Diabetes-Associated Myelopoiesis Drives Stem Cell Mobilopathy Through an OSM-P66shc Signaling Pathway

Diabetes Volume 68, June 2019 1303

Diabetes-Associated Myelopoiesis Drives Stem Cell Mobilopathy Through an OSM-p66Shc Signaling Pathway

Mattia Albiero,1,2 Stefano Ciciliot,1 Serena Tedesco,1,2 Lisa Menegazzo,1 Marianna D’Anna,1,2 Valentina Scattolini,1,2 Roberta Cappellari,1,2 Gaia Zuccolotto,3,4 Antonio Rosato,3,4 Andrea Cignarella,2 Marco Giorgio,5,6 Angelo Avogaro,2 and Gian Paolo Fadini1,2

Diabetes 2019;68:1303–1314 | https://doi.org/10.2337/db19-0080

Diabetes impairs the mobilization of hematopoietic Diabetes is associated with low-grade inflammation, which stem/progenitor cells (HSPCs) from the bone marrow contributes to chronic complications (1,2). A skewed dif- (BM), which can worsen the outcomes of HSPC trans- ferentiation of common myeloid progenitors (CMPs) transplantation and of diabetic complications. In this study, lates hyperglycemia into production of proinflammatory we examined the oncostatin M (OSM)–p66Shc pathway cells (3). Such enhanced myelopoiesis propagates inflam- as a mechanistic link between HSPC mobilopathy and mation from the bone marrow (BM) to the adipose and the excessive myelopoiesis. We found that streptozotocin- vasculature, leading to insulin resistance and atherosclerosis COMPLICATIONS induced diabetes in mice skewed hematopoiesis toward (3,4). In parallel, mobilization of hematopoietic stem/ the myeloid lineage via hematopoietic-intrinsic p66Shc. progenitor cells (HSPCs) from the BM to peripheral blood The overexpression of Osm resulting from myelopoiesis (PB) after stimulation with granulocyte colony-stimulating prevented HSPC mobilization after granulocyte colony- factor (G-CSF) is impaired in murine (5,6) and human stimulating factor (G-CSF) stimulation. The intimate diabetes (7,8), a condition termed mobilopathy (9). link between myelopoiesis and impaired HSPC mobilization after G-CSF stimulation was confirmed in human We herein hypothesize that myelopoiesis and mobilopa- diabetes. Using cross-transplantation experiments, we thy, described as two distinct pathological features of the found that deletion of p66Shc in the hematopoietic diabetic BM, are instead mechanistically linked. Disentan- or nonhematopoietic system partially rescued defective gling the processes linking myelopoiesis to mobilopathy has HSPC mobilization in diabetes. Additionally, p66Shc me- relevant clinical implications. First, pharmacologic mobili- diated the diabetes-induced BM microvasculature remod- zation of HSPCs is the gold standard for HSPC transplan- eling. Ubiquitous or hematopoietic restricted Osm deletion tation (10), and failure to collect robust numbers of HSPCs phenocopied p66Shc deletion in preventing diabetes- can delay engraftment, thereby worsening the outcome of associated myelopoiesis and mobilopathy. Mechanisti- patients with diabetes undergoing transplantation (5). Sec- cally, we discovered that OSM couples myelopoiesis to ond, reduction of circulating HSPCs in patients with di- mobilopathy by inducing Cxcl12 in BM stromal cells via nonmitochondrial p66Shc. Altogether, these data indicate abetes predicts the future development of micro- and that cell-autonomous activation of the OSM-p66Shc path- macrovascular complications (11,12). Glucose control effec- way leads to diabetes-associated myelopoiesis, whereas tively prevents myelopoiesis and partially rescues HSPC its transcellular hematostromal activation links myelopoi- mobilization (3,13), but many patients fail to achieve esis to mobilopathy. Targeting the OSM-p66Shc pathway necessary glucose targets. Therefore, disconnecting is a novel strategy to disconnect mobilopathy from mye- mobilopathy from myelopoiesis can provide a direct ther- lopoiesis and restore normal HSPC mobilization. apeutic strategy to restore normal HSPC mobilization.

1Veneto Institute of Molecular Medicine (VIMM), Padova, Italy Received 24 January 2019 and accepted 15 March 2019 2 – Department of Medicine DIMED, University of Padova, Padova, Italy This article contains Supplementary Data online at http://diabetes 3 Department of Surgery, Oncology and Gastroenterology, University of Padova, .diabetesjournals.org/lookup/suppl/doi:10.2337/db19-0080/-/DC1. Padova, Italy M.A. and S.C. contributed equally to this study. 4Istituto Oncologico Veneto (IOV)-IRCCS, Padova, Italy 5European Institute of Oncology (IEO), Milan, Italy © 2019 by the American Diabetes Association. Readers may use this article as 6Department of Biomedical Sciences, University of Padova, Padova, Italy long as the work is properly cited, the use is educational and not for proﬁt, and the work is not altered. More information is available at http://www.diabetesjournals Corresponding author: Gian Paolo Fadini, [email protected] or gianpaolo .org/content/license. [email protected] 1304 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

