Changes in SOX9 and RUNX2 protein activity correlate to the health state of human primary chondrocytes

Abstract

Purpose: SOX9 and RUNX2 are the master transcription factors involved in cartilage and bone development respectively. Altered SOX9 and RUNX2 activity is implicated in osteoarthritis (OA) pathophysiology and hypertrophic differentiation of chondrocytes. We have previously shown that SOX9 and RUNX2 protein activity can be measured by Transcription Factor - Fluorescence Recovery After Photobleaching (TF-FRAP). To decipher the underlying signaling mechanism in OA pathophysiology, we aim to use TF-FRAP to correlate SOX9 and RUNX2 mobility to the health state of the chondrocytes. We hypothesize that the mobility of SOX9 increases with OA progression, while the mobility of RUNX2 decreases in chondrocyte hypertrophy. We measured the mobility and correlated protein activity of these transcription factors at the single cell level by TF-FRAP in healthy, preserved and OA chondrocytes. Methods: Healthy human primary chondrocytes (hPCs) were purchased from Articular Engineering, USA. Human preserved and OA hPCs were isolated from patients undergoing total knee replacement therapy. hPCs were seeded on glass coverslips and transfected with either SOX9-mGFP or eGFP-RUNX2. TF-FRAP was performed in at least 40 cells per condition in two healthy donors and three donors for each preserved and OA chondrocytes. TF-FRAP measurements were done using a Nikon A1 confocal microscope (Japan). In TF-FRAP, a small circular region (Ø 25 pixels) of the nucleus was bleached using a high intensity laser and the recovery of the fluorescence in the bleach spot is recorded. For every TF-FRAP measurement, 25 pre-bleach images and 260 post-bleach images were acquired at 4 frames per second for 60 seconds, with a frame size of 256x256 pixels. Higher immobile fraction (i.e., the fraction of transcription factors bound to DNA) and higher half-time recovery (longer DNA interaction times) were associated with a higher transcriptional activity. For all donors, early passage (≤3) cells were used for TF-FRAP and qPCR measurements. To correlate protein mobility and DNA binding with the protein activity, we quantified the expression levels of SOX9 target genes (ACAN and COL2A) and RUNX2 target genes (MMP13 and COL10) by qPCR. Results: In the healthy hPCs, the SOX9 immobile fraction and half-time recovery were 60% (SD ±6%) and 14.2 sec (SD ±3.2 sec) respectively. In preserved hPCs, the SOX9 immobile fraction and half-time to recover was significantly lower, ranging from 50% - 44% (SD ±7% - ±9.7%), and 14.2 - 13.3 sec (SD ±3.9 - ±4.5 sec) respectively. In OA hPCs, the SOX9 immobile fraction and half-time recovery was significantly lower than in the healthy and preserved hPCs, ranging from 48.6% - 42.8% (SD ±7.3% - ±8.4%), and half-time to recover ranged from 13.3 - 12.2 sec (SD ±3.9 - ±5.3 sec). RUNX2 immobile fraction and half-time to recover data spread was large in healthy, preserved and OA hPCs. In healthy hPCs, immobile fraction and half-time to recover were ranging from 33.6% - 47.9% (SD 19.4% - 20.7%) and 13.97 sec - 14.46 sec (SD 7.9 - 10.1 sec) respectively. In the preserved hPCs, the immobile fraction and half-time to recover were ranging from 38.8% - 47.0 % (SD 19.7% - 22.2%) and 11.7 sec - 14.8 sec (SD 6.8 - 8.2 sec) respectively. In OA hPCs, the immobile fraction and half-time to recover ranged from 33.0% - 41.02% (SD 20.3% - 25.7%) and 11.2 sec - 12.9 sec (SD 7.4 - 9.1 sec) respectively. We have an initial non-quantitative indication for a distinct cell population that, irrespective of the health state, shows decreased RUNX2 mobility, which could be indicative of a hypertrophic cell state. We will show the quantification and segregation. RT-qPCR measurements show decreased SOX9 and increased RUNX2 target gene expression during OA progression. Conclusions: Our data show that the SOX9 and RUNX2 transcriptional activity is differentially regulated depending on the health state of hPCs. If we assume a correlation between mobility and transcription factor activity, then we can say that SOX9 protein activity decreases with increasing OA progression. RUNX2 mobility data showed large spread especially in the preserved and OA donors possibly due to cells were in the different health state in the OA joint. Our findings at the single cell level provide novel insights into SOX9 and RUNX2 protein activity with respect to OA progression.