IL-27 receptor-regulated stress myelopoiesis drives abdominal aortic aneurysm development

Abdominal aortic aneurysm (AAA) is a prevalent life-threatening disease, where aortic wall degradation is mediated by accumulated immune cells. Although cytokines regulate inflammation within the aorta, their contribution to AAA via distant alterations, particularly in the control of hematopoietic stem cell (HSC) differentiation, remains poorly defined. Here we report a pathogenic role for the interleukin-27 receptor (IL-27R) in AAA, as genetic ablation of IL-27R protects mice from the disease development. Mitigation of AAA is associated with a blunted accumulation of myeloid cells in the aorta due to the attenuation of Angiotensin II (Ang II)-induced HSC expansion. IL-27R signaling is required to induce transcriptional programming to overcome HSC quiescence and increase differentiation and output of mature myeloid cells in response to stress stimuli to promote their accumulation in the diseased aorta. Overall, our studies illuminate how a prominent vascular disease can be distantly driven by a cytokine-dependent regulation of bone marrow precursors.

A bdominal aortic aneurysm (AAA) is a cardiovascular disease (CVD) characterized by abdominal aorta dilatation caused by accumulated immune cells, inflammation, and degradation of the medial layer followed by aortic rupture and bleeding, which is often fatal [1][2][3][4] . The current standard of care for AAA is limited to surgery at the later stages of disease progression. Thus, a better understanding of its mechanisms, particularly the inflammatory nature of AAA, is urgently needed. Various risk factors are associated with AAA pathogenesis 5 , including elevated blood pressure that is mediated by activation of the renin-angiotensin system (RAS) and upregulation of Angiotensin II (Ang II). Accordingly, a model of AAA that employs long-term infusion of Ang II in atherosclerosis-prone mice recapitulates many aspects of human AAA, including elevated blood pressure, dependence on higher Ang II levels, and gradual recruitment of immune cells 6,7 . Moreover, the disease in this model is exacerbated by hypercholesterolemia, one of the additional risk factors for AAA 8 . The importance of RAS/Ang II in connection between blood pressure and AAA in humans is supported by clinical trials data where blood pressure-reducing medications targeting RAS/Ang II circuit significantly decreases the incidence of aortic dissection 9 . However, Ang II acts via AT1a receptors (AT1aRs), which are broadly expressed on different cell types in the aortic wall, and therefore "hypertensionindependent" functions of Ang II in AAA have also been widely observed [10][11][12][13][14] .
Chronic inflammation presented as infiltration of various immune cells is a key feature and driver of AAA 2,4,15,16 . Neutrophils, monocytes, and macrophages are bone marrow (BM)-derived myeloid cells that play important roles in immunity and tissue repair 17 . These cells also contribute to the aortic inflammation and vessel destruction associated with AAA 4,18-20 . Their output from hematopoietic stem and progenitor cells (HSPCs) in the BM can be influenced by environmental stimuli and stresses. During infection or inflammatory insult, a variety of cytokines and danger-associated molecular patterns stimulate HSPCs in BM to rapidly increase the production and systemic release of innate immune cells [21][22][23] . Interestingly, HSPCs are also responsive to Ang II 24 , which drives their expansion in mice 25 . Although alterations in myelopoiesis have been reported to play a role in AAA progression 4,18 , the key cytokine signaling factors that control HSPC fate and hematopoiesis in AAA have yet to be determined. Particularly, the contribution of Ang II and cytokine signaling in the regulation of HSPC proliferation and myeloid cell differentiation to AAA pathogenesis has not been examined.
Interleukin (IL)-27 is a member of IL-6/IL-12 cytokine superfamily that regulates various hematopoietic and non-hematopoietic cells in infectious diseases and autoimmunity [26][27][28][29][30][31] . The role of IL-27 signaling in AAA and its contribution to BM cell production in response to chronic stress stimuli has not been investigated.
Here we demonstrate that genetic inactivation of IL-27 receptor (IL-27R) protects mice from AAA. Mechanistically, we show that IL-27R signaling is essential to drive HSPC proliferation, differentiation, and BM output. These findings illustrate how IL-27R signaling acts distantly to control AAA development by cooperating with stress-induced factors in the BM to accelerate stress hematopoiesis and promote the production of myeloid cells that are subsequently recruited to the aortic wall mediating its destruction.

Results
IL-27R signaling promotes Ang II-induced AAA development. Inactivation of IL-27R exacerbates atherosclerosis [29][30][31] and leads to the development of certain abdominal aortic lesions, which typically are rare in atherosclerotic mice. As atherosclerosis and AAA share some common underlying chronic inflammatory mechanisms 5 , we initially investigated whether the lesions in IL-27R-deficient mice may be an incipient source of AAA formation, and that IL-27R deficiency would increase inflammation and promote AAA.