Recent studies highlight that murine and human di- The protocols were approved by the local ethical committee abetes cause BM microvascular remodeling (14) and au- and conducted in accordance with the Declaration of Hel- tonomic neuropathy (6,15), both of which can affect HSPC sinki as revised in 2000. Cross-sectional data on the asso- traffic (16,17). We previously found that BM denervation ciation between myeloid bias and circulating HSPCs were in diabetic mice accounts for impaired response to G-CSF derived from two previous studies that had been approved and is mediated by p66Shc (6). Unlike p46 and p52, p66Shc by the local ethics committee (6,25). Total and differential functions both as an adaptor protein for membrane white blood cell (WBC) counts were determined in the same receptors and a redox enzyme. Upon phosphorylation at laboratory for both studies, and CD34+ HSPC levels were Ser36, p66Shc translocates to the mitochondrial intermem- quantified by flow cytometry relative to the WBC count. brane space where it catalyzes the production of hydrogen Details are given in the previous publications (6,25). The peroxide (18), contributing to processes linked to oxidative study for G-CSF–induced mobilization was approved by the stress, including diabetic complications (19,20). local ethics committee and is registered in ClinicalTrials Besides sympathetic nervous system activation, deple- .gov (NCT01102699). This was a prospective, parallel-group tion of BM macrophages is a key event in the mobilization study of direct BM stimulation with G-CSF in subjects with cascade induced by G-CSF, because macrophage paracrine and without diabetes. Specific methods for quantifying blood activity sustains CXCL12 production (21). We have iden- cells and HSPCs were given in the previous publication (7). tified oncostatin M (OSM) as the macrophage-derived Informed consent was obtained from all participants. soluble factor that induces Cxcl12 expression in stromal cells, thereby antagonizing HSPC mobilization (22). OSM Animal Models is a cytokine of the interleukin 6 family, which signals via Diabetes was induced in 2-month-old mice by a single in- mitogen-activated protein kinase (MAPK) and the Janus traperitoneal injection of 175 mg/kg streptozotocin (STZ). kinase (JAK)/STAT pathways, leading to pleiotropic func- Blood glucose was measured using a FreeStyle glucometer tions, including modulation of inflammation and bone (Abbott, Abbott Park, IL). HSPC mobilization was induced by formation (23,24). In murine diabetes, excess BM macro- subcutaneous injection of 200 mg/kg/day G-CSF daily for 4 phages result in persistent OSM signaling, inability to days. Three-month-old mice were treated with vehicle or switch off CXCL12 levels after G-CSF, and impaired carrier free recombinant mouse OSM (495-MO/CF; R&D HSPC mobilization (22). Thus, OSM represents a candidate Systems, Minneapolis, MN) at 0.5 mg per injection every 6 h link between myelopoiesis and mobilopathy. In view of the for 48 h before analysis was performed. Total WBC count similar benefits of p66Shc deletion and OSM inhibition on was performed using the CELL-DYN Emerald hematology the diabetic stem cell mobilopathy (6,22), we have hypoth- analyzer (Abbott) on fresh EDTA-treated mouse blood. esized that the two pathways are mechanistically con- nected. In the current study, we therefore examined the Mouse Embryonic Fibroblast Transduction interplay between OSM and p66Shc in determining the Mouse embryonic fibroblasts (MEFs) were isolated from 2 2 link between myelopoiesis and mobilopathy observed in E13.5 p66Shc / mice after digestion with trypsin (Corn- experimental and human diabetes. ing) and cultured with DMEM and 10% FBS. PINCO retroviral particles were produced from the amphotropic RESEARCH DESIGN AND METHODS packaging cell line Phoenix. Cells were infected with an Mice empty vector, a vector encoding mouse full-length p66Shc, C57BL/6J wild-type (Wt) mice were purchased from The a vector encoding the mutants p66ShcS36A (S→Asub- Jackson Laboratory and established as a colony since 2001. stitution at position 36) and p66ShcQQ (EE→QQ sub- 2/2 p66Shc mice were originally obtained from Pelicci’s stitutions at positions 132–133). P3 MEFs were infected laboratory (European Institute of Oncology, Milan, Italy), with three rounds of infection with Polybrene Infection/ a colony was established at our facility in 2010, and mice Transfection Reagent (Sigma-Aldrich), followed by 96 h of have been backcrossed on the C57BL/6J background selection with 2 mg/mL puromycin. Experiments were 2/2 for .10 generations. Osm mice on the C57BL/6J performed with p4 or p5 cells. background were obtained from GlaxoSmithKline (Steven- age, U.K.), and a colony was established in 2015. For all the BM Transplantation experiments, we used sex- and age-matched animals. As- Recipient mice (3 months old) were treated with a myeloablative signment of mice to treatments or experimental groups dose of total body irradiation of 10 Gy, split in two doses of was based on a computer-generated random sequence. All 5 Gy 3 h apart and followed by an intravenous injection of 7 animal studies were approved by the Venetian Institute of BM cells from donor mice (4 3 10 /each) isolated by Molecular Medicine Animal Care and Use Committee and flushing femurs and tibias with sterile ice-cold PBS. by the Italian Health Ministry. CFU Assay Humans BM cells (3 3 104) were plated in 35-mm Petri dishes Individuals with and without diabetes were recruited at the containing 1 mL methylcellulose-based medium Metho- University Hospital of Padova Division of Metabolic Diseases. Cult supplemented with 1% penicillin/streptomycin. After diabetes.diabetesjournals.org Albiero and Associates 1305 red blood cell lysis, 25 mL/well PB was plated in 24-well Slides were mounted with an antifade aqueous mounting plates containing 0.5 mL MethoCult supplemented with medium. Images were taken with a Leica DM5000B mi- 1% penicillin/streptomycin. Colony formation was scored croscope, equipped with a DFC300 FX CCD camera, or with after 10 days of culture. When required, murine recombi- Cytell (GE Healthcare, Milan, Italy). Images were then nant S100A8/9 heterodimer (BioLegend) was mixed with processed with Fiji/ImageJ 1.50 software (National Insti- the MethoCult medium. tutes of Health, Bethesda, MD) or with Adobe Photoshop CS2 9.0.2 software (Adobe Systems, San Jose, CA). Flow Cytometry Flow cytometry was performed on BM cells or EDTA- Morphometric Measurements treated PB. BM cells were isolated by flushing femurs Vessel size and shape were measured using Fiji/ImageJ and tibias with ice-cold MACS Separation Buffer (Miltenyi 1.50 software. Briefly, random 500 mm2 fields from the Biotec GmbH, Gladbach, Germany) through a 40-mm cell epiphyseal and diaphyseal region (three each, at least) of strainer. Then 100 mL PB or BM cells was incubated with the samples were analyzed. Vessel structure was visualized antibodies for 15 min in the dark at room temperature. by laminin staining, and regions of interest were manually After red blood cell lysis, samples were resuspended in outlined in Fiji/ImageJ. Area and shape parameters, such as 200 mL PBS, and data were acquired with a FACSCanto (BD circularity, were recorded. The bivariate distribution of area Biosciences) cytometer, followed by analysis using FlowJo and circularity was visualized using the Bivariate Kernel software (Tree Star). Density Estimation 1.0.9 in R 3.1 software. BM innervation was determined by Tyr-HO staining. Arterioles were iden- BM-Derived Mesenchymal Stem Cells tified in the whole femur section, and diameters of arterioles Murine BM-derived mesenchymal stem cells (BM-MSCs) were measured with Fiji/ImageJ. were isolated by flushing the BM of 3-month-old mice and cultured in minimum essential medium-a containing 10% Molecular Biology FBS, glutamine (2 mmol/L), and penicillin-streptomycin. RNA was isolated from flushed BM or cells by using QIAzol Passage 3–6 was used in all experiments. For gene expres- or with Total RNA Purification Micro Kit (Norgen Biotek) sion analysis, cells were treated with murine recombinant and quantified with a NanoDrop 2000 Spectrophotometer OSM (R&D Systems) for 48 h in serum-free media. (Thermo Fisher Scientific, Waltham, MA). cDNA was syn- thesized using the SensiFAST cDNA Synthesis Kit (Bioline, Tissue Processing London, U.K.). Quantitative PCR was performed using the Femur bones were fixed in 4% paraformaldehyde and SensiFAST SYBR Lo-ROX Kit (Bioline) via a QuantStudio decalcified. Bones were then washed with PBS, embedded 5 Real-Time PCR System (Thermo Fisher Scientific). A list in Killik cryostat medium (Bio-Optica, Milan, Italy), and of primers can be found in Supplementary Table 1. frozen in liquid nitrogen–cooled 2-methylbutane (Sigma- Aldrich). Longitudinal 10-mm-thick femur sections were Statistical Analysis obtained with a Leica CM 1950 cryostat (Leica Biosystems Continuous data are expressed as mean 6 SEM, whereas S.r.l., Milan, Italy), placed on Superfrost Plus slides categorical data are presented as the percentage. Normal- (J1800AMNZ; Gerhard Menzel GmbH, Braunschweig, ity was checked using the Kolmogorov-Smirnov test, and Germany), and stored at 280°C. nonnormal data were log-transformed before analysis. Comparison between two or more groups was performed Protein Phosphorylation by Flow Cytometry using the Student t test and ANOVA for normal variables Confluent BM-MSCs were treated with SCH772984 or the Mann-Whitney U test and Kruskal-Wallis test for (4 nmol/L) (Cayman Chemical) overnight or with Stattic nonnormal variables that could not be log-transformed. (2.5 mmol/L) (Selleckchem) for 1 h before adding recombi- Bonferroni adjustment was used to account for multiple nant OSM (R&D Systems) for 30 min in serum-free media. testing. Linear correlations were checked using the Pear- Cells were detached by scraping and incubated with PE son r coefficient. Statistical analysis was accepted at P , mouse anti-Stat3 (pY705) or PE mouse anti-extracellular 0.05. Statistical analysis was performed using GraphPad signal–regulated kinase (ERK)1/2 (pT202/pY204) (both Prism 6, Matlab, and SPSS 21 software. from BD Biosciences) in Perm Buffer III (BD Biosciences) according to the manufacturer’s instructions. RESULTS Mobilopathy Associates With Myelopoiesis in Immunohistochemistry Experimental Diabetes Femur sections were air dried for 20 min and then in- We first evaluated whether myelopoiesis and mobilopathy cubated with blocking solution. Sections were incubated coexist in murine diabetes. We found that STZ-induced with primary antibody: anti-laminin (1:50 for 4 days), diabetic mice had an approximately twofold expansion of anti–tyrosine-hydroxylase (Tyr-OH) (1:200 for 4 days), PB granulocytes compared with nondiabetic mice (P , and anti-CD150 (1:50 for 3 days). Sections were then 0.001) (Fig. 1A and B and Supplementary Fig. 1), resulting washed with PBS and incubated with secondary antibodies. in a strikingly six times higher granulocyte-to-lymphocyte 1306 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

