Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 May 1.
Published in final edited form as: Arterioscler Thromb Vasc Biol. 2016 Mar 17;36(5):908–918. doi: 10.1161/ATVBAHA.116.307373

Mesenchymal Stem Cells Attenuate NADPH Oxidase-Dependent High Mobility Group Box 1 Production and Inhibit Abdominal Aortic Aneurysms

Ashish K Sharma 1, Morgan D Salmon 1, Guanyi Lu 1, Gang Su 1, Nicolas H Pope 1, Joseph R Smith 1, Mark L Weiss 1, Gilbert R Upchurch Jr 1
PMCID: PMC4861899  NIHMSID: NIHMS765143  PMID: 26988591

Abstract

Objective

Abdominal aortic aneurysm (AAA) formation is characterized by inflammation, smooth muscle activation and matrix degradation. This study tests the hypothesis that macrophage-produced high mobility group box1 (HMGB1) production is dependent on NADPH oxidase (Nox2) which leads to increase in IL-17 production resulting in AAA formation and that treatment with human mesenchymal stem cells (MSCs) can attenuate this process thereby inhibiting AAA formation.

Approach and Results

Human aortic tissue demonstrated a significant increase in HMGB1 expression in AAA patients compared to controls. An elastase-perfusion model of AAA demonstrated a significant increase in HMGB1 production in C57BL/6 (wild type; WT) mice, which was attenuated by MSC treatment. Furthermore, anti-HMGB1 antibody treatment of WT mice attenuated AAA formation, IL-17 production and immune cell infiltration compared to elastase-perfused WT mice on day 14. Elastase-perfused Nox2−/y mice demonstrated a significant attenuation of HMGB1 and IL-17 production, cellular infiltration, matrix metalloproteinase activity and AAA formation compared to WT mice on day 14. In vitro studies showed that elastase-treated macrophages from WT mice, but not Nox2−/y mice, produced HMGB1, which was attenuated by MSC treatment. The production of macrophage-dependent HMGB1 involved Nox2 activation and superoxide anion production, which was mitigated by MSC treatment.

Conclusions

These results demonstrate that macrophage-produced HMGB1 leads to aortic inflammation and acts as a trigger for CD4+ T cell produced IL-17 during AAA formation. HMGB1 release is dependent on Nox2 activation, which can be inhibited by MSCs leading to attenuation of proinflammatory cytokines, especially IL-17, and protection against AAA formation.

Keywords: abdominal aortic aneurysms, HMGB1, inflammation, stem cells, IL-17

Introduction

Abdominal aortic aneurysm (AAA) is a significant medical problem with a high mortality and a leading cause of death in elderly men1. Currently, there are no directed therapies for aortic aneurysms which if left untreated leads to an increased risk of aortic rupture. The pathobiology of AAAs remains poorly understood, but likely involves vascular inflammation characterized by leukocyte infiltration into the aortic wall associated with the production of many pro-inflammatory cytokines, as part of the innate inflammatory response2, 3. It is during this stage that matrix degrading enzymes are released leading to aortic wall elastin and collagen degradation, thereby causing AAA formation4.

Recent studies have shown that mesenchymal stem cells (MSCs) can effectively attenuate vascular inflammation and AAA formation5, 6. In a recent study using a murine elastase-perfused model of AAA, we demonstrated that CD4+ T cell-produced IL-17 is a key proinflammatory cytokine regulating aortic inflammation, which can be attenuated by treatment with human MSCs7. However the exact signaling mechanism of MSC-mediated protection remains to be defined. Recent studies have demonstrated the ability of high mobility group box 1 (HMGB1), a damage associated molecular pattern molecule, to be rapidly secreted by activated macrophages and stimulate IL-17 production8. Additionally, blockade of HMGB1 has been shown to suppress development of AAA formation in an animal model9. Thus, in the present study we investigated the role of the cross-talk between macrophages and T cells, via HMGB1 and IL-17 secretion respectively, as a pivotal axis important in vascular inflammation during AAA formation, which can be downregulated by MSCs.

Since reactive oxygen species in vascular tissue can regulate inflammation, we investigated the role of NADPH oxidase in contributing to AAA formation via modulation of HMGB1 and IL-17 production. The NADPH oxidase enzyme complex consists of transmembrane (e.g. NOX1-5 and p22phox) and cytosolic (p47phox, p67phox and Rac) subunits which assemble at the cell membrane and reduce molecular oxygen to superoxide anion by transferring electrons from NADPH. In particular, the NOX2 isoform in phagocytic macrophages, involve translocation and binding of the cytosolic subunits to the gp91phox/p22phox catalytic complex which is facilitated and organized by p47phox. Previously, Miller et al. have shown that NADPH oxidase activity and its p47phox subunit are significantly upregulated in human aneurysm tissue10. Moreover, deletion of p47phox has been shown to attenuate oxidative stress and AAA formation in Ang II-infused apoE−/− mice11.

In the present study, we demonstrate that Nox2−/y mice have significantly decreased macrophage-dependent HMGB1 production and AAA formation. Furthermore, our findings demonstrate that human MSCs can attenuate NADPH oxidase activity by decreasing superoxide anion production in macrophages leading to the mitigation of HMGB1 production. These results underline the importance of NADPH oxidase-dependent release of HMGB1 by macrophages, which can upregulate IL-17 production and vascular inflammation, as well as loss of smooth muscle integrity leading to AAA formation. The mechanistic signaling pathway of MSC-mediated protection involves downregulation of Nox2 activation in macrophages to inhibit HMGB1 secretion, CD4+ T cell-produced IL-17 leading to decreased inflammation and vascular remodeling, thereby offering protection against AAA.