To evaluate the potential role of IL-27R signaling in AAA, we first employed a well-characterized mouse model of AAA driven by chronic Ang II infusion (800 ng kg −1 min −1 ) using a surgically implanted osmotic mini-pump into mice on Apoe −/− background 6,32 . To exclude any differences in genetics or microbiota, we used cage-mate and littermate controls. As hypercholesterolemia promotes AAA development 8,33 , male and female Apoe −/− , Apoe −/− Il27ra +/− or Apoe −/− Il27ra −/− mice were fed a Western diet (WD) for 8 weeks followed by Ang II pump implantation. Four weeks later, mice were assessed for abdominal aorta bulging and AAA development (Fig. 1a). Ang II infusion induced AAA formation in IL-27R-proficient Apoe −/− and Apoe −/− Il27ra +/− mice, whereas unexpectedly the incidence of AAA was markedly reduced in IL-27R-deficient Apoe −/− mice ( Fig. 1b-f). Blood pressure was elevated in response to Ang II infusion, but IL-27R regulated AAA independent of effects on blood pressure. Body weight also remained unchanged by IL-27R deficiency ( Supplementary Fig. 1a, b). Both male and female Apoe −/− and Apoe −/− Il27ra +/− mice developed larger AAAs with visual hemorrhages in the artery wall compared with their Apoe −/− Il27ra −/− counterparts (Fig. 1b, c and Supplementary Fig. 1c). Verhoeff-Van Gieson staining showed extensive disruption and degradation of elastic lamina in the aortas of both Apoe −/− and Apoe −/− Il27ra +/− mice, but not in Apoe −/− Il27ra −/− mice (Fig. 1d). Female Apoe −/− and Apoe −/− Il27ra +/− mice (Fig. 1e) developed slightly lower rates of AAA than did their male counterparts (Fig. 1f); however, the incidence of AAA was reduced by IL-27R deficiency in both genders (Fig. 1e, f). Although both Apoe −/− and Apoe −/− Il27ra +/− control mice experienced significant sudden AAA-related mortality in the Ang II model, 100% of Apoe −/− Il27ra −/− mice remained alive throughout the experiment (Fig. 1g, h). Pathological severity index, which was calculated based on the level of aortic wall degradation and immune infiltrate 34 , showed that both female and male Apoe −/− and Apoe −/− Il27ra +/− mice displayed more advanced stages of AAA (IV stage) compared with IL-27R-deficient Apoe −/− mice, where AAA progression, if any, was restricted to the early stages (I-II) ( Supplementary Fig. 1d). The effect of IL-27R deficiency on AAA development was also confirmed in another AAA model 35 induced by topical application of elastase combined with administration of 0.2% β-aminopropionitrile (BAPN) in drinking water ( Supplementary Fig. 2).
To ensure that our observations were not restricted to the dietary factors introduced by the WD, we also performed experiments on a cohort of mice fed a regular chow diet. As expected, incidence of AAA was significantly lower in Apoe −/− and Apoe −/− Il27ra +/− male mice fed the chow diet compared with WD-fed groups and AAA formation was not detected at all in females fed with chow diet. However, IL-27R deficiency still rendered male mice to be less susceptible to AAA induction ( Supplementary Fig. 3). Collectively, our data demonstrate that IL-27R signaling promotes AAA in two distinct in vivo models of AAA.
Thus, Ang II infusion promoted the expansion and accumulation of IL-27R-sufficient myeloid cells in suprarenal aorta. In contrast, IL-27R deficiency limited the expansion of circulating cells and reduced the accumulation of myeloid cells, particularly monocytes and neutrophils in AAA, affecting the expression of myeloid-derived chemokines, cytokines, and key enzymes in the area of AAA formation.
In line with previous observations that Ang II receptor (AT1aR) is expressed on HSPCs 24 and Ang II infusion promotes both proliferation and "myeloid-biased" differentiation of HSCs 25 , we found that Ang II infusion strongly increased the proportion of proliferating cells, marked in vivo by bromodeoxyuridine (BrdU) incorporation, in IL-27R-sufficient but not -deficient animals, although PBS-infused mice did not display any significant differences ( Fig. 3e and Supplementary Fig. 5b). Cell cycle analysis showed higher proportion of IL-27R-deficient cells in G 0 /G 1 and less in G 2 /M phase of cell cycle compared with IL-27R-sufficient cells (Fig. 3f). These data suggest that IL-27R signaling is required to potentiate the proliferative effect of Ang II on HSPCs and is involved into the regulation of HSPC cell cycle.