Figure 1—Mobilopathy and myelopoiesis in experimental diabetes. Panels A–G report the comparison between diabetic (n = 16) and nondiabetic control (n = 12) mice in total WBC counts (A), absolute counts of lymphocyte (Lympho), monocytes (Mono), and granulocytes (Granulo) (B), as well as the G-to-L ratio (C). D: Comparison of FACS-defined CMPs and GMPs. E: Results of the CFU assay from BM cells. GEMM, granulocyte-erythroid-macrophage-MK colonies; GM, granulocyte-macrophage colonies; M, macrophage colonies; G, granulocyte colonies. F: Percentages of BM macrophages over total BM cells in diabetic vs. control mice in the unstimulated and G-CSF–stimulated conditions. G: Gene expression of Osm in the BM of diabetic vs. control mice. *P , 0.05 for the comparison between diabetic and control mice. Panels H–J illustrate HSPC mobilization in diabetic vs. nondiabetic mice. HSPCs, defined as LKS, were quantified before and after G-CSF administration and are reported as fold change from baseline in nondiabetic control mice (n = 20) and in diabetic mice (n = 20). H: Individual lines, indicating single mice, are shown along with the average fold change (95% CI) for each time point and the respective P values. I: Comparison of the fold change in LKS cell levels after G-CSF between control and diabetic mice. J:Comparisonofthe percentage of mice achieving a mobilization response of at least 1.5-fold in the diabetic vs. nondiabetic control condition. Histograms indicate mean 6 SEM. Box plots indicate median with interquartile range, and whiskers indicate range.

(G-to-L) ratio (P , 0.001) (Fig. 1C). The BM of diabetic The profound degree of mobilization impairment allowed mice contained higher numbers of granulocyte-monocyte us to use the G-CSF mobilization assay as a robust readout progenitors (GMPs) at the expense of CMPs (Fig. 1D). As for mobilopathy in subsequent mouse experiments. a consequence, the clonogenic assay of BM cells showed an increased output of macrophage and granulocyte colonies Mobilopathy Associates With Myelopoiesis in Human from diabetic versus nondiabetic mice (2.2 times and 2.8 Diabetes times, respectively) (Fig. 1E). The diabetic BM showed We then checked whether myelopoiesis and mobilopathy excess macrophages both in basal unstimulated conditions occurred simultaneously in human diabetes. We first ana- and after G-CSF stimulation (Fig. 1F). These cells are lyzed cross-sectional data of two studies wherein circulating known to produce OSM (22), and Osm gene expression WBC types and levels of CD34+ HSPCs were determined in in the BM of diabetic mice was indeed upregulated 7.7 the same sample (6,25) (Supplementary Table 2). In a pooled times compared with that in nondiabetic mice (P = 0.006) cohort of 344 subjects, the patients with diabetes (n = 108; (Fig. 1G). Altogether, these data are consistent with ex- 74% type 2) displayed 25% lower levels of HSPCs and a 24% aggerated myelopoiesis and myeloid bias in diabetic mice. higher neutrophil-to-lymphocyte (N-to-L) ratio than indi- As the resulting overproduction of OSM can hamper viduals without diabetes (Fig. 2A). Higher N-to-L ratio and mobilization (22), we evaluated whether mobilopathy lower CD34+ HSPCs remained significantly associated with occurred in the same mice. Preliminary to this, we verified diabetes after adjusting for age, sex, BMI, hypertension, 2 that the baseline PB level of HSPCs (Lin c-Kit+Sca-1+ lipids, coronary artery disease, and retinopathy (Supplemen- [LKS] cells) was nonsignificantly different in diabetic tary Table 3). We also found a significant inverse correlation versus nondiabetic mice (Supplementary Fig. 2). After between the N-to-L ratio and the steady-state level of PB G-CSF was administered for 4 days, the HSPC level in- HSPCs (r = 20.28) (Fig. 2B). Considering glucose control as creased by 4.38 times in nondiabetic but not in diabetic a continuous variable in the entire cohort, we found a sig- H fi mice (Fig. 1 ). This difference in the fold change of the ni cant inverse correlation between HbA1c and HSPC levels PB-LKS cell level between diabetic and nondiabetic mice (r = 20.23; P , 0.001) (Fig. 2C) and a direct correlation was highly significant (Fig. 1I): 80% of nondiabetic mice vs. between HbA1c and the N-to- L ratio (r = 0.21; P , 0.001) 10% of diabetic mice achieved a mobilization response of (Fig. 2D), which persisted (both with P , 0.01) after adjust- at least 1.5-fold (Fig. 1J). The colony-forming unit (CFU) ing for the above-mentioned confounders. These data sug- assay from PB cells confirmed the absence of functional gest that myeloid bias is linked to a reduction in HSPC levels HSPC mobilization in diabetes (Supplementary Fig. 3A). in human diabetes, possibly driven by hyperglycemia. diabetes.diabetesjournals.org Albiero and Associates 1307

