Selective inhibition of receptor activator of NF-κB ligand (RANKL) in hematopoietic cells improves outcome after experimental myocardial infarction

Selective inhibition of receptor activator of NF-κB ligand (RANKL) in hematopoietic cells... The RANK (receptor activator of nuclear factor κB)/RANKL (RANK ligand)/OPG (osteoprotegerin) axis is activated after myocardial infarction (MI), but its pathophysiological role is not well understood. Here, we investigated how global and cell compartment-selective inhibition of RANKL affects cardiac function and remodeling after MI in mice. Global RANKL inhibition was achieved by treatment of human RANKL knock-in (huRANKL-KI) mice with the monoclonal antibody AMG161. huRANKL-KI mice express a chimeric RANKL protein wherein part of the RANKL molecule is humanized. AMG161 inhibits human and chimeric but not murine RANKL. To dissect the pathophysiological role of RANKL derived from hematopoietic and mesenchymal cells, we selectively exchanged the hematopoietic cell compart- ment by lethal irradiation and across-genotype bone marrow transplantation between wild-type and huRANKL-KI mice, exploiting the specificity of AMG161. After permanent coronary artery ligation, mice were injected with AMG161 or an isotype control antibody over 4 weeks post-MI. MI increased RANKL expression mainly in cardiomyocytes and scar- infiltrating cells 4 weeks after MI. Only inhibition of RANKL derived from hematopoietic cellular sources, but not global or mesenchymal RANKL inhibition, improved post-infarct survival and cardiac function. Mechanistically, hema- topoietic RANKL inhibition reduced expression of the pro-inflammatory cytokine IL-1ß in the cardiac cellular infiltrate. In conclusion, inhibition of RANKL derived from hematopoietic cellular sources is beneficial to maintain post-ischemic cardiac function by reduction of pro-inflammatory cytokine production. Key messages & Experimental myocardial infarction (MI) augments cardi- ac RANKLexpressioninmice. Olena Andrukhova passed away before the submission of the final & RANKL expression is increased in cardiomyocytes and version of this manuscript. scar-infiltrating cells after MI. Electronic supplementary material The online version of this article & Global or mesenchymal cell RANKL inhibition has no (https://doi.org/10.1007/s00109-018-1641-x) contains supplementary influence on cardiac function after MI. material, which is available to authorized users. & Inhibition of RANKL derived from hematopoietic cells improves heart function post-MI. * Reinhold G. Erben & Hematopoietic RANKL inhibition reduces pro- Reinhold.Erben@vetmeduni.ac.at inflammatory cytokines in scar-infiltrating cells. Institute of Physiology, Pathophysiology and Biophysics, Department of Biomedical Research, University of Veterinary . . . Keywords Myocardial infarction RANKL Inflammation Medicine Vienna, Veterinaerplatz 1, 1210 Vienna, Austria Osteoprotegerin VetCore, University of Veterinary Medicine Vienna, Vienna, Austria Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine III and Center for Healthy Aging, Introduction Technische Universität Dresden, Dresden, Germany Amgen Inc., Thousand Oaks, CA, USA Improvements in the prevention and treatment of cardiovas- cular diseases have significantly changed the epidemiology of Present address: Phylon Pharma Services, Newbury Park, CA, USA 560 J Mol Med (2018) 96:559–573 myocardial infarction (MI). Whereas early survival has been fully inhibited by AMG161 [13]. Based on our previous find- improved and the severity of infarctions has declined progres- ing that only endothelial and hematopoietic but not stromal sively, the overall incidence of MI and the long-term survival precursors engraft after transplantation of unfractionated bone has remained constant [1, 2]. Long-term mortality after MI is marrow into lethally irradiated rats [14] and mice [15] and the still high, partly due to progression of the disease to heart fact that AMG161 exclusively inhibits human but not murine failure (HF), a condition which requires improvement in treat- RANKL, we used a reconstitution model in which we selec- ment options. Thus, a better understanding of the molecular tively exchanged the hematopoietic cell compartment to dis- mechanisms involved in HF progression is necessary for de- sect the pathophysiological role of RANKL derived from he- velopment of novel therapeutic strategies to prevent heart fail- matopoietic and mesenchymal cellular sources in the develop- ure post-MI. In this context, an increasing body of evidence ment of cardiac dysfunction after ischemia. We found that indicates that signaling through the RANK (receptor activator inhibition of RANKL derived from hematopoietic cellular of nuclear factor κB)/RANKL (RANK ligand)/OPG sources had beneficial effects on cardiac function after MI. (osteoprotegerin) axis might be involved in the pathophysiol- ogy of cardiovascular diseases. RANKL was first discovered as a cytokine which drives macrophage maturation to osteo- Methods clasts through its signaling receptor RANK, while OPG in- hibits RANKL signaling acting as a soluble decoy receptor of Animals RANKL [3]. Besides its role in bone remodeling, baseline levels of soluble RANKL predict the risk of cardiovascular All animal procedures were undertaken in accordance with events such as myocardial infarction and stroke [4]. current guidelines for animal care and welfare and were ap- Furthermore, it was suggested that RANKL may be important proved by the Ethical Committees of the University of for destabilizing atherosclerotic plaques [4, 5], whereas OPG Veterinary Medicine Vienna and of the Austrian Federal may prevent blood vessel calcification in atherosclerosis [6]. Ministry of Science, Research and Economy. All animals Notably, the myocardial OPG/RANKL ratio is significantly were kept in groups of two to seven at 22–24 °C and a 12-h lower in clinical and experimental heart failure, due to light/12-h dark cycle with free access to tap water and a com- disproportionally enhanced RANKL expression [7]. mercial rodent diet (Sniff™). The generation of the However, it is currently not known which exact role huRANKL-KI transgenic mouse line was described in detail RANKL plays during the post-ischemic myocardial healing elsewhere [13]. huRANKL-KI mice were obtained from and during transition to HF. Amgen Inc. and were backcrossed to C57BL/6 genetic back- The complex NF-κB signaling initiated by binding of ground for a minimum of six generations. Heterozygous mice RANKL to its receptor RANK can be, depending of the were bred, and the resulting wild-type (wt) and homozygous cellular context, detrimental or protective [8]. It was sug- huRANKL-KI animals were genotyped from tail biopsies by gested that RANKL promotes myocardial inflammation PCR analysis of genomic DNA. during acute cardiac overload [9] and favors adverse remod- eling by matrix degradation after acute myocardial infarc- Lethal irradiation and bone marrow transplantation tion [7]. In addition, RANKL was reported to act as a pro- inflammatory cytokine on hepatocytes as shown by the Male wt and huRANKL-KI mice at 6 to 8 weeks of age were finding that hepatic RANKL deletion counteracted hepatic used as bone marrow (BM) donors. Mice were killed by ex- insulin resistance by reducing NF-κB signaling [10]. On the sanguination from the abdominal vena cava under ketamine/ other hand, increased RANKL signaling reduced the infarct xylazine anesthesia (70/7 mg/kg i.p.). Unfractionated BM was size in stroke and hepatic ischemia, by acting in an anti- isolated from femora, tibias, and humeri by centrifugation at inflammatory and pro-cell survival manner [11, 12]. 800×g for 5 min at room temperature. The cells suspended in Here, we aimed to investigate the effect of RANKL inhi- PBS were filtered through a 40-μm cell strainer and were re- bition on post-ischemic cardiac function and structure. We suspended in PBS to contain 4 × 10 cells per milliliter. inhibited RANKL using the human monoclonal IgG1 anti- Male 16-week-old wt and transgenic huRANKL-KI recip- RANKL antibody AMG161. AMG161 selectively inhibits ient mice were lethally irradiated with a single dose of 10 human but not murine RANKL, by binding to a peptide se- Gray, using a linear accelerator (6MV, Primus, Siemens). quence encoded by exon 5 of the human RANKL gene. To Four hours after the irradiation, 4 × 10 of freshly prepared block RANKL in vivo, we used a transgenic mouse line (hu- unfractionated BM cells were injected into a lateral tail vein man RANKL knock-in, huRANKL-KI) where exon 5 of the of the recipient mice. Irradiated wt mice received the BM cells murine RANKL gene was replaced by the human sequence of huRANKL-KI donors, and irradiated huRANKL-KI mice [13]. The chimeric RANKL protein expressed by huRANKL- received BM cells of wt donors. We previously demonstrated KI mice is capable of inducing bone resorption, while being that this protocol efficiently reconstitutes cells of the J Mol Med (2018) 96:559–573 561 hematopoietic origin with a chimerism greater than 90% as Transthoracic Doppler echocardiography analyzed by flow cytometry, 4 weeks post-transplantation [15]. To avoid infections during the aplastic phase, irradiated Left ventricular (LV) function was non-invasively assessed animals were daily subcutaneously treated with an antibiotic 7 days and 3 weeks after MI in mice under isofluorane anesthe- (enrofloxacin, 10 mg/kg) over 7 days. Animals were left to sia using a 14-MHz linear transducer (Siemens S2000). Short- recover for 4 weeks before they were subjected to sham sur- axis M-mode recordings of LV were obtained from a left gery or to myocardial infarction. parasternal acoustic window. LV dimensions in systole and di- astole were measured through the largest diameter of the LV at the level of papillary muscles, and fractional shortening (FS) Myocardial infarction was calculated. Diastolic LV function was evaluated using the pulsed-wave Doppler recording of trans-mitral blood flow ve- Twenty-week-old male huRANKL-KI and wt mice were locities in the apical four-chamber view. A minimum of four anesthetized with ketamine/medetomidine (100/0.25 mg/ cardiac cycles were averaged for each measured parameter. kg i.p.) anesthesia. Endotracheal intubation was performed after disappearance of the paw pinch reflex. Animals were Central arterial and cardiac pressure measurement ventilated with a tidal volume of 200 μL and a frequency of 210 breathing cycles per minute using a small animal ventila- Pressure was assessed using a SPR-671NR pressure catheter tor (MiniVentTyp 845, Hugo Sachs Elektronik-Harvard (1.4F, Millar Instruments, Houston, TX, USA). The catheter Apparatus GmbH). Permanent ligation of the left descending was inserted into the ascending aorta via the carotid artery and coronary artery was performed after a left lateral thoracotomy. central arterial pressure was measured under 1.0% isofluorane Analgesic (buprenorphine 0.25 mg/kg s.c.) and antibiotic anesthesia. Thereafter, the catheter was advanced into the LV (enrofloxacin, 10 mg/kg s.c.) were injected for 4 and 5 days, for measurement of cardiac pressure parameters. Pressure was respectively. Sham animals underwent the same procedure but recorded over 5 min and traces were analyzed using LabChart without the arterial ligation. 