Materials and Methods

Materials and Methods are available in the online-only Data Supplement

Results

HMGB1 Expression is Increased in Human AAA

Protein levels of HMGB1 were significantly elevated in aortic tissue from AAA patients compared to controls (50.7±4.07 vs. 26.87±2.52 ng/ml; p<0.05; Figure 1A). Human aortic explants from male AAA patients were cultured in vitro with or without MSCs and transiently exposed to elastase. After 24hrs, a significant increase in HMGB1 expression was observed in cell culture supernatants after elastase treatment compared to controls (57.22± 4.24 vs. 15.06±2.63 ng/ml; Figure 1B) which was significantly attenuated by MSC treatment (23.22±2.35 ng/ml). Human aortic explants treated with elastase also demonstrated a significant increase in MMP2 and MMP9 activity which was attenuated by MSC treatment (Figure 1C–E). These results indicate that HMGB1 may play an important pro-inflammatory role in human AAA and that MSCs have the ability to mitigate HMGB1 production in human aortic tissue.

Figure 1.

Figure 1

Open in a new tab

Increased HMGB1 protein expression in human AAA. A, Human aortic tissue demonstrated an increased expression of HMGB1 in AAA patients (n=16) compared to controls (n=8). B, Human aortic explants in culture treated with transient elastase treatment for 5 min and analyzed after 24 hrs showed a significant increase in HMGB1 production, which was significantly mitigated by MSC treatment (n=8/group). C–E, Gelatin zymography of human aortic explant tissue and subsequent quantification of optical density (O.D.) demonstrates a significantly increased level of MMP2 and MMP9 activity compared to controls and was attenuated by co-cultures with MSCs (n=3–5/group). Mean +/− S.E.; *p<0.05 vs. other groups.

MSCs Inhibit HMGB1 and AAA Formation in Murine Elastase Perfusion Model

Using the elastase perfusion model, aortic diameter was measured in WT mice treated with or without MSC treatment. Human umbilical cord MSCs were isolated and characterized as described in Methods (Supplemental Figure S1). Elastase-perfused WT mice had a significant increase in aortic diameter compared to heat-inactivated elastase controls (134.9±10.14 vs. 50.98±3%; Figure 2A). There was no significant difference in aortic diameter in WT mice controls treated with or without MSCs. There was a significant decrease in aortic diameter on day 14 in elastase-perfused mice treated with MSCs compared to elastase perfused mice alone (99.39±4.84 vs. 134.9±10.14%; p=0.02).

Figure 2.

Figure 2

Open in a new tab

MSCs attenuate aortic diameter and HMGB1 expression in murine AAA model. A, Infrarenal mouse aortas were perfused with elastase (0.4U/mL) or heat-inactivated elastase (control), and aortic diameter was measured on day 14. A multifold increase in aortic diameter observed in elastase-perfused WT mice compared to controls was significantly attenuated by MSC treatment (n=8/group). B, A significant, multifold increase in HMGB1 expression was observed in aortic tissue of elastase-perfused WT mice compared to controls. Treatment of elastase-perfused WT mice with MSC significantly attenuated HMGB1 expression compared to elastase-perfused mice alone (n=8/group). C–E, Gelatin zymography of murine aortic tissue and subsequent quantification of optical density (O.D.) demonstrates a significantly increased level of MMP2 and MMP9 activity compared to controls and was attenuated by treatment with MSCs (n=3–5/group). Mean +/− S.E.; *p<0.05 vs. Control; #p<0.05 vs. Elastase.

A significant increase in HMGB1 expression was observed in aortic tissue from elastase-perfused WT mice compared to controls on day 14 (79.6±3.67 vs. 10.58±1.64 ng/ml; p<0.05; Figure 2B). Treatment of WT mice with MSCs significantly attenuated HMGB1 expression in the elastase-perfused murine aortic tissue compared to elastase-perfused mice alone (33.49±3.67 vs. 79.6±3.67 ng/ml; p<0.001). Moreover, aortic tissue from elastase-perfused WT mice treated with MSCs showed a significant attenuation in MMP2 and MMP9 activity compared to elastase-perfusion alone (Figure 2C–E). These results suggest an important role of MSCs in the mitigation of aortic inflammation, matrix degradation and AAA formation.

Macrophages Play a Key Role in AAA Formation via HMGB1 Production

Treatment of WT mice with anti-HMGB1 antibody significantly attenuated the elastase-perfused increase of aortic diameter at day 14 in WT mice compared to elastase-perfused mice alone (70.43±6.34 vs. 93.63±2.85% respectively; p<0.05; Figure 3A). There was no significant difference in aortic diameter in WT mice controls treated with or without anti-HMGB1 antibody. A significant attenuation of IL-17 was observed in aortic tissue of anti-HMGB1 antibody treated elastase-perfused WT mice compared to elastase-perfused mice alone (47.64±5.28 vs. 118.5±10.15 pg/ml respectively; p<0.05; Figure 3B). Comparative histology and immunostaining of aortic tissue on day 14 showed a significant decrease in T lymphocytes (CD3+), macrophages (Mac-2), and neutrophils (Ly6G+) as well as decrease in elastic fiber disruption in elastase-perfused WT mice treated with anti-HMGB1 antibody compared to elastase-perfused mice (Figure 3C). Immunostaining with another macrophage-specific marker (i.e. CD68) and HMGB1 in aortic tissue from elastase-perfused WT mice on day 14 also demonstrated co-localization of HMGB1 with macrophages (Supplemental Figure S2). The respective IgG control antibodies showed negligible immunostaining for macrophages and HMGB1 (Supplemental Figure S3).

Figure 3.