Figure 2—Myelopoiesis and mobilopathy in human diabetes. A: Comparison of circulating CD34+ HSPCs and the N-to-L ratio in a pooled cohort of patients without diabetes (n = 236) and with diabetes (n = 108). *P , 0.05. B: Linear correlation between HSPC levels and the N-to-L ratio. Regression coefficients are reported for the entire cohort (along with P values) and for the patients without and with diabetes separately. Linear correlation between HbA1c and HSPC levels (C) or the N-to-L ratio (D): the regression line with its 95% CI is shown along with the regression coefficients and P values. E: Comparison between patients with and without diabetes in the increase (fold change) in HSPC levels after G-CSF and in the baseline N-to-L ratio. *P , 0.05. F: Linear correlation between the baseline N-to-L ratio and the increase (fold change) in HSPC levels after G-CSF stimulation. Regression coefficients are reported for the entire cohort (along with the P value) and for the patients without and with diabetes separately. Histograms indicate mean 6 SEM.

Second, we evaluated whether myeloid bias was asso- Deletion of p66Shc Prevents Diabetes-Associated ciated with mobilopathy by analyzing data from a previous Myelopoiesis and Mobilopathy prospective study wherein patients with and without Having shown that myelopoiesis and mobilopathy concur diabetes (n = 43) received low-dose G-CSF to test HSPC in murine and human diabetes, we explored the mecha- mobilization (7). The fold change in G-CSF–induced nisms driving their association. We first focused on HSPC levels versus baseline was significantly lower, p66Shc, which we previously showed to be responsible and the pre–G-CSF N-to-L ratio tended to be higher for diabetes-associated BM denervation and mobilopathy (P = 0.06) in patients with diabetes versus those without (6). BM p66Shc gene expression was more than twofold diabetes (Fig. 2E). Remarkably, there was a significant higher in diabetic versus control mice (P = 0.003) (Sup- inverse correlation between the N-to-L ratio and HSPC plementary Fig. 4), consistent with prior data in mice and mobilization (r = 20.32; P =0.03)(Fig.2F). These humans (26,27). In the nondiabetic condition, we found no results cannot prove causality, because secondary differences in WBC count and subtypes, G-to-L ratio, BM analyses of previously collected cohort data can be sub- colonies, and CMPs/GMPs, as well as BM macrophages 2 2 jected to bias and prone to false-positive signals. None- between Wt and p66Shc / mice (Fig. 3A–G). However, in 2 2 theless, we confirm that myelopoiesis and mobilopathy p66Shc / mice, diabetes did not increase PB granulocytes are associated in human diabetes as they are in murine counts, and granulocytes were significantly lower than in diabetes. Wt diabetic mice (Fig. 3D and Supplementary Fig. 5), as 1308 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

Figure 3—p66Shc deletion protects from diabetes-induced myelopoiesis and mobilopathy. Myelopoiesis was evaluated in nondiabetic and diabetic Wt and p66Shc2/2 mice (n . 10/group, unless specified) by comparing PB WBC count (A), WBC types (B–D), the G-to-L ratio (E), the BM cell CFU assay (F), FACS-defined BM CMPs and GMPs (G), and the percentage of BM macrophages (n = 5/group) (H). GEMM, granulocyte-erythroid-macrophage-MK colonies; GM, granulocyte-macrophage colonies; M, macrophage colonies; G, granulocyte colonies. *P , 0.05 diabetic vs. control (Ctrl); †P , 0.05 p66Shc2/2 vs. Wt. I and J: CFU assay performed using BM cells from Wt (I)orp66Shc2/2 (J) mice, which were stimulated ex vivo with 2 mg/mL S100A8/9. *P , 0.05 for the comparison with untreated control cells. K: Percentage of FACS-defined GMP in the BM of nondiabetic Wt and p66Shc2/2 mice treated with vehicle or S100A8/9 (20 mg/kg twice a day for 3 days). L: Mobilization of HSPCs was evaluated in p66Shc2/2 diabetic and nondiabetic mice (n = 5/group) by enumerating circulating LKS cells. *P , 0.05 in post–G-CSF vs. baseline. M: Schematic representation of the generation of hematopoietic and nonhematopoietic p66Shc2/2 mice. N: The fold change with 95% CI of LKS cells (calculated as post–G-CSF divided by pre–G-CSF levels) in Wt diabetic mice, p66Shc2/2 diabetic mice, and diabetic mice with crossed BMT (n =4–5/group).*Significantly different from 1.0, denoted by the dashed line indicating no effect. O: The change in the G-to-L ratio induced by diabetes is plotted for Wt mice, p66Shc2/2 mice, and mice with crossed BMT. The annotation under panels N and O indicates the genotype of the host (receiver mice) or the BM donor mice. KO, knockout. *Significantly different from 1.0, denoted by the dashed line.

2 2 was the G-to-L ratio (Fig. 3E). Furthermore, the diabetes- were completely prevented in p66Shc / mice. As a result, induced increase in myeloid CFUs (Fig. 3F), CMP/GMP the surge in BM expression of Osm observed in diabetic Wt imbalance (Fig. 3G), and excess BM macrophages (Fig. 3H) mice, which derives from BM macrophages (22), was diabetes.diabetesjournals.org Albiero and Associates 1309

Figure 4—p66Shc deletion ameliorates BM microvascular remodeling in diabetes. A: BM sections were stained with Hoechst (total cellularity) and anti-laminin to evaluate the microvasculature: based on specific thresholds, vascular items were scored as arterioles, sinusoids, or small vessels of capillary caliber. B: The four aligned panels report the numbers of any vessel, arterioles, sinusoids, and capillary-size structures per field in diabetic and nondiabetic Wt and p66Shc2/2 mice. *P , 0.05 diabetic vs. nondiabetic; †P , 0.05 vs. Wt. C: Kernel density plot of vascular items scored based on size (x-axis) and circularity (y-axis): the area of the plot identified by the dashed box contains large irregular items (likely sinusoids), which was reduced by diabetes in Wt but not in p66Shc2/2 mice. D: BM sections were stained with Hoechst, anti-laminin (blood vessels), and anti–Tyr-OH, a marker of sympathetic nerve fibers. A representative example from a nondiabetic Wt mouse is shown to illustrate the pattern of Tyr-OH staining. Number of innervated arterioles/field (E) and the fraction of innervated arterioles (F) over the total number of arterioles. Histograms indicate mean 6 SEM, with superimposed individual data points (n = 5/group). *P , 0.05 vs. nondiabetic control.