7 software and a blood pressure module. Animals were killed 4 weeks after MI by exsanguination from the abdominal vena cava under ketamine/xylazine anesthe- sia (70/7 mg/kg i.p.). This time point was chosen in order to be Infarct size measurement able to document a robust decline in cardiac functional param- eters after MI. Serum samples, hearts, aortas, bones (femora, L4 Hearts excised at necropsy were washed from blood with PBS vertebra), and bone marrow were flash frozen and stored at − and fixed using 40% ethanol for at least 48 h. Dehydrated and 80 °C until assayed, or processed for histological analysis. paraffin-embedded hearts were sectioned at 5 μm thickness and werestainedwithMasson’s trichrome staining to visualize mus- cle and infarct tissue. At least nine standardized sections between Global and compartment-selective RANKL inhibition basis and apex of the heart were evaluated using AMIRA soft- by AMG161 ware. Infarct area was calculated as percentage of total LV area. Animals were randomized in a blinded fashion to the treatment with the anti-RANKL antibody AMG161 or an isotype control Gene expression analysis antibody (anti-keyhole limpet hemocyanin, KLH). Both AMG161 and the control antibody are humanized IgG1 anti- Total RNA from LV tissue was isolated using TRI Reagent® bodies. All antibodies were dissolved in A5su buffer containing Solution (Invitrogen). The concentration, purity, and quality 0.004% Tween 20 and were kindly provided by Amgen Inc. were determined spectrophotometrically (NanoDrop 2000; AMG161 or control antibody treatment (both at 10 mg/kg) Thermo Scientific) and by 2100 Bioanalyzer (Agilent started post-operatively and was continued twice weekly for Technologies). One microgram of RNA was reverse transcribed the duration of 4 weeks. Global inhibition of RANKL was (High Capacity cNDA Reverse Transcription Kit; Applied achieved by treatment of homozygous huRANKL-KI mice with Biosciences). Quantitative RT-PCR was performed on a Vii7 AMG161. To inhibit RANKL derived from cells of hematopoi- device (Applied Biosystems®) using the 5× Hot Firepol® Eva etic origin, wt mice were lethally irradiated, reconstituted with Green kit (Solis Biodyne). To exclude amplification of the ge- bone marrow from homozygous huRANKL-KI donors, and nomic DNA, primers were designed as exon spanning and their subsequently treated with AMG161. Vice versa, administration sequence is available upon request. A product melting curve of AMG161 to lethally irradiated homozygous huRANKL-KI analysis was performed to exclude primer dimerization and non- mice which were previously reconstituted with bone marrow specific amplification. All samples were measured in duplicate from wt mice, blocked RANKL from mesenchymal cell andexpressionvalueswerenormalizedtoornithinedecarbox- sources. The study design is graphically presented in Fig. 1. ylase antizyme-1 (Oaz1)mRNA. 562 J Mol Med (2018) 96:559–573 Fig. 1 Study design of global and compartment-selective RANKL inhibition. Irradiation and bone marrow transplantation (BMT) were performed in 16- week-old mice. Sham surgery or myocardial infarction (MI) were performed in 20-week-old mice Serum and urine biochemistry Immunofluorescence Serum phosphorus, calcium, alkaline phosphatase, and urinary Sections were prepared for staining as described for immunohis- creatinine were analyzed using a Cobas c111 analyzer (Roche). tochemistry. For co-staining of RANKL and CD3, primary anti- Total urinary deoxypyridinoline (DPD) concentrations were body against RANKL (rabbit polyclonal IgG, 1:100 in blocking assessed by a commercially available ELISA (MicroVue DPD solution, Santa Cruz Biotechnology) and anti-CD3 (Monoclonal EIA kit, Quidel) after acid hydrolysis. Rat IgG, 1:70 in blocking solution, R&D Systems) were incu- bated overnight at 4 °C. After washing, secondary anti-rabbit Alexa Fluor 555 (Molecular Probes, 1:500) and anti-rat Alexa pQCT analysis Fluor 594 (Thermo Fisher, 1:500) were incubated at room tem- perature for 60 min. Co-immunostaining of RANKL and cardiac Bone specimens were collected at necropsy and stored in 70% troponin T was performed in two steps. Firstly, primary anti- ethanol until analysis of mineral density using a XCT RANKL antibody was incubated overnight as described above. Research M+ pQCT device (Stratec Medizintechnik). Subsequently, biotinylated secondary anti-rabbit antibody (Vector, 1:400) was incubated at room temperature for 1 h, followed by incubation with Streptavidin-Alexa 546 conjugate Immunohistochemistry (Molecular Probes) for 30 min. After washing, primary rabbit anti-mouse cardiac troponin T antibody (St John’s Laboratory) Antigen retrieval was performed by heating the de-paraffinized and secondary anti-rabbit Alexa Fluor 647 (Thermo Fisher) an- cardiac sections to 100 °C for 15 min in citrate buffer (pH 6). tibodies were sequentially incubated for 1 h at room temperature. Sections were then treated with 0.1% Triton X-100 for 5 min at Immunostainings where primary antibodies were omitted served room temperature to permeabilize cell membranes, and for fur- as negative control. Co-immunostaining of IL-1ß and CD45 was ther 30 min with blocking solution containing 10% goat serum performed by incubating primary antibodies (rabbit anti-mouse and 0.02% Triton X in PBS to prevent unspecific antibody IL1ß, Abcam, 1:200; rat anti-moue CD45, BD Pharmingen, binding. Primary antibody against RANKL (rabbit polyclonal 1:40) overnight at 4 °C followed by incubation with secondary IgG, 1:200 in blocking solution, Santa Cruz Biotechnology), IL- anti-rabbit Alexa Fluor 555 and anti-rat Alexa Fluor 594 at room 1ß (goat polyclonal, 1:500 in blocking solution, R&D Systems), temperature for 60 min. Co-staining of IL-1ß and troponin T was and CD68 (rat monoclonal, 1:100, Bio Rad) were incubated performed in a two-step reaction as above described. Nuclei were overnight at 4 °C. After washing, secondary biotinylated anti- stained with DAPI (4′,6-diamidino-2-phenylindole). All sections bodies (Vector) were added and incubated for 60 min at room were imaged on a LSM 880 Airyscan confocal microscope. To temperature. Signal was developed by incubation with avoid cross talk, Alexa Fluor 555 and Alexa Fluor 594 fluoro- streptavidin-peroxidase (Vector) followed by 3-amino-9-ethyl chromes were exited at 514 and 633 nm, respectively. carbazol (AEC) or DAB staining. J Mol Med (2018) 96:559–573 563 Statistics Results Data are presented as mean ± SEM. Statistical analysis Cardiac ischemia/reperfusion injury activates was performed using GraphPad Prism 6. The data were RANK-RANKL-OPG axis in mice analyzed by two-sided t test (two groups) or one-way analysis of variance (ANOVA) followed by Bonferroni’s In line with data reported by Ueland et al. [7]inrats, wefound multiple comparison test (> 2 groups). P values of 0.05 that myocardial infarction (MI) activated the myocardial or less were considered significant. RANK/RANKL/OPG axis also in mice (Fig. 2). Left ventricular Fig. 2 Cardiac activation of the RANK-RANKL-OPG axis after myocardial infarction (MI) in mice. a Gene expression analysis in the left ventricle (LV), 4 weeks after MI (n =4 per group). *p < 0.05 vs. sham. b Immunohistochemical anti- RANKL staining in paraffin sec- tions of the LV in sham and MI mice, 4-weeks after MI. Upper left panel: negative (neg co) per- formed by omitting the primary anti-RANKL antibody. Upper right panel: sham control. Lower left panel: positive RANKL staining in cardiomyocytes (CM) and infiltrating cells in peri- ischemic LV region. Lower right panel: strong RANKL staining in remaining CM and infiltrating cells of the infarcted region. Bar = 100 μm. c Immunofluorescent co-staining of CD3 and RANKL in cardiac par- affin sections, 4 weeks after MI. RANKL co-localizes with some CD3 cells (red arrows), but is mainly expressed by CD3- negative cells with a fibroblast- like morphology (white arrows) in the infarct region. d Co- localization of RANKL and tro- ponin T in the infarct border re- gion, 4 weeks after MI. Bar = 20 μmin c and d 564 J Mol Med (2018) 96:559–573 Table 1 Basic characteristics, Sham Sham MI MI femoral BMD, blood parameters, and urinary deoxypyridinoline Co Ab AMG161 Co Ab AMG161 excretion after global RANKL inhibition in huRANKL-KI mice, Body weight (g) 29.7 ± 1.1 29.1 ± 0.7 30.6 ± 0.6 30.2 ± 0.6 4 weeks after surgery Lung/body weight ratio (mg/g) 5.1 ± 0.1 5.3 ± 0.1 5.2 ± 0.2 5.0 ± 0.1 Heart/body weight ratio (mg/g) 4.1 ± 0.1 4.2 ± 0.1 4.7 ± 0.2* 4.8 ± 0.1 3 §§ Femoral metaph. total BMD (mg/cm ) 478 ± 4 527 ± 16* 469 ± 10 523 ± 17 Serum P (mmol/L) 2.53 ± 0.12 2.48 ± 0.21 2.94 ± 0.21 2.99 ± 0.31 Serum Ca (mmol/L) 2.11 ± 0.04 2.02 ± 0.04 2.14 ± 0.06 2.13 ± 0.05 Serum ALP (U/L) 41.1 ± 2.1 35.2 ± 3.2 41.6 ± 3.7 32.5 ± 0.8 Urinary DPD/creatinine (nM/mM) 8.2 ± 1.9 1.6 ± 1.4 10.7 ± 3.5 2.9 ± 1.1 n =5–10 per group *p < 0.05 vs. sham + control antibody (Co Ab) p <0.05 vs. sham + AMG161 p <0.05 vs. MI + Co Ab §§ p <0.01 vs. MI + Co Ab (LV) mRNA abundance of Rankl significantly increased, (Fig. 3). Global inhibition of RANKL after MI by treatment of 4 weeks after MI (Fig. 2a). OPG and Rank gene expression also huRANKL-KI mice with AMG161 did not have any statistically tended to increase in the LVafter MI, but this effect did not reach significant effects on survival, heart/body weight ratio, infarct statistical significance (Fig. 2a). Immunohistochemical analysis size, or cardiac functional parameters compared to isotype control showed that RANKL expression was mainly induced in antibody-treated huRANKL-KI MI mice (Fig. 3 and Table 1). cardiomyocytes adjacent to the infarct region as well as in the cellular infiltrate within the infarct (Fig. 2b). Further analysis of Hematopoietic, but not mesenchymal, RANKL the infarct region by immunofluorescent imaging revealed that inhibition reduced pro-inflammatory cytokine RANKL co-localized with some CD3-positive T lymphocytes, production in the left ventricle of MI mice but more abundant RANKL expression was present in fibroblast-like cells, cardiomyocytes, and blood vessels (Fig. Depending on the cell type, RANKL can exert both pro- and 2c, d and Suppl. Fig. S1). These findings suggest that cells other anti-inflammatory effects [11, 16] and the lack of therapeutic than lymphocytes are the main RANKL source in the infarcted effect of global RANKL inhibition may be caused by oppos- myocardium. Cardiomyocytes in remote myocardium did not ing RANKL actions in the myocardium vs. the cellular infil- express RANKL (Suppl. Fig. S1). trate. Thus, we next asked the question whether selective blockade of hematopoietic and mesenchymal RANKL might have positive therapeutic effects. Global RANKL inhibition by AMG161 lacks beneficial effect in murine myocardial infarction model To selectively block hematopoietic RANKL, we lethally irradiated wt mice and reconstituted them with bone marrow The exact pathophysiological role of increased cardiac RANKL from huRANKL-KI mice to replace their hematopoietic com- partment with cells responsive to AMG161, using a previous- after cardiac ischemia is not known. Because RANKL was re- ported to promote inflammation and matrix degradation [7, 9], ly established protocol [15]. To prove that the expected RANKL form was expressed by cells of hematopoietic origin we hypothesized that inhibition of RANKL could improve the post-infarct outcome after MI. To test this hypothesis, we in- duced MI in huRANKL-KI mice and subsequently treated the Fig. 3 Global RANKL inhibition by AMG161 does not influence mice with the monoclonal anti-human RANKL antibody outcome in huRANKL-KI mice after MI. a Kaplan-Meier survival curves AMG161. Biological activity ofAMG161was confirmedby after MI (n =22–24 per group). b Infarct size measured by planimetry after Masson’s trichrome staining (n =11–13 per group). c Representative significantly increased femoral BMD, as well as suppressed Masson’s trichrome-stained cardiac cross-sections, 4-weeks after sham or serum alkaline phosphatase and urinary DPD excretion in MI surgery. d Cardiac function and LV diameters measured by echocar- AMG161-treated huRANKL-KI mice (Table 1). However, diography (n =10–24 per group). e Representative M-mode echocardio- treatment with AMG161 had no effect on calcium or phosphate grams, obtained 3 weeks after MI. f Cardiac parameters measured by intra-cardiac catheterization (n =5–14 per group). LVIDd left ventricle serum levels (Table 1). internal diameter in diastole, LVIDs left ventricle internal diameter in Induction of MI led to a significant deterioration of cardiac systole, MAP mean arterial pressure, dP/dt maximal rate of left ventricle function in huRANKL-KI mice, as evidenced by reduced frac- pressure rise. *p < 0.05 vs. sham + control antibody (Co Ab); **p <0.01 tional shortening, dilation of the LV in both systole and diastole, vs. sham + Co Ab; ***p < 0.001 vs. sham + Co Ab; p < 0.05 vs. sham + ### AMG161; p < 0.001 vs. sham + AMG161 and diminished contractile cardiac function as assessed by dP/dt J Mol Med (2018) 96:559–573 565 after bone marrow transfer, we analyzed splenic Rankl gene Rankl in their spleens, respectively (Suppl. Fig. S2). In con- expression in non-irradiated mice as well as after bone marrow trast, after lethal irradiation and vice versa reconstitution, chi- transfer. As expected, non-irradiated wt mice did not express meric Rankl gene was abundantly expressed in the spleen of chimeric Rankl, and huRANKL-KI mice did not express wt reconstituted wt mice, whereas reconstituted huRANKL-KI 566 J Mol Med (2018) 96:559–573 mice expressed the mouse wt Rankl gene in their spleens Surprisingly, hematopoietic RANKL inhibition improved (Suppl. Fig. S2), indicating successful exchange of the hema- post-infarct survival as well as cardiac function as shown by a topoietic compartment. significant rise in fractional shortening, lower diastolic and J Mol Med (2018) 96:559–573 567 Fig. 4 Inhibition of RANKL derived from hematopoietic (left column) but not from mesenchymal cellular sources (right column) improves survival and cardiac function post-MI. Kaplan-Meier sur- vival curves after MI (n =22–31 per group), echocardiographic pa- rameters measured 3 weeks after surgery (n =16–17 per group), and infarct size measured by planimetry after Masson’s trichrome staining (n =8–14 per group) in sham and MI huRANKL-KI mice treated with AMG161 or control antibody (Co Ab), 4 weeks after MI. LVIDd left ventricular internal diameter in diastole, LVIDs left ventricular inter- nal diameter in systole. **p < 0.01 and ***p < 0.001 vs. sham + Co # ### § Ab; p <0.05 and p < 0.001 vs. sham + AMG161; p <0.05 and §§ p < 0.01vs. MI +CoAb systolic endocardial LV diameters, and enhanced LV contractil- ity in AMG161 vs. control Ab-treated mice (Fig. 4 and Table 2). In contrast, inhibition of RANKL derived from the mesenchy- mal cell compartment in AMG161-treated huRANKL-KI mice reconstituted with wt bone marrow did not have any effect on survival or post-infarct cardiac function (Fig. 4 and Table 2). Infarct area was