Figure 3

Figure 3

Open in a new tab

HMGB1 inhibition leads to decreased AAA formation. A, Treatment of elastase-perfused WT mice with anti-HMGB1 antibody significantly attenuates the increase in aortic diameter observed in elastase-perfused WT mice on day 14. B, IL-17 production in the aortic tissue on day 14 is attenuated in elastase-perfused WT mice treated with anti-HMGB1 antibody compared to elastase-perfused WT mice alone. C, Comparative histology and immunohistochemistry performed on day 14 indicates that WT mice treated with anti-HMGB1 antibody have a marked decrease in CD3+ T cell, neutrophil (PMN) and macrophage (Mac-2) infiltration, as well as decrease in elastic fiber disruption (Verhoeff-Van Gieson staining for elastin) compared to elastase-perfused WT mice. Arrows indicate areas of immunostaining. Scale bars represent 50 μM. n=5 mice/group; *p<0.05 vs. Control, #p<0.05 vs. Elastase+IgG.

NADPH Oxidase 2 (Nox2) Mediates HMGB1 Production and AAA Formation

One feature of AAA pathogenesis is inflammation and vascular remodeling which can be regulated by reactive oxygen species and oxidative stress. To determine if oxidative stress via NADPH oxidase can independently modulate the activation of HMGB1, Nox2−/y mice were exposed to elastase-perfusion AAA model. A significant attenuation of aortic diameter was observed in the elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice (83.14±9.33 vs. 162.6±17.91% respectively; p<0.05; Figure 4A). There was no significant difference in aortic diameter in Nox2−/y mice controls compared to WT mice controls. Aortic tissue from elastase-perfused Nox2−/y mice demonstrated a significant attenuation in HMGB1 expression (Figure 4B) as well as MMP2 and MMP9 activity (Figure 4C–E) compared to elastase-perfused WT mice.

Figure 4.

Figure 4

Figure 4

Figure 4

Open in a new tab

AAA formation is attenuated in Nox2−/y mice. A, Elastase-perfused Nox2−/y mice have a significantly decreased aortic diameter compared to elastase-perfused WT mice on day 14 (n=5–9/group). B, HMGB1 expression is significantly attenuated in aortic tissue of elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice on day 14 (n=5/group). C–E, Gelatin zymography of murine aortic tissue and subsequent quantification of optical density (O.D.) demonstrates a significant attenuation of MMP2 and MMP9 expression in elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice (n=3–5/group). F, Pro-inflammatory cytokine production, especially IL-17, in the aortic tissue on day 14 is attenuated in elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice (n=5–8/group). G, A marked decrease in CD3+ T cell, neutrophil (PMN) and macrophage (Mac-2) infiltration as well as decrease in elastic fiber disruption (Verhoeff-Van Gieson staining for elastin) was observed in elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice (n=5/group). Arrows indicate areas of immunostaining. Scale bars represent 50 μM. Mean+/− S.E.; *p<0.05 vs. WT Control; #p<0.05 vs. WT Elastase.

Furthermore, aortic tissue from Nox2−/y mice displayed a significant attenuation of pro-inflammatory cytokines, including IL-17, compared to elastase-perfused WT mice (Figure 4F). Comparative histology and immunohistological analysis of aortic tissue also revealed a significant attenuation of inflammatory cell infiltration (CD3+ T cells, macrophages and neutrophils) and a decrease in elastic fiber disruption in elastase-perfused Nox2−/y mice compared to elastase-perfused WT mice (Figure 4G). HMGB1 expression was also markedly attenuated in the aortic tissue of elastase-perfused Nox2−/y mice compared to WT mice on days 3, 7 and 14 by immunofluorescence via confocal microscopy (Figure 5). These results demonstrate that aortic diameter is significantly reduced in elastase-perfused Nox2−/y mice and this is associated with concomitant mitigation of HMGB1 and IL-17 production, matrix degradation and inflammatory cell infiltration.

Figure 5.

Figure 5

Figure 5

Open in a new tab

A, Macrophage-produced HMGB1 expression is attenuated in Nox2−/y mice. Confocal immunohistochemistry performed on aortic tissue from elastase-perfused Nox2−/y and WT mice on days 3, 7 and 14 with cell nuclei (4′,6-diamidino-2-phenylindole [DAPI]), smooth muscle cells (SM α-actin), macrophages (Mac-2) and HMGB1. Elastase-perfused Nox2−/y mice demonstrated decreased expression of HMGB1 immunostaining compared to WT mice. B, Co-localization of HMGB1 expression with macrophages in aortic tissue of elastase-perfused WT mice on day 14 is decreased in elastase-perfused Nox2−/y mice. Areas of immunostaining are highlighted in squares. Scale bars represent 10 μM. n=5 mice/group.

Mesenchymal Stem Cells Suppress Macrophage and CD4+ T Cell Activation To Attenuate HMGB1 and IL-17 Production

To determine the signaling events in Nox2-dependent mechanism of macrophage produced HMGB1 and to decipher the crosstalk between macrophages and CD4+ T cells via HMGB1/IL-17 axis, in vitro experiments were conducted. Primary CD11b+ macrophages (1×105 cells) were grown in culture with or without MSCs and transiently (5 min) exposed to elastase (0.4 U/mL). Cells were washed with PBS and fresh media was added. HMGB1 measurement in culture supernatants was performed after 24hr. A significant increase in HMGB1 production was observed in elastase-treated WT macrophages which were attenuated by co-cultures with MSCs (29.6±3.4 vs. 12.8±2.9 ng/ml, respectively; p<0.05; Figure 6A). Moreover, elastase-treated macrophages from Nox2−/y mice showed a significantly decreased expression of HMGB1 compared to elastase-treated WT macrophages (11.4±2.4 vs. 29.6±3.4 ng/ml, respectively). There was no additional attenuation of HMGB1 levels by MSC treatment in elastase-perfused Nox2−/y macrophages compared to elastase-treatment alone (7.3±2.3 vs. 11.4±2.4 ng/ml, respectively). This data suggests a prominent role of Nox2 in macrophage-dependent HMGB1 production.

