Faculty Bibliography

Background: Wnt/beta-catenin signaling pathway is a key regulator in bone formation and maintaining bone hemostasis. Wnt signaling upon activation leads to stabilization of the transcriptional regulator beta-catenin and its further nuclear localization. Osteoblast differentiation and proliferation are regulated by a number of local and systemic factors, among which adenosine receptors are prominent during bone development. We recently reported cross-talk between Wnt/beta-catenin pathway and A2aR in fibroblasts; here we seek to determine whether there is similar cross-talk between Wnt signaling and the purinergic adenosine A2A receptor in osteoblasts. Since nuclear translocation of beta-catenin protein is a critical intermediate step in the Wnt signaling pathway, we studied the effect of A2aR signaling on nuclear and cytosolic beta-catenin levels in osteoblasts.
Method(s):We used an osteoblast cell line (MC3T3- E1) as well as primary osteoblasts derived frommesenchymal stem cells ofmice. The cells were treated with CGS21680, a selective A2aR agonist, at doses ranging from 0 to 10muM, and for varying incubation periods from 0 to 240 minutes. Using western blot analysis, the levels of total beta- catenin and phosphorylated beta-catenin at Ser552 (active component) were measured. Additionally, we measured beta-catenin levels in the nuclear extracts of the cells, both before and after A2aR activation. We also analyzed nuclear translocation of p-Ser552 beta-catenin using immunofluorescence (IF) staining.
Result(s): We observed a significant increase (p<0.05), in total beta-catenin (169+/-50% control, n=5) and p-Ser552 beta-catenin (253+/-122%, n=5) levels in MC3T3-E1 cells treated with 1 muM CGS21680 compared to control at 15 minutes following A2aR activation. The immunofluorescence staining results revealed enhanced nuclear accumulation of p-Ser552 beta-catenin by approximately 40% among the osteoblastoid cells treated with CGS21680. Similarly, western blot assays showed a significant increase (p<0.05) in the nuclear translocation of phosphorylated beta-catenin at Ser552 following administration of A2aR agonist inMC3T3-E1 cells (153+/-37%, n=4), as well as osteoblasts derived from mesenchymal stem cells (148+/-31%, n=4).
Conclusion(s): These results demonstrate cross-talk between A2aR and Wnt/beta-catenin signaling pathway in osteoblasts. Moreover, our results suggest that A2aR activation can bypass blockade of Wnt/Frizzled interactions at the cell surface and thereby maintain bone homeostasis

Background:We have previously reported that endogenously produced adenosine, interacting with A2AR, is a critical autocrine factor for maintenance of chondrocyte and cartilage homeostasis and intra-articular injections of liposomal preparations of adenosine inhibit progression of OA in a posttraumatic OA (PTOA) model in rats. We therefore determined whether intra-articular injection of a more selective A2AR agonist could also prevent progression and possibly reverse OA in this model and in the obesity related OA model in mice.
Method(s): PTOA was induced in SD rats following rupture of the anterior cruciate ligament (ACL) by application of external force to the knee. Starting 4 weeks after injury, when OA has already progressed, knees were injected with 100ul of saline, empty liposomes (LIPO) or liposomes containing CGS21680 (LIPO-CGS) every 10 days (6 injections) before sacrifice. The cartilage volume in OA and normal knees was measured by microCTafter staining with hexabrix (40%). Chondrocytes were isolated from neonatal mice and cultured, only first passage chondrocytes were studied. For the obesity-OA model, C57Bl6 mice (5-6/group, 12 weeks old) were fed a 60%fat diet (HFF mice). After 3 months, when OAwas present, mice received intrarticular knee injection (10 mul) of LIPO, LIPOCGS or liposomal adenosine (LIPO-Ado) every 10 days for 4 injections before sacrifice.
Result(s): Injection of LIPO-CGS but not saline or LIPO, significantly reduced swelling of affected rat knees (p<0.001). Surprisingly, there was an increase in tibial and femoral cartilage volume in normal knees treated with intra-articular injections of LIPO-CGS but not LIPO or saline (47% increase in tibia and 22% in femur). More importantly, intra-articular injections of LIPO-CGS, but not LIPO or saline, increased tibial and femoral cartilage volume inOAknees, as compared to normal knees and completely abrogated the histologic evidence of OA as well (OARSI score for CGS21680 0.66+/-0.33 vs 4.55+/-0.82 in the vehicle group and 3.90+/-0.89 in the saline group). There wasmarked chondrocyte proliferation in the deep cartilage of knees of rats treated with LIPO-CGS (Ki67 immunofluorescence). Similarly, LIPO-CGS reversed the OA changes in the obesity related OA model. HFF mice had an OARSI score of 5.17+/-1.84. Treatments with LIPO-Ado and lipo-CGS decreased OA severity (OARSI score 1.33+/-0.81 and 1.83+/-0.98, respectively, p<0.001 vs untreated). A2AR stimulation increased TGF-beta immunostaining in LIPO-CGS-injected joints and increased TGF-beta production by cultured neonatal murine chondrocytes with increased SMAD2/3 phosphorylation and diminished RUNX2 expression.
Conclusion(s): These results demonstrate that intra-articular injection of a long-acting A2AR agonist stimulates chondrocyte and cartilage regeneration, likely by a TGF-beta-dependent mechanism. More importantly, these results indicate that treatment with an A2AR agonist can reverse OA in both traumatic and obesity-related OA