To evaluate the effect of IL-27R signaling on the clonogenic and differentiation capacities of HSPCs, we performed ex vivo colony-formation assays. Lineage-depleted (Lin − ) BM HSPCs ( Supplementary Fig. 5c) isolated from Apoe −/− Il27ra −/− and Apoe −/− Il27ra +/− mice that had been infused with Ang II for 4 weeks were cultured in M3434 media under myeloid differentiation conditions. We found that IL-27R-deficient HSPCs showed a marked decrease in total numbers of colonies formed compared with IL-27R-sufficient cells (Fig. 4a, b). Conversely, in vitro recombinant (rec) IL-27 stimulation promoted myeloid colony formation, especially when combined with Ang II (Fig. 4c, d). Importantly, pharmacological inhibition of JAK1/2 kinases, which are immediately downstream of IL-27R, with ruxolitinib ablated this stimulating effect (Fig. 4d).
Overall, these data suggest that IL-27R signaling potentiates the Ang II-induced HSPC proliferation and differentiation, and "stress" myelopoiesis during AAA development.
IL-27R-deficient HSCs fail to induce AAA. Due to the inherent variability of AAA model, the effect on BM hematopoiesis may be mediated by direct role of IL-27R signaling or influenced by the size and incidence of AAA lesion development, thereby warranting studies where IL-27R-proficient and -deficient cells co-exist within the same animal. To evaluate whether a cell-intrinsic lack of IL-27R signals render HSPCs unable to expand and contribute to AAA in an overall IL-27R-sufficient environment, we conducted competitive BM transplantation (BMT) analyses. BM cells isolated from congenically marked Apoe −/− CD45.1 + (further referred as wild type (WT)) or Apoe −/− Il27ra −/− CD45.2 + (further referred as Il27ra −/− ) mice were mixed in different ratios including 90%:10%, 50%:50%, and 10%:90%, respectively, and transplanted into lethally irradiated Apoe −/− CD45.1 + IL-27R-sufficient recipients. Mice were placed on WD for 4 weeks after BM transfer and maintained on WD for another 8 weeks, allowing for a total of 12 weeks of BM reconstitution. The efficiency of introduction of correct BM mixes was confirmed by fluorescence-activated cell sorting (FACS) for CD45.1/CD45.2 ratio in the blood of naive unchallenged mice 4 weeks after BMT ( Supplementary Fig. 6a). After 8 weeks of WD feeding, mice were implanted with Ang II pump to induce AAA and assess the effect of Ang II on the BM cells of different IL-27R status and their role in AAA (Fig. 5a). Interestingly, 86% of mice transplanted with a 90%wt:10%Il27ra −/− mix of BM developed late stages of AAA and some died due to AAA rupture, whereas only 17% of mice transplanted with a 10%wt:90%Il27ra −/− mix of BM developed AAA, arguing for cell autonomous pathogenic role of IL-27R. The incidence of AAA in mice transplanted with mix of 50% wt:50%Il27ra −/− BM was 38% (Fig. 5b, c). Chimeric recipient mice, which received 50%wt:50%Il27ra −/− BM mix and infused with Ang II, were used to compare side by side the development and accumulation of IL-27R-sufficient and IL-27R-deficient myeloid cells in IL-27R-sufficient environment. Both WT and Il27ra −/− donors cells exhibited~50% cell ratio in the peripheral blood 4 weeks after transplantation and before the Ang II infusion and AAA induction ( Supplementary Fig. 6a). However, the analysis of HSPCs compartment in the BM of chimeric mice infused with Ang II revealed the reduction of Il27ra −/− LSK, HPC-1, and more mature myeloid progenitors compared with WT-derived cells within the same mouse (Fig. 5d, e). Conversely, the percentage of LT-HSC derived from Il27ra −/− BM was elevated (Fig. 5d), consistent with more quiescent state of IL-27R-deficient cells. In line with limited competitive expansion of IL-27R-deficient precursors, the percentage of mature peripheral Apoe −/− Il27ra −/− (CD45.2)-derived monocytes and neutrophils were also diminished in the spleen and Fig. 2 Limited accumulation of myeloid cells in AAA of IL-27R-deficient mice. Single-cell suspensions of suprarenal aortas/AAA of Apoe −/− Il27ra +/− (n = 6) or Apoe −/− Il27ra −/− (n = 8) mice fed with WD and infused with Ang II were stained for Live/Dead, CD45, CD11b, CD11c, Ly6G, Ly6C, and analyzed by FACS. a Left: representative FACS plots of CD45 + cells. Numbers indicate percentage of cells in each gate. a Right: absolute number of live CD45 + cells. b Representative FACS plots and absolute number of CD45 + live CD11b + , CD11b + CD11c + , and CD11c + cells. c Representative FACS plots and absolute number of live CD11b + Ly6C hi , CD11b + Ly6G + (Neutrophils), and CD11b + Ly6C low cells. **p < 0.01, ***p < 0.005, unpaired Student's t-test (two-tailed). d Immunofluorescent confocal analysis of localization and abundance of CD11b + and Ly6G + (Neutrophils) cells in suprarenal aortas/AAA of Apoe −/− Il27ra +/− and Apoe −/− Il27ra −/− mice infused with Ang II. L-lumen, M-media, A-adventitia, AAA-abdominal aortic aneurysm. Representative images from three independent experiments. Relative gene expression of chemokines (e), cytokines (f), and MMPs (g) in suprarenal aortas/AAA of Apoe −/− Il27ra −/− mice were normalized to Rpl32 expression and then normalized to gene expression in control Apoe −/− Il27ra +/− mice. **p < 0.01, ***p < 0.005, Wilcoxon's signed-rank test. h Leukocytes, monocytes, and neutrophils counts were evaluated by VetScan HM5 in the blood of Apoe −/− Il27ra +/− (n = 5) or Apoe −/− Il27ra −/− (n = 5) mice prior to and on day 5 after pump implantation. Data are mean ± SEM from at least three independent experiments. *p < 0.05. **p < 0.01, ***p < 0.005, unpaired Student's t-test (two-tailed)  (Fig. 5f, g). Of note, the number of circulating myeloid cells of IL-27R-sufficient origin was elevated in response to Ang II infusion compared with PBS control, whereas significantly less prominent effect was observed in IL-27R-deficient cells (Supplementary Fig. 6b). The analysis of suprarenal aorta/AAA revealed preferential accumulation of monocytes and neutrophils derived from WT (IL-27R-sufficient)  Apoe -/-Il27ra progenitors, whereas the percentage of accumulated cells derived from Il27ra −/− BM was diminished ( Fig. 5h and Supplementary  Fig. 6c). This indicates that when WT and IL-27R-deficient BM progenitors compete in the same animals, WT myeloid cells are preferentially produced and therefore are overrepresented in AAA lesions. Importantly, no difference in cell trafficking/ recruitment was observed between the genotypes once mature IL-27R-sufficient and IL-27R-deficient myeloid cells were administered intravenously (i.v.), thereby bypassing the stage of myeloid cell expansion in the BM (Supplementary Fig. 7). This indicates that primary defect of IL-27R deficiency is in their inability to properly expand in the BM and not in the recruitment of mature cells into the AAA site. IL-27R deficiency was previously reported to promote atherosclerosis in Apoe −/− mice 31 . Here we found that Ang II infusion accelerated the disease in IL-27R sufficient, but did not further increase atherosclerosis development in IL-27R-deficient mice ( Supplementary Fig. 8a-c), eliminating the phenotypic difference between genotypes in Ang II-treated conditions. The analysis of myeloid cells accumulation in 50%wt:50%Il27ra −/− BM chimeric mice revealed that myeloid cells in the aortic arch were mostly cells of WT Apoe −/− (CD45.1) origin, whereas Apoe −/− Il27ra −/− (CD45.2) cells were significantly underrepresented (Supplementary Figs. 6c and 8d). Thus, enhanced BM output of IL-27Rsufficient cells upon Ang II treatment drives augmented atherosclerosis, whereas Ang II is unable to cause the same enhancement of atherosclerosis in IL-27R-deficient hosts (Supplementary Fig. 8d). As a result, in Ang II-untreated conditions, Apoe −/− Il27ra −/− mice develop more atherosclerosis in comparison with Apoe −/− Il27ra +/− controls 31 . However, because of Ang II-driven increase in atherosclerosis in Apoe −/− Il27ra +/− but not in Apoe −/− Il27ra −/− mice, Apoe −/− Il27ra +/− , and Apoe −/− Il27ra −/− Ang II-infused mice develop equally high levels of atherosclerosis. Fig. 3 IL-27R deficiency leads to limited expansion of HSCs in response to Ang II. BM was obtained from Apoe −/− Il27ra +/− (n = 7) and Apoe −/− Il27ra −/− (n = 5), or Apoe −/− Il27ra +/− (n = 12) and Apoe −/− Il27ra −/− (n = 12) mice fed with WD for overall period of 12 weeks and infused with PBS or Ang II for last 4 weeks of