Figure 6.

Figure 6

Figure 6

Open in a new tab

MSCs inhibit NADPH oxidase activation in macrophages to attenuate HMGB1 production. A, CD11b+ macrophages from WT and Nox2−/y mice were exposed to transient exposure with elastase (0.4U/ml for 5 min) and HMGB1 secretion was measured in cell culture supernatants after 24hr. A multifold increase in HMGB1 production was observed in WT macrophages compared to controls and was significantly attenuated by co-cultures with MSCs. Also, elastase-treated Nox2−/y macrophages had a significant decrease in HMGB1 expression compared to elastase-treated WT macrophages. n=8/group; *p<0.05 vs. Control; #p<0.05 vs. WT Elastase. B, NADPH oxidase activity was measured by quantification of superoxide anion production in cell culture supernatants of WT and Nox2−/y derived-macrophages. Elastase-treated WT macrophages displayed a significant increase in superoxide anion production compared to controls and this was attenuated by MSCs. However, there was no significant induction of NADPH oxidase activation in WT macrophages treated with apocynin or in elastase-treated Nox2−/y derived-macrophages. n=8/group; *p<0.05 vs. Control; #p<0.05 vs. Elastase. C, CD4+ T cells purified from WT mice were grown in culture with or without MSCs and treated with either recombinant (r)HMGB1 (10 ng/ml), transient elastase-treatment or both. Cell culture supernatants were analyzed for IL-17 expression. rHMGB1 or elastase treatment stimulate CD4+ T cells to produce IL-17 which is significantly attenuated by MSCs. n=8/group; *p<0.05 vs. control; #p<0.05 vs. elastase alone; ##p<0.05 vs. rHMGB1; **p<0.05 vs. elastase and δp<0.05 vs. elastase+rHMGB1. D, IL-17 production was significantly enhanced upon conditioned media transfer (CMT) from elastase-exposed WT macrophages (MΦElastase) to WT CD4+ T cells compared to controls. Co-culture of MSCs with elastase-exposed macrophages blocked the enhancement of IL-17 production after CMT to CD4+ T cells. Pretreatment of elastase-exposed macrophages with anti-HMGB1 (1 μg/ml) significantly attenuated IL-17 production after CMT to CD4+ T cells. *P<0.05 vs. Control; #P<0.05 vs.MΦElastase→ CD4+ T; n = 8/group. E, Schematic diagram of signaling events during MSC-mediated attenuation of AAA formation. Hypothetical schematic depicts the crosstalk between macrophages and CD4+ T cells during AAA formation. Upon activation, macrophages produce HMGB1 which is dependent on Nox2 activation. Once secreted, HMGB1 can further activate CD4+ T cells to produce IL-17 likely via receptor for advanced glycation end products (RAGE) or toll-like receptors (TLRs) and initiate a proinflammatory signaling cascade leading to aortic inflammation, matrix degradation and AAA formation. Treatment with MSCs inhibited superoxide anion production to attenuate macrophage-produced HMGB1 and subsequent CD4+ T cell-produced IL-17 thereby mitigating inflammation and AAA formation. MSCs can also inhibit CD4+ T cell-produced IL-17 via its direct actions on T cells which may involve Nox2-dependent mechanisms.

Furthermore, NADPH oxidase activity was assessed by measuring superoxide anion production in primary CD11b+ macrophages from WT mice. Elastase-treated WT macrophages had a significant increase in superoxide anion production which was attenuated by co-culture of macrophages with MSCs (Figure 6B). Elastase-exposed WT macrophages treated with apocynin and elastase-treated Nox2−/y macrophages displayed a significant decrease in superoxide anion production compared to elastase-treated WT macrophages. These results demonstrate that MSCs can downregulate Nox2 activation, which then leads to decreased HMGB1 production.

CD4+ T cells were also separately treated with recombinant HMGB1 (10 ng/ml) for 24hr and IL-17 production in the supernatants was measured. Elastase or HMGB1 treatment of CD4+ T cells significantly increased the production of IL-17, which was significantly attenuated by co-culture of CD4+ T cells with MSCs (Figure 6C). There was no production of either IL-17 by macrophages or HMGB1 production by CD4+ T cells after elastase treatment (Supplemental Figure S4). Also, recombinant IL-17 treatment of CD11b+ macrophages did not induce HMGB1 production (Supplemental Figure S4), while recombinant HMGB1 treatment of CD4+ T cells induces IL-17 production (Figure 6C).

Conditioned media transfer (CMT) experiments were performed to evaluate the potential crosstalk between macrophages and CD4+ T cells via HMGB1-mediated signaling (Figure 6D). CMT from elastase-exposed WT macrophages to WT CD4+T cells resulted in a significant enhancement of IL-17 production compared to controls. Co-cultures of MSCs with elastase-exposed macrophages and subsequent CMT to WT CD4+ T cells significantly attenuated the CD4+ T cell-produced IL-17 production. Similarly, treatment of elastase-exposed macrophages with ant-HMGB1 antibody and subsequent CMT to WT CD4+ T cells blocked the CD4+ T cell-dependent IL-17 production (Figure 6D). These results signify the importance of the crosstalk between macrophages and CD4+ T cells via HMGB1 and IL-17 to induce aortic inflammation, which is markedly downregulated by MSCs via their actions on both macrophages and CD4+ T cells (Figure 6E).