Background: Patients with inflammatory arthritis and psoriasis are at greater risk for developing atherosclerotic plaques and associated cardiovascular diseases. Apremilast (APR), a PDE4 inhibitor used to treat psoriasis and psoriatic arthritis, was tested for effects in models of cholesterol efflux, foam cell formation, and atherosclerosis.
Method(s): A RAW264.7 murine macrophage cell line was infected with lentiviruses expressing short hairpin RNA (shRNA) to silence PKA, EPAC1, or EPAC2. The effect of APR on foam cell formation was tested following treatment with interferon-gamma (0.5 U/muL) for 24 hours and then treatment with an LDL (50 mug/mL) with/without APR (10 mumol/L) for another 48 hours. Cells were stained with Oil Red O and then cells containing lipid droplets were counted. Cholesterol efflux was measured after treatment with bodipy-cholesterol for 1 hour, equilibration buffer for 18 hours, and HDL (20 mug/mL) and ApoA1 (10 mug/mL with/without APR) for 4 hours.
Result(s): APR treatment reduced foam cell formation in RAW264.7 cells stably expressing EPAC1 and EPAC2 shRNA by 43 +/- 5% and 42 +/- 1%, respectively, while in cells expressing PKA shRNA, foam cell formation increased slightly (14 +/- 6%; P <.001, analysis of variance [ANOVA]). APR treatment enhanced HDL- and ApoA1-induced cholesterol efflux from RAW264.7 cells (2.35 +/- 0.03% vs. 1.73 +/- 0.08% [P <.001; n = 4] and 9.74 +/- 0.32% vs. 3.96 +/- 0.77% [P <.05; n = 3], respectively). APR treatment also enhanced HDL-induced efflux from RAW264.7 cells expressing EPAC1 and PKA but not EPAC2 shRNA (0.737 +/- 0.6% vs. 1.19 +/- 0.04% of control; P <.05, ANOVA); APR treatment enhanced ApoA1-induced cholesterol efflux from RAW264.7 cells stably expressing scrambled shRNA but not EPAC1, EPAC2, and PKA shRNA (2.62 +/- 0.15%, 1.55 +/- 0.22%, and 1.44 +/- 0.04%, respectively, vs. 0.95 +/- 0.15% of control; P <.01, P <.05, and P <.01, ANOVA). To determine the relevance of these findings to atherosclerosis, LDLR-/- mice were fed a Western diet +/- APR. A reduction in Oil Red O-stained lipid and CD68+ macrophages (P <.05, n = 3 or n = 4 per group) was seen in the plaques of animals treated with APR.
Conclusion(s): These results suggest that APR treatment: 1) promotes HDL-induced cholesterol efflux through an EPAC2-dependent mechanism, and ApoA1-induced cholesterol efflux through EPAC1-, EPAC2-, PKA-dependent mechanisms; 2) inhibits foam cell formation only through PKA-dependent mechanisms; and 3) reduces atherosclerotic plaque macrophages in a murine model of atherosclerosis.
Copyright