2 2 absent in p66Shc / mice (Supplementary Fig. 6). The link performed cross-transplantation experiments as illus- between hyperglycemia and myelopoiesis has been attrib- trated in Fig. 3M. We conﬁrmed that BM transplantation uted to the accumulation of advanced glycation end prod- (BMT) did not impinge on G-CSF responsiveness, as non- ucts (AGEs) and the receptor for the AGE (RAGE) ligand diabetic but not diabetic Wt mice that received BMTs from S100A8/9 (3). Interestingly, RAGE signaling has been Wt mice were able to mobilize functional HSPCs (Supple- previously linked with downstream p66Shc activation mentary Fig. 3B and C). Wt mice that received BMTs from 2 2 2 2 (28). We found that S100A8/9 potentiated myelopoiesis p66Shc / mice (p66Shc / →Wt BMT) and were rendered in vitro by BM cells of Wt mice, evidenced by a 1.75-fold diabetic showed a partial restoration of HSPC increase and a 2.0-fold increase in macrophage and granulocyte after G-CSF stimulation. An almost identical, but partial, 2 2 colonies, respectively (Fig. 3I). However, such effect was improvement was observed in diabetic p66Shc / mice 2 2 2 2 completely abolished in p66Shc / BM cells (Fig. 3J). who received BMTs from Wt mice (Wt→p66Shc / BMT), In vivo treatment of nondiabetic Wt mice with S100A8/ and in both cases, the fold change in HSPC count was lower 2 2 9 increased GMP and myeloid cell colonies, but such effect than that in ubiquitous p66Shc / diabetic mice (Fig. 3N). 2 2 was not observed in p66Shc / mice (Fig. 3K and Supple- In contrast, while diabetes increased granulocyte counts 2 2 2 2 mentary Fig. 7). Together, these data indicate that p66Shc and the G-to-L ratio in Wt→p66Shc / , p66Shc / →Wt is required for the effects of hyperglycemia on myelopoi- BMT mice were largely protected from elevation of the esis, possibly by preventing the activity of S100 proteins. G-to-L ratio induced by diabetes (Fig. 3O and Supplemen- In agreement with our previous study (6), p66Shc de- tary Fig. 9). letion partially rescued HSPC mobilization in diabetic Knowing that myeloid-biased HSPCs reside in mega- mice, as indicated by the 1.9 times increase in LKS cell karyocytic (MK) niches (29), we analyzed MK density by counts after G-CSF administration (Fig. 3L). The CFU assay staining BM sections with anti-CD150. As previously showed that mobilized HSPCs were functionally compe- noted by others (30), MKs were increased by 60% in tent, as G-CSF increased PB hematopoietic colonies both in diabetic compared with nondiabetic Wt mice (P = 2 2 diabetic and nondiabetic p66Shc / mice (Supplementary 0.004). However, such an effect was not observed in 2 2 Fig. 8). To dissect the hematopoietic-intrinsic and -extrinsic p66Shc / mice (Supplementary Fig. 10), providing a fur- roles of p66Shc in regulating HSPC mobilization, we ther explanation for the protection of ubiquitous and 1310 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

Figure 5—Osm deletion protects from diabetes-induced myelopoiesis and mobilopathy. A: Mobilization of HSPCs, deﬁned as LKS cells, induced by G-CSF in nondiabetic and in diabetic Osm2/2 mice (n = 5). *P , 0.05 vs. baseline. B: Percentage of BM macrophages over total BM cellularity in diabetic and nondiabetic Wt (same as Fig. 1F) and Osm2/2 mice in unstimulated (Unst.) and G-CSF–stimulated conditions (n =8–10/group). *P , 0.05 vs. control; †P , 0.05 vs. unstimulated; ‡P , 0.05 vs. Wt. Total WBC count (C) and G-to-L ratio (D) in nondiabetic and STZ diabetic Wt and Osm2/2 mice (n . 10/group). Statistics marks as in panel B. E: Schematic representation of the BMT experiment to generate hematopoietic-restricted Osm-deleted mice. F: Mobilization of HSPCs in nondiabetic and diabetic hematopoietic-restricted Osm2/2 mice (n =5).*P , 0.05 vs. baseline. Total WBC count (G) and G-to-L ratio (H) in diabetic and nondiabetic Wt mice with Osm2/2 BM. Histograms indicate mean 6 SEM with superimposed individual data points, where appropriate.

2 2 hematopoietic-restricted p66Shc / mice from diabetes- distribution according to size and circularity, we found that induced myelopoiesis. diabetes also led to a reduction of larger irregular vessels, Altogether, these data indicate that hematopoietic- likely sinusoids, that was prevented by p66Shc deletion (Fig. intrinsic and -extrinsic mechanisms are responsible for 4C). We then evaluated sympathetic innervation of the BM the rescue of HSPC mobilization by p66Shc deletion, and found that Tyr-OH+ sympathetic terminals were almost whereas prevention of myelopoiesis is hematopoietic cell exclusively located close to arteriolar walls (Fig. 4D). The intrinsic. percentage of innervated arterioles was significantly reduced by more than twofold in diabetic versus nondiabetic Wt 2 2 p66Shc Deletion Improves Diabetes-Induced BM mice, but not in p66Shc / mice (P , 0.001) (Fig. 4E and F). Microvascular Remodeling Taken together, these findings indicate that deletion of The partial restoration of G-CSF–induced HSPC mobilization p66Shc protects BM from microvascular remodeling, which 2/2 in Wt→p66Shc BMT diabetic mice suggested that deletion can explain the partial rescue of HSPC mobilization in of p66Shc exerted protective effects on the BM stroma nonhematopoietic p66Shc-deleted mice. against hyperglycemic damage. We previously reported that 2 2 p66Shc / mice were protected from BM sympathectomy Osm Deletion Phenocopies p66Shc Deletion induced by diabetes (6). We herein characterized microvas- The hematopoietic cell–intrinsic mechanism whereby 2 2 cular BM remodeling in p66Shc / versus Wt diabetic and G-CSF exerts its mobilizing activity relies on suppression nondiabetic mice. In the peculiar BM microcirculation, the of BM macrophages (21). This pathway is independent nutrient arterial system drains into sinusoids with capillary- from the stromal effect of G-CSF through nerve terminals, size vessels, and the irregular sinusoid lumen is occasionally as mice sympathectomized by 6-OH dopamine showed compressed to capillary caliber (31). Using an unbiased auto- a normal post–G-CSF suppression of BM macrophages instructed procedure to score BM vessels (Fig. 4A), we found despite being unable to mobilize HSPCs (Supplementary that the total vascular density and numbers of arterioles and Fig. 11). This finding indicates that both hematopoietic sinusoids were similar, but a significant 2.5-fold reduction and nonhematopoietic effects of G-CSF are required to in capillary-size structures was noted in diabetic versus non- yield a full HSPC mobilizing response and justifies the 2 2 diabetic Wt mice (P = 0.006) (Fig. 4B). Remarkably, the partial recovery of mobilization in Wt↔p66Shc / cross- density of BM capillary-sizevesselswasnotreducedin transplanted animals. 2 2 p66Shc / diabetic versus nondiabetic mice and was higher We previously demonstrated that antibody-mediated 2 2 in p66Shc / versus Wt diabetic mice (P = 0.03). With OSM neutralization allowed HSPC mobilization in diabetic a more detailed morphometric analysis of BM blood vessel mice by relieving the brake of CXCL12 produced by stromal diabetes.diabetesjournals.org Albiero and Associates 1311