Discussion

The present study demonstrates that Nox2 is a key regulator of HMGB1 production and subsequent inflammation in an animal model of AAA and this can be mitigated by treatment with MSCs. An increased expression of HMGB1 was observed in aortic tissue from human male AAA patients, and further investigation into the importance of HMGB1 and its induction of other pro-inflammatory cytokines, i.e. IL-17, were then conducted using in vivo and in vitro studies. Aortic diameter and pro-inflammatory cytokine production was significantly attenuated in aortic tissue of elastase-perfused WT mice treated with anti-HMGB1 antibody. Furthermore, elastase-perfused Nox2−/y mice demonstrated a significant attenuation of aortic diameter, HMGB1 expression, proinflammatory cytokine production, matrix degradation, as well as decreased immune cell infiltration and preservation of aortic morphology. In vitro studies confirmed that elastase treatment modulates activation of macrophages to produce HMGB1 in a Nox2-dependent manner. Moreover, MSCs downregulated superoxide anion production by mitigation of Nox2 activation to inhibit HMGB1 production in elastase-treated macrophages. The ability of macrophage-produced HMGB1 to amplify IL-17 generation by CD4+T cells demonstrates the critical signaling crosstalk between these immune cells which can be markedly reduced by immunomodulation of the Nox2 signaling pathway by MSCs.

Recent studies demonstrate an important role for macrophage-produced HMGB1 in vascular inflammation and its ability to rapidly cause cellular inflammation12, 13. HMGB1 has been shown to act as a danger signal which mediates crosstalk between immune cells resulting in initiation of inflammation and injury14, 15. Immune cells, like macrophages and monocytes, actively release HMGB1 in response to exogenous (i.e. bacterial endotoxin) or endogenous host stimuli (TNF-α, IFN-γ or H2O2)16–18. Upon release, it can bind to several receptors including receptor for advanced glycation end products (RAGE), toll-like receptors (TLR)-2 and TLR-4, thereby triggering cell signaling pathways involving MAPK, NF-κB or PI3K/Akt to mediate cell migration, activation, proliferation and differentiation8, 19, 20. Recent studies have elucidated an increased expression of HMGB1 in aortic atherosclerosis in human and animal studies underlining the importance of HMGB1 and its receptors in regulating aortic inflammation21, 22. In the present study, we defined a crucial aspect of macrophage-mediated aortic inflammation via HMGB1 which upregulates CD4+ T cell-dependent IL-17 to initiate a signaling cascade resulting in AAA formation. HMGB1/TLR-4 signaling has also been shown to activate NADPH oxidase activation in circulating neutrophils to mediate hemorrhagic shock/resuscitation23. However, in our in vivo and in vitro studies, we demonstrated a critical link between Nox2 activation and HMGB1 release in macrophages. Thus, ROS generation via Nox2 can lead to increased HMGB1 production in activated, viable immune cells and upon release, HMGB1 retains its proinflammatory cytokine activity to further induce inflammation.

ROS production is a major mediator of the inflammatory cascade which is responsible for the pathogenesis of vascular inflammation and AAA formation10, 24, 25. One of the key enzymes for ROS production is the NADPH oxidase complex which is a highly regulated membrane-bound enzyme complex that catalyzes the production of superoxide by the single electron reduction of oxygen using NADPH as the electron donor. In particular, a major contributor to redox imbalance is H2O2, which is converted from superoxide and can induce both active and passive release of HMGB1 from macrophages. Previous findings have established a key role for H2O2-dependent oxidative stress in inducing active HMGB1 release in macrophage/monocyte cultures via an MAPK- and chromosome region maintenance 1 (CRM1)-dependent mechanism17. Our present study suggests that Nox2 activation regulates HMGB1 release in macrophages, which is likely mediated by the p47phox subunit of the NADPH oxidase complex. In fact, Aoki et al. demonstrated that p47phox was mainly expressed in macrophages and upregulated during cerebral aneurysm formation26. On the other hand, a recent study by Kigawa et al showed that NADPH oxidase deficiency can exacerbate angiotensin-II induced AAA formation in mice27. This unexpected occurrence of hyperinflammation by Nox2 deficiency is contrary to our findings where elastase-perfused Nox2−/y mice were significantly protected from AAA. The apparent contrast of Nox2 signaling in aortic aneurysms between the two studies could be attributed to differences in the mouse models as Kigawa et al. induced aneurysmal formation in atherogenic mice which were fed a high cholesterol diet compared to the elastase-perfusion model used in our study. The complex heterogeneity between different Nox isoforms and the deleterious role of free radicals in T cells and macrophages may account for the phenotypic differences in Nox2−/y mice between the two models of AAA. However, our findings are in conjunction with previous studies wherein the deletion of p47phox subunit of NADPH oxidase results in protection from cerebral and aortic aneurysm formation as shown in animal models using p47phox−/− mice11, 26. Our study demonstrates the ability of HMGB1 to induce pro-inflammatory cytokines during the pathogenesis of AAA formation and specifically decipher the crosstalk between macrophages and CD4+ T cells, as Nox2-dependent HMGB1 augmented CD4+T cell-produced IL-17, which leads to vascular remodeling and AAA formation. Importantly, we present evidence that human MSCs have the unique ability to immunomodulate Nox2 activation and downregulate HMGB1 and IL-17 production which is associated with decrease in experimental AAA formation.

The potent anti-inflammatory and immunosuppressive properties of MSCs have been widely used as a cell-based therapy to treat various diseases like graft-versus-host disease, myocardial infarction and stroke28–31. We and others have recently demonstrated the remarkable ability of MSCs to attenuate AAA formation in animal models, as well as to migrate to the site of aortic inflammation5–7, 32, 33. The paracrine effects of MSCs have been postulated to be mediated by various soluble factors, such as prostaglandin E2 (PGE2), which inhibits actions on T lymphocytes, TGF-β and IL-10, which can inhibit the proinflammatory potential of other immune cells and mitigate aortic smooth muscle cell activation and remodeling7, 34–36. However, a direct effect of MSCs on immune cell activation via Nox2 during the pathogenesis of AAA remains undefined. In this study, we present evidence of an antioxidant effect of MSCs to inhibit NADPH oxidase activation which can lead to inhibition of macrophage-produced HMGB1. The mechanism of MSC-mediated inhibition of Nox2 is postulated to be mediated via cell-cell contact or paracrine effects through upregulation of PGE2, keratinocyte growth factor (KGF), hepatocyte growth factor (HGF), TGF-β or IL-10.

