage zone analyzed Probe ID bta-miR-221 bta-miR-222 hsa-miR-34c-3p hsa-miR-574-5p hsa-miR-374b hsa-miR-140-5p hsa-miR-411 hsa-miR-130a hsa-miR-155 hsa-miR-23a M1 superficial zone vs. M4 superficial zone Cartilage zone analyzed Probe ID bta-miR-221 bta-miR-222 hsa-miR-126 hsa-miR-143 hsa-miR-145 hsa-miR-34c-3p hsa-miR-155 hsa-miR-495 hsa-miR-376a hsa-miR-181a Cartilage sample A Cartilage sample B Log2 ratio M1 superficial M1 middle Superficial/middle 3,229.53 2,524.47 2,099.08 1,732.88 2,393.41 1,553.91 1,739.44 1,136.82 2,677.12 51,060.78 206.72 350.17 362.29 6,140.02 851.05 4,032.21 4,090.37 2,766.62 975.6 2,3748.42 4.04 2.85 2.55 ⫺1.77 1.49 ⫺1.37 ⫺1.31 ⫺1.28 1.24 1.23 M1 superficial M4 superficial M1/M4 superficial 5,097.17 4,256.4 15,539.59 4,090.81 10,018.33 1,173.87 4,381.34 984.6 1,250.4 3,229.99 452.1 494.34 2,739.21 803.25 1,972.54 299.99 1,132.04 3,550.88 4,065.49 1,164.43 3.42 3.08 2.60 2.35 2.34 2.07 1.95 ⫺1.80 ⫺1.70 1.54 * Values are from microarray experiments examining microRNA (miRNA) levels in the anterior medial femoral condyle (M1) superficial zone versus the M1 middle zone, as well as levels in the M1 superficial zone versus the posterior medial femoral condyle (M4) superficial zone. The normalized microarray hybridization values are given for each 2-color competition experiment (P ⬍ 0.01). For each dual-sample experiment, a P value calculation was performed, and a list of miRNA candidates was generated based on their differential expression between the 2 samples. The log2 ratios are based on the hybridization values for each condition. The top 10 miRNA with the greatest absolute log2 ratios, indicating the greatest differential expression between the test conditions, are listed for each comparison. The miRNA candidates were also screened to include only those for which the greater of the 2 compared hybridization values was at least ⬎1,000 (cutoff value). where contact pressures are minimal (5). The second experiment investigated miRNA expression in cartilage from the M1 superficial zone compared with cartilage from the M1 middle zone. The LC Sciences miRNA microarray platform was synthesized with all known human/mouse/rat and bovine miRNA in version 10.1 of the miRBase database. Non-bovine probes were included on the array, since the database contained only a limited number of verified bovine miRNA sequences. Based on our preliminary bioinformatics analysis (Dunn W, et al: unpublished observations) a number of verified human/mouse/rat miRNA sequences were found in the bovine genome sequence with perfect or significant sequence identity, but which were not in the miRBase bovine miRNA database. LC Sciences performed the labeling of the miRNA, hybridization onto the array, and subsequent analysis of the data, generating a list of differentially expressed miRNA based on the log2 ratios between the competing conditions. Table 1 shows a modified list of those differentially expressed miRNA. The top 10 candidates with the greatest differential expression and for which the hybridization value was ⬎1,000 (cutoff value) for at least 1 of the 2 compared samples are shown. A signal value of 1,000 on the array is the approximate threshold for which miRNA can be detected by quantitative RT-PCR. The top 2 differentially expressed miRNA on both arrays were miR-221 and miR-222. Findings of quantitative RT-PCR analysis of miRNA expression. We then selected miRNA-221 and miR-222, along with several other miRNA, for further quantification using quantitative RT-PCR techniques. The same comparison conditions were used for the quantitative RT-PCR analyses as for the microarray experiments. We compared miRNA expression in cartilage from the M1 superficial zone versus the M4 superficial zone and from the M1 superficial zone versus the M1 middle zone. The cartilage samples used for the quantitative RT-PCR were taken from a separate harvest than those used for the array experiments, and total MicroRNA EXPRESSION PROFILES IN BOVINE ARTICULAR CARTILAGE 2337 Figure 2. Quantitative reverse transcription–polymerase chain reaction (RT-PCR) analysis of microRNA (miRNA) expression and the average fold change. Quantitative RT-PCR was used to quantify selected miRNA expressed in bovine articular cartilage chondrocytes. The anterior medial femoral condyle (M1) and posterior medial femoral condyle (M4) values are from RNA that was extracted directly from sections of the articular cartilage plugs. Tissue culture monolayer samples were extracted from superficial layer chondrocytes that were plated as a monolayer in cell culture. The ‚Ct values are the threshold cycle values normalized against GAPDH; ‚‚Ct values can be calculated from the ‚Ct values to obtain the fold difference in miRNA expression under 2 different conditions. For example, the ‚Ct value for miR-221 in the M1 superficial zone is 