Figure 6—Signaling of OSM requires p66Shc. A: Hypothetical model wherein recruitment of p66Shc to OSMR, instead of migration to mitochondria (M), is required for OSM to regulate Cxcl12 expression. N, nucleus. B: Gene expression of Cxcl12 in BM-MSCs isolated from Wt or p66Shc2/2 mice and treated with OSM (30 ng/mL) or vehicle (control) (n = 5/condition). *P , 0.05 vs. control. C: Gene expression of Cxcl12 in MEFs isolated from p66Shc2/2 mice and transfected with an empty vector or vector encoding for Wt p66Shc (p66WT), serine 36 mutated p66Shc (p66S36A), or catalytically inactive p66Shc (p66qq) and treated with OSM or vehicle (control) (n = 4/condition). *P , 0.05 vs. control. D: Phosphorylation of ERK1/2 (p-ERK1/2) on threonine 202 or tyrosine 204 and phosphorylation of STAT3 (p-STAT3) on tyrosine 705 was evaluated by flow cytometry in Wt and p66Shc2/2 MSCs treated with vehicle (Ctrl) or mouse OSM (mOSM) with and without an inhibitor of ERK (SCH772984) or STAT (Stattic), respectively (a representative experiment of three replicates is shown). E: Signaling model wherein p66Shc recruited to the OSMR is required for ERK activation, but not for JAK-STAT activation via gp130/OSMR, although both ERK and STAT are required for OSM to induce Cxcl12. F: Gene expression of Cxcl12 in the BM of Wt, Osm2/2, and p66Shc2/2 mice treated with vehicle (control) or OSM (0.5 mg every 6 h for 48 h).*P , 0.05 vs. control. G: Levels of HSPCs, defined as LKS cells, in Wt, Osm2/2,andp66Shc2/ 2 mice treated with vehicle (control) or OSM. *P , 0.05 vs. control. H: Fold change of the G-to-L ratio in Wt, Osm2/2,andp66Shc2/2 mice treated with vehicle (control) or OSM. *P , 0.05 vs. control. Histograms indicate mean 6 SEM with superimposed individual data points for each experiment. cells (22). Consistent with the notion that OSM retains a macrophage-derived paracrine factor but it is also re- 2 2 HSPCs in the BM niche, nondiabetic Osm / mice dis- quired for accumulation of BM macrophages in a para- played higher HSPC levels in unstimulated PB than Wt crine-autocrine loop, as already seen by others in the heart 2 2 mice, which was not further increased by diabetes (Sup- (32). Indeed, diabetic Osm / mice had normal WBC plementary Fig. 2), and HSPC mobilization in diabetic mice counts (Fig. 5C) but significantly lower levels of PB gran- was partially rescued toward normal levels (2.2 times) by ulocytes compared with Wt diabetic mice (Supplementary genetic Osm deletion (P = 0.01) (Fig. 5A). In addition, we Fig. 12), and the G-to-L ratio was restored toward the observed a marked (;80%) reduction of BM macrophages levels seen in nondiabetic mice (Fig. 5D). To avoid the 2 2 in Osm / mice, both in the diabetic and nondia- confounding factor of the absence of nonhematopoietic 2 2 betic condition, which was further suppressed by G-CSF OSM in ubiquitous Osm / mice, we transplanted BM cells 2 2 (Fig. 5B). This result suggests that OSM is not only from Osm / mice into Wt mice and induced diabetes 1312 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