In summary, our results highlight the mechanistic aspect of MSC therapy for AAA treatment via its direct effects on macrophage-dependent inflammation. The initiation of aortic inflammation by key regulatory cytokines i.e. macrophage-produced HMGB1 and CD4+ T cell-produced IL-17, can lead to immune cell infiltration, upregulation of MMPs, aortic smooth cell remodeling, thereby causing AAA formation. The ability of MSCs to modulate the pivotal initial inflammatory triggers is essential to attenuate the molecular pathogenesis of AAA formation. The multiple effects of MSCs on targeting the Nox2 pathway in macrophages and inhibition of the HMGB1/IL-17 axis demonstrates the multi-factorial capacities of MSCs as a non-surgical therapeutic strategy for treatment of aortic aneurysms. Additional studies are needed to decipher the impact of MSC-mediated upregulation of specific paracrine factors in the modulation of Nox subtypes on AAA formation.

Supplementary Material

1
2

Significance.

Abdominal aortic aneurysm (AAA) remains a significant clinical problem in the elderly population. Although most patients are asymptomatic, the risk of mortality increases due to aortic rupture upon expansion of the aneurysm. Currently, the only definitive treatment is surgical intervention and a therapeutic approach to treat AAAs remains to be defined. In this study, we determined that high mobility group box 1 (HMGB1), a proinflammmatory cytokine secreted by macrophages, regulates AAA formation via upregulation of CD4+ T cell-dependent IL-17 production. Furthermore, we observed that HMGB1 production is dependent on NADPH oxidase complex as Nox2−/y mice demonstrate a significant reduction in HMGB1 production and AAA formation. Finally, our results further define a key role for human mesenchymal stem cells (MSCs) in decreasing HMGB1 production via inhibition of Nox2 activation, thereby leading to inhibition of AAA growth.

Acknowledgments

The authors would like to thank Anthony Herring and Cynthia Dodson at the University of Virginia for help with maintaining animal colonies.

Funding

This work was supported by National Institute of Health grant RO1 HL081629 (GRU).