Background/Purpose: The most common cause of total joint replacement revision surgeries is loosening of the implant due to loss of bone around the prosthesis.Wear particles shed from the prosthesis plays a critical role by increasing local inflammation and osteoclast number and activity, ultimately causing osteolysis.We have previously reported that an A2A adenosine receptor selective agonist (CGS21680, CGS) prevents osteolysis in wear particle-induced osteolysis model in mice. Frequent administration requirements and potential toxicity make it a less than optimal treatment for inflammatory osteolysis.We therefore generated and tested a novel alendronate-CGS conjugate (MRS7216) that specifically localizes to bone targeting the agonist to the site of tissue injury and thereby diminishing the frequency of administration and curtailing systemic side effects.
Method(s): The conjugate was synthesized from CGS by sequential activation of the carboxylic acid moiety and reacting with the appropriate amino acid under basic conditions. A PEG6 linker was incorporated to alendronic acid by direct coupling. Osteolysis in 6-8-week-old C57BL/6J mice was induced by surgical implantation of 3mg of ultrahigh-molecular-weightpolyethylene particles over the calvaria. Mice received a weekly 10mg/kg intraperitoneal dose of MRS7216 conjugate, starting at the time of surgery. Other groups of mice were treated with equivalent weekly doses of alendronate-PEG6 (AlenP) or saline respectively. An additional control group underwent sham surgery. After 2 weeks, animals were sacrificed and microCT and histology analyses were performed. The studies were approved by the Institutional Animal Care and Use Committee of NYU School of Medicine.
Result(s): Receptor binding studies demonstrate that the Ki for CGS, 7216 conjugates and the control AlenP molecules were 21.5 nM, 69.2 nMand >10,000 nM respectively, indicating thatMRS7216 efficiently binds the A2A adenosine receptor. MicroCT studies showed that mice treated with weekly doses of 7216 had a significant reduction in bone damage of 40% (p=0.04) compared to saline treated mice. In contrast, AlenP molecules did not prevent bone erosion. Histological analysis of TRAP stained samples showed a significant decrease of osteoclast number/high-power field (HPF) of 55% (p=0.03) in AlenP treated mice compared to the saline treated group. The osteoclast depletion was more dramatic inMRS7216 treated group with an 81% reduction of osteoclasts number/HPF (p= 0.002). Additionally alkaline phosphatase staining in MRS7216 treated group, showed a significant increase in osteoblast number/HPF compared to saline (55%, p=0.01) and to AlenP group (45%, p=0.03).
Conclusion(s): Alendronate-CGS conjugates represent a novel therapeutic approach to prevent osteolysis and prosthetic failure in patients with prosthetic joints

Background/Purpose: Osteopenia and fragility fractures have been associated with HIV infection. Tenofovir, one of the most commonly used antivirals in HIV, also leads to increases in bone catabolism markers and decreased bone mineral density (BMD) in children and young adults. In murine models and human cell lines, Tenofovir inhibits ATP release and decreases extracellular adenosine levels. Adenosine, acting at its adenosine A2A and A2B receptors, inhibits osteoclast formation, and increasing local adenosine concentration with Dipyridamole, an agent that blocks adenosine cellular uptaken, stimulates new bone formation as well as rhBMP-2 by an A2A receptor-dependent effect. We hypothesized that Tenofovir regulates bone resoprtion by diminishing endogenous adenosine levels and determined whether Dipyridamole could counteract the deleterious effects of Tenofovir on bone.
Method(s): Male C57Bl/6 mice were treated as follows: IP injection of saline (control), Tenofovir 75mg/Kg/day, Dipyridamole 25mg/Kg/day, combination Tenofovir/Dipyridamole (n=10, 4 weeks). Female C57Bl/6 mice were ovariectomized and treated as follow: sham (no surgery), saline (control), Tenofovir 75mg/Kg/day, Dipyridamole 25mg/Kg/ day, combination Tenofovir/Dipyridamole (n=10, 5 weeks). Weekly weight was annotated. DXA scanning was performed before sacrifice. Calcein/AlizarinRed-labelling of newly formed bone was used, and long bones were prepared for microCT/ histology.
Result(s): Male mice treated with Tenofovir lost nearly 10% of body weight (p<0.001). DXA scanning showed a decrease in BMD in mice treated with Tenofovir that was reversed with Dipyridamole. microCT revealed decreased BMD and diminished trabecular bone in Tenofovir-treated mice and reversal by Dipyridamole treatment. TRAP-staining showed increased osteoclasts in Tenofovir-treated mice (p<0.005) an effect reversed by Dipyridamole. Similar results were obtained for Cathepsin K and CD68. RANKL-positive-cells were increased in Tenofovir-treated mice whereas OPG-positive-cells decreased, and both effects were reversed by Dipyridamole. In the case of female OVX mice, Tenofovir treatment also produced a decreased in body weight (p<0.05) that was reversed with Dipyridamole. DXA scanning showed decreased BMD in Tenofovir-treated mice and microCT revealed diminished trabecular bone, similar to findings in male mice. Similar results were found for Cathepsin K, CD68, RANKL and OPG-positive-cells.
Conclusion(s): These results suggest that treatment with agents that increase local adenosine concentrations, like Dipyridamole, might prevent bone loss following Tenofovir treatment