concentration of OSM was chosen based on a previous dose-effect curve (22). In contrast to p46 and p52, p66Shc acts as both an adaptor protein for signaling cascades and a mitochondrial redox protein (18). To dissect whether mitochondrial function of p66Shc was required for OSM signaling and 2 2 Cxcl12 induction, we transfected p66Shc / MEFs with an → empty vector or vectors encoding Wt p66, 36Ser Ala mutated p66 (which cannot translocate to mitochondria), or a catalytically inactive p66 (p66qq), and then treated MEFs 2 2 with OSM: Cxcl12 induction by OSM in p66Shc / MEFs was rescued by expression of Wt, Ser36 mutated, or catalytically inactive p66Shc but not empty vector (Fig. 6C), suggesting that the adaptor function, and not the mitochondrial function, of p66Shc was required for OSM signaling. In addition, we found that activation of ERK by OSM was abolished in BM-derived stromal cells from 2 2 p66Shc / mice, while activation of STAT3 was unaffected (Fig. 6D). This set of experiments is in line with the model E Figure 7—Schematic representation of the link between myelopoi- depicted in Fig. 6 , where p66Shc is recruited to OSMR esis and mobilopathy exerted by the OSM-p66Shc signaling path- and cooperates to activate the MAPK pathway, which, way. PMNs, polymorphonuclear cells; MF, macrophages; M, along with STAT3 activation via gp130, is needed to induce “ ” mitochondrion; N, nucleus. Red bullets marked with + denote Cxcl12. stimulatory effects. Finally, to gather in vivo evidence that OSM signaling requires p66Shc, we treated mice with systemic OSM injections. Gene expression of Cxcl12 in the BM was 4 weeks later (Fig. 5E). After 4 weeks of diabetes, we tested significantly induced in Wt (4.5 times; P = 0.01) and in 2 2 2 2 HSPC mobilization after G-CSF treatment and found Osm / mice (7.9 times; P = 0.01) but not in p66Shc / that hematopoietic-restricted Osm deletion rescued HPSC mice (0.8 times; P = 0.77) (Fig. 7F). In parallel, the more mobilization in diabetic mice toward normal levels (5.2 than twofold higher levels of HSPCs observed in the 2 2 2 2 times; P = 0.03) (Fig. 5F). In addition, hematopoietic- steady-state basal condition in Osm / and in p66Shc / restricted knockout of Osm largely prevented the surge mice could be significantly suppressed by OSM injection in 2 2 2 2 in granulocyte levels (Fig. 5G and Supplementary Fig. 12) Osm / mice (P = 0.04) but not in p66Shc / mice (Fig. 6G). and in the G-to-L ratio induced by diabetes (Fig. 5H). These These data support the concept that regulation of HSPC data indicated that Osm deletion prevented hyperglyce- trafficking by OSM via Cxcl12 requires p66Shc. Further- mia-induced myelopoiesis and mobilopathy in a hemato- more, injection of OSM increased circulating granulocytes 2 2 poietic cell–intrinsic manner. and reduced lymphocytes in both Wt and Osm / mice, thereby increasing twofold the G-to-L ratio, but this effect 2 2 p66Shc Is Required for the Stem Cell–Retaining Activity was absent in p66Shc / mice (Fig. 6H), demonstrating that of OSM the effect of OSM on myelopoiesis is also dependent on 2 2 2 2 Since Osm / mice phenocopied p66Shc / mice in protect- p66Shc. ing against diabetes-induced myelopoiesis and mobilopathy, we hypothesized that OSM signaling required downstream p66Shc. OSM signals through heterodimers of DISCUSSION the OSM receptor (OSMR) and gp130, which elicit intra- Defective HSPC mobilization in response to G-CSF is cellular events leading to activation of the MAPK and JAK- a consistent finding in experimental and human diabetes STAT3/5 pathways (33). Shc proteins cooperate with (8), but the underlying causes are incompletely under- other adaptor proteins to transduce membrane receptor stood. Our new data indicate that mobilopathy is inti- signals to MAPK. We previously found that both STAT3 mately linked with myelopoiesis, an underlying driver of and MAPK are required for Cxcl12 induction by OSM in diabetes-associated inflammation. We herein show a novel BM stromal cells (22). Here, we hypothesized that the OSM-p66Shc signaling pathway that is overactive in di- adaptor function of p66Shc is required for OSMR signal abetes against HSPC mobilization via mechanisms that are transduction to MAPK to induce Cxcl12 (model shown in hematopoietic cell intrinsic and extrinsic (Fig. 7). OSM is Fig. 6A). We found that the ability of OSM (30 ng/mL; produced by myeloid inflammatory cells that are exceed- ;1 mmol/L) to stimulate Cxcl12 gene expression in ingly present in the diabetic BM as part of the enhanced BM-derived stromal cells (3.6 times) was completely myelopoiesis induced by hyperglycemia (22). In turn, OSM abolished in the absence of p66Shc (Fig. 6B). The signal transduction is activated in BM stromal cells via diabetes.diabetesjournals.org Albiero and Associates 1313 nonmitochondrial p66Shc to induce CXCL12 production, M.A, S.C., S.T., L.M., M.D’A., V.S., R.C., G.Z., A.R., A.C., M.G, A.A., and G.P.F. thereby retaining HSPCs in the BM niche (Fig. 6). Notably, reviewed and edited the manuscript. G.P.F. is the guarantor of this work and, as p66Shc also mediates microvascular remodeling of the di- such, had full access to all the data in the study and takes responsibility for the abetic BM that can jeopardize HSPC traffic (Fig. 4). G-CSF integrity of the data and the accuracy of the data analysis. exerts its mobilizing function by acting on hematopoietic cells References and on the BM stroma (21). Remarkably, both hemato- 1. Donath MY, Shoelson SE. Type 2 diabetes as an inflammatory disease. Nat poietic and nonhematopoietic p66Shc deletion was needed Rev Immunol 2011;11:98–107 to restore HSPC mobilization response to G-CSF in diabetes (Fig. 2. Devaraj S, Glaser N, Griffen S, Wang-Polagruto J, Miguelino E, Jialal I. 3). Hematopoietic-restricted p66Shc deletion partially res- Increased monocytic activity and biomarkers of inflammation in patients with type cued mobilization in diabetic mice, along with inhibition 1 diabetes. Diabetes 2006;55:774–779 of diabetes-induced MK expansion and myelopoiesis. 3. Nagareddy PR, Murphy AJ, Stirzaker RA, et al. Hyperglycemia promotes Hyperglycemia-driven myelopoiesis arises from the myelopoiesis and impairs the resolution of atherosclerosis. Cell Metab 2013; 17:695–708 skewed hematopoiesis stimulated by RAGE ligands (3), an 4. Nagareddy PR, Kraakman M, Masters SL, et al. Adipose tissue macrophages effect that we found requires p66Shc. At the same time, – Osm promote myelopoiesis and monocytosis in obesity. Cell Metab 2014;19:821 835 hematopoietic deletion prevented diabetes-induced mye- 5. Ferraro F, Lymperi S, Méndez-Ferrer S, et al. Diabetes impairs hematopoietic lopoiesis, and the ability of OSM to stimulate myelopoiesis also stem cell mobilization by altering niche function. Sci Transl Med 2011;3:104ra101 2/2 required p66Shc. The striking similarities between Osm 6. Albiero M, Poncina N, Tjwa M, et al. Diabetes causes bone marrow auto- 2/2 and p66Shc mice are indeed consistent with the notion that nomic neuropathy and impairs stem cell mobilization via dysregulated p66Shc and OSM couples myelopoiesis with mobilopathy via p66Shc. Sirt1. Diabetes 2014;63:1353–1365 These data together indicate that activation of the OSM- 7. Fadini GP, Albiero M, Vigili de Kreutzenberg S, et al. Diabetes impairs stem cell p66Shc pathway drives diabetes-associated myelopoiesis and proangiogenic cell mobilization in humans. Diabetes Care 2013;36:943–949 in a cell-autonomous way, whereas its transcellular hemato- 8. Fadini GP, DiPersio JF. Diabetes mellitus as a poor mobilizer condition. Blood – stromal activation links myelopoiesis to mobilopathy. Rev 2018;32:184 191 9. DiPersio JF. Diabetic stem-cell “mobilopathy”. N Engl J Med 2011;365:2536–2538 Understanding HSPC mobilization unresponsiveness to 10. Schmitz N, Linch DC, Dreger P, et al. Randomised trial of filgrastim-mobilised G-CSF has clinical implications for patients undergoing peripheral blood progenitor cell transplantation versus autologous bone-marrow HSPC collection for transplantation purposes (8). Thus, transplantation in lymphoma patients. Lancet 1996;347:353–357 interrupting the OSM-p66Shc pathway provides a thera- 11. Rigato M, Bittante C, Albiero M, Avogaro A, Fadini GP. Circulating progenitor peutic strategy in conditions of poor HSPC mobilization, cell count predicts microvascular outcomes in type 2 diabetic patients. J Clin like diabetes. Diabetic stem cell mobilopathy precedes Endocrinol Metab 2015;100:2666–2672 reduction of steady-state PB HSPCs in human diabetes 12. Fadini GP, Rigato M, Cappellari R, Bonora BM, Avogaro A. Long-term (7), which in turn has been linked with worsening of diabetic prediction of cardiovascular outcomes by circulating CD34+ and CD34+CD133+ complications (11,12). Mobilopathy preceded the reduction stem cells in patients with type 2 diabetes. Diabetes Care 2017;40:125–131 of PB HSPCs also in diabetic mice (Fig. 1 and Supplementary 13. Fadini GP, Sartore S, Schiavon M, et al. Diabetes impairs progenitor cell Fig. 2). However, since our new data reveal a causal link mobilisation after hindlimb ischaemia-reperfusion injury in rats. Diabetologia 2006;49:3075–3084 between myelopoiesis and mobilopathy, future studies 14. Spinetti G, Cordella D, Fortunato O, et al. Global remodeling of the vascular should better clarify whether diabetes outcomes are more fl stem cell niche in bone marrow of diabetic patients: implication of the microRNA- related to alterations in blood in ammatory cells, circulat- 155/FOXO3a signaling pathway. Circ Res 2013;112:510–522 ing stem cells, or stem cell mobilization. 15. Dang Z, Maselli D, Spinetti G, et al. Sensory neuropathy hampers noci- In summary, we provide evidence that an overactive ception-mediated bone marrow stem cell release in mice and patients with di- OSM-p66Shc pathway couples diabetes-associated myelo- abetes. Diabetologia 2015;58:2653–2662 poiesis with HSPC mobilopathy. In addition to rescuing 16. Itkin T, Gur-Cohen S, Spencer JA, et al. Distinct bone marrow blood vessels HSPC mobilization, tackling this pathway in the BM could differentially regulate haematopoiesis. Nature 2016;532:323–328 provideanewavenuefortheimprovementsofthe 17. Katayama Y, Battista M, Kao WM, et al. Signals from the sympathetic nervous diabetes-related inflammation and complication risk. system regulate hematopoietic stem cell egress from bone marrow. Cell 2006; 124:407–421 18. Giorgio M, Migliaccio E, Orsini F, et al. Electron transfer between cytochrome Funding. The study was supported by the following grants to G.P.F.: European c and p66Shc generates reactive oxygen species that trigger mitochondrial ap- Foundation for the Study of Diabetes Novartis 2013 grant and Lilly 2016 grant, optosis. Cell 2005;122:221–233 Ministry of University and Education Progetti di Rilevante Interesse Nazionale 19. Cosentino F, Francia P, Camici GG, Pelicci PG, Lüscher TF, Volpe M. Final (PRIN) grant 2015, Italian Diabetes Society/Lilly grant 2017, and Fondazione common molecular pathways of aging and cardiovascular disease: role of the Cariplo 2016-0922. p66Shc protein [published correction appears in Arterioscler Thromb Vasc Biol Duality of Interest. M.A., S.C., and G.P.F. are the inventors of a patent, held 2008;28:e154]. Arterioscler Thromb Vasc Biol 2008;28:622–628 by the University of Padova, on the use of pharmacologic OSM inhibition to allow 20. Vono R, Fuoco C, Testa S, et al. Activation of the pro-oxidant PKCbII-p66Shc stem cell mobilization. No other potential conflicts of interest relevant to this article signaling pathway contributes to pericyte dysfunction in skeletal muscles of were reported. patients with diabetes with critical limb ischemia. Diabetes 2016;65:3691–3704 Author Contributions. M.A, S.C., S.T., L.M., V.S., R.C., G.Z., A.R., A.C., 21. Chow A, Lucas D, Hidalgo A, et al. Bone marrow CD169+ macrophages and M.G. performed the research. M.A., S.C., S.T., V.S., and G.P.F. analyzed the promote the retention of hematopoietic stem and progenitor cells in the mes- data. M.A., S.C., A.A., and G.P.F. designed the research and wrote the manuscript. enchymal stem cell niche. J Exp Med 2011;208:261–271 1314 OSM-p66Shc Regulates Mobilopathy/Myelopoiesis Diabetes Volume 68, June 2019