Nonstandard Abbreviations and Acronyms

AAA

abdominal aortic aneurysms

MSC

mesenchymal stem cells

HMGB1

high mobility group box 1

IL-17

interleukin-17

NOX

nicotinamide adenine dinucleotide phosphate oxidase

TNF

tumor necrosis factor

MMP

matrix metalloproteinase

RAGE

receptor for advanced glycation end products

Footnotes

Disclosures

None

References

  • 1.Bengtsson H, Sonesson B, Bergqvist D. Incidence and prevalence of abdominal aortic aneurysms, estimated by necropsy studies and population screening by ultrasound. Ann N Y Acad Sci. 1996;800:1–24. doi: 10.1111/j.1749-6632.1996.tb33294.x. [DOI] [PubMed] [Google Scholar]
  • 2.Ailawadi G, Eliason JL, Upchurch GR., Jr Current concepts in the pathogenesis of abdominal aortic aneurysm. J Vasc Surg. 2003;38:584–8. doi: 10.1016/s0741-5214(03)00324-0. [DOI] [PubMed] [Google Scholar]
  • 3.Anidjar S, Dobrin PB, Eichorst M, Graham GP, Chejfec G. Correlation of inflammatory infiltrate with the enlargement of experimental aortic aneurysms. J Vasc Surg. 1992;16:139–47. doi: 10.1067/mva.1992.35585. [DOI] [PubMed] [Google Scholar]
  • 4.Aziz F, Kuivaniemi H. Role of matrix metalloproteinase inhibitors in preventing abdominal aortic aneurysm. Ann Vasc Surg. 2007;21:392–401. doi: 10.1016/j.avsg.2006.11.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Schneider F, Saucy F, de Blic R, Dai J, Mohand F, Rouard H, Ricco JB, Becquemin JP, Gervais M, Allaire E. Bone marrow mesenchymal stem cells stabilize already-formed aortic aneurysms more efficiently than vascular smooth muscle cells in a rat model. Eur J Vasc Endovasc Surg. 2013;45:666–72. doi: 10.1016/j.ejvs.2013.03.007. [DOI] [PubMed] [Google Scholar]
  • 6.Turnbull IC, Hadri L, Rapti K, Sadek M, Liang L, Shin HJ, Costa KD, Marin ML, Hajjar RJ, Faries PL. Aortic implantation of mesenchymal stem cells after aneurysm injury in a porcine model. J Surg Res. 2011;170:e179–88. doi: 10.1016/j.jss.2011.05.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Sharma AK, Lu G, Jester A, Johnston WF, Zhao Y, Hajzus VA, Saadatzadeh MR, Su G, Bhamidipati CM, Mehta GS, Kron IL, Laubach VE, Murphy MP, Ailawadi G, Upchurch GR., Jr Experimental abdominal aortic aneurysm formation is mediated by IL-17 and attenuated by mesenchymal stem cell treatment. Circulation. 2012;126:S38–45. doi: 10.1161/CIRCULATIONAHA.111.083451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Sharma AK, LaPar DJ, Stone ML, Zhao Y, Kron IL, Laubach VE. Receptor for advanced glycation end products (RAGE) on iNKT cells mediates lung ischemia-reperfusion injury. Am J Transplant. 2013;13:2255–67. doi: 10.1111/ajt.12368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kohno T, Anzai T, Kaneko H, Sugano Y, Shimizu H, Shimoda M, Miyasho T, Okamoto M, Yokota H, Yamada S, Yoshikawa T, Okada Y, Yozu R, Ogawa S, Fukuda K. High-mobility group box 1 protein blockade suppresses development of abdominal aortic aneurysm. J Cardiol. 2012;59:299–306. doi: 10.1016/j.jjcc.2012.01.007. [DOI] [PubMed] [Google Scholar]
  • 10.Miller FJ, Jr, Sharp WJ, Fang X, Oberley LW, Oberley TD, Weintraub NL. Oxidative stress in human abdominal aortic aneurysms: a potential mediator of aneurysmal remodeling. Arterioscler Thromb Vasc Biol. 2002;22:560–5. doi: 10.1161/01.atv.0000013778.72404.30. [DOI] [PubMed] [Google Scholar]
  • 11.Thomas M, Gavrila D, McCormick ML, Miller FJ, Jr, Daugherty A, Cassis LA, Dellsperger KC, Weintraub NL. Deletion of p47phox attenuates angiotensin II-induced abdominal aortic aneurysm formation in apolipoprotein E-deficient mice. Circulation. 2006;114:404–13. doi: 10.1161/CIRCULATIONAHA.105.607168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Lotze MT, Tracey KJ. High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Nat Rev Immunol. 2005;5:331–42. doi: 10.1038/nri1594. [DOI] [PubMed] [Google Scholar]
  • 13.Schiraldi M, Raucci A, Munoz LM, Livoti E, Celona B, Venereau E, Apuzzo T, De Marchis F, Pedotti M, Bachi A, Thelen M, Varani L, Mellado M, Proudfoot A, Bianchi ME, Uguccioni M. HMGB1 promotes recruitment of inflammatory cells to damaged tissues by forming a complex with CXCL12 and signaling via CXCR4. J Exp Med. 2012;209:551–63. doi: 10.1084/jem.20111739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Li J, Wang H, Mason JM, Levine J, Yu M, Ulloa L, Czura CJ, Tracey KJ, Yang H. Recombinant HMGB1 with cytokine-stimulating activity. J Immunol Methods. 2004;289:211–23. doi: 10.1016/j.jim.2004.04.019. [DOI] [PubMed] [Google Scholar]
  • 15.Yanai H, Ban T, Wang Z, Choi MK, Kawamura T, Negishi H, Nakasato M, Lu Y, Hangai S, Koshiba R, Savitsky D, Ronfani L, Akira S, Bianchi ME, Honda K, Tamura T, Kodama T, Taniguchi T. HMGB proteins function as universal sentinels for nucleic-acid-mediated innate immune responses. Nature. 2009;462:99–103. doi: 10.1038/nature08512. [DOI] [PubMed] [Google Scholar]
  • 16.Rendon-Mitchell B, Ochani M, Li J, Han J, Wang H, Yang H, Susarla S, Czura C, Mitchell RA, Chen G, Sama AE, Tracey KJ, Wang H. IFN-gamma induces high mobility group box 1 protein release partly through a TNF-dependent mechanism. J Immunol. 2003;170:3890–7. doi: 10.4049/jimmunol.170.7.3890. [DOI] [PubMed] [Google Scholar]
  • 17.Tang D, Shi Y, Kang R, Li T, Xiao W, Wang H, Xiao X. Hydrogen peroxide stimulates macrophages and monocytes to actively release HMGB1. J Leukoc Biol. 2007;81:741–7. doi: 10.1189/jlb.0806540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Ivanov S, Dragoi AM, Wang X, Dallacosta C, Louten J, Musco G, Sitia G, Yap GS, Wan Y, Biron CA, Bianchi ME, Wang H, Chu WM. A novel role for HMGB1 in TLR9-mediated inflammatory responses to CpG-DNA. Blood. 2007;110:1970–81. doi: 10.1182/blood-2006-09-044776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Yu M, Wang H, Ding A, Golenbock DT, Latz E, Czura CJ, Fenton MJ, Tracey KJ, Yang H. HMGB1 signals through toll-like receptor (TLR) 4 and TLR2. Shock. 2006;26:174–9. doi: 10.1097/01.shk.0000225404.51320.82. [DOI] [PubMed] [Google Scholar]