not affected by RANKL inhibition in any of the investigated groups (Fig. 4), suggesting that mechanisms other than those regulating cardiomyocyte cell death are responsible for the protective effects on survival and cardiac function seen after inhibition of hematopoietic cell-derived RANKL. To shed more light on the intriguing finding that inhibition of hematopoietic, but not of mesenchymal or global, RANKL had these beneficial effects after MI, we measured inflammatory cell infiltration and the mRNA abundance of pro-inflammatory cy- tokines in the LV. Because hematopoietic RANKL inhibition protected against post-ischemic LV chamber dilation, we hy- pothesized that inflammatory signaling pathways initiated by RANKL secreted from cells of hematopoietic origin may drive adverse post-ischemic remodeling of the LV. We first examined whether RANKL inhibition altered the post-ischemic infiltration with macrophages or lymphocytes. Immunohistochemical anal- Fig. 5 Global RANKL inhibition slightly reduces CD68 macrophage ysis of CD68-positive macrophages in the infarct region re- abundance in the left ventricle, 4 weeks after MI. a Representative images vealed a trend for reduced macrophage infiltration after global, of immunohistochemical stainings of CD68-expressing macrophages in the mesenchymal, and hematopoietic RANKL inhibition (Fig. 5). infarct region after global, mesenchymal, and hematopoietic RANKL inhi- However, this effect reached statistical significance only after bition. Bar = 50 μm. b Quantification of CD68 macrophages in the whole infarct region, presented as cell number per image. n=4–5 mice per group. global RANKL inhibition (Fig. 5). In contrast, the LV mRNA *p<0.05 vs. MI + Co Ab abundance of the lymphocyte-specific genes CD3 remained Table 2 Hemodynamic variables Hematopoietic RANKL inhibition Mesenchymal RANKL inhibition measured invasively using intra- cardiac catheter, 4 weeks after MI MI + Co Ab MI + AMG161 MI + Co Ab MI + AMG161 Systolic P (mmHg) 86.95 ± 7.5 93.18 ± 2.0 80.03 ± 3.3 84.57 ± 3.6 Diastolic P (mmHg) 60.17 ± 4.9 64.81 ± 1.7 53.87 ± 3.1 59.78 ± 4.0 MAP (mmHg) 76.57 ± 5.9 78.15 ± 1.6 66.49 ± 3.1 72.30 ± 4.0 dP/dt (mmHg/s) 4416 ± 546 6744 ± 571* 4912 ± 281 5178 ± 602 max EDP (mmHg) 8.76 ± 1.9 8.19 ± 1.7 8.27 ± 2.4 9.46 ± 3.1 Tau (ms) 2.33 ± 0.3 1.91 ± 0.3 2.26 ± 0.4 1.78 ± 0.2 n =4–6per group MAP mean arterial pressure, dP/dt maximal rate of left ventricle pressure rise, EDP end-diastolic pressure, Tau left ventricular relaxation time constant *p < 0.05 vs. MI + Co Ab 568 J Mol Med (2018) 96:559–573 unchanged after global, mesenchymal, or hematopoietic inhibition in surviving cardiomyocytes within the ischemic RANKL inhibition (Fig. 6). Interestingly, however, the post- zone. To characterize further the cellular source of IL-1ß, we ischemic rise in LV gene expression of IL-1ß and Mmp-9,but performed co-immunostaining of IL-1ß with CD45 and cardiac not of TNFα, was significantly reduced after hematopoietic troponin T (Fig. 8 and Suppl. Fig. S5 and S6). Although some RANKL blockade (Fig. 6 and Suppl. Fig. S3). In contrast, mes- CD45 cells in the infarct region expressed IL-1ß, the majority enchymal RANKL inhibition did not change the LVexpression of IL-1ß-expressing cells had a fibroblast-like morphology (Fig. pattern of pro-inflammatory genes after MI, and global RANKL 8 and Suppl. Fig. S5). Troponin-positive cardiomyocytes in the inhibition even enhanced IL-1ß mRNA expression in the post- border zone of the infarct also clearly expressed IL-1ß (Suppl. ischemic LV (Fig. 6). Intriguingly, mRNA expression of the Fig. S6), but their IL-1ß expression, in contrast to non- resolving M2 macrophage markers Mrc-1, Arg-1,and Ym-1 cardiomyocyte cells (Fig. 8), was not downregulated by hema- was profoundly reduced in the post-ischemic LV after hemato- topoietic RANKL inhibition. Altogether, these findings suggest poietic RANKL blockade, but not after global or mesenchymal that RANKL derived from scar-infiltrating cells of hematopoi- RANKL inhibition (Suppl. Fig. S4). In addition, the MI-induced etic origin is an important pro-inflammatory stimulus whose upregulation of Arg-1 and Ym-1 was lower in huRANKL-KI inhibition can be beneficial for post-ischemic recovery. mice relative to mice with a wt background (Suppl. Fig. S4). To further characterize the changes in IL-1ß protein expres- sion induced by RANKL blockade, we used immunohisto- Discussion chemical analysis of the infarct region. As shown in Fig. 7, inhibition of global or hematopoietic RANKL decreased IL-1ß RANKL plays a pivotal role in bone remodeling and in immu- expression in the non-cardiomyocyte cell compartment of the nity [3, 17, 18] and may also be an important signaling molecule scar, whereas IL-1ß expression was not influenced by RANKL in diseases affecting the cardiovascular system. We show here Fig. 6 Hematopoietic but not global or mesenchymal RANKL inhibition hematopoietic RANKL inhibition. Gene expression is presented as fold downregulates IL-1β mRNA expression in the left ventricle after MI. Left increase compared to sham + control antibody (Co Ab) group. n =3–9per ventricular expression of IL-1ß (upper panels), TNFα (middle panels), group. *p < 0.05 and ***p < 0.001 vs. sham + Co Ab; p < 0.05 vs. sham and CD3 mRNA (lower panels) after global, mesenchymal, or + AMG161; p <0.05 vs. MI +Co Ab J Mol Med (2018) 96:559–573 569 Fig. 7 Hematopoietic and global but not mesenchymal RANKL inhibition suppresses IL-1β pro- tein expression in infiltrating cells post-MI. a Representative images of immunohistochemical anti-IL- 1ß staining in the infarct region, 4 weeks after MI. Bar = 100 μm. b Semi-quantitative analysis of IL-1ß expression in cardiomyocytes and non- cardiomyocyte cells. IL-1ß stain- ing intensity is shown as the mean of intensity ratios between AMG161 and control antibody (Co Ab) treatment with 95% confidence intervals. Samples stained in the same experiment were paired (Co Ab and AMG161-treated). *p <0.05 by ratio paired t-test that experimental MI induced LV RANKL expression in mice. after MI in huRANKL-KI mice. Reason for this discrepancy Global RANKL inhibition or inhibition of RANKL derived may lay in reduced in vivo activity of chimeric RANKL in from mesenchymal cellular sources using a monoclonal anti- transgenic animals compared to that of murine RANKL in wt RANKL antibody did not significantly alter cardiac function mice. Although chimeric RANKL and wt murine RANKL have post-MI, whereas specific inhibition of RANKL derived from the same affinity to the murine RANK receptor, the hematopoietic cellular sources improved post-ischemic cardiac osteoclastogenic potency of chimeric RANKL was slightly low- function, reduced mortality, and downregulated post-ischemic er than that of murine wt RANKL [13]. It is conceivable that the production of inflammatory cytokines. lack of myocardial MMP-9 upregulation and the reduced upreg- In the current study, we used a monoclonal antibody ulation of some M2 macrophage markers after MI in AMG161 to block RANKL. Because AMG161 selectively huRANKL-KI mice may be explained by this fact. However, blocks human RANKL, but not murine RANKL, we used the promoter and splicing regions of the Rankl gene are intact huRANKL-KI mice expressing a humanized RANKL protein in huRANKL-KI mice, and thus, regulation of RANKL levels in this study. Although hematopoietic RANKL inhibition was during pathological conditions such as MI is expected to be beneficial for cardiac function after MI in WT mice, global comparable to that of wt mice. Another possibility for the dis- RANKL inhibition did not significantly change cardiac function crepant findings after global and hematopoietic RANKL 570 J Mol Med (2018) 96:559–573 Fig. 8 Immunofluorescent co- staining of IL-1β andCD45in the infarct region after global, hema- topoietic, and mesenchymal RANKL inhibition. IL-1β is co- expressed by some CD45 cells, but abundantly expressed in fibroblast-like cells within the in- farct region, 4 weeks post-MI. Note the striking reduction of IL- 1β staining after hematopoietic RANKL inhibition (right middle panel). Representative images of n = 2 mice per group. Bar = 20 μm inhibition may be the fact that global RANKL inhibition was Timing of RANKL signaling and its cellular source may be performed in non-irradiated huRANKL-KI mice, whereas hema- important determinants of its effect during/after ischemic injury. topoietic RANKL inhibition was performed in irradiated and The tissue response to increased RANKL levels during acute reconstituted mice. However, since the inflammatory response ischemia may be different from prolonged RANKL signaling post-MI was almost identical in irradiated (mesenchymal during tissue repair and remodeling. Signaling downstream of RANKL inhibition) and non-irradiated huRANKL-KI mice the RANK receptor in the myocardium probably involves (global RANKL inhibition), it is highly unlikely that this was a NF-κB[9]. NF-κB activation during acute injury may promote major influencing factor. A third possibility is that the hemato- cell survival and suppress apoptotic signaling [8]. Indeed, poietic output of the bone marrow is regulated by RANKL- models of brain and liver ischemia demonstrated an important driven osteoclastogenesis [19, 20]. Therefore, treatment of non- role of acute RANKL signaling in cell survival and limitation irradiated and irradiated huRANKL-KI mice with AMG161 may of the final infarct size [11, 12]. In contrast, a recent study modulate inflammatory responses through inhibition of osteo- reported beneficial effects of RANKL inhibition during cardiac clastogenesis and altered output of bone marrow-derived cells. ischemia on infarct size, 24 h after reperfusion [16]. However, However, we found no major and consistent differences in lym- in the current study, infarct size after global RANKL inhibition phocyte or macrophage infiltration of the infarct between was not changed, 4 weeks after MI. Thus, we can exclude huRANKL-KI and wt mice, arguing against a major difference increased vulnerability to ischemia after RANKL inhibition. in the bone marrow output of inflammatory/anti-inflammatory On the other hand, prolonged NF-κB activation through cells. RANKL may promote inflammation and adverse cardiac J Mol Med (2018) 96:559–573 571 remodeling [8]. We hypothesize that during post-ischemic re- RANKL reduces transcription of the pro-inflammatory cytokine modeling, the cellular source of RANKL determines the effect IL-1β in the myocardium, and more specifically in scar- of RANKL in the ischemic myocardium. Our results support infiltrating inflammatory cells. This is in line with our finding the notion that RANKL produced by cells of hematopoietic of improved cardiac function in mice with reduced IL-1β levels origin, but not by cardiomyocytes, contributes to maladaptive after hematopoietic RANKL blockade. processes, deteriorating cardiac function after myocardial in- RANKL signaling receptor RANK is present in both neutro- farction. In this context, it is tempting to speculate that cells of phils [27] and macrophages [28, 29], which makes both cellular hematopoietic origin, in contrast to cells of mesenchymal ori- population responsive to RANKL. Role of RANKL in neutro- gin, do not produce enough OPG to inhibit excess RANKL phil infiltration and MMP-9 secretion was recently demonstrated signaling. This hypothesis is supported by findings that osteo- early after MI [16]. However, in our long-term study, it is more blasts but not activated lymphocytes secrete OPG in the culture probable that beneficial effects seen after hematopoietic RANKL medium, and that OPG mRNA was not detected in human T inhibition are mediated by macrophages which infiltrate ische- lymphocytes or monocytes [5]. Hence, the local increase in mic myocardium long term after MI [30]. Although we did not RANKL secretion from BM-derived cells may be a driving observe changes in macrophage number in the infarct region force of inflammation in the heart after MI, whereas RANKL after hematopoietic RANKL blockade, markers of the reparative produced by cardiomyocytes may be inhibited by concurrently M2 subset were profoundly reduced in the myocardium after increased cardiomyocytic OPG secretion. This notion may ex- hematopoietic blockade of RANKL, 4 weeks post-MI. This find- plain why inhibition of hematopoietic RANKL has beneficial ing may suggest that inhibition of hematopoietic RANKL may effects post-MI, whereas selective inhibition of mesenchymal alter the time course of resolution of inflammation. Further ex- RANKL does not influence post-ischemic cardiac remodeling. periments are needed to address the question of how RANKL A puzzling observation in our study was that post-ischemic inhibition affects resolution of inflammation after MI. LV IL-1ß mRNA abundance was promoted after global Our finding that global RANKL inhibition did not increase RANKL inhibition, but reduced when only hematopoietic infarct size or deteriorate cardiacfunctionafter MI mayhave RANKL was inhibited. There are several possible explanations important clinical implications. RANKL is the molecular target for this finding. Global RANKL inhibition using AMG161 may