RENAL-CARDIAC-VASCULAR

CCL2-Dependent Macrophage Recruitment Is Critical for Mineralocorticoid Receptor-Mediated Cardiac Fibrosis, Inflammation, and Blood Pressure Responses in Male Mice J.Z. Shen, J. Morgan, G.H. Tesch, P.J. Fuller, and M.J. Young Prince Henry’s Institute of Medical Research (J.Z.S., J.M., P.J.F., M.J.Y.); Departments of Medicine (J.Z.S., G.H.T., P.J.F., M.J.Y.) and Physiology (M.J.Y.), Monash University; and Department of Nephrology (G.H.T.), Monash Medical Centre, Clayton 3168, Victoria, Australia

Recent studies show that mice with selective deletion of the mineralocorticoid receptor (MR) in macrophages are protected from mineralocorticoid-induced cardiac fibrosis and hypertension without altering cardiac macrophage accumulation. However, it is unclear whether preventing macrophages from entering cardiac tissue would provide similar or additional protection in this disease setting. Therefore, we examined mineralocorticoid-induced cardiovascular disease in mice lacking the CCL2 gene (encoding monocyte chemoattractant protein-1), which have a markedly reduced capacity to recruit proinflammatory tissue macrophages. Male wild-type (WT) and CCL2null mice were treated for 8 days or 8 weeks with either vehicle (control, CON) or deoxycorticosterone (DOC). At both time points, there was a significant reduction in DOC-induced macrophage recruitment (50% at 8 d and 75% at 8 wk) in the heart with a corresponding suppression of cardiac inflammatory markers in the CCL2-null mice. CCL2-null mice given DOC/salt also displayed 35% less cardiac fibrosis at 8 weeks vs WT DOC. Absence of recruited macrophages in CCL2-null mice promotes greater collagen breakdown by matrix metalloproteinase-9 in the heart and also leads to significantly reduced cardiac fibroblast and myofibroblast numbers. Systolic blood pressure (BP) after DOC/salt was significantly lower in CCL2-null than for WT mice. In the aorta at 8 weeks, MR-responsive gene expression remained intact. However, macrophage-mediated proinflammatory gene expression was reduced in the CCL2-null mice and may account for differential regulation of BP. Our data thus demonstrate an important role for CCL2-dependent macrophage recruitment in MR-dependent cardiac inflammation and remodeling and in the regulation of systolic BP. (Endocrinology 155: 1057–1066, 2014)

M

ineralocorticoid receptor (MR) signaling plays an important mediating role in cardiovascular disease entities, such as cardiac arrhythmia, and heart failure as well as hypertension (1). Similarly, in experimental animals, activation of MR in the presence of salt generates early oxidative stress and inflammatory responses followed by adverse cardiac tissue remodeling and fibrosis (2– 4), leading to subsequent cardiovascular dysfunction; all of which may be reversed with MR antagonist treatment (5, 6).

Macrophages are specialized phagocytes with pleiotropic capabilities, including defense against foreign organisms, promoting healing and repair, and immunomodulation (7). Macrophage MR activation by aldosterone or deoxycorticosterone (DOC) promotes an inflammatory phenotype, which includes synthesis of cytokines capable of amplifying tissue oxidative stress and inflammation (8). The recruitment of these macrophages thus plays a major role in cardiovascular tissue inflammation and, subsequently, in repair and fibrosis, when a chronic inflamma-

ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2014 by the Endocrine Society Received August 19, 2013. Accepted December 23, 2013. First Published Online January 15, 2014

Abbreviations: Ang II, angiotensin II; CCR, C-C chemokine receptor; CD3, cluster of differentiation 3; COL3, collagen 3; CON, control; CTGF, connective tissue growth factor; DOC, deoxycorticosterone; FSP, fibroblast-specific protein; MMP, matrix metalloproteinase; MR, mineralocorticoid receptor; Nox2, NADPH oxidase 2; RANTES, Regulated on Activation, Normal T Cell Expressed and Secreted; SBP, systolic blood pressure; ␣SMA, ␣-smooth muscle actin; WT, wild type.