11.2, and the ‚Ct value for miR-221 in the M4 superficial zone is 9.0. The ‚‚Ct value is calculated as 11.2 – 9.0 ⫽ 2.2. Therefore, there is a 22.2 ⫽ 4.59-fold greater amount of miR-221 in cartilage from the M1 superficial zone than in cartilage from the M4 superficial zone. RNA from several knees were pooled for these analyses. The quantitative RT-PCR results correlated well with the array data, with miR-221 and miR-222 being the top 2 differentially expressed miRNA of the miRNA that were analyzed. There was a 27-fold greater amount of miR-221 and miR-222 in the M1 superficial zone cartilage as compared with the M1 middle zone cartilage. The quantitative RT-PCR data also demonstrated that there was a 4-fold greater amount of miR-221 and miR-222 in the M1 superficial zone cartilage than in the M4 superficial zone cartilage (Figure 2). In addition to the miRNA that were up-regulated in the anterior weight-bearing M1 region of cartilage, quantitative RT-PCR identified miRNA that were down-regulated. In particular, miR-148a was downregulated ⬃5-fold in cartilage from the M1 superficial zone as compared with that from the M4 superficial zone (Figure 2). The quantitative RT-PCR data for miRNA levels in the superficial zone cartilage sections were compared with data from superficial zone chondrocytes that had been extracted from their cartilage matrix and plated as a monolayer in cell culture medium for 2 days. Loss of the extracellular matrix environment and growth in cell culture conditions significantly altered the expression of several miRNA species. Levels of miRNA-126 were 274-fold lower in the monolayer chon- drocytes than in the M1 superficial zone intact cartilage plug. It is noteworthy that miR-145 and miR-335 were present at 111-fold and 55-fold lower levels in the monolayer chondrocytes, respectively. Interestingly, miR-221 and miR-222 levels were also decreased in the monolayer chondrocytes, but not as dramatically as the decreases observed for miR-126, miR-145, or miR-335. Levels of miRNA-221 and miR-222 in the monolayer chondrocytes were similar to those found in the M1 middle zone, which represented an ⬃27-fold decrease from levels in the M1 superficial zone (Figure 2). The mean fold difference in miRNA expression levels between the M1 superficial zone of bovine articular cartilage and other zones or regions of articular cartilage were calculated from the average ‚Ct values shown in Figure 2. These additional data are available upon request from the author. DISCUSSION Although the function of many miRNA have yet to be elucidated, several identified in this study as being differentially expressed in bovine articular cartilage have previously been reported in other biologic contexts (7–9). Both miRNA-221 and miR-222 have previously been identified as playing a role in cancer cell prolifer- 2338 ation through their targeting of p27, a known cell cycle inhibitor (8–10). Thus, these miRNA may play a role in chondrocyte proliferation by possibly inhibiting p27 in the chondrocytes found in the anterior higher weightbearing regions of articular cartilage. There is evidence of increased cell division activity in the superficial zone of cartilage as compared with the middle and deep zones, as well as evidence of a stem cell progenitor population in the superficial zone (11,12). The upregulation of miR-221 and miR-222 observed in this study may be part of the stem cell population signature. In addition, the findings in the anterior weightbearing area with maximum contact pressure (M1 superficial zone) as compared with the posterior non–weightbearing area with minimum contact pressure (M4 superficial zone) suggest that greater mechanical stress may lead to greater chondrocyte proliferation, possibly through a mechanism of suppressing inhibitors of proliferation via miRNA. This implies that these miRNA may be examples of mechanical stress–induced miRNA that are up-regulated in response to greater weight bearing and maximum contact pressures. Previous studies in plants, such as the California poplar tree, have identified stress-induced plant miRNA, but this would be the first example of miRNA regulated in response to mechanical stress in a mammalian system (13). MicroRNA-148a, which was down-regulated in the M1 superficial zone cartilage as compared with the M4 superficial zone, has also been implicated in acute myeloid leukemia. Its expression was found to be inversely correlated with the expression of the BAALC gene, whose up-regulated expression is thought to be associated with early markers of progenitor cell phenotypes (14). Thus, lower miR-148a expression may correlate with higher levels of early progenitor cell markers that indicate a phenotype that favors cell proliferation over differentiation. Again, this is consistent with our observations, since lower miR-148a levels