behind one of the most effective osteoporosis treatments today, leave larger amounts of OPG available for other signaling path- the treatment with the anti-RANKL antibody denosumab. The ways, including binding of TRAIL. Furthermore, it was reported typical osteoporosis patient is of advanced age and usually at that RANKL may actually reduce both innate [11, 21]and adap- high risk to suffer co-morbidities, and there is a rising awareness tive immune responses [22]insomemodels. Forexample,ithas that cardiovascular diseases and osteoporosis might be been shown that RANKL expression in keratinocytes can drive pathophysiologically linked diseases. A large 3-year placebo- controlled trial of denosumab in postmenopausal women with formationofregulatoryT cells[22]. In analogy, RANKL ex- pression on cardiomyocytes may downregulate the inflammato- low bone mass showed no treatment-related effects on the inci- ry response post-MI. However, since inhibition of mesenchymal dence of cardiovascular events, coronary heart disease, or atrial RANKL had no influence on LV function or gene expression, it fibrillation, with a trend towards reduced all-cause mortality in is unlikely that the latter scenarioistrue.Itmayalsobeimportant the denosumab arm [31]. However, this clinical trial population, to note in this context that the increased LV IL-1ß mRNA ex- postmenopausal women, was not selected for increased CV risk pression observed after global RANKL inhibition in factors, and it may therefore be reassuring that the current study huRANKL-KI MI mice was not evident in immunohistochem- showed no untoward effects of global RANKL inhibition on ical analyses of the scar region (Figs. 7 and 8). Therefore, it is post-MI survival or cardiac function. unclear whether the increased IL-1ß gene transcription after Our study has shown that the upregulation of RANKL in the global RANKL inhibition fully translates into augmented IL- post-ischemic myocardium mainly involves fibroblast-like 1ß secretion at the protein level. On the other hand, the down- cells, blood vessels, and surviving cardiomyocytes. We and regulation of IL-1ß mRNA abundance in the LV after hemato- others showed earlier that a substantial amount of endothelial poietic RANKL inhibition was confirmed at the protein level by cells and fibroblasts/myofibroblasts in the heart is donor-derived immunohistochemistry. after bone marrow transplantation [14, 32]. Therefore, it is con- Secreted IL-1β has a critical role in the post-ischemic remod- ceivable that some beneficial effects after hematopoietic eling by stimulating inflammatory cell accumulation, inflamma- RANKL blockade are attributable to endothelial RANKL tory cytokine production, myofibroblast differentiation, extracel- blockade. Future studies need to address the question how the lular matrix degradation, and collagen production [23, 24]. availability of RANKL/RANK molecules on specific cell types Moreover, IL-1β depresses cardiac contractility by reducing regulates cell-cell interactions and the immune/inflammatory 2+ L-type Ca currents in neonatal and adult ventricular tissue response in the course of tissue repair after MI. cardiomyocytes [25], and by inhibiting their β-adrenergic re- Improved insight into these mechanisms may eventually open up new possibilities for the treatment of MI patients. sponse [26]. We show here that inhibition of hematopoietic 572 J Mol Med (2018) 96:559–573 Acknowledgements Open access funding provided by University of Dysregulated osteoprotegerin/RANK ligand/RANK axis in clinical Veterinary Medicine Vienna. We thank Alexandra Petric for excellent and experimental heart failure. Circulation 111:2461–2468 technical assistance, and Ute Zeitz and Miriam Kleiter for the help with 8. Gordon JW, Shaw JA, Kirshenbaum LA (2011) Multiple facets of the animal experiments. NF-κB in the heart: to be or not to NF-κB. Circ Res 108:1122–1132 9. Ock S, Ahn J, Lee SH, Park H, Son JW, Oh JG, Yang DK, Lee WS, Kim HS, Rho J, Oh GT, Abel ED, Park WJ, Min JK, Kim J (2012) Author Contributions S.S., L.C.H, P.J.K, and R.G.E. conceived and de- Receptor activator of nuclear factor-B ligand is a novel inducer of signed the study. S.S., O.A., K.F., S.H., N.L., U.R., S.S., and C.B. per- myocardial inflammation. Cardiovasc Res 94:105–114 formed the experiments and analysed the data. S.S., L.C.H, P.J.K, and 10. Kiechl S, Wittmann J, Giaccari A, Knoflach M, Willeit P, Bozec A, R.G.E. wrote the manuscript. R.G.E. accepts responsibility for the integ- Moschen AR, Muscogiuri G, Sorice GP, Kireva T, Summerer M, rity and validity of the data collected and analysed. Wirtz S, Luther J, Mielenz D, Billmeier U, Egger G, Mayr A, Oberhollenzer F, Kronenberg F, Orthofer M, Penninger JM, Funding information This work was supported by a grant from the Meigs JB, Bonora E, Tilg H, Willeit J, Schett G (2013) Blockade Austrian Science Fund (FWF P 21904-B11) to R.G.E. of receptor activator of nuclear factor-κB (RANKL) signaling im- proves hepatic insulin resistance and prevents development of dia- betes mellitus. Nat Med 19:358–363 Compliance with ethical standards 11. Shimamura M, Nakagami H, Osako MK, Kurinami H, Koriyama H, Zhengda P, Tomioka H, Tenma A, Wakayama K, Morishita R (2014) All animal procedures were undertaken in accordance with current guide- OPG/RANKL/RANK axis is a critical inflammatory signaling system in lines for animal care and welfare and were approved by the Ethical ischemic brain in mice. Proc Natl Acad Sci U S A 111:8191–8196 Committees of the University of Veterinary Medicine Vienna and of the 12. Sakai N, Van Sweringen HL, Schuster R et al (2012) Receptor acti- Austrian Federal Ministry of Science, Research and Economy. vator of nuclear factor-κB ligand (RANKL) protects against hepatic ischemia/reperfusion injury in mice. Hepatology 55:888–897 Conflict of interest PJK is a former Amgen employee who owns 13. Kostenuik PJ, Nguyen HQ, McCabe J, Warmington KS, Kurahara Amgen stock and serves as a consultant and contract medical writer for C, Sun N, Chen C, Li L, Cattley RC, van G, Scully S, Elliott R, Amgen. The other authors declare no conflicts of interest. Grisanti M, Morony S, Tan HL, Asuncion F, Li X, Ominsky MS, Stolina M, Dwyer D, Dougall WC, Hawkins N, Boyle WJ, Simonet WS, Sullivan JK (2009) Denosumab, a fully human monoclonal Open Access This article is distributed under the terms of the Creative antibody to RANKL, inhibits bone resorption and increases BMD Commons Attribution 4.0 International License (http:// in knock-in mice that express chimeric (murine/human) RANKL. J creativecommons.org/licenses/by/4.0/), which permits unrestricted use, Bone Miner Res Off J Am Soc Bone Miner Res 24:182–195 distribution, and reproduction in any medium, provided you give 14. Odörfer KI, Egerbacher M, Unger NJ, Weber K, Jamnig A, appropriate credit to the original author(s) and the source, provide a link Lepperdinger G, Kleiter M, Sandgren EP, Erben RG (2011) to the Creative Commons license, and indicate if changes were made. Hematopoietic bone marrow cells participate in endothelial, but not epithelial or mesenchymal cell renewal in adult rats. J Cell Mol Med 15:2232–2244 15. Streicher C, Heyny A, Andrukhova O, Haigl B, Slavic S, Schüler C, References Kollmann K, Kantner I, Sexl V, Kleiter M, Hofbauer LC, Kostenuik PJ, Erben RG (2017) Estrogen regulates bone turnover by targeting RANKL expression in bone lining cells. Sci Rep 7:6460 1. Kostis WJ, Deng Y, Pantazopoulos JS, Moreyra AE, Kostis JB, for 16. Carbone F, Crowe LA, Roth A, Burger F, Lenglet S, the Myocardial Infarction Data Acquisition System (MIDAS14) Braunersreuther V, Brandt KJ, Quercioli A, Mach F, Vallée JP, Study Group (2010) Trends in mortality of acute myocardial infarc- Montecucco F (2016) Treatment with anti-RANKL antibody re- tion after discharge from the hospital. Circ Cardiovasc Qual duces infarct size and attenuates dysfunction impacting on Outcomes 3:581–589 neutrophil-mediated injury. J Mol Cell Cardiol 94:82–94 2. Roger VL (2010) Myocardial infarction outcomes: Bthe times, they 17. Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, are a-changin...^. Circ Cardiovasc Qual Outcomes 3:568–570 Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, 3. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Galibert L (1997) A homologue of the TNF receptor and its ligand Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, enhance T-cell growth and dendritic-cell function. Nature 390:175–179 Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, . Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ Morony S, Oliveira-dos-Santos AJ, van G, Itie A, Khoo W, (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast Wakeham A, Dunstan CR, Lacey DL, Mak TW, Boyle WJ, differentiation and activation. Cell 93:165–176 Penninger JM (1999) OPGL is a key regulator of osteoclastogene- 4. Kiechl S, Schett G, Schwaiger J, Seppi K, Eder P, Egger G, Santer sis, lymphocyte development and lymph-node organogenesis. P, Mayr A, Xu Q, Willeit J (2007) Soluble receptor activator of Nature 397:315–323 nuclear factor-kappa B ligand and risk for cardiovascular disease. 19. Kollet O, Dar A, Shivtiel S, Kalinkovich A, Lapid K, Sztainberg Y, Circulation 116:385–391 Tesio M, Samstein RM, Goichberg P, Spiegel A, Elson A, Lapidot T 5. Sandberg WJ, Yndestad A, Øie E et al (2006) Enhanced T-cell ex- (2006) Osteoclasts degrade endosteal components and promote mobi- pression of RANK ligand in acute coronary syndrome: possible role in lization of hematopoietic progenitor cells. Nat Med 12:657–664 plaque destabilization. Arterioscler Thromb Vasc Biol 26:857–863 20. Cho K-A, Joo S-Y, Han H-S, Ryu KH, Woo SY (2010) Osteoclast 6. Bucay N, Sarosi I, Dunstan CR, Morony S, Tarpley J, Capparelli C, activation by receptor activator of NF-kappaB ligand enhances the Scully S, Tan HL, Xu W, Lacey DL, Boyle WJ, Simonet WS (1998) mobilization of hematopoietic progenitor cells from the bone mar- Osteoprotegerin-deficient mice develop early onset osteoporosis row in acute injury. Int J Mol Med 26:557–563 and arterial calcification. Genes Dev 12:1260–1268 21. Maruyama K, Takada Y, Ray N et al (2006) Receptor activator of NF- 7. Ueland T, Yndestad A, Øie E, Florholmen G, Halvorsen B, Frøland kappa B ligand and osteoprotegerin regulate proinflammatory cytokine SS, Simonsen S, Christensen G, Gullestad L, Aukrust P (2005) production in mice. J Immunol (Baltimore, Md 1950) 177:3799–3805 J Mol Med (2018) 96:559–573 573 22. Loser K, Mehling A, Loeser S, Apelt J, Kuhn A, Grabbe S, Staub C, Ratib O, Dallegri F, Schindler TH, Montecucco F (2012) Receptor activator of NF-κB ligand (RANKL) increases the release Schwarz T, Penninger JM, Beissert S (2006) Epidermal RANKL controls regulatory T-cell numbers via activation of dendritic cells. of neutrophil products associated with coronary vulnerability. Nat Med 12:1372–1379 Thromb Haemost 107:124–139 23. Sager HB, Heidt T, Hulsmans M, Dutta P, Courties G, Sebas M, 28. Hsu H, Lacey DL, Dunstan CR, Solovyev I, Colombero A, Timms Wojtkiewicz GR, Tricot B, Iwamoto Y, Sun Yet al (2015) Targeting E, Tan HL, Elliott G, Kelley MJ, Sarosi I, Wang L, Xia XZ, Elliott interleukin-1β reduces leukocyte production after acute myocardial R, Chiu L, Black T, Scully S, Capparelli C, Morony S, Shimamoto infarction. Circulation. https://doi.org/10.1161/ G, Bass MB, Boyle WJ (1999) Tumor necrosis factor receptor CIRCULATIONAHA.115.016160 family member RANK mediates osteoclast differentiation and acti- 24. Bujak M, Dobaczewski M, Chatila K, Mendoza LH, Li N, Reddy vation induced by osteoprotegerin ligand. Proc Natl Acad Sci U S A A, Frangogiannis NG (2008) Interleukin-1 receptor type I signaling 96:3540–3545 critically regulates infarct healing and cardiac remodeling. Am J 29. Antonios JK, Yao Z, Li C, Rao AJ, Goodman SB (2013) Pathol 173:57–67 Macrophage polarization in response to wear particles in vitro. 25. El Khoury N, Mathieu S, Fiset C (2014) Interleukin-1β reduces L- Cell Mol Immunol 10:471–482 type Ca2+ current through protein kinase Cϵ activation in mouse 30. Swirski FK, Nahrendorf M (2013) Leukocyte behavior in atheroscle- heart. J Biol Chem 289:21896–21908 rosis, myocardial infarction, and heart failure. Science 339:161–166 26. Gulick T, Chung MK, Pieper SJ, Lange LG, Schreiner GF (1989) 31. Cummings SR, Martin JS, McClung MR et al (2009) Denosumab Interleukin 1 and tumor necrosis factor inhibit cardiac myocyte beta- for prevention of fractures in postmenopausal women with osteo- adrenergic responsiveness. Proc Natl Acad Sci U S A 86:6753–6757 porosis. N Engl J Med 361:756–765 27. Quercioli A, Mach F, Bertolotto M, Lenglet S, Vuilleumier N, 32. Ogawa M, LaRue AC, Drake CJ (2006) Hematopoietic origin of Galan K, Pagano S, Braunersreuther V, Pelli G, Pistoia V, Bianchi fibroblasts/myofibroblasts: its pathophysiologic implications. G, Cittadini G, Viviani GL, Pende A, Roux-Lombard P, Thomas A, Blood 108:2893–2896 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Molecular Medicine Springer Journals

Selective inhibition of receptor activator of NF-κB ligand (RANKL) in hematopoietic cells improves outcome after experimental myocardial infarction

Free
15 pages

Loading next page...