doi: 10.1210/en.2013-1772

Endocrinology, March 2014, 155(3):1057–1066

endo.endojournals.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

1057

1058

Shen et al

Macrophage and Cardiovascular Disease

tory state results from sustained MR activation. Previously, we (9, 10), and subsequently others (8), demonstrated the crucial role of macrophage MR signaling in the development of cardiac fibrosis and hypertension in both the DOC/salt model and the aldosterone-independent NG-nitro-l-arginine methyl ester/salt model of cardiovascular disease. We hypothesize that recruited macrophages are essential in promoting MR-mediated cardiovascular disease. This study aims to determine whether CCL2-directed macrophage recruitment plays a direct role in MR-dependent (DOC/salt) cardiac fibrosis and hypertension and to identify the mechanisms involved by interrogating and comparing the early inflammatory responses at 8 days and the expanded inflammatory, fibrotic, and hypertensive responses at the later 8-week time points in wild-type (WT) mice and mice lacking in the CCL2 gene (CCL2-null). The CCL2 gene encodes monocyte chemoattractant protein-1, the major monocyte chemokine required for tissue macrophage recruitment, which is typically released by injured tissues. The CCL2-null mouse is a well-characterized mouse model used to study the role of macrophages in a variety of pathologic states (11, 12). Although there are other chemokine-receptor systems that are able to recruit macrophages, CCL2 and its corresponding receptor C-C chemokine receptor (CCR) have previously been shown to be the principal mechanism for facilitating and directing macrophage infiltration into target tissues/organs (13, 14), particularly proinflammatory macrophages (15). Our data demonstrate a fundamental role for recruited macrophages in promoting cardiovascular tissue inflammation and remodeling. We have shown that proinflammatory cytokines generated by inflammatory macrophages promote MR-dependent tissue inflammation and injury. Our data further highlight the significant downstream effects of macrophage regulation of inflammation and fibrosis via other key cell types, such as T cells and fibroblasts; arguing for a complex interaction in the pathogenesis of cardiac fibrosis and hypertension.

Materials and Methods CCL2-null mice All procedures involving animals were approved by the Monash University Animal Ethics and Biosafety Committees. CCL2-null mice (129Sv/ X C57Bl/6)F1 were created by targeted gene disruption (13) and backcrossed more than 12 times onto a C57Bl/6 background; control mice were of C57Bl/6 background purchased from the Walter and Eliza Hall Institute animal facility (Clive and Vera Ramaciotti Laboratories). CCL2null mice are viable, breed normally, and show no morphological abnormalities.

Endocrinology, March 2014, 155(3):1057–1066

DOC/salt treatment model Male mice, approximately 8 weeks of age, of both genotypes (n ⫽ 10 per group) were uninephrectomized and given either vehicle or DOC (Sigma-Aldrich Co), administered via a sc slow release 21-day pellet (7-mg pellet replaced every 3 wk) embedded dorsally, as described in published literature (9, 16). Mice were maintained on standard chow and 0.9% NaCl plus 0.4% KCl solution and killed after 8 days or 8 weeks of treatment (total of 8 groups, cohort of 80 animals for the entire study).

Systolic blood pressure (SBP) SBP was measured by tail-cuff plethysmography at weeks 0, 4, and 8 as described previously in the literature (ITTC Life Science) (9, 16). Prewarmed, trained mice were subjected to 3 consecutive BP readings at 4 and 8 weeks. Recordings were used for between group analyses if no greater than 5-mm Hg variation is present.

Tissue collection The animals were killed by CO2 asphyxiation at 8 days or 8 weeks, and this was followed by the collection of an arterial blood sample, the heart and the aorta. The heart and aorta were immediately halved; one half into 4% paraformaldehyde fixation for histology, and the apex was snap frozen in liquid nitrogen for RNA extraction and quantitative RT-PCR.

