Poster 217: Local and Systemic Comparison of Senescent Profiles in Acute Knee Injury

OBJECTIVES: Synovial fluid profiles in the setting of intra-articular ligament injury remains largely undefined and may play a significant role in graft incorporation and the development of fibrosis and PTOA. Senescence, a characteristic of cellular aging, has been recently linked to the acceleration of osteoarthritis (OA), but has yet to be examined in the setting of acute knee injury. To our knowledge, the presence of senescent profiles, including senescent cells and their senescence associated secretory phenotype (SASP) factors, have not been evaluated following an acute intraarticular ligament injury in humans. The purpose of this study was to prospectively identify, measure and compare senescent profiles between synovial fluid and peripheral blood mononuclear cells (PBMCs) in patients that have sustained an acute knee injury within 48 hours. METHODS: Eight subjects between 18-65 years of age who presented within 0-48 hours of an acute knee effusion with a confirmed ACL or posterior cruciate ligament tears on magnetic resonance imaging were prospectively enrolled in the study. Synovial fluid (4-30 mL) and peripheral blood (˜15 mL) samples were collected in anticoagulant at the time of initial evaluation, processed using RBC lysing buffer and/or separation gradient (Lympoprep(TM), Stem Cell Technologies) following manufacturer instructions, and then their senescent phenotype and SASP profile was quantified using flow cytometry (Guava EasyCyte, Luminex), multiplex (Luminex 200®) and qRT-PCR. Senescent cell burden was quantified in triplicate per sample using C(12)FDG (senescence stain)(1), bafilomycin (reduces false positives), and DRAQ7(®) (viability stain) by flow cytometry (Guava EasyCyte, Luminex). Cells were gated to singlets and live/dead (viability) groups, then live cells were gated for senescence. Three distinct senescent cell populations were assessed including non-senescent (C(12)FDG-), bright (++, highly fluorescent), dim (+, moderately fluorescent) and total (dim+bright) cells are analyzed in percentages and count. We subsequently analyzed matrix metalloproteinases (MMPs; MMP-1,-2, -3, -9, -12, -13) and tissue inhibitors of MMPs (TIMPs; TIMP-1, -2) that have previously been described as part of the SASP complex(2) using the Luminex 200® multiplexing technology according to manufacturer’s guidelines. Cellular expression of cell cycle regulators (p16, p21, p53) established as senescent markers and proinflammatory SASPs (IL-1β, IL-6, IL-8) were quantified using RT-PCR . Statistical analysis was performed using two-way ANOVA using Graphpad Prism followed by Tukey-post hoc and paired Wilcoxon Signed Rank testing. Results were reported as median and quartiles [1(st) quartile, 4(th) quartile]. RESULTS: Patient demographics and injury details were collected and reported on 7 subjects, including age (median 43.0, range 21-61), gender (3 males, 4 females), knee laterality (3 right, 4 left), injury type (6 ACL, 1 PCL). Time from injury to synovial fluid aspiration/ venipuncture was 1.14 days (range 0-2 days). Using C(12)FDG to quantify senescent cells we found the total number of C(12)FDG was significantly different in synovial fluid (33.0 % [26.1, 66.6]) compared to PBMCs (31 % [27.1, 38.3]) (Fig. 1C, p<0.05). C(12)FDG subsets, bright (C(12)FDG(++)) and dim (C(12)FDG(+)) were not statistically different across tissue source (Fig. 1A&B). All MMP analytes (MMP-1,-2, -3, -9, -12, -13) were detectable in synovial fluid, plasma and serum samples (Fig. 2D). MMP-3 was significantly elevated in synovial fluid (median: 435404 pg/mL [349605.7, 581710]) compared to plasma (median: 12537 pg/mL [10970.9, 43637.1]) and serum samples (median: 15700.8 pg/mL [13231.1, 38228.1]) (Fig. 2A, p<0.05). TIMP-1 was also significantly higher in synovial fluid (median: 915507.4 pg/mL [77861.5, 1034542.9]) compared to plasma TIMP-1 (median: 40648.3 pg/mL [38887.8, 50075.9]) (Fig. 2B, p<0.05), but not significantly different to serum TIMP-1 concentrations (median: 90737.8 pg/mL [73246.1, 102032.2]). TIMP-2 concentrations in synovial fluid (median: 49333.80 pg/mL [23445.7, 76458.8]) were not statistically different compared to plasma (median: 32636.3 pg/mL [29153.7, 38075.8]) and serum samples (median: 46954.3 pg/mL [41807.9, 55572.6]) (Fig. 2C). For gene expression, a significant difference was observed in synovial fluid IL-8 compared to PBMCs (Fig. 3A, p<0.05), while there were no significant difference in IL-1β and IL-6 inflammatory genes (Fig. 3A) or p16, p21, and p53 senescent genes (Fig. 3B). CONCLUSIONS: Our results suggest that there is a significant presence of senescent cells and SASP within the synovial fluid local environment compared to the systemic environment within hours after an acute knee injury (within 48 hours) involving ligament tears. Of note, we did not find elevated senescence associated gene expression within the acute knee injury setting. Interesting, we found that there are two distinct senescent cell populations (bright and dim C(12)FDG+ populations) in synovial fluid samples, similar to that of senescent PBMC samples. This preliminary data may have a role in identifying strategies to modify the acute environment within the synovial fluid either at the time of acute ligament injury or during reconstruction surgery. To this end, reducing senescent cells, and downstream activation of inflammatory and degradative factor expression has the potential to decrease the rates of inflammation and scar tissue formation and improve graft incorporation and healing, with the goal to prevent the progression of PTOA. A larger sample size and comparing 48 hour senescent profiles to post-surgical time points will be necessary to improve our understanding of the potential changes in these profiles when comparing the systemic and local environments in an acute knee injury setting.