 
/lp/springer_journal/selective-inhibition-of-receptor-activator-of-nf-b-ligand-rankl-in-q73LR2ZOnH
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Biomedicine; Molecular Medicine; Human Genetics; Internal Medicine
ISSN
0946-2716
eISSN
1432-1440
D.O.I.
10.1007/s00109-018-1641-x
Publisher site
See Article on Publisher Site

Abstract

The RANK (receptor activator of nuclear factor κB)/RANKL (RANK ligand)/OPG (osteoprotegerin) axis is activated after myocardial infarction (MI), but its pathophysiological role is not well understood. Here, we investigated how global and cell compartment-selective inhibition of RANKL affects cardiac function and remodeling after MI in mice. Global RANKL inhibition was achieved by treatment of human RANKL knock-in (huRANKL-KI) mice with the monoclonal antibody AMG161. huRANKL-KI mice express a chimeric RANKL protein wherein part of the RANKL molecule is humanized. AMG161 inhibits human and chimeric but not murine RANKL. To dissect the pathophysiological role of RANKL derived from hematopoietic and mesenchymal cells, we selectively exchanged the hematopoietic cell compart- ment by lethal irradiation and across-genotype bone marrow transplantation between wild-type and huRANKL-KI mice, exploiting the specificity of AMG161. After permanent coronary artery ligation, mice were injected with AMG161 or an isotype control antibody over 4 weeks post-MI. MI increased RANKL expression mainly in cardiomyocytes and scar- infiltrating cells 4 weeks after MI. Only inhibition of RANKL derived from hematopoietic cellular sources, but not global or mesenchymal RANKL inhibition, improved post-infarct survival and cardiac function. Mechanistically, hema- topoietic RANKL inhibition reduced expression of the pro-inflammatory cytokine IL-1ß in the cardiac cellular infiltrate. In conclusion, inhibition of RANKL derived from hematopoietic cellular sources is beneficial to maintain post-ischemic cardiac function by reduction of pro-inflammatory cytokine production. Key messages & Experimental myocardial infarction (MI) augments cardi- ac RANKLexpressioninmice. Olena Andrukhova passed away before the submission of the final & RANKL expression is increased in cardiomyocytes and version of this manuscript. scar-infiltrating cells after MI. Electronic supplementary material The online version of this article & Global or mesenchymal cell RANKL inhibition has no (https://doi.org/10.1007/s00109-018-1641-x) contains supplementary influence on cardiac function after MI. material, which is available to authorized users. & Inhibition of RANKL derived from hematopoietic cells improves heart function post-MI. * Reinhold G. Erben & Hematopoietic RANKL inhibition reduces pro- Reinhold.Erben@vetmeduni.ac.at inflammatory cytokines in scar-infiltrating cells. Institute of Physiology, Pathophysiology and Biophysics, Department of Biomedical Research, University of Veterinary . . . Keywords Myocardial infarction RANKL Inflammation Medicine Vienna, Veterinaerplatz 1, 1210 Vienna, Austria Osteoprotegerin VetCore, University of Veterinary Medicine Vienna, Vienna, Austria Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine III and Center for Healthy Aging, Introduction Technische Universität Dresden, Dresden, Germany Amgen Inc., Thousand Oaks, CA, USA Improvements in the prevention and treatment of cardiovas- cular diseases have significantly changed the epidemiology of Present address: Phylon Pharma Services, Newbury Park, CA, USA 560 J Mol Med (2018) 96:559–573 myocardial infarction (MI). Whereas early survival has been fully inhibited by AMG161 [13]. Based on our previous find- improved and the severity of infarctions has declined progres- ing that only endothelial and hematopoietic but not stromal sively, the overall incidence of MI and the long-term survival precursors engraft after transplantation of unfractionated bone has remained constant [1, 2]. Long-term mortality after MI is marrow into lethally irradiated rats [14] and mice [15] and the still high, partly due to progression of the disease to heart fact that AMG161 exclusively inhibits human but not murine failure (HF), a condition which requires improvement in treat- RANKL, we used a reconstitution model in which we selec- ment options. Thus, a better understanding of the molecular tively exchanged the hematopoietic cell compartment to dis- mechanisms involved in HF progression is necessary for de- sect the pathophysiological role of RANKL derived from he- velopment of novel therapeutic strategies to prevent heart fail- matopoietic and mesenchymal cellular sources in the develop- ure post-MI. In this context, an increasing body of evidence ment of cardiac dysfunction after ischemia. We found that indicates that signaling through the RANK (receptor activator inhibition of RANKL derived from hematopoietic cellular of nuclear factor κB)/RANKL (RANK ligand)/OPG sources had beneficial effects on cardiac function after MI. (osteoprotegerin) axis might be involved in the pathophysiol- ogy of cardiovascular diseases. RANKL was first discovered as a cytokine which drives macrophage maturation to osteo- Methods clasts through its signaling receptor RANK, while OPG in- hibits RANKL signaling acting as a soluble decoy receptor of Animals RANKL [3]. Besides its role in bone remodeling, baseline levels of soluble RANKL predict the risk of cardiovascular All animal procedures were undertaken in accordance with events such as myocardial infarction and stroke [4]. current guidelines for animal care and welfare and were ap- Furthermore, it was suggested that RANKL may be important proved by the Ethical Committees of the University of for destabilizing atherosclerotic plaques [4, 5], whereas OPG Veterinary Medicine Vienna and of the Austrian Federal may prevent blood vessel calcification in atherosclerosis [6]. Ministry of Science, Research and Economy. All animals Notably, the myocardial OPG/RANKL ratio is significantly were kept in groups of two to seven at 22–24 °C and a 12-h lower in clinical and experimental heart failure, due to light/12-h dark cycle with free access to tap water and a com- disproportionally enhanced RANKL expression [7]. mercial rodent diet (Sniff™). The generation of the However, it is currently not known which exact role huRANKL-KI transgenic mouse line was described in detail RANKL plays during the post-ischemic myocardial healing elsewhere [13]. huRANKL-KI mice were obtained from and during transition to HF. Amgen Inc. and were backcrossed to C57BL/6 genetic back- The complex NF-κB signaling initiated by binding of ground for a minimum of six generations. Heterozygous mice RANKL to its receptor RANK can be, depending of the were bred, and the resulting wild-type (wt) and homozygous cellular context, detrimental or protective [8]. It was sug- huRANKL-KI animals were genotyped from tail biopsies by gested that RANKL promotes myocardial inflammation PCR analysis of genomic DNA. during acute cardiac overload [9] and favors adverse remod- eling by matrix degradation after acute myocardial infarc- Lethal irradiation and bone marrow transplantation tion [7]. In addition, RANKL was reported to act as a pro- inflammatory cytokine on hepatocytes as shown by the Male wt and huRANKL-KI mice at 6 to 8 weeks of age were finding that hepatic RANKL deletion counteracted hepatic used as bone marrow (BM) donors. Mice were killed by ex- insulin resistance by reducing NF-κB signaling [10]. On the sanguination from the abdominal vena cava under ketamine/ other hand, increased RANKL signaling reduced the infarct xylazine anesthesia (70/7 mg/kg i.p.). Unfractionated BM was size in stroke and hepatic ischemia, by acting in an anti- isolated from femora, tibias, and humeri by centrifugation at inflammatory and pro-cell survival manner [11, 12]. 800×g for 5 min at room temperature. The cells suspended in Here, we aimed to investigate the effect of RANKL inhi- PBS were filtered through a 40-μm cell strainer and were re- bition on post-ischemic cardiac function and structure. We suspended in PBS to contain 4 × 10 cells per milliliter. inhibited RANKL using the human monoclonal IgG1 anti- Male 16-week-old wt and transgenic huRANKL-KI recip- RANKL antibody AMG161. AMG161 selectively inhibits ient mice were lethally irradiated with a single dose of 10 human but not murine RANKL, by binding to a peptide se- Gray, using a linear accelerator (6MV, Primus, Siemens). quence encoded by exon 5 of the human RANKL gene. To Four hours after the irradiation, 4 × 10 of freshly prepared block RANKL in vivo, we used a transgenic mouse line (hu- unfractionated BM cells were injected into a lateral tail vein man RANKL knock-in, huRANKL-KI) where exon 5 of the of the recipient mice. Irradiated wt mice received the BM cells murine RANKL gene was replaced by the human sequence of huRANKL-KI donors, and irradiated huRANKL-KI mice [13]. The chimeric RANKL protein expressed by huRANKL- received BM cells of wt donors. We previously demonstrated KI mice is capable of inducing bone resorption, while being that this protocol efficiently reconstitutes cells of the J Mol Med (2018) 96:559–573 561 hematopoietic origin with a chimerism greater than 90% as Transthoracic Doppler echocardiography analyzed by flow cytometry, 4 weeks post-transplantation [15]. To avoid infections during the aplastic phase, irradiated Left ventricular (LV) function was non-invasively assessed animals were daily subcutaneously treated with an antibiotic 7 days and 3 weeks after MI in mice under isofluorane anesthe- (enrofloxacin, 10 mg/kg) over 7 days. Animals were left to sia using a 14-MHz linear transducer (Siemens S2000). Short- recover for 4 weeks before they were subjected to sham sur- axis M-mode recordings of LV were obtained from a left gery or to myocardial infarction. parasternal acoustic window. LV dimensions in systole and di- astole were measured through the largest diameter of the LV at the level of papillary muscles, and fractional shortening (FS) Myocardial infarction was calculated. Diastolic LV function was evaluated using the pulsed-wave Doppler recording of trans-mitral blood flow ve- Twenty-week-old male huRANKL-KI and wt mice were locities in the apical four-chamber view. A minimum of four anesthetized with ketamine/medetomidine (100/0.25 mg/ cardiac cycles were averaged for each measured parameter. kg i.p.) anesthesia. Endotracheal intubation was performed after disappearance of the paw pinch reflex. Animals were Central arterial and cardiac pressure measurement ventilated with a tidal volume of 200 μL and a frequency of 210 breathing cycles per minute using a small animal ventila- Pressure was assessed using a SPR-671NR pressure catheter tor (MiniVentTyp 845, Hugo Sachs Elektronik-Harvard (1.4F, Millar Instruments, Houston, TX, USA). The catheter Apparatus GmbH). Permanent ligation of the left descending was inserted into the ascending aorta via the carotid artery and coronary artery was performed after a left lateral thoracotomy. central arterial pressure was measured under 1.0% isofluorane Analgesic (buprenorphine 0.25 mg/kg s.c.) and antibiotic anesthesia. Thereafter, the catheter was advanced into the LV (enrofloxacin, 10 mg/kg s.c.) were injected for 4 and 5 days, for measurement of cardiac pressure parameters. Pressure was respectively. Sham animals underwent the same procedure but recorded over 5 min and traces were analyzed using LabChart without the arterial ligation. 