Background/Purpose: The Wnt/beta-catenin signaling pathway plays a key role in regulating bone formation and maintaining bone hemostasis. Wnt activates a pathway that leads to stabilization of beta-catenin and its translocation to the nucleus. Osteoblast differentiation and proliferation are also regulated by adenosine receptors, among other signals. We recently reported that A2aR signaling promotes Wnt/b-catenin signaling in fibroblasts via activation of Akt and p38MAPK. In the present study we sought to determine whether there is a similar interaction between these pathways in osteoblasts.
Method(s): We studied murine osteoblast cell line (MC3T3-E1) and primary osteoblasts derived from bone marrow-derived mesenchymal stem cells of mice. The cells were treated with CGS21680, a selective A2aR agonist, at doses ranging from 0 to 10muM, and for varying incubation periods up to 240 minutes. Levels of phosphorylated beta-catenin at Ser552 (p-Ser552), a beta-catenin isoform with enhanced transcriptional activity, were measured by Western Blot assays before and after A2aR activation. We also analyzed nuclear translocation of p-Ser552 beta-catenin in the osteoblastoid cell line and primary cell cultures using immunofluorescence (IF) staining. Cellular levels of activated AKT were measured by immunoblotting assays before and following administration of CGS21680.
Result(s): We observed a significant increase in p-Ser552 beta-catenin levels in the osteoblastoid cells treated with 1 muM CGS21680 compared to the control, starting at 15 minutes following A2aR activation (253+/-122%, p<0.05, n=5). Western blot analysis showed a significant increase in nuclear translocation of p-Ser552 beta-catenin at 15 minutes after treatment with A2aR agonist in MC3T3-E1 cells (153+/-37%, p<0.05, n=4), and primary osteoblasts (148+/-31%, p<0.05, n=4). Similarly, immunofluorescence revealed approximately a 40% increase in nuclear accumulation of p-Ser552 beta-catenin in CGS21680- treated MC3T3-E1 cells as well as in primary osteoblasts. We also found a significant increase in the levels of phosphorylated AKT at Ser473 among osteoblastoid cells following A2aR stimulation (203+/-47%, p<0.05, n=4).
Conclusion(s): These findings demonstrate cross-talk between A2aR and Wnt/b-catenin signaling pathways in osteoblasts. Moreover, our results suggest that A2aR activation can bypass blockade of Wnt ligands at the cell surface and thereby maintain bone homeostasis

Background/Purpose: Low dose weekly MTX remains the anchor drug for treatment of Rheumatoid Arthritis. The principal mechanism by which MTX suppresses inflammation in Rheumatoid Arthritis is thought to be enhanced adenosine release from cells which suppresses inflammation by stimulating adenosine receptors on T cells, macrophages and other inflammatory cells (Nature Rev Rheumatol 13:41, 2017). Many patients do not respond to low dose MTX and studies in mice suggest that MTX resistance may be due to inadequately increased adenosine release (Clin Exp Rheumatol 31:433, 2013). Because adenosine is primarily taken up by cells from the extracellular space via the nucleoside transporter ent1 we asked whether an agent that blocks adenosine uptake could enhance the effect of MTX in the treatment of RA. We therefore carried out an open label 1 month study, adding an inhibitor of adenosine uptake via ent1, ticagrelor (a P2Y12 inhibitor that is approved for inhibition of platelet aggregation to prevent severe cardiovascular events) (Nat Rev Cardiology12:156,2014), to patients who were poorly controlled with low dose methotrexate therapy for RA. (NCT02874092) Methods: Patients (5 female/1 male, mean age 49.6 years) who all met ACR criteria for RA and had active disease, as defined by DAS28 (ESR) > 3.6 and who were on stable doses of MTX monotherapy (for a minimum of 12 weeks), were recruited from the Bellevue Hospital Center Arthritis Clinic. Patients had no known contraindication to ticagrelor and had no history of coronary artery disease. After giving informed consent patients entered an open label protocol in which they were administered Ticagrelor (90mg) twice daily for one month in addition to their stable dose of MTX. Disease activity was reassessed and change in activity from the start of the trial was noted. This study was approved by the NYULMCBellevue IRB.
Result(s): Five of six patient achieved an improvement in their DAS28(ESR) >0.6. Half of the patients (3 patients) achieved a reduction in DAS28 (ESR) >1.2, 2 patients achieved a reduction in DAS28 (ESR) >0.6 but less than 1.2 and 1 patient showed no improvement. Four of six patients had a reduction in their tender joints and all had a reduction in swollen joints. No patients reported any adverse reactions, including excessive bleeding.
Conclusion(s): The results of this small open label trial suggest that treatment with ticagrelor enhances the effect of MTX on RA and may be a useful addition to the therapeutic armamentarium. Moreover, these results offer further support for the hypothesis that enhanced adenosine release at inflamed sites mediates the anti-inflammatory effects of MTX therapy. The limitations of this trial include the fact that it was an open label trial in a small group of patients but the results support further study of ticagrelor in combination with MTX. (Table Presented)