feeding, respectively. a Representative dot plots of live LSK (Lin − Sca-1 + c-kit + ) and myeloid progenitors (Lin − Sca-1 − c-kit + ). b Representative dot plots of live LT-HSC (Lin − Sca-1 + c-kit + CD150 + CD48 − ) and HPC-1 (Lin − Sca-1 + c-kit + CD48 + CD150 − ) cells. c Average percentage of live LSK, LT-HSC, HPC-1, myeloid progenitors, and CMP (Lin − Sca-1 − c-kit + CD16/CD32 − CD34 + ). d Absolute number of live LSK, LT-HSC, HPC-1, myeloid progenitors, and CMP. e Proliferation of HSPCs from Apoe −/− Il27ra +/− (n = 8) or Apoe −/− Il27ra −/− (n = 6) mice fed with WD and infused with Ang II or PBS as determined by BrdU incorporation. Percentage of live BrdU-positive LSK and myeloid progenitors, including LT-HSC, HPC-1, and CMP. Data are mean ± SEM from two independent experiments. f Percentage of progenitor cells from Apoe −/− Il27ra +/− (n = 6) or Apoe −/− Il27ra −/− (n = 6) mice infused with Ang II in G 0 /G 1 and G 2 /M phase of cell cycle, analyzed for BrdU inclusion, and DAPI staining. Data are mean ± SEM from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.005, unpaired Student's t-test (two-tailed) Apoe Overall, our data suggest that Ang II-driven expansion of HSCs and myelopoiesis facilitates myeloid cell accumulation and AAA, and this process is dependent on IL-27R signaling.
These results suggest that distinct transcriptional profile regulated by IL-27R signaling is required to maintain the balance between quiescence, proliferation and differentiation in early hematopoietic precursors challenged with stress stimuli, as responsiveness of HSPCs to Ang II is blunted in the absence of IL-27R 39,40 . These data are consistent with our findings of reduced BrdU incorporation, cell accumulation in G 0 /G 1 cell cycle phase, and limited BM output of IL-27Rdeficient cells.
Along with the quiescence gene signature and upregulation of p53 gene targets uncovered by IPA Upstream Regulator Analysis, we found that IL-27R-deficient HSPCs sorted from Ang IIinfused mice were characterized by elevated levels of p21/Waf1 protein, a prominent target of p53 pathway (Fig. 7a-c). To assess whether the observed effect on p21 expression was dependent upon Ang II and IL-27 action, we isolated HSPCs from Apoe −/− mice and stimulated them in vitro with Ang II, recIL-27 or their combination (Fig. 7d). Indeed in vitro stimulation of HSPCs with Ang II and recIL-27 downregulated p21 protein levels (Fig. 7e, f) 4), or Apoe −/− Il27ra −/− (n = 3) mice fed with WD for 12 weeks and infused with Ang II or PBS for last 2 weeks of feeding, and subjected to whole transcriptome analysis (RNA-seq). a Overall number of genes whose expression is changed between Apoe −/− and Apoe −/− Il27ra −/− mice either treated with Ang II (587 genes) or with PBS (control, 16 genes). b Deregulated upstream regulators whose targets were significantly overrepresented (p < 0.05, Fisher's exact test) among genes uniquely affected by genetic IL-27R deficiency. c Various pathways significantly deregulated (p < 0.05, GSEA) by IL-27R deficiency, as determined by Reactome pathway enrichment analysis. Changed pathways include cell cycle, proliferation, and division in LT-HCS after Ang II infusion ( # p = 0.056, GSEA) proliferation, it was unable to induce downregulation of cell cycle inhibitors by itself and IL-27R signaling was essential to allow Ang II-stimulated progenitors to escape negative control of quiescence expression program and to start proliferation and differentiation.
Several signaling pathways could participate in the regulation of p21 expression, including interferon (IFN) and Myc pathways, both of which are enriched in Ang II-treated IL-27R-sufficient cells (Fig. 6b). In addition, p53 pathway, one of the key upstream bona fide regulators of p21 expression, was strongly affected by IL-27R
Collectively, these data suggest that HSPCs depends on IL-27R signaling to control the expression of various regulators of quiescence/proliferation balance, including p53-dependent regulators such as p21/Waf1. IL-27R signaling is required for cell cycle entry in response to Ang II-induced stimulation. This enhances myelopoiesis and BM myeloid cell output, overall contributing to AAA progression (Fig. 7i).