7 software and a blood pressure module. Animals were killed 4 weeks after MI by exsanguination from the abdominal vena cava under ketamine/xylazine anesthe- sia (70/7 mg/kg i.p.). This time point was chosen in order to be Infarct size measurement able to document a robust decline in cardiac functional param- eters after MI. Serum samples, hearts, aortas, bones (femora, L4 Hearts excised at necropsy were washed from blood with PBS vertebra), and bone marrow were flash frozen and stored at − and fixed using 40% ethanol for at least 48 h. Dehydrated and 80 °C until assayed, or processed for histological analysis. paraffin-embedded hearts were sectioned at 5 μm thickness and werestainedwithMasson’s trichrome staining to visualize mus- cle and infarct tissue. At least nine standardized sections between Global and compartment-selective RANKL inhibition basis and apex of the heart were evaluated using AMIRA soft- by AMG161 ware. Infarct area was calculated as percentage of total LV area. Animals were randomized in a blinded fashion to the treatment with the anti-RANKL antibody AMG161 or an isotype control Gene expression analysis antibody (anti-keyhole limpet hemocyanin, KLH). Both AMG161 and the control antibody are humanized IgG1 anti- Total RNA from LV tissue was isolated using TRI Reagent® bodies. All antibodies were dissolved in A5su buffer containing Solution (Invitrogen). The concentration, purity, and quality 0.004% Tween 20 and were kindly provided by Amgen Inc. were determined spectrophotometrically (NanoDrop 2000; AMG161 or control antibody treatment (both at 10 mg/kg) Thermo Scientific) and by 2100 Bioanalyzer (Agilent started post-operatively and was continued twice weekly for Technologies). One microgram of RNA was reverse transcribed the duration of 4 weeks. Global inhibition of RANKL was (High Capacity cNDA Reverse Transcription Kit; Applied achieved by treatment of homozygous huRANKL-KI mice with Biosciences). Quantitative RT-PCR was performed on a Vii7 AMG161. To inhibit RANKL derived from cells of hematopoi- device (Applied Biosystems®) using the 5× Hot Firepol® Eva etic origin, wt mice were lethally irradiated, reconstituted with Green kit (Solis Biodyne). To exclude amplification of the ge- bone marrow from homozygous huRANKL-KI donors, and nomic DNA, primers were designed as exon spanning and their subsequently treated with AMG161. Vice versa, administration sequence is available upon request. A product melting curve of AMG161 to lethally irradiated homozygous huRANKL-KI analysis was performed to exclude primer dimerization and non- mice which were previously reconstituted with bone marrow specific amplification. All samples were measured in duplicate from wt mice, blocked RANKL from mesenchymal cell andexpressionvalueswerenormalizedtoornithinedecarbox- sources. The study design is graphically presented in Fig. 1. ylase antizyme-1 (Oaz1)mRNA. 562 J Mol Med (2018) 96:559–573 Fig. 1 Study design of global and compartment-selective RANKL inhibition. Irradiation and bone marrow transplantation (BMT) were performed in 16- week-old mice. Sham surgery or myocardial infarction (MI) were performed in 20-week-old mice Serum and urine biochemistry Immunofluorescence Serum phosphorus, calcium, alkaline phosphatase, and urinary Sections were prepared for staining as described for immunohis- creatinine were analyzed using a Cobas c111 analyzer (Roche). tochemistry. For co-staining of RANKL and CD3, primary anti- Total urinary deoxypyridinoline (DPD) concentrations were body against RANKL (rabbit polyclonal IgG, 1:100 in blocking assessed by a commercially available ELISA (MicroVue DPD solution, Santa Cruz Biotechnology) and anti-CD3 (Monoclonal EIA kit, Quidel) after acid hydrolysis. Rat IgG, 1:70 in blocking solution, R&D Systems) were incu- bated overnight at 4 °C. After washing, secondary anti-rabbit Alexa Fluor 555 (Molecular Probes, 1:500) and anti-rat Alexa pQCT analysis Fluor 594 (Thermo Fisher, 1:500) were incubated at room tem- perature for 60 min. Co-immunostaining of RANKL and cardiac Bone specimens were collected at necropsy and stored in 70% troponin T was performed in two steps. Firstly, primary anti- ethanol until analysis of mineral density using a XCT RANKL antibody was incubated overnight as described above. Research M+ pQCT device (Stratec Medizintechnik). Subsequently, biotinylated secondary anti-rabbit antibody (Vector, 1:400) was incubated at room temperature for 1 h, followed by incubation with Streptavidin-Alexa 546 conjugate Immunohistochemistry (Molecular Probes) for 30 min. After washing, primary rabbit anti-mouse cardiac troponin T antibody (St John’s Laboratory) Antigen retrieval was performed by heating the de-paraffinized and secondary anti-rabbit Alexa Fluor 647 (Thermo Fisher) an- cardiac sections to 100 °C for 15 min in citrate buffer (pH 6). tibodies were sequentially incubated for 1 h at room temperature. Sections were then treated with 0.1% Triton X-100 for 5 min at Immunostainings where primary antibodies were omitted served room temperature to permeabilize cell membranes, and for fur- as negative control. Co-immunostaining of IL-1ß and CD45 was ther 30 min with blocking solution containing 10% goat serum performed by incubating primary antibodies (rabbit anti-mouse and 0.02% Triton X in PBS to prevent unspecific antibody IL1ß, Abcam, 1:200; rat anti-moue CD45, BD Pharmingen, binding. Primary antibody against RANKL (rabbit polyclonal 1:40) overnight at 4 °C followed by incubation with secondary IgG, 1:200 in blocking solution, Santa Cruz Biotechnology), IL- anti-rabbit Alexa Fluor 555 and anti-rat Alexa Fluor 594 at room 1ß (goat polyclonal, 1:500 in blocking solution, R&D Systems), temperature for 60 min. Co-staining of IL-1ß and troponin T was and CD68 (rat monoclonal, 1:100, Bio Rad) were incubated performed in a two-step reaction as above described. Nuclei were overnight at 4 °C. After washing, secondary biotinylated anti- stained with DAPI (4′,6-diamidino-2-phenylindole). All sections bodies (Vector) were added and incubated for 60 min at room were imaged on a LSM 880 Airyscan confocal microscope. To temperature. Signal was developed by incubation with avoid cross talk, Alexa Fluor 555 and Alexa Fluor 594 fluoro- streptavidin-peroxidase (Vector) followed by 3-amino-9-ethyl chromes were exited at 514 and 633 nm, respectively. carbazol (AEC) or DAB staining. J Mol Med (2018) 96:559–573 563 Statistics Results Data are presented as mean ± SEM. Statistical analysis Cardiac ischemia/reperfusion injury activates was performed using GraphPad Prism 6. The data were RANK-RANKL-OPG axis in mice analyzed by two-sided t test (two groups) or one-way analysis of variance (ANOVA) followed by Bonferroni’s In line with data reported by Ueland et al. [7]inrats, wefound multiple comparison test (> 2 groups). P values of 0.05 that myocardial infarction (MI) activated the myocardial or less were considered significant. RANK/RANKL/OPG axis also in mice (Fig. 2). Left ventricular Fig. 2 Cardiac activation of the RANK-RANKL-OPG axis after myocardial infarction (MI) in mice. a Gene expression analysis in the left ventricle (LV), 4 weeks after MI (n =4 per group). *p < 0.05 vs. sham. b Immunohistochemical anti- RANKL staining in paraffin sec- tions of the LV in sham and MI mice, 4-weeks after MI. Upper left panel: negative (neg co) per- formed by omitting the primary anti-RANKL antibody. Upper right panel: sham control. Lower left panel: positive RANKL staining in cardiomyocytes (CM) and infiltrating cells in peri- ischemic LV region. Lower right panel: strong RANKL staining in remaining CM and infiltrating cells of the infarcted region. Bar = 100 μm. c Immunofluorescent co-staining of CD3 and RANKL in cardiac par- affin sections, 4 weeks after MI. RANKL co-localizes with some CD3 cells (red arrows), but is mainly expressed by CD3- negative cells with a fibroblast- like morphology (white arrows) in the infarct region. d Co- localization of RANKL and tro- ponin T in the infarct border re- gion, 4 weeks after MI. Bar = 20 μmin c and d 564 J Mol Med (2018) 96:559–573 Table 1 Basic characteristics, Sham Sham MI MI femoral BMD, blood parameters, and urinary deoxypyridinoline Co Ab AMG161 Co Ab AMG161 excretion after global RANKL inhibition in huRANKL-KI mice, Body weight (g) 29.7 ± 1.1 29.1 ± 0.7 30.6 ± 0.6 30.2 ± 0.6 4 weeks after surgery Lung/body weight ratio (mg/g) 5.1 ± 0.1 5.3 ± 0.1 5.2 ± 0.2 5.0 ± 0.1 Heart/body weight ratio (mg/g) 4.1 ± 0.1 4.2 ± 0.1 4.7 ± 0.2* 4.8 ± 0.1 3 §§ Femoral metaph. total BMD (mg/cm ) 478 ± 4 527 ± 16* 469 ± 10 523 ± 17 Serum P (mmol/L) 2.53 ± 0.12 2.48 ± 0.21 2.94 ± 0.21 2.99 ± 0.31 Serum Ca (mmol/L) 2.11 ± 0.04 2.02 ± 0.04 2.14 ± 0.06 2.13 ± 0.05 Serum ALP (U/L) 41.1 ± 2.1 35.2 ± 3.2 41.6 ± 3.7 32.5 ± 0.8 Urinary DPD/creatinine (nM/mM) 8.2 ± 1.9 1.6 ± 1.4 10.7 ± 3.5 2.9 ± 1.1 n =5–10 per group *p < 0.05 vs. sham + control antibody (Co Ab) p <0.05 vs. sham + AMG161 p <0.05 vs. MI + Co Ab §§ p <0.01 vs. MI + Co Ab (LV) mRNA abundance of Rankl significantly increased, (Fig. 3). Global inhibition of RANKL after MI by treatment of 4 weeks after MI (Fig. 2a). OPG and Rank gene expression also huRANKL-KI mice with AMG161 did not have any statistically tended to increase in the LVafter MI, but this effect did not reach significant effects on survival, heart/body weight ratio, infarct statistical significance (Fig. 2a). Immunohistochemical analysis size, or cardiac functional parameters compared to isotype control showed that RANKL expression was mainly induced in antibody-treated huRANKL-KI MI mice (Fig. 3 and Table 1). cardiomyocytes adjacent to the infarct region as well as in the cellular infiltrate within the infarct (Fig. 2b). Further analysis of Hematopoietic, but not mesenchymal, RANKL the infarct region by immunofluorescent imaging revealed that inhibition reduced pro-inflammatory cytokine RANKL co-localized with some CD3-positive T lymphocytes, production in the left ventricle of MI mice but more abundant RANKL expression was present in fibroblast-like cells, cardiomyocytes, and blood vessels (Fig. Depending on the cell type, RANKL can exert both pro- and 2c, d and Suppl. Fig. S1). These findings suggest that cells other anti-inflammatory effects [11, 16] and the lack of therapeutic than lymphocytes are the main RANKL source in the infarcted effect of global RANKL inhibition may be caused by oppos- myocardium. Cardiomyocytes in remote myocardium did not ing RANKL actions in the myocardium vs. the cellular infil- express RANKL (Suppl. Fig. S1). trate. Thus, we next asked the question whether selective blockade of hematopoietic and mesenchymal RANKL might have positive therapeutic effects. Global RANKL inhibition by AMG161 lacks beneficial effect in murine myocardial infarction model To selectively block hematopoietic RANKL, we lethally irradiated wt mice and reconstituted them with bone marrow The exact pathophysiological role of increased cardiac RANKL from huRANKL-KI mice to replace their hematopoietic com- partment with cells responsive to AMG161, using a previous- after cardiac ischemia is not known. Because RANKL was re- ported to promote inflammation and matrix degradation [7, 9], ly established protocol [15]. To prove that the expected RANKL form was expressed by cells of hematopoietic origin we hypothesized that inhibition of RANKL could improve the post-infarct outcome after MI. To test this hypothesis, we in- duced MI in huRANKL-KI mice and subsequently treated the Fig. 3 Global RANKL inhibition by AMG161 does not influence mice with the monoclonal anti-human RANKL antibody outcome in huRANKL-KI mice after MI. a Kaplan-Meier survival curves AMG161. Biological activity ofAMG161was confirmedby after MI (n =22–24 per group). b Infarct size measured by planimetry after Masson’s trichrome staining (n =11–13 per group). c Representative significantly increased femoral BMD, as well as suppressed Masson’s trichrome-stained cardiac cross-sections, 4-weeks after sham or serum alkaline phosphatase and urinary DPD excretion in MI surgery. d Cardiac function and LV diameters measured by echocar- AMG161-treated huRANKL-KI mice (Table 1). However, diography (n =10–24 per group). e Representative M-mode echocardio- treatment with AMG161 had no effect on calcium or phosphate grams, obtained 3 weeks after MI. f Cardiac parameters measured by intra-cardiac catheterization (n =5–14 per group). LVIDd left ventricle serum levels (Table 1). internal diameter in diastole, LVIDs left ventricle internal diameter in Induction of MI led to a significant deterioration of cardiac systole, MAP mean arterial pressure, dP/dt maximal rate of left ventricle function in huRANKL-KI mice, as evidenced by reduced frac- pressure rise. *p < 0.05 vs. sham + control antibody (Co Ab); **p <0.01 tional shortening, dilation of the LV in both systole and diastole, vs. sham + Co Ab; ***p < 0.001 vs. sham + Co Ab; p < 0.05 vs. sham + ### AMG161; p < 0.001 vs. sham + AMG161 and diminished contractile cardiac function as assessed by dP/dt J Mol Med (2018) 96:559–573 565 after bone marrow transfer, we analyzed splenic Rankl gene Rankl in their spleens, respectively (Suppl. Fig. S2). In con- expression in non-irradiated mice as well as after bone marrow trast, after lethal irradiation and vice versa reconstitution, chi- transfer. As expected, non-irradiated wt mice did not express meric Rankl gene was abundantly expressed in the spleen of chimeric Rankl, and huRANKL-KI mice did not express wt reconstituted wt mice, whereas reconstituted huRANKL-KI 566 J Mol Med (2018) 96:559–573 mice expressed the mouse wt Rankl gene in their spleens Surprisingly, hematopoietic RANKL inhibition improved (Suppl. Fig. S2), indicating successful exchange of the hema- post-infarct survival as well as cardiac function as shown by a topoietic compartment. significant rise in fractional shortening, lower diastolic and J Mol Med (2018) 96:559–573 567 Fig. 4 Inhibition of RANKL derived from hematopoietic (left column) but not from mesenchymal cellular sources (right column) improves survival and cardiac function post-MI. Kaplan-Meier sur- vival curves after MI (n =22–31 per group), echocardiographic pa- rameters measured 3 weeks after surgery (n =16–17 per group), and infarct size measured by planimetry after Masson’s trichrome staining (n =8–14 per group) in sham and MI huRANKL-KI mice treated with AMG161 or control antibody (Co Ab), 4 weeks after MI. LVIDd left ventricular internal diameter in diastole, LVIDs left ventricular inter- nal diameter in systole. **p < 0.01 and ***p < 0.001 vs. sham + Co # ### § Ab; p <0.05 and p < 0.001 vs. sham + AMG161; p <0.05 and §§ p < 0.01vs. MI +CoAb systolic endocardial LV diameters, and enhanced LV contractil- ity in AMG161 vs. control Ab-treated mice (Fig. 4 and Table 2). In contrast, inhibition of RANKL derived from the mesenchy- mal cell compartment in AMG161-treated huRANKL-KI mice reconstituted with wt