  • 20.Tsung A, Klune JR, Zhang X, Jeyabalan G, Cao Z, Peng X, Stolz DB, Geller DA, Rosengart MR, Billiar TR. HMGB1 release induced by liver ischemia involves Toll-like receptor 4 dependent reactive oxygen species production and calcium-mediated signaling. J Exp Med. 2007;204:2913–23. doi: 10.1084/jem.20070247. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Inoue K, Kawahara K, Biswas KK, Ando K, Mitsudo K, Nobuyoshi M, Maruyama I. HMGB1 expression by activated vascular smooth muscle cells in advanced human atherosclerosis plaques. Cardiovasc Pathol. 2007;16:136–43. doi: 10.1016/j.carpath.2006.11.006. [DOI] [PubMed] [Google Scholar]
  • 22.Kalinina N, Agrotis A, Antropova Y, DiVitto G, Kanellakis P, Kostolias G, Ilyinskaya O, Tararak E, Bobik A. Increased expression of the DNA-binding cytokine HMGB1 in human atherosclerotic lesions: role of activated macrophages and cytokines. Arterioscler Thromb Vasc Biol. 2004;24:2320–5. doi: 10.1161/01.ATV.0000145573.36113.8a. [DOI] [PubMed] [Google Scholar]
  • 23.Fan J, Li Y, Levy RM, Fan JJ, Hackam DJ, Vodovotz Y, Yang H, Tracey KJ, Billiar TR, Wilson MA. Hemorrhagic shock induces NAD(P)H oxidase activation in neutrophils: role of HMGB1-TLR4 signaling. J Immunol. 2007;178:6573–80. doi: 10.4049/jimmunol.178.10.6573. [DOI] [PubMed] [Google Scholar]
  • 24.Guzik B, Sagan A, Ludew D, Mrowiecki W, Chwala M, Bujak-Gizycka B, Filip G, Grudzien G, Kapelak B, Zmudka K, Mrowiecki T, Sadowski J, Korbut R, Guzik TJ. Mechanisms of oxidative stress in human aortic aneurysms–association with clinical risk factors for atherosclerosis and disease severity. Int J Cardiol. 2013;168:2389–96. doi: 10.1016/j.ijcard.2013.01.278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Usui F, Shirasuna K, Kimura H, Tatsumi K, Kawashima A, Karasawa T, Yoshimura K, Aoki H, Tsutsui H, Noda T, Sagara J, Taniguchi S, Takahashi M. Inflammasome activation by mitochondrial oxidative stress in macrophages leads to the development of angiotensin II-induced aortic aneurysm. Arterioscler Thromb Vasc Biol. 2015;35:127–36. doi: 10.1161/ATVBAHA.114.303763. [DOI] [PubMed] [Google Scholar]
  • 26.Aoki T, Nishimura M, Kataoka H, Ishibashi R, Nozaki K, Hashimoto N. Reactive oxygen species modulate growth of cerebral aneurysms: a study using the free radical scavenger edaravone and p47phox(−/−) mice. Lab Invest. 2009;89:730–41. doi: 10.1038/labinvest.2009.36. [DOI] [PubMed] [Google Scholar]
  • 27.Kigawa Y, Miyazaki T, Lei XF, Nakamachi T, Oguchi T, Kim-Kaneyama JR, Taniyama M, Tsunawaki S, Shioda S, Miyazaki A. NADPH oxidase deficiency exacerbates angiotensin II-induced abdominal aortic aneurysms in mice. Arterioscler Thromb Vasc Biol. 2014;34:2413–20. doi: 10.1161/ATVBAHA.114.303086. [DOI] [PubMed] [Google Scholar]
  • 28.Mei SH, Haitsma JJ, Dos Santos CC, Deng Y, Lai PF, Slutsky AS, Liles WC, Stewart DJ. Mesenchymal stem cells reduce inflammation while enhancing bacterial clearance and improving survival in sepsis. Am J Respir Crit Care Med. 2010;182:1047–57. doi: 10.1164/rccm.201001-0010OC. [DOI] [PubMed] [Google Scholar]
  • 29.Panfilov IA, de Jong R, Takashima S, Duckers HJ. Clinical study using adipose-derived mesenchymal-like stem cells in acute myocardial infarction and heart failure. Methods Mol Biol. 2013;1036:207–12. doi: 10.1007/978-1-62703-511-8_16. [DOI] [PubMed] [Google Scholar]
  • 30.Le Blanc K, Frassoni F, Ball L, Locatelli F, Roelofs H, Lewis I, Lanino E, Sundberg B, Bernardo ME, Remberger M, Dini G, Egeler RM, Bacigalupo A, Fibbe W, Ringden O, Developmental Committee of the European Group for B. Marrow T. Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versus-host disease: a phase II study. Lancet. 2008;371:1579–86. doi: 10.1016/S0140-6736(08)60690-X. [DOI] [PubMed] [Google Scholar]
  • 31.Honmou O, Houkin K, Matsunaga T, Niitsu Y, Ishiai S, Onodera R, Waxman SG, Kocsis JD. Intravenous administration of auto serum-expanded autologous mesenchymal stem cells in stroke. Brain. 2011;134:1790–807. doi: 10.1093/brain/awr063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hashizume R, Yamawaki-Ogata A, Ueda Y, Wagner WR, Narita Y. Mesenchymal stem cells attenuate angiotensin II-induced aortic aneurysm growth in apolipoprotein E-deficient mice. J Vasc Surg. 2011;54:1743–52. doi: 10.1016/j.jvs.2011.06.109. [DOI] [PubMed] [Google Scholar]
  • 33.Davis JP, Salmon M, Pope NH, Lu G, Su G, Sharma AK, Ailawadi G, Upchurch GR., Jr Attenuation of aortic aneurysms with stem cells from different genders. J Surg Res. 2015;199:249–58. doi: 10.1016/j.jss.2015.04.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Singer NG, Caplan AI. Mesenchymal stem cells: mechanisms of inflammation. Annu Rev Pathol. 2011;6:457–78. doi: 10.1146/annurev-pathol-011110-130230. [DOI] [PubMed] [Google Scholar]
  • 35.Kronsteiner B, Wolbank S, Peterbauer A, Hackl C, Redl H, van Griensven M, Gabriel C. Human mesenchymal stem cells from adipose tissue and amnion influence T-cells depending on stimulation method and presence of other immune cells. Stem Cells Dev. 2011;20:2115–26. doi: 10.1089/scd.2011.0031. [DOI] [PubMed] [Google Scholar]
  • 36.Duffy MM, Pindjakova J, Hanley SA, McCarthy C, Weidhofer GA, Sweeney EM, English K, Shaw G, Murphy JM, Barry FP, Mahon BP, Belton O, Ceredig R, Griffin MD. Mesenchymal stem cell inhibition of T-helper 17 cell- differentiation is triggered by cell-cell contact and mediated by prostaglandin E2 via the EP4 receptor. Eur J Immunol. 2011;41:2840–51. doi: 10.1002/eji.201141499. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS765143-supplement-1.pdf (135.7KB, pdf)
2
NIHMS765143-supplement-2.pdf (513.9KB, pdf)

RESOURCES