Discussion
The function of IL-27R signaling was extensively investigated in various infectious models 27,28 and an anti-inflammatory role IL-27R has been demonstrated in atherosclerosis [29][30][31] . Moreover, some of the IL-27R-deficient mice employed in the atherosclerosis studies developed small abdominal aorta lesions, grossly similar to potential incipient AAA sites, raising the possibility that IL-27R might also restrict spontaneous AAA development. The role of IL-27R in AAA, another vascular pathology with a clear role of yet unidentified inflammatory cytokine-driven mechanisms, however, has never been assessed. Here we made the unanticipated observation that mice with genetic ablation of IL-27R were largely protected from AAA. This correlated with diminished accumulation of monocytes and neutrophils in suprarenal aortas, which is a key early event in AAA 4,18-20 . The expression of myeloid cell-derived cytokines and chemokines was also decreased in the AAA lesions of Apoe −/− Il27ra −/− mice infused with Ang II. Although homing mechanisms play an important role in the regulation of immune cell accumulation at the site of inflammation (AAA) 53 , the recruitment of a large amount of immune cells into rapidly developing AAA lesions may also require an increased BM output, known as "stressinduced" BM myelopoiesis. The unifying regulators required for the AAA development, vessel inflammation, and "stress myelopoiesis," however, remained unknown. In earlier studies, Ang II was shown to induce the mobilization of splenic monocytes, whereas splenectomy and subsequent reduction of circulating monocytes suppresses AAA development 18 . Moreover, Ang II was shown to act directly on HSCs in BM and induce their self-renewal and rapid amplification of myeloid progenitors with subsequent production of mature myeloid cells 25 . In our model we found that Ang II causes the expansion of various HSCs and progenitor cells in the BM of IL-27R-sufficient (Apoe −/− or Apoe −/− Il27ra +/− ) mice, resulting in elevated numbers of circulating leukocytes. A surprising observation of our studies was, however, that inactivation of IL-27R signaling significantly blunted Ang II-driven expansion of HSCs and progenitor cells, and BM output, indicating a critical role of IL-27R signaling in potentiating Ang IIinduced myelopoiesis, contributing to AAA development.
Several cytokines were shown to regulate the inflammatory milieu within the aortic wall in AAA [54][55][56][57][58][59] . However, the distant contribution of cytokines to AAA pathogenesis through systemic control of the immune response (e.g., through blood cell production due to hematopoietic stem cell (HSC) activation by stress factors (diet and high blood pressure tightly linked with elevated levels of Ang II) has not been previously explored. Here we establish IL-27R signaling as a critical regulator of Ang II-driven proliferation and differentiation of HSCs, essential for Ang IIinduced stress myelopoiesis in a non-infectious, chronic vascular injury model. Given the role of myeloid cells in host defense and tissue repair 17 , this cytokine-driven induction of HSCs may be a common mechanism regulating the rapid need for increased BM output during various pathophysiological processes.
HSCs are maintained in a resting quiescent state in specialized hypoxic niches in the BM. In response to changes in microenvironment, they can exit this quiescent state and rapidly proliferate and differentiate into different blood cell types 23 . Maintenance of quiescence is crucial for preventing stem cell pool exhaustion, overproduction of mature cells, and the development of hematopoietic malignancies 60 . HSC quiescence is therefore tightly regulated by many factors including IFN cytokine signaling and miRNAs (miR-21, miR-223, and let-7) 44,61,62 . In chronic diseases, such as AAA, prolonged exposure to "stress" factors such as high-fat diet and Ang II affecting BM cells 25,63 may provoke HSCs to exit quiescence and activate stress myelopoiesis, which serves as an additional pathogenic mechanism driving AAA. Here we establish IL-27R signaling as an in vivo regulator of HSC function in a non-infection model of vascular injury and AAA. This role of IL-27 is in line with the studies where IL-27 transgenic overexpression causes myeloproliferation in mice, and malaria infection drives the BM response and infection-induced myelopoiesis in a IFN/IL-27-dependent manner 64 .