bone marrow did not have any effect on survival or post-infarct cardiac function (Fig. 4 and Table 2). Infarct area was not affected by RANKL inhibition in any of the investigated groups (Fig. 4), suggesting that mechanisms other than those regulating cardiomyocyte cell death are responsible for the protective effects on survival and cardiac function seen after inhibition of hematopoietic cell-derived RANKL. To shed more light on the intriguing finding that inhibition of hematopoietic, but not of mesenchymal or global, RANKL had these beneficial effects after MI, we measured inflammatory cell infiltration and the mRNA abundance of pro-inflammatory cy- tokines in the LV. Because hematopoietic RANKL inhibition protected against post-ischemic LV chamber dilation, we hy- pothesized that inflammatory signaling pathways initiated by RANKL secreted from cells of hematopoietic origin may drive adverse post-ischemic remodeling of the LV. We first examined whether RANKL inhibition altered the post-ischemic infiltration with macrophages or lymphocytes. Immunohistochemical anal- Fig. 5 Global RANKL inhibition slightly reduces CD68 macrophage ysis of CD68-positive macrophages in the infarct region re- abundance in the left ventricle, 4 weeks after MI. a Representative images vealed a trend for reduced macrophage infiltration after global, of immunohistochemical stainings of CD68-expressing macrophages in the mesenchymal, and hematopoietic RANKL inhibition (Fig. 5). infarct region after global, mesenchymal, and hematopoietic RANKL inhi- However, this effect reached statistical significance only after bition. Bar = 50 μm. b Quantification of CD68 macrophages in the whole infarct region, presented as cell number per image. n=4–5 mice per group. global RANKL inhibition (Fig. 5). In contrast, the LV mRNA *p<0.05 vs. MI + Co Ab abundance of the lymphocyte-specific genes CD3 remained Table 2 Hemodynamic variables Hematopoietic RANKL inhibition Mesenchymal RANKL inhibition measured invasively using intra- cardiac catheter, 4 weeks after MI MI + Co Ab MI + AMG161 MI + Co Ab MI + AMG161 Systolic P (mmHg) 86.95 ± 7.5 93.18 ± 2.0 80.03 ± 3.3 84.57 ± 3.6 Diastolic P (mmHg) 60.17 ± 4.9 64.81 ± 1.7 53.87 ± 3.1 59.78 ± 4.0 MAP (mmHg) 76.57 ± 5.9 78.15 ± 1.6 66.49 ± 3.1 72.30 ± 4.0 dP/dt (mmHg/s) 4416 ± 546 6744 ± 571* 4912 ± 281 5178 ± 602 max EDP (mmHg) 8.76 ± 1.9 8.19 ± 1.7 8.27 ± 2.4 9.46 ± 3.1 Tau (ms) 2.33 ± 0.3 1.91 ± 0.3 2.26 ± 0.4 1.78 ± 0.2 n =4–6per group MAP mean arterial pressure, dP/dt maximal rate of left ventricle pressure rise, EDP end-diastolic pressure, Tau left ventricular relaxation time constant *p < 0.05 vs. MI + Co Ab 568 J Mol Med (2018) 96:559–573 unchanged after global, mesenchymal, or hematopoietic inhibition in surviving cardiomyocytes within the ischemic RANKL inhibition (Fig. 6). Interestingly, however, the post- zone. To characterize further the cellular source of IL-1ß, we ischemic rise in LV gene expression of IL-1ß and Mmp-9,but performed co-immunostaining of IL-1ß with CD45 and cardiac not of TNFα, was significantly reduced after hematopoietic troponin T (Fig. 8 and Suppl. Fig. S5 and S6). Although some RANKL blockade (Fig. 6 and Suppl. Fig. S3). In contrast, mes- CD45 cells in the infarct region expressed IL-1ß, the majority enchymal RANKL inhibition did not change the LVexpression of IL-1ß-expressing cells had a fibroblast-like morphology (Fig. pattern of pro-inflammatory genes after MI, and global RANKL 8 and Suppl. Fig. S5). Troponin-positive cardiomyocytes in the inhibition even enhanced IL-1ß mRNA expression in the post- border zone of the infarct also clearly expressed IL-1ß (Suppl. ischemic LV (Fig. 6). Intriguingly, mRNA expression of the Fig. S6), but their IL-1ß expression, in contrast to non- resolving M2 macrophage markers Mrc-1, Arg-1,and Ym-1 cardiomyocyte cells (Fig. 8), was not downregulated by hema- was profoundly reduced in the post-ischemic LV after hemato- topoietic RANKL inhibition. Altogether, these findings suggest poietic RANKL blockade, but not after global or mesenchymal that RANKL derived from scar-infiltrating cells of hematopoi- RANKL inhibition (Suppl. Fig. S4). In addition, the MI-induced etic origin is an important pro-inflammatory stimulus whose upregulation of Arg-1 and Ym-1 was lower in huRANKL-KI inhibition can be beneficial for post-ischemic recovery. mice relative to mice with a wt background (Suppl. Fig. S4). To further characterize the changes in IL-1ß protein expres- sion induced by RANKL blockade, we used immunohisto- Discussion chemical analysis of the infarct region. As shown in Fig. 7, inhibition of global or hematopoietic RANKL decreased IL-1ß RANKL plays a pivotal role in bone remodeling and in immu- expression in the non-cardiomyocyte cell compartment of the nity [3, 17, 18] and may also be an important signaling molecule scar, whereas IL-1ß expression was not influenced by RANKL in diseases affecting the cardiovascular system. We show here Fig. 6 Hematopoietic but not global or mesenchymal RANKL inhibition hematopoietic RANKL inhibition. Gene expression is presented as fold downregulates IL-1β mRNA expression in the left ventricle after MI. Left increase compared to sham + control antibody (Co Ab) group. n =3–9per ventricular expression of IL-1ß (upper panels), TNFα (middle panels), group. *p < 0.05 and ***p < 0.001 vs. sham + Co Ab; p < 0.05 vs. sham and CD3 mRNA (lower panels) after global, mesenchymal, or + AMG161; p <0.05 vs. MI +Co Ab J Mol Med (2018) 96:559–573 569 Fig. 7 Hematopoietic and global but not mesenchymal RANKL inhibition suppresses IL-1β pro- tein expression in infiltrating cells post-MI. a Representative images of immunohistochemical anti-IL- 1ß staining in the infarct region, 4 weeks after MI. Bar = 100 μm. b Semi-quantitative analysis of IL-1ß expression in cardiomyocytes and non- cardiomyocyte cells. IL-1ß stain- ing intensity is shown as the mean of intensity ratios between AMG161 and control antibody (Co Ab) treatment with 95% confidence intervals. Samples stained in the same experiment were paired (Co Ab and AMG161-treated). *p <0.05 by ratio paired t-test that experimental MI induced LV RANKL expression in mice. after MI in huRANKL-KI mice. Reason for this discrepancy Global RANKL inhibition or inhibition of RANKL derived may lay in reduced in vivo activity of chimeric RANKL in from mesenchymal cellular sources using a monoclonal anti- transgenic animals compared to that of murine RANKL in wt RANKL antibody did not significantly alter cardiac function mice. Although chimeric RANKL and wt murine RANKL have post-MI, whereas specific inhibition of RANKL derived from the same affinity to the murine RANK receptor, the hematopoietic cellular sources improved post-ischemic cardiac osteoclastogenic potency of chimeric RANKL was slightly low- function, reduced mortality, and downregulated post-ischemic er than that of murine wt RANKL [13]. It is conceivable that the production of inflammatory cytokines. lack of myocardial MMP-9 upregulation and the reduced upreg- In the current study, we used a monoclonal antibody ulation of some M2 macrophage markers after MI in AMG161 to block RANKL. Because AMG161 selectively huRANKL-KI mice may be explained by this fact. However, blocks human RANKL, but not murine RANKL, we used the promoter and splicing regions of the Rankl gene are intact huRANKL-KI mice expressing a humanized RANKL protein in huRANKL-KI mice, and thus, regulation of RANKL levels in this study. Although hematopoietic RANKL inhibition was during pathological conditions such as MI is expected to be beneficial for cardiac function after MI in WT mice, global comparable to that of wt mice. Another possibility for the dis- RANKL inhibition did not significantly change cardiac function crepant findings after global and hematopoietic RANKL 570 J Mol Med (2018) 96:559–573 Fig. 8 Immunofluorescent co- staining of IL-1β andCD45in the infarct region after global, hema- topoietic, and mesenchymal RANKL inhibition. IL-1β is co- expressed by some CD45 cells, but abundantly expressed in fibroblast-like cells within the in- farct region, 4 weeks post-MI. Note the striking reduction of IL- 1β staining after hematopoietic RANKL inhibition (right middle panel). Representative images of n = 2 mice per group. Bar = 20 μm inhibition may be the fact that global RANKL inhibition was Timing of RANKL signaling and its cellular source may be performed in non-irradiated huRANKL-KI mice, whereas hema- important determinants of its effect during/after ischemic injury. topoietic RANKL inhibition was performed in irradiated and The tissue response to increased RANKL levels during acute reconstituted mice. However, since the inflammatory response ischemia may be different from prolonged RANKL signaling post-MI was almost identical in irradiated (mesenchymal during tissue repair and remodeling. Signaling downstream of RANKL inhibition) and non-irradiated huRANKL-KI mice the RANK receptor in the myocardium probably involves (global RANKL inhibition), it is highly unlikely that this was a NF-κB[9]. NF-κB activation during acute injury may promote major influencing factor. A third possibility is that the hemato- cell survival and suppress apoptotic signaling [8]. Indeed, poietic output of the bone marrow is regulated by RANKL- models of brain and liver ischemia demonstrated an important driven osteoclastogenesis [19, 20]. Therefore, treatment of non- role of acute RANKL signaling in cell survival and limitation irradiated and irradiated huRANKL-KI mice with AMG161 may of the final infarct size [11, 12]. In contrast, a recent study modulate inflammatory responses through inhibition of osteo- reported beneficial effects of RANKL inhibition during cardiac clastogenesis and altered output of bone marrow-derived cells. ischemia on infarct size, 24 h after reperfusion [16]. However, However, we found no major and consistent differences in lym- in the current study, infarct size after global RANKL inhibition phocyte or macrophage infiltration of the infarct between was not changed, 4 weeks after MI. Thus, we can exclude huRANKL-KI and wt mice, arguing against a major difference increased vulnerability to ischemia after RANKL inhibition. in the bone marrow output of inflammatory/anti-inflammatory On the other hand, prolonged NF-κB activation through cells. RANKL may promote inflammation and adverse cardiac J Mol Med (2018) 96:559–573 571 remodeling [8]. We hypothesize that during post-ischemic re- RANKL reduces transcription of the pro-inflammatory cytokine modeling, the cellular source of RANKL determines the effect IL-1β in the myocardium, and more specifically in scar- of RANKL in the ischemic myocardium. Our results support infiltrating inflammatory cells. This is in line with our finding the notion that RANKL produced by cells of hematopoietic of improved cardiac function in mice with reduced IL-1β levels origin, but not by cardiomyocytes, contributes to maladaptive after hematopoietic RANKL blockade. processes, deteriorating cardiac function after myocardial in- RANKL signaling receptor RANK is present in both neutro- farction. In this context, it is tempting to speculate that cells of phils [27] and macrophages [28, 29], which makes both cellular hematopoietic origin, in contrast to cells of mesenchymal ori- population responsive to RANKL. Role of RANKL in neutro- gin, do not produce enough OPG to inhibit excess RANKL phil infiltration and MMP-9 secretion was recently demonstrated signaling. This hypothesis is supported by findings that osteo- early after MI [16]. However, in our long-term study, it is more blasts but not activated lymphocytes secrete OPG in the culture probable that beneficial effects seen after hematopoietic RANKL medium, and that OPG mRNA was not detected in human T inhibition are mediated by macrophages which infiltrate ische- lymphocytes or monocytes [5]. Hence, the local increase in mic myocardium long term after MI [30]. Although we did not RANKL secretion from BM-derived cells may be a driving observe changes in macrophage number in the infarct region force of inflammation in the heart after MI, whereas RANKL after hematopoietic RANKL blockade, markers of the reparative produced by cardiomyocytes may be inhibited by concurrently M2 subset were profoundly reduced in the myocardium after increased cardiomyocytic OPG secretion. This notion may ex- hematopoietic blockade of RANKL, 4 weeks post-MI. This find- plain why inhibition of hematopoietic RANKL has beneficial ing may suggest that inhibition of hematopoietic RANKL may effects post-MI, whereas selective inhibition of mesenchymal alter the time course of resolution of inflammation. Further ex- RANKL does not influence post-ischemic cardiac remodeling. periments are needed to address the question of how RANKL A puzzling observation in our study was that post-ischemic inhibition affects resolution of inflammation after MI. LV IL-1ß mRNA abundance was promoted after global Our finding that global RANKL inhibition did not increase RANKL inhibition, but reduced when only hematopoietic infarct size or deteriorate cardiacfunctionafter MI mayhave RANKL was inhibited. There are several possible explanations important clinical implications. RANKL is the molecular target for this finding. Global RANKL inhibition using AMG161 may behind one of the most effective osteoporosis treatments today, leave larger amounts of OPG available for other signaling path- the treatment with the anti-RANKL antibody denosumab. The ways, including binding of TRAIL. Furthermore, it was reported typical osteoporosis patient is of advanced age and usually at that RANKL may actually reduce both innate [11, 21]and adap- high risk to suffer co-morbidities, and there is a rising awareness tive immune responses [22]insomemodels. Forexample,ithas that cardiovascular diseases and osteoporosis might be been shown that RANKL expression in keratinocytes can drive pathophysiologically linked diseases. A large 3-year placebo- controlled trial of denosumab in postmenopausal women with formationofregulatoryT cells[22]. In analogy, RANKL ex- pression on cardiomyocytes may downregulate the inflammato- low bone mass showed no treatment-related effects on the inci- ry response post-MI. However, since inhibition of mesenchymal dence of cardiovascular events, coronary heart disease, or atrial RANKL had no influence on LV function or gene expression, it fibrillation, with a trend towards reduced all-cause mortality in is unlikely that the latter scenarioistrue.Itmayalsobeimportant the denosumab arm [31]. However, this clinical trial population, to note in this context that the increased LV IL-1ß mRNA ex- postmenopausal women, was not selected for increased CV risk pression observed after global RANKL inhibition in factors, and it may therefore be reassuring that the current study huRANKL-KI MI mice was not evident in immunohistochem- showed no untoward effects of global RANKL inhibition on ical analyses of the scar region (Figs. 7 and 8). Therefore, it is post-MI survival or cardiac function. unclear whether the increased IL-1ß gene transcription after Our study has shown that the upregulation of RANKL in the global RANKL inhibition fully translates into augmented IL- post-ischemic myocardium mainly involves fibroblast-like 1ß secretion at the protein level. On the other hand, the down- cells, blood vessels, and surviving cardiomyocytes. We and regulation of IL-1ß mRNA abundance in the LV after hemato- others showed earlier that a substantial amount of endothelial poietic RANKL inhibition was confirmed at the protein level by cells and fibroblasts/myofibroblasts in the heart is donor-derived immunohistochemistry. after bone marrow transplantation [14, 32]. Therefore, it is con- Secreted IL-1β has a critical role in the post-ischemic remod- ceivable that some beneficial effects after hematopoietic eling by stimulating inflammatory cell accumulation, inflamma- RANKL blockade are attributable to endothelial RANKL tory cytokine production, myofibroblast differentiation, extracel- blockade. Future studies need to address the question how the lular matrix degradation, and collagen production [23, 24]. availability of RANKL/RANK molecules on specific cell types Moreover, IL-1β depresses cardiac contractility by reducing regulates cell-cell interactions and the immune/inflammatory 2+ L-type Ca currents in neonatal and adult ventricular tissue response in the course of tissue repair after MI. cardiomyocytes [25], and by inhibiting their β-adrenergic re- Improved insight into these mechanisms may eventually open up new possibilities for the treatment of MI patients. sponse [26]. We show here that inhibition of hematopoietic 572 J Mol Med (2018) 96:559–573 Acknowledgements Open access funding provided by University of Dysregulated osteoprotegerin/RANK ligand/RANK axis in clinical Veterinary Medicine Vienna. We thank Alexandra Petric for excellent and experimental heart failure. Circulation 111:2461–2468 technical assistance, and Ute Zeitz and Miriam Kleiter for the help with 8. Gordon JW, Shaw JA, Kirshenbaum LA (2011) Multiple facets of the animal experiments. NF-κB in the heart: to be or not to NF-κB. Circ Res 108:1122–1132 9. Ock S, Ahn J, Lee SH, Park H, Son JW, Oh JG, Yang DK, Lee WS, Kim HS, Rho J, Oh GT, Abel ED, Park WJ, Min JK, Kim J (2012) Author Contributions S.S., L.C.H, P.J.K, and R.G.E. conceived and de- Receptor activator of nuclear factor-B ligand is a novel inducer of signed the study. S.S., O.A., K.F., S.H., N.L., U.R., S.S., and C.B. per- myocardial inflammation. Cardiovasc Res 94:105–114 formed the experiments and analysed the data. S.S., L.C.H, P.J.K, and 10. Kiechl S, Wittmann J, Giaccari A, Knoflach M, Willeit P, Bozec A, R.G.E. wrote the manuscript. R.G.E. accepts responsibility for the integ- Moschen AR, Muscogiuri G, Sorice GP, Kireva T, Summerer M, rity and validity of the data collected and analysed. Wirtz S, Luther J, Mielenz D, Billmeier U, Egger G, Mayr A, Oberhollenzer F, Kronenberg F, Orthofer M, Penninger JM, Funding information This work was supported by a grant from the Meigs JB, Bonora E, Tilg H, Willeit J, Schett G (2013) Blockade Austrian Science Fund (FWF P 21904-B11) to R.G.E. of receptor activator of nuclear factor-κB (RANKL) signaling im- proves hepatic insulin resistance and prevents development of dia- betes mellitus. Nat Med 19:358–363 Compliance with ethical standards 11. Shimamura M, Nakagami H, Osako MK, Kurinami H, Koriyama H, Zhengda P, Tomioka H, Tenma A, Wakayama K, Morishita R (2014) All animal procedures were undertaken in accordance with current guide- OPG/RANKL/RANK axis is a critical inflammatory signaling system in lines for animal care and welfare and were approved by the Ethical ischemic brain in mice. Proc Natl Acad Sci U S A 111:8191–8196 Committees of the University of Veterinary Medicine Vienna and of the 12. Sakai N, Van Sweringen HL, Schuster R et al (2012) Receptor acti- Austrian Federal Ministry of Science, Research and Economy. vator of nuclear factor-κB ligand (RANKL) protects against hepatic ischemia/reperfusion injury in mice. Hepatology 55:888–897 Conflict of interest PJK is a former Amgen employee who owns 13. Kostenuik PJ, Nguyen HQ, McCabe J, Warmington KS, Kurahara Amgen stock and serves as a consultant and contract medical writer for C, Sun N, Chen C, Li L, Cattley RC, van G, Scully S, Elliott R, Amgen. The other authors declare no conflicts of interest. Grisanti M, Morony S, Tan HL, Asuncion F, Li X, Ominsky MS, Stolina M, Dwyer D, Dougall WC, Hawkins N, Boyle WJ, Simonet WS, Sullivan JK (2009) Denosumab, a fully human monoclonal Open Access This article is distributed under the terms of the Creative antibody to RANKL, inhibits bone resorption and increases BMD Commons Attribution 4.0 International License (http:// in knock-in mice that express chimeric (murine/human) RANKL. J creativecommons.org/licenses/by/4.0/), which permits unrestricted use, Bone Miner Res Off J Am Soc Bone Miner Res 24:182–195 distribution, and reproduction in any medium, provided you give 14. Odörfer KI, Egerbacher M, Unger NJ, Weber K, Jamnig A, appropriate credit to the original author(s) and the source, provide a link Lepperdinger G, Kleiter M, Sandgren EP, Erben RG (2011) to the Creative Commons license, and indicate if changes were made. Hematopoietic bone marrow cells participate in endothelial, but not epithelial or mesenchymal cell renewal in adult rats. J Cell Mol Med 15:2232–2244 15. Streicher C, Heyny A, Andrukhova O, Haigl B, Slavic S, Schüler C, References Kollmann K, Kantner I, Sexl V, Kleiter M, Hofbauer LC, Kostenuik PJ, Erben RG (2017) Estrogen regulates bone turnover by targeting RANKL expression in bone lining cells. Sci Rep 7:6460 1. Kostis WJ, Deng Y, Pantazopoulos JS, Moreyra AE, Kostis JB, for 16. Carbone F, Crowe LA, Roth A, Burger F, Lenglet S, the Myocardial Infarction Data Acquisition System (MIDAS14) Braunersreuther V, Brandt KJ, Quercioli A, Mach F, Vallée JP, Study Group (2010) Trends in mortality of acute myocardial infarc- Montecucco F (2016) Treatment with anti-RANKL antibody re- tion after discharge from the hospital. Circ Cardiovasc Qual duces infarct size and attenuates dysfunction impacting on Outcomes 3:581–589 neutrophil-mediated injury. J Mol Cell Cardiol 94:82–94 2. Roger VL (2010) Myocardial infarction outcomes: Bthe times, they 17. Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, are a-changin...^. Circ Cardiovasc Qual Outcomes 3:568–570 Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, 3. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Galibert L (1997) A homologue of the TNF receptor and its ligand Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, enhance T-cell growth and dendritic-cell function. Nature 390:175–179 Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, . Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ Morony S, Oliveira-dos-Santos AJ, van G, Itie A, Khoo W, (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast Wakeham A, Dunstan CR, Lacey DL, Mak TW, Boyle WJ, differentiation and activation. Cell 93:165–176 Penninger JM (1999) OPGL is a key regulator of osteoclastogene- 4. Kiechl S, Schett G, Schwaiger J, Seppi K, Eder P, Egger G, Santer sis, lymphocyte development and lymph-node organogenesis. P, Mayr A, Xu Q, Willeit J (2007) Soluble receptor activator of Nature 397:315–323 nuclear factor-kappa B ligand and risk for cardiovascular disease. 19. Kollet O, Dar A, Shivtiel S, Kalinkovich A, Lapid K, Sztainberg Y, Circulation 116:385–391 Tesio M, Samstein RM, Goichberg P, Spiegel A, Elson A, Lapidot T 5. Sandberg WJ, Yndestad A, Øie E et al (2006) Enhanced T-cell ex- (2006) Osteoclasts degrade endosteal components and promote mobi- pression of RANK ligand in acute coronary syndrome: possible role in lization of hematopoietic progenitor cells. Nat Med 12:657–664 plaque destabilization. Arterioscler Thromb Vasc Biol 26:857–863 20. Cho K-A, Joo S-Y, Han H-S, Ryu KH, Woo SY (2010) Osteoclast 6. Bucay N, Sarosi I, Dunstan CR, Morony S, Tarpley J, Capparelli C, activation by receptor activator of NF-kappaB ligand enhances the Scully S, Tan HL, Xu W, Lacey DL, Boyle WJ, Simonet WS (1998) mobilization of hematopoietic progenitor cells from the bone mar- Osteoprotegerin-deficient mice develop early onset osteoporosis row in acute injury. Int J Mol Med 26:557–563 and arterial calcification. Genes Dev 12:1260–1268 21. Maruyama K, Takada Y, Ray N et al (2006) Receptor activator of NF- 7. Ueland T, Yndestad A, Øie E, Florholmen G, Halvorsen B, Frøland kappa B ligand and osteoprotegerin regulate proinflammatory cytokine SS, Simonsen S, Christensen G, Gullestad L, Aukrust P (2005) production in mice. J Immunol (Baltimore, Md 1950) 177:3799–3805 J Mol Med (2018) 96:559–573 573 22. Loser K, Mehling A, Loeser S, Apelt J, Kuhn A, Grabbe S, Staub C, Ratib O, Dallegri F, Schindler TH, Montecucco F (2012) Receptor activator of NF-κB ligand (RANKL) increases the release Schwarz T, Penninger JM, Beissert S (2006) Epidermal RANKL controls regulatory T-cell numbers via activation of dendritic cells. of neutrophil products associated with coronary vulnerability. Nat Med 12:1372–1379 Thromb Haemost 107:124–139 23. Sager HB, Heidt T, Hulsmans M, Dutta P, Courties G, Sebas M, 28. Hsu H, Lacey DL, Dunstan CR, Solovyev I, Colombero A, Timms Wojtkiewicz GR, Tricot B, Iwamoto Y, Sun Yet al (2015) Targeting E, Tan HL, Elliott G, Kelley MJ, Sarosi I, Wang L, Xia XZ, Elliott interleukin-1β reduces leukocyte production after acute myocardial R, Chiu L, Black T, Scully S, Capparelli C, Morony S, Shimamoto infarction. Circulation. https://doi.org/10.1161/ G, Bass MB, Boyle WJ (1999) Tumor necrosis factor receptor CIRCULATIONAHA.115.016160 family member RANK mediates osteoclast differentiation and acti- 24. Bujak M, Dobaczewski M, Chatila K, Mendoza LH, Li N, Reddy vation induced by osteoprotegerin ligand. Proc Natl Acad Sci U S A A, Frangogiannis NG (2008) Interleukin-1 receptor type I signaling 96:3540–3545 critically regulates infarct healing and cardiac remodeling. Am J 29. Antonios JK, Yao Z, Li C, Rao AJ, Goodman SB (2013) Pathol 173:57–67 Macrophage polarization in response to wear particles in vitro. 25. El Khoury N, Mathieu S, Fiset C (2014) Interleukin-1β reduces L- Cell Mol Immunol 10:471–482 type Ca2+ current through protein kinase Cϵ activation in mouse 30. Swirski FK, Nahrendorf M (2013) Leukocyte behavior in atheroscle- heart. J Biol Chem 289:21896–21908 rosis, myocardial infarction, and heart failure. Science 339:161–166 26. Gulick T, Chung MK, Pieper SJ, Lange LG, Schreiner GF (1989) 31. Cummings SR, Martin JS, McClung MR et al (2009) Denosumab Interleukin 1 and tumor necrosis factor inhibit cardiac myocyte beta- for prevention of fractures in postmenopausal women with osteo- adrenergic responsiveness. Proc Natl Acad Sci U S A 86:6753–6757 porosis. N Engl J Med 361:756–765 27. Quercioli A, Mach F, Bertolotto M, Lenglet S, Vuilleumier N, 32. Ogawa M, LaRue AC, Drake CJ (2006) Hematopoietic origin of Galan K, Pagano S, Braunersreuther V, Pelli G, Pistoia V, Bianchi fibroblasts/myofibroblasts: its pathophysiologic implications. G, Cittadini G, Viviani GL, Pende A, Roux-Lombard P, Thomas A, Blood 108:2893–2896

Journal

Journal of Molecular MedicineSpringer Journals

Published: May 8, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off