RIA hormones The plasma concentrations of aldosterone, corticosterone, and angiotensin II (Ang II) were determined using ImmunChem Double Antibody RIA kits (MP Biomedicals) as per manufacturer’s instructions (9, 10).

Histological and immunohistochemical analyses Cardiac collagen was stained with 0.1% Sirius Red in saturated picric acid (Sigma-Aldrich) and subjected to systematic digital analysis of entire sections using Analytical Imaging Station software package (version 4.0 Beta 1.5; Imaging Research, Inc). Cardiac interstitial collagen content was quantified as a percentage of total myocardial area, excluding blood vessels. Tissue macrophages, T cells, neutrophils, fibroblasts, and myofibroblasts were detected by immunohistochemistry using primary antibodies: antigalectin 3 (Mac-2) antibody (1:500; eBioscience), anti-cluster of differentiation 3 (CD3) antibody (1: 150; Abcam), antineutrophil antibody (NIMP-R14) (1:50; Abcam), antifibroblast-specific protein (FSP)1 (S100A4) antibody (1:100; Abcam), and anti-␣-smooth muscle actin (␣SMA) antibody (1:50; Abcam), respectively, on 5-␮m heart sections. Investigators were blinded to the identities of the slides during analysis. Infiltrating Mac-2 positive macrophages and FSP1 positive fibroblasts were quantified by an optical dissector method using Computer-Assisted Stereological Toolbox (C.A.S.T.GRID) software package, version 1.10 (Olympus DKA/s), which randomly select a constant number of fields (n ⫽ 40) and provides a value for the average number of cells per frame (826 890 ␮m2) as previously described (9, 16). CD3 positive T cells and ␣SMA positive myofibroblasts were manually counted from alternate fields via light microscopy at ⫻20 and expressed as the average number of cells per view. Sections were preheated for 60 minutes in a 60°C oven and then either treated with trypsin for 10 minutes or boiled in citrate buffer for at least 5 minutes for

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2013-1772

antigen retrieval and incubated overnight with antibodies for TGF␤1 (1:400; Santa Cruz Biotechnology, Inc), connective tissue growth factor (CTGF) (1:400; Abcam), collagen 3 (COL3) (1:100; Abcam), or the relevant negative IgG as control at equivalent concentration. Appropriate biotinylated secondary antibodies, followed by ABC complex (Vectastain, Vector Laboratories) were subsequently applied. Incubation with 3,3⬘-diaminobenzidine (Sigma-Aldrich, Co) followed by counterstain with hematoxylin allowed tissue visualization. Assessment of immunostaining was performed by 2 investigators via light microscopy at ⫻20, using a semiquantitative scoring of 0, 1, 2, and 3, where 0 represents negative staining and 3 for strongly positive staining.

Quantitative RT-PCR Heart tissue was homogenized with a QIAGEN Tissue Lyser (QIAGEN). Total RNA was subsequently isolated using the QIAGEN RNAeasy Mini kit followed by Dnase treatment and removal with Ambion DNA free (Life Technologies). First strand cDNA synthesis from 400 ng of total RNA was performed using the SuperScript III First-Strand kit (Invitrogen, Life Technologies). PCRs were carried out with the primer sets listed in Supplemental Table 1, published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org. Quantitative PCR amplification was performed on Applied Biosystems 7900HT Fast Real-Time PCR System using SYBR Green reaction mix and was analyzed using SDS Automation Controller software (version 2.3; Applied Biosystems, Life Technologies) and normalized to glyceraldehyde 3-phosphate dehydrogenase or 18S mRNA levels. Relative quantification of changes in gene expression was calculated using the formula 2 (⫺⌬⌬Ct). Data presented are representative of at least 2 separate reverse transcriptase and quantitative PCR experiments.

endo.endojournals.org

1059

and baseline SBP (Supplemental Tables 2 and 3). The previously reported changes in plasma corticosterone, aldosterone, and Ang II were observed with DOC/salt treatment in both genotypes at 8 days and returned to vehicle-treated levels at 8 weeks (Supplemental Tables 2 and 3). Macrophages regulate inflammation in DOC/saltmediated cardiac pathology Inflammatory response at 8 days and 8 weeks Modest increases in inflammatory cell recruitment after DOC/salt treatment in WT mice at 8 days did not reach significance (Figure 1, A and B), but at 8 weeks, a significant recruitment of Mac-2 positive macrophages and CD3 positive T cells was observed as previously described (5) (50% increase in macrophage and 85% increase in T cell number; Figure 2, A and B, and Supplemental Figure 1). Absence of CCL2 was associated with an overall reduction in cardiac inflammatory cell infiltrate in controltreated mice. In response to DOC/salt treatment, there was a markedly suppressed recruitment of both macrophages (⫺50% at 8 d and ⫺70% at 8 wk) and T cells (⫺50% at both 8 d and 8 wk) (WT DOC vs CCL2-null DOC in Figures 1, A and B, and 2, A and B). Longer exposure to normal saline (8 d vs 8 wk) led to an observed significant overall increase in macrophage infiltrate in controltreated mice, both WT and CCL2-null without a change in T cell infiltrate. No significant neutrophil immunostaining was detected at either 8 days or 8 weeks (data not shown).

Matrix metalloproteinase (MMP)2/MMP9 activity The homogenized tissue was assayed with a Molecular Probes EnzChek Gelatinase/Collagenase Assay kit (Molecular Probes). Total protein was determined with the Pierce BCA Protein Assay kit (Thermo Scientific), and a total of 70 ␮g per sample was used. Individual samples were made up to a total volume of 200 ␮L with reaction buffer and 10 ␮L of DQ fluorescent gelatin, the substrate for gelatinases (tissue MMP2 and MMP9). Twenty micromoles per liter of EDTA were added to separate samples and inhibited MMP2/MMP9 activity by 30% (data not shown). Fluorescent activity was recorded on the Agilereader (Agilent) with a 515-nm filter at 1, 2, and 24 hours.

Statistics All data sets were analyzed by two-way ANOVA with Tukey’s multiple comparison post hoc tests to identify significant effects between groups (GraphPad Prism version 6.0a; GraphPad Software). The mean difference was considered significant at P ⬍ .05. All data are reported as mean ⫾ SEM.

Results CCL2-null mice CCL2-null mice displayed normal baseline characteristics, including fertility; body, heart and kidney weights;

Loss of CCL2 reduces the cardiac tissue proinflammatory response We observed a decrease in immune cell-regulated proinflammatory gene expression in hearts from CCL2-null mice compared with WT at both 8 days and 8 weeks, consistent with the reduced inflammatory cell infiltrate. Specifically, CCL2-null mice exhibited a significant reduction in mRNA levels for TNF␣ and Regulated on Activation, Normal T Cell Expressed and Secreted (RANTES) expression at 8 days compared with WT mice (Figure 1, C and D). In contrast, DOC/salt treatment significantly elevated inflammatory chemokine receptor CCR5 and osteopontin gene expression above control at 8 days (Supplemental Table 4), supporting a role for functional MR signaling in other cell types. At 8 weeks, the well-characterized markers of DOC/ salt cardiac responses NADPH oxidase 2 (Nox2), CCR5, and osteopontin (16, 17) were elevated in DOC-treated WT mice but not in CCL2-null mice (Figure 2, C and D, and Supplemental Table 4). Similarly, CCL2 mRNA levels were significantly increased by DOC/salt but were absent in the CCL2 mice as expected (Supplemental Figure 2).

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

Shen et al

A

Macrophage and Cardiovascular Disease

Macrophage number Ma ac-2+ cells per fram me

4

*

3 2 1 0

WT WT CON DOC

*

1.0 0.5

1.0

0.5

0.0

D

RANTES 1.5

CCL2 CCL2 -null -null CON DOC

P < 0.05 for CCL2-null vs. WT P = 0.33 for DOC vs. CON P = 0.33 for interaction

*

1.0

0.5

0.0

WT WT CON DOC

CCL2 CCL2 -null -null CON DOC

WT WT CON DOC

P < 0.05 for CCL2-null vs. WT P=0 0.83 83 ffor DOC vs. CON P = 0.11 for interaction

mRNA/18S

mRNA/18S

1.5

*

1.5

0.0

CCL2 CCL2 -null -null CON DOC

TNF

C

2-way ANOVA

T cells number 2.0

P < 0.05 for CCL2-null vs. WT P=0 0.14 14 ffor DOC vs. CON P = 0.13 for interaction

2-way ANOVA

Endocrinology, March 2014, 155(3):1057–1066

B CD3+ T cells per fram me

1060

WT WT CON DOC

CCL2 CCL2 -null -null CON DOC

P < 0.05 for CCL2-null vs. WT P = 0.26 for DOC vs. CON P = 0.10 for interaction

Figure 1. Markers of cardiac inflammation at 8 days. Treatment groups as follows. WT CON, untreated WT mice; WT DOC, WT mice treated with DOC; CCL2-null CON, untreated CCL2-null mice; CCL2-null DOC, CCL2-null mice treated with DOC. A, Quantitation of Mac2⫹ macrophage infiltration. B, Quantitation of CD3⫹ T cells infiltration. C, Gene expression of inflammatory cytokine TNF␣. D, Gene expression of chemokine RANTES. All data are analyzed by two-way ANOVA with results displayed below each figure. *, P ⬍ .05 by Tukey’s multiple comparison post hoc tests. Mean ⫾ SEM; n ⫽ 6 –10. CON, control.

Rho guanosine diphosphate dissociation inhibitor 2 expression, a crucial protein for macrophage superoxide production, also showed significantly lower expression in the CCL2-null mice, consistent with the macrophage infiltration profile (Supplemental Table 4). Thus, the sustained absence of macrophages at 8 weeks appears to limit the inflammatory expression profile in CCL2-null mice. Cardiac tissue macrophages regulate fibrosis but not hypertrophy in DOC/salt-mediated cardiac pathology Cardiac fibrosis at 8 days and 8 weeks No changes in interstitial fibrosis levels were detected across genotype and treatment groups at 8 days (Figure 3A) as previously described (3, 9). At 8 weeks, a significant increase in collagen content (30% increase) was detected in DOC/salt-treated WT mice compared with all other groups (Figure 4A and Supplemental Figure 3A). Consistent with these data were the increased numbers of FSP1 positive (cardiac fibroblasts) and interstitial ␣SMA posi-

tive cells (myofibroblasts) at 8 weeks (Figure 4, C and D, and Supplemental Figure 3, B and C). MMP2/MMP9-dependent remodeling at 8 days and 8 weeks MMP2/MMP9 activity is an important determinant of collagen breakdown, and it is significantly reduced by DOC treatment as compared with mice genotype (Figures 3B and 4B). Consistent with our previous study (16), significantly reduced cardiac MMP2/MMP9 activity was observed in DOC/salttreated WT but not in CCL2-null mice at 8 weeks (Figure 4B). Changes in MMP2/MMP9 activity with DOC treatment are consistent with the reduced ratio of MMP9/tissue inhibitors of metalloproteinase-1 mRNA in response to DOC at 8 days and at 8 weeks (Supplemental Table 5). These data support a role for MR regulation of MMP2 and MMP9 at 8 days and a regulatory role for macrophages at 8 weeks.

The role of CCL2 in DOC/salt-mediated cardiac remodeling at 8 days and 8 weeks Profibrotic markers CTGF, COL3, and TGF␤1 gene expression were significantly increased in WT mice by DOC/salt treatment compared with control treatment at both 8 days and 8 weeks (Figure 3C and Supplemental Table 5 and Supplemental Figure 4). Although equivalent responses were found for DOC-treated CCL2null mice at 8 weeks, COL3 expression was notably lower in CCL2-null mice at 8 days, suggesting that early tissue expression of COL3 is dependent upon macrophage signaling (Figure 3D). CTGF and TGF␤1 immunostaining was detected in the cardiomyocytes and the vessel wall and was significantly increased by DOC/salt treatment in both genotypes, illustrating active MR signaling in the vessel wall and accounting for, in part, the changes in mRNA levels reported above (Supplemental Figures 4, Supplemental Figure 5, A and B, and Supplemental Table 5). FSP1 and ␣SMA mRNA levels were also lower in CCL2-null mice vs WT (Supplemental Table 5), which is consistent with FSP1 positive and ␣SMA positive cell numbers, respectively, and are thus are more representative of the cardiac collagen levels seen at 8 weeks.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2013-1772

endo.endojournals.org

1061

DOC/salt treatment for 8 weeks in WT but not CCL2-null mice (Figure * * * * 2.5 5 5B). CCL2-null mice also displayed 2.0 4 significantly lower IL-23 and osteo1.5 3 pontin mRNA levels (Figure 5, B and 1.0 2 C). Although elevated, vascular 0.5 1 TNF␣ gene expression in WT mice 0.0 0 did not reach significance (P ⫽ .06) WT WT CCL2 CCL2 WT WT CCL2 CCL2 CON DOC -null -null (Supplemental Table 6). This is conCON DOC -null -null CON DOC CON DOC sistent with the expression of the P < 0.05 for CCL2-null vs. WT P < 0.05 for CCL2-null vs. WT macrophage gene, Mac-2, which 2-way P < 0.05 for interaction P < 0.05 for DOC vs. CON ANOVA was significantly increased (P ⬍ .05) P = 0.23 for DOC vs. CON P = 0.056 for interaction after DOC/salt treatment in WT Nox2 D CCR5 C aorta but not CCL2-null aorta (Sup* * 20 2.0 2.5 plemental Figure 7A). Gene expres* 2.0 1.5 sion for the T-cell marker, CD3, sim1.5 ilarly showed a significant increase 1.0 1.0 in WT mice given DOC/salt com0.5 pared with CCL2-null mice (Supple0.5 mental Figure 7B). Endothelial 0.0 0.0 WT WT CCL2 CCL2 WT WT CCL2 CCL2 activation markers (intercellular CON DOC -null -null CON DOC -null -null CON DOC CON DOC adhesion molecule, endothelin 1) P < 0.05 for DOC vs. CON P < 0.05 for DOC vs. CON and vascular remodeling markers 2-way P = 0.80 for CCL2-null vs. WT P = 0.09 for CCL2-null vs. WT ANOVA (CTGF, plasminogen activator inhibP < 0.05 for interaction P < 0.05 for interaction itor-1, and platelet-derived growth Figure 2. Markers of cardiac inflammation at 8 weeks. Treatment groups as per Figure 1. A, Quantitation of Mac 2 ⫹ macrophage infiltration. B, Quantitation of CD3⫹ T cells infiltration. C, factor receptor beta) were equivaGene expression of chemokine receptor CCR5. D, Gene expression of oxidative stress marker lently regulated by DOC/salt treatNox2. All data are analyzed by two-way ANOVA with results displayed below each figure. *, P ⬍ ment in WT and CCL2-null mice, .05 by Tukey’s multiple comparison post hoc tests. Mean ⫾ SEM; n ⫽ 6 –10. consistent with direct MR signaling The role of CCL2 in cardiac hypertrophy responses in the vessel wall, independent of macrophage recruitment Heart weight/body mass was used as the index of car(Supplemental Figures 8 and 9, and Supplemental Table diac hypertrophy and was significantly increased by DOC/ 7). The inflammatory gene expression profile detected in salt treatment compared with control treatment at 8 weeks the aorta highlights the synergistic role of recruited macindependent of genotype. This finding was supported by rophages and T cells in mediating vascular inflammation. B-type natriuretic peptide gene expression (a marker of In the kidney, DOC/salt treatment in WT mice significardiac dilatation and hypertrophy) at 8 weeks (Supplecantly increased macrophage markers CD68 and Mac-2 mental Figure 6). As previously described, at 8 days, there over control-treated WT, whereas a modest increase in was no significant effect of DOC (3) or genotype on carMac-2 alone was seen in CCL2-null mice given DOC/salt diac hypertrophy (Supplemental Table 2). (Supplemental Figure 10, A and B). Given that the T cells Macrophages regulate SBP, inflammation, and marker (CD3) in the kidney also showed a significant inremodeling in DOC/salt-mediated vascular crease in WT mice given DOC/salt but not in CCL2-null pathology mice (Supplemental Figure 10C), this suggests that a difference in renal inflammatory profile may also influence Systolic BP DOC/salt treatment for 8 weeks significantly elevated SBP response to DOC/salt. mRNA/18S m

B

T cells number

CD3+ T cells per frame

Mac-2 2+ cells per frame

Macrophage number

mRNA/18S m

A

SBP in the WT mice above control (15%). In contrast, no significant increase in BP was observed with DOC/salt treatment in CCL2-null mice (Figure 5A).

Discussion

DOC-mediated vascular and renal inflammatory responses depend upon CCL2 signaling Aortic expression of the macrophage-regulated proinflammatory gene, IL-23, was significantly increased with

This present study demonstrates that tissue macrophage infiltration is essential for cardiovascular inflammation, remodeling, and hypertension in a model of MR-dependent injury. Loss of CCL2-dependent macrophage recruit-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

1062

Shen et al

Macrophage and Cardiovascular Disease

Endocrinology, March 2014, 155(3):1057–1066

A key role for the macrophage in MR-dependent cardiac 110 0.8 inflammation 100 0.6 We have shown that loss of macrophage recruitment dramatically re90 0.4 duces early MR-dependent, tissue in80 0.2 flammation. MR activation in the 10 vessel wall triggers a vascular injury 0 00 0.0 WT WT CCL2 CCL2 WT WT CCL2 CCL2 response with endothelial activation CON DOC -null -null -null CON DOC -null and recruitment of inflammatory cells CON DOC CON DOC (2). Macrophages respond to MR sigP < 0.05 for DOC vs. CON P = 0.13 for DOC vs. CON naling by adopting an inflammatory 2-way 0.76 6 for o CC CCL2-null u vs. s WT CCL2-null u vs. s WT P=0 0.053 053 for o CC P=0 ANOVA phenotype, whereas deletion of the P = 0.99 for interaction P = 0.94 for interaction macrophage MR results in loss of the proinflammatory phenotype and no COL3 CTGF C D change in macrophage alternate acti* * 3 * * 2.0 vation markers (10). Consistent with these data, the present study demon1.5 2 strated an early reduction in cardiac 1.0 levels of the T cell inflammatory 1 marker, RANTES, and the macro0.5 phage-regulated inflammatory marker, 0 0.0 TNF␣, in CCL2-null mice at 8 days, WT WT CCL2 CCL2 WT WT CCL2 CCL2 CON DOC -null -null CON DOC -null -null whichmayinpartreflectthelossofmacCON DOC CON DOC rophage recruitment. At 8 weeks, MRdependent Nox2 and osteopontin exP < 0.05 for DOC vs. CON 2-way P < 0.05 for DOC vs. CON P = 0.54 for CCL2-null vs. WT P < 0.05 for CCL2-null vs. WT ANOVA pression remained significantly lower in P = 0.096 for interaction P < 0.05 for interaction CCL2-null mice. We also detected a sigFigure 3. Cardiac fibrosis and markers of tissue remodeling in the heart at 8 days. Treatment nificant increase in CCL2 expression groups as per Figure 1. A, Cardiac fibrosis at 8 days. B, MMP2/MMP9 activity in whole-heart with DOC/salt in WT hearts. Our data tissue. C, Gene expression of tissue remodeling marker CTGF. D, Gene expression of perivascular tissue remodeling marker COL3. All data are analyzed by two-way ANOVA with results displayed thus demonstrate a central role for probelow each figure. *, P ⬍ .05 by Tukey’s multiple comparison post hoc tests. Mean ⫾ SEM; n ⫽ inflammatory macrophages and for 6 –10. CCL2-dependent cardiac inflammation in DOC/salt cardiac fibrosis (10). The role for T cells in the DOC/ ment reduces the tissue inflammatory response in the ⫹ DOC/salt model (5, 9) from 8 days and fibrosis and hy- salt model has not been fully elucidated, although CD4 pertension at 8 weeks. This study also supports a key role T helper 2 cells have been implicated in other fibrotic disfor macrophage MR signaling in tissue inflammation and ease models (18). The greater increase in T cell infiltrate in remodeling. Although not focusing on a specific change in WT mice with 8-week exposure to DOC/salt may reflect macrophage phenotype, our current study directly dem- an adaptive immune response and suggests a chronic pathogenic response as seen in other inflammatory cononstrates the central role for CCL2-directed macrophages ditions. Given that T cells also express CCR2, these data in both the inflammatory component and the pathogenic may reflect a role for CCL2 and/or tissue macrophages in remodeling process in response to mineralocorticoid treatT cell recruitment (19). The lack of a macrophage-mediment, which translates into cardiac fibrosis and hypertenated inflammatory response in the CCL2-null mice would sion (Supplemental Figure 11). Previously, we have shown be expected to lead to a fall in T cell chemokine production that macrophage dysfunction associated with deletion of and thus T cell recruitment (20). That we found a promthe macrophage MR attenuates the response to DOC/salt inent reduction in expression of the T cell chemokine treatment despite retention of macrophage recruitment RANTES supports a role for CCL2 and/or inflammatory (9). This study demonstrates the central role of macro- macrophages in promoting T cell infiltration in the tissue phage recruitment per se with the DOC/salt model despite inflammation response. A potential limitation of the study the macrophages having intact MR signaling. is that the WT mice were not littermate controls of the

B

MMP2/MMP9 activity

% activity

Fibrosis

mRNA/18S

mRNA/18S

% area cardiac collagen

A

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2013-1772

endo.endojournals.org

1063

pathology of cardiac tissue remodeling is a change in the tissue MMP * * * (MMP2/MMP9) activity that regu0.8 100 lates collagen degradation and turn0.6 90 over. An increase in cardiac MMP 0.4 activity was regarded as likely to pro80 0.2 10 mote collagen degradation in the 0 0.0 myocardium (22, 23). The reduced CCL2 CCL2 WT WT CCL2 CCL2 WT WT MMP2/MMP9 activity in the DOC/ CON DOC -null -null CON DOC -null -null CON DOC CON DOC salt-treated WT mice favors increased collagen deposition, and this 2-way P < 0.05 for CCL2-null vs. WT P < 0.05 for DOC vs. CON change in MMP2/MMP9 activity ANOVA P = 0.18 for DOC vs. CON P = 0.43 for CCL2-null vs. WT P < 0.05 for interaction P = 0.11 for interaction was absent in the DOC-treated CCL2-null mice. In some studies, however, aldosterone was shown to FSP1 positive cells -SMA positive cells C D 5 increase tissue MMP2/MMP9 activ0.4 * * * ity (17). How tissue macrophages 4 0.3 regulate net tissue MMP activity is 3 not clear. We postulate that macro0.2 2 phage-induced inflammation is 0.1 1 likely to play a role in the regulation of both tissue MMP and tissue inhib0 0.0 WT WT CCL2 CCL2 WT WT CCL2 CCL2 itors of metalloproteinase-1 expresCON DOC null null -null -null CON DOC -null -null CON DOC sion and activity. CON DOC Reduced tissue fibrosis was assoP < 0.05 for CCL2-null vs. WT P < 0.05 for CCL2-null vs. WT 2-way ciated with fewer fibroblast and P < 0.