By performing RNA-sequencing analysis of purified LT-HSCs we found that IL-27R-deficient LT-HSCs were characterized by a quiescence transcriptional profile, even following exposure to Ang II. Lack of IL-27R signaling rendered LT-HSCs unable to upregulate the expression of genes involved in cell division, proliferation, and differentiation normally induced by Ang II. Specifically, we found that the loss of IL-27R signaling in Fig. 7 Elevated expression of p53 and p21 in IL-27R-deficient HSPCs. a Scheme of experiment. b Lin − HSPCs isolated from WD-fed Apoe −/− Il27ra +/− (n = 4) and Apoe −/− Il27ra −/− (n = 4) mice infused with PBS or Apoe −/− Il27ra +/− (n = 5) and Apoe −/− Il27ra −/− (n = 4) mice infused with Ang II for 4 weeks were lysed and subjected to western blotting with p21, p53, and β-actin antibody. Each lane corresponds to individual mouse. c Quantification of WB analysis for p21 and p53, normalized to β-actin. d Scheme of experiment. e WB on Lin − HSPCs from naive Apoe −/− Il27ra +/− mice (n = 4) stimulated in vitro with Ang II, IL-27, or both for 24 h. f Quantification of WB analysis for p21. g Scheme of experiment. h Lin − HSPCs from BM of Apoe −/− (n = 4), Apoe −/− Il27ra −/− (n = 4), or Apoe −/− Il27ra −/− Trp53 d/d (n = 5) mice were sorted and plated in M3434 media under myeloid conditions with or without Ang II. GM colony formation was accessed on day 6. Data are mean ± SEM from two independent experiments. *p < 0.05, **p < 0.01, ***p < 0.005, unpaired Student's t-test (two-tailed). i Scheme describing the role of IL-27R signaling in regulation of Ang II-induced myelopoiesis and AAA. Elevation of Ang II in AAA provides a stimulus for the activation of "stress" myelopoiesis via regulation of gene expression controlling proliferation and differentiation in IL-27R-sufficient LT-HSCs. Lack of IL-27R signaling renders Ang II-primed cells unable to overcome quiescence state and significantly reduces proliferation of progenitors and myeloid cell bone marrow output, therefore reducing cell accumulation in AAA lesions and AAA progression LT-HSCs most prominently inhibited pathways involved into the regulation of proliferation (Myc, E2F, and CCND1) and control of HSC expansion (IFN, STAT1). Conversely, inhibitors of proliferation were induced by IL-27R deficiency, particularly p53, miR-21, mir-223, and let-7 [46][47][48][49][50] . Many of these regulators have also been previously implicated into the control HSCs fate. Indeed, miR-21 was shown to suppress hematopoiesis via activation of transforming growth factor-β (TGFβ) signaling 46 . MiR-223 has been also reported to attenuate hematopoietic cell proliferation 47 , whereas let-7 was shown to regulate HSC fate by controlling self-renewal, proliferation, quiescence, and differentiation via inhibition of TGFβ pathway 65,66 . Moreover, let-7 family members were shown to repress cell cycle regulators (Cyclin D1) and negatively regulate Myc expression 50,67 . These previous observations are in line with our transcriptomics results. TGFβ, p53, and IFN pathways may converge at the level of p21 expression, a key negative regulator of cell proliferation, whose levels in HSPCs are regulated by IL-27R signaling. In agreement with pronounced changes in transcriptional profile, Apoe −/− Il27ra −/− HSCs demonstrated limited in vivo proliferation and cell cycle progression. Taken together, global gene expression profiling of LT-HSCs revealed that IL-27R signaling is required for HSC response to Ang II stimulation and potentiates stress-induced myelopoiesis. In the absence of this cytokine signaling, HSCs maintain quiescence even when AAA-inducing Ang II-driven stress is applied.
Collectively, our data establish an unexpected role of IL-27R signaling in AAA and suggest that Ang II, a stress factor that is constantly present in AAA, perhaps, in combination with other factors such as WD, promotes HSPCs proliferation and reveals a previously unexplored requirement for IL-27R signaling in Ang II-driven HSCs expansion and hematopoiesis during AAA development. Although our study is focused on the role of IL-27R signaling in hematopoietic compartment, IL-27R may also regulate AAA development via control of non-hematopoietic cells, representing an interesting subject for future work. Current work, however, strongly suggests that IL-27 can be a potential testable target for prevention and treatment of AAA and other forms of vascular injury, which require HSC BM mobilization for its full pathology. Therefore, overstimulation of this otherwise antiinflammatory signaling pathway can actually be pathogenic in AAA due to its ability to modulate hematopoiesis.

Methods
Mice. Il27ra −/− (JAX #018078) mice were crossed to Apoe −/− (JAX#002052) mice to obtain Apoe −/− , Apoe −/− Il27ra +/− , and Apoe −/− Il27ra −/− mice. C57BL/6 CD45.1 (JAX#002014) mice were crossed to Apoe −/− mice to obtain CD45.1 Apoe −/− mice. All mice were on C57BL/6 background. Mice were bred and housed under specific pathogen-free conditions in an AAALAC-approved barrier facility at Fox Chase Cancer Center (FCCC). The genotyping was performed by standard PCR protocols. Animal numbers for each specific analysis are given in the figure legends. Apoe −/− , Apoe −/− Il27ra +/− , and Apoe −/− Il27ra −/− mice were fed a WD (Teklad 88137) for 10 or 12 weeks beginning at 8 weeks after birth and subcutaneously implanted with Ang II containing pump for the last 2 or 4 weeks of feeding, after which mice were killed and AAA formation was analyzed. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) at FCCC and performed in compliance with all relevant ethical regulations for animal research.
