INTRODUCTION: Macrophage infiltration in the synovial membrane (SM) and intra-articular fat pads (FP) is common in osteoarthritis (OA) development, and can contribute to catabolic and anabolic cytokine and protease production, which contributes significantly to OA symptoms. However, whether macrophages are appropriate targets for therapy in OA is unclear, as macrophages can also promote tissue repair. The purpose of this study is to characterize the timeline and phenotype of macrophages in SM and FP in a translationally relevant murine model of post-traumatic OA. We hypothesized that by analyzing macrophage populations by two separate approaches, cellular phenotype and gene expression analysis, we could confirm the precise temporal role and characteristics of infiltrating macrophages while OA is developing. METHODS: All animal research was conducted with IACUC approval from the University of Pennsylvania and the CMC VA Medical Center. C57BL/6 male mice (10-12 wks old) were subjected to destabilization of medial meniscus (DMM) on the right hind leg, and the left hind leg was un-operated. Mice were sacrificed 4 and 8 weeks post-surgery, and SM/FP dissected for cellular analysis. Tissues from 4 knees were pooled, cells isolated enzymatically, and stained with the Live/Dead™ Fixable Violet Dead Cell Stain Kit (Invitrogen) and the following antibodies: CD45-PerCP Cy5.5, CD11c-Super Bright 645, F4/80-APC, iNOS-Alexa Fluor 488, CD206-PE. Multicolor flow cytometry was performed and data analyzed with FlowJo software. After gating on single live cells, F4/80+ (general macrophage marker) and CD11c+ (expressed by dendritic cells, monocytes and macrophages) cells were expressed as percent of the CD45+ population. iNOS (expressed by M1-type inflammatory macrophages) and CD206 (expressed by M2 reparative macrophages) expression was then characterized on F4/80 and CD11c expressing cells. To determine whether gene expression of macrophage-related transcripts reflects cytometric analysis, additional groups of mice were sacrificed pre-DMM (baseline), and post-DMM at 4 and 8 weeks. SM/FP tissues from 4 knees were dissected and pooled for each sample to obtain adequate mRNA. cDNA was synthesized by routine methods, and mRNA transcripts amplified using the QX200TM Droplet Digital PCR System (BioRad). Primers for macrophage markers (CD68, F4/80 and CD11c) as well as M1 (iNOS, CCR7) and M2 macrophage products (CD206 and CD163) were used, and transcript levels expressed relative to TATA-Box binding protein (TBP) transcript numbers. RESULTS: Gating on the CD45+ population, two main populations of cells were defined by F4/80 and CD11c expression: A CD11c+ F4/80- population (reflective of dendritic cell phenotype), and a CD11c+ F4/80+ population (reflective of macrophage phenotype). Percentages of both populations were significantly increased in DMM-operated compared to un-operated joints at 4 weeks but not at 8 weeks (Fig 1A&B, F4/80- CD11c+, DMM: 11.84±1.46, un-operated: 4.55±0.98, p=0.009. F4/80+ CD11c+, DMM: 15.28 ± 2.55, un-operated: 4.66±2.06, p=0.017). Compared to the un-operated side, CD206+ macrophages (F4/80+ CD11c+) and dendritic cells (F4/80- CD11c+) were significantly lower proportionally in DMM-operated limbs at 4 weeks, and this trend was sustained at 8 weeks in the F4/80+ CD11c+ population (Fig 1C). Percentage of iNOS + cells were slightly elevated in the F4/80+ CD11c+ macrophage population at 8 weeks (p=0.02) post-DMM, but overall numbers of cells were small. We next tested whether similar phenotypic changes post-DMM could be detected at the mRNA level. We measured multiple markers of macrophage lineage (F4/80, CD68, and CD11c) and phenotype (M1: iNOS, CCR7; M2: CD206, CD163) by qPCR. CD11c and CD68 expression levels were increased on the DMM-operated side at 4 weeks post-DMM compared to the un-operated side (CDllc: 9.8-fold higher; CD68: 9.7-fold higher than un-operated, all p<0.05), and CD68 remained elevated (3.3-fold higher, p<0.05) at 8 weeks post-DMM. Of the M1-markers, mRNA levels for iNOS and CCR7 did not differ between DMM- and un-operated sides at 4 or 8 weeks. However, levels of iNOS were highly upregulated compared to baseline at 4 weeks (2000-fold higher, p<0.05), while levels of CCR7 mRNA were upregulated compared to baseline at both 4 (6.6-fold) and 8 weeks (7-fold). Transcripts for the M2 marker CD163 were lower on the DMM-operated side at 4 weeks (30-fold lower) and 8 weeks (35- fold lower than un-operated, both p<0.05), but no difference in levels of CD206 mRNA were observed at any time point. DISCUSSION: Interestingly, time-dependent increases in both F4/80+CD11c+ macrophages and F4/80-CD11c+ dendritic cells can be detected in SM/FP tissues after DMM surgery, peaking at 4 weeks. These changes in cell proportions were reflected by changes at the transcriptional level of select but not all macrophage markers. Overall proportions of CD206+ cells were much higher in the macrophage population than iNOS+ cells. Significant decreases compared to the un-operated side at 4 and 8 weeks post-DMM in macrophages expressing CD206 suggest early and sustained suppression of the M-2 macrophage phenotype. Although iNOS expressing macrophages were slightly increased at 8 weeks post-DMM, this reflected expression on very small proportions of cells. Importantly, not all phenotypic markers measured at the mRNA level reflected changes in these markers on macrophage and dendritic cell populations. Specifically, F4/80 and CD206 transcripts were not modulated despite clear differences in proportions of cells expressing these markers, while iNOS mRNA expression was highly upregulated more than other markers despite minimal expression on measured cells. It is possible that iNOS mRNA changes reflect other unmeasured cells in these tissues, or that post-transcriptional or translational mechanisms are more important in controlling cellular expression of this gene. It is quite revealing that mRNA levels of other M1 and M2 phenotypic markers, specifically CCR7 and CD163, more closely reflected phenotypic changes to macrophage populations detected by cytometry in this model. These discrepancies point to the importance of confirming transcriptional analysis at the cellular level and demonstrate that conclusions cannot be drawn solely from one measurement of gene expression. SIGNIFICANCE/CLINICAL RELEVANCE: This study demonstrates clearly divergent macrophage signatures in the early stages of OA development in the DMM model, and is the first to document suppression of M2 macrophage proportions post-DMM. These observations will be insightful to plan timing of interventions to better understand the pathologic roles of macrophage subtypes in OA. Specifically, this data suggests that enhancement of M2-reparative macrophage responses, rather than blockade of M1-inflammatory macrophages, could be a novel therapeutic strategy for treatment of post-traumatic OA.
No datasets are available for this submission.
No license information is available for this submission.