were found in the M1 region as compared with the M4 region and since the M1 cartilage bears comparatively more weight and, thus, more cell proliferation is stimulated. The present investigation on the differential expression of miRNA in articular cartilage yielded novel findings of the differential expression of miRNA between the different zones and in the anterior weightbearing regions of articular cartilage. The juvenile calf cartilage used in the current study is immature cartilage and may not be directly comparable to adult human articular cartilage. This investigation identified several miRNA of interest that were differentially expressed and perhaps related to regulation of chondrocyte prolif- DUNN ET AL eration. Furthermore, miR-221 and miR-222 were identified as mechanosensitive stress–induced miRNA, and represent a novel discovery in a mammalian articular cartilage. This insight into the role of miRNA in the biology of articular cartilage will provide a basis for further systematic investigation of articular cartilage in health as well as in disease, such as in the osteoarthritis, a degenerative disease associated with aging. Since the completion of our analyses, Iliopoulous et al (15) reported their finding of a 16-miRNA osteoarthritis gene signature from their studies comparing osteoarthritic and nondiseased human cartilage. It is noteworthy that the miRNA identified in bovine cartilage in our investigation are novel as compared with those identified in human osteoarthritic cartilage. In conclusion, miRNA-221 and miRNA-222 appear to be expressed in cartilage of the anterior weight-bearing superficial zone and may therefore represent targets of mechanotransduction in articular cartilage. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Reddi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Dunn, DuRaine, Reddi. Acquisition of data. Dunn, DuRaine. Analysis and interpretation of data. Dunn, Reddi. REFERENCES 1. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004;116:281–97. 2. Suh MR, Lee Y, Kim JY, Kim SK, Moon SH, Lee JY, et al. Human embryonic stem cells express a unique set of microRNAs. Dev Biol 2004;270:488–98. 3. Lagos-Quintana M, Rauhut R, Lendeckel W, Tuschl T. Identification of novel genes coding for small expressed RNAs. Science 2001;294:853–8. 4. Lagos-Quintana M, Rauhut R, Yalcin A, Meyer J, Lendeckel W, Tuschl T. Identification of tissue-specific microRNAs from mouse. Curr Biol 2002;12:735–9. 5. Neu CP, Khalafi A, Komvopoulos K, Schmid TM, Reddi AH. Mechanotransduction of bovine articular cartilage superficial zone protein by transforming growth factor ␤ signaling. Arthritis Rheum 2007;56:3706–14. 6. Chen C, Ridzon DA, Broomer AJ, Zhou Z, Lee DH, Nguyen JT, et al. Real-time quantification of microRNAs by stem-loop RTPCR. Nucleic Acids Res 2005;33:e179. 7. Visone R, Russo L, Pallante P, De Martino I, Ferraro A, Leone V, et al. MicroRNAs (miR)-221 and miR-222, both overexpressed in human thyroid papillary carcinomas, regulate p27Kip1 protein levels and cell cycle. Endocr Relat Cancer 2007;14:791–8. 8. Medina R, Zaidi SK, Liu CG, Stein JL, van Wijnen AJ, Croce CM, et al. MicroRNAs 221 and 222 bypass quiescence and compromise cell survival. Cancer Res 2008;68:2773–80. 9. Le Sage C, Nagel R, Egan DA, Schrier M, Mesman E, Mangiola A, et al. Regulation of the p27(Kip1) tumor suppressor by MicroRNA EXPRESSION PROFILES IN BOVINE ARTICULAR CARTILAGE 10. 11. 12. 13. miR-221 and miR-222 promotes cancer cell proliferation. EMBO J 2007;26:3699–708. Yeh N, Miller JP, Gaur T, Capellini TD, Nikolich-Zugich J, de la Hoz C, et al. Cooperation between p27 and p107 during endochondral ossification suggests a genetic pathway controlled by p27 and p130. Mol Cell Biol 2007;27:5161–71. Dowthwaite GP, Bishop JC, Redman SN, Khan IM, Rooney P, Evans DJ, et al. The surface of articular cartilage contains a progenitor cell population. J Cell Sci 2004;117:889–97. Hattori S, Oxford C, Reddi AH. Identification of superficial zone articular chondrocyte stem/progenitor cells. Biochem Biophys Res Commun 2007;358:99–103. Lu S, Sun YH, Shi R, Clark C, Li L, Chiang VL. Novel and 2339 mechanical stress-responsive microRNAs in Populus trichocarpa that are absent from Arabidopsis. Plant Cell 2005;17:2186–203. 14. Langer C, Radmacher MD, Ruppert AS, Whitman SP, Paschka P, Mrozek K, et al. High BAALC expression associates with other molecular prognostic markers, poor outcome, and a distinct gene-expression signature in cytogenetically normal patients younger than 60 years with acute myeloid leukemia: a Cancer and Leukemia Group B (CALGB) study. Blood 2008; 111:5371—9. 15. Iliopoulos D, Malizos KN, Oikonomou P, Tsezou A. Integrative microRNA and proteomic approaches identify novel osteoarthritis genes and their collaborative metabolic and inflammatory networks. PLoS ONE 2008;3:e3740.