Models of AAA. Ang II model. Ang II containing osmotic mini-pumps were prepared and implanted as previously described 6 . Briefly, mice were anesthetized and osmotic mini-pumps (Alzet 2004, 10389-18) loaded with Ang II (800 ng/kg/ min; Calbiochem, 2787322) were surgically implanted subcutaneously in the midscapular area over the shoulder blade followed by closing the wound with clips. AAA formation was analyzed after 14 or 28 days of Ang II infusion.
Topical Elastase/BAPN model was performed according to previously described protocol 35 . Ten microliters (7.6 mg protein/ml, 4 U/mg) of active elastase (Sigma Aldrich, 1002730705) or 10 μl of heat de-activated (100°C for 30 min) elastase (sham) was applied on top of the infrarenal abdominal aorta of C57BL/6 or Il27ra −/− mice for 5 min followed by washing of the exposed area with saline solution. Mice were given drinking water containing 0.2% BAPN fumarate salt (Sigma Aldrich, A3134) starting 2 days prior the surgery and elastase treatment until the end of study (day 14 after Elastase treatment), when AAA development was analyzed.
Blood pressure measurements. Systolic blood pressure was measured on conscious mice using tail cuff system (CODA Non-Invasive Blood Pressure Monitor, Kent Scientific Corporation) 4 weeks after infusion of Ang II, according to the manufacturer's protocol. Briefly, mice were placed in the holder maintained on the Animal Warming Platform. "Occlusion Cuff" and "VPR Cuff" were placed near the base of the tail. The cuffs were attached to the CODA Controller. Twenty cycles of measurement were performed and accepted cycles were automatically displayed in the spreadsheet format within the CODA application. Data were processed in Microsoft Excel by calculation of the average and standard deviation.
Histology and immunofluorescence. For histological analysis, suprarenal aortas (with or without AAA) were isolated and embedded in Tissue-Tek O.C.T. (Optimal Cutting Temperature) compound (Sakura Finetek) and stored at −80°C. Five micrometer serial frozen sections were cut and stained with hematoxylin (Sigma, SLBX9844) and eosin (ThermoScientific, 287569). All images were acquired with Nikon Eclipse 80i microscope.
Verhoeff-Van Gieson staining of elastic fibers. Frozen, 5 μm sections of suprarenal aortas (with or without AAA) were cut and staining of elastic fibers was performed. Frozen sections were hydrated followed by staining in Verhoeff's solution for 1 h. Afterwards, slides were incubated in 2% ferric chloride for 2 min and treated with 5% sodium thiosulfate for 1 min. Sections were counterstained in Van Gieson's solution for 5 min and dehydrated (all reagents were from Electron Microscopy Sciences). Images were acquired with Nikon Eclipse 80i microscope.
Peripheral blood cellularity. After killing, peripheral blood was collected by cardiac puncture to K 2 EDTA BD Microcontainer (BD Biosciences). White blood counts were obtained by VetScan HM5 analyzer (Abaxis).
Flow cytometry. Cells were isolated from the aortas (with or without AAA or aortic arch) and spleen from mice infused with Ang II and were analyzed by flow cytometry. Mice were killed by CO 2 inhalation and the aortas were perfused with PBS containing 2% heparin to remove all traces of blood. Suprarenal aortas were isolated, cut into small pieces, and incubated in a cocktail of digestion enzymes containing hyaluronidase (120 U/ml) (Sigma, H3506), collagenase I (450 U/ml) (Sigma, C0130), collagenase XI (250 U/ml) (Sigma, C7657), and DNAse I (120 U/ml) (Sigma, D4263) in Hank's buffered salt solution (HBSS) for 55 min at 37°C with gentle shaking. After incubation, cell suspensions were filtered through a 70 μm cell strainer and were stained with CD45-PerCP Blood was collected by cardiac puncture and erythrocytes were lysed by incubation with red blood cell (RBC) lysis buffer (15 mM NH 4 Cl, 0.1 mM NaHCO 3 , 0.1 mM Na 2 -EDTA) for 5 min at RT and washed with PBS. The spleen and blood were stained with the above mentioned antibody mix for mature immune cells.