22. Albiero M, Poncina N, Ciciliot S, et al. Bone marrow macrophages contribute cells: relationship to oxidative stress. J Clin Endocrinol Metab 2005;90:1130– to diabetic stem cell mobilopathy by producing oncostatin M. Diabetes 2015;64: 1136 2957–2968 28. Cai W, He JC, Zhu L, Chen X, Striker GE, Vlassara H. AGE-receptor-1 23. West NR, Hegazy AN, Owens BMJ, et al.; Oxford IBD Cohort Investigators. counteracts cellular oxidant stress induced by AGEs via negative regulation of Oncostatin M drives intestinal inﬂammation and predicts response to tumor p66shc-dependent FKHRL1 phosphorylation. Am J Physiol Cell Physiol 2008;294: necrosis factor-neutralizing therapy in patients with inﬂammatory bowel disease. C145–C152 Nat Med 2017;23:579–589 29. Pinho S, Marchand T, Yang E, Wei Q, Nerlov C, Frenette PS. Lineage-biased 24. Guihard P, Danger Y, Brounais B, et al. Induction of osteogenesis in hematopoietic stem cells are regulated by distinct niches. Dev Cell 2018;44:634–641.e4 mesenchymal stem cells by activated monocytes/macrophages depends on 30. Kraakman MJ, Lee MK, Al-Sharea A, et al. Neutrophil-derived S100 calcium- oncostatin M signaling. Stem Cells 2012;30:762–772 binding proteins A8/A9 promote reticulated thrombocytosis and atherogenesis in 25. Fadini GP, Bonora BM, Marcuzzo G, et al. Circulating stem cells associate diabetes. J Clin Invest 2017;127:2133–2147 with adiposity and future metabolic deterioration in healthy subjects. J Clin En- 31. De Bruyn PP, Breen PC, Thomas TB. The microcirculation of the bone docrinol Metab 2015;100:4570–4578 marrow. Anat Rec 1970;168:55–68 26. Costantino S, Paneni F, Mitchell K, et al. Hyperglycaemia-induced epigenetic 32. Lörchner H, Pöling J, Gajawada P, et al. Myocardial healing requires Reg3b- changes drive persistent cardiac dysfunction via the adaptor p66Shc. Int J Cardiol dependent accumulation of macrophages in the ischemic heart. Nat Med 2015;21: 2018;268:179–186 353–362 27. Pagnin E, Fadini G, de Toni R, Tiengo A, Calò L, Avogaro A. Diabe- 33. Tanaka M, Miyajima A. Oncostatin M, a multifunctional cytokine. Rev Physiol tes induces p66shc gene expression in human peripheral blood mononuclear Biochem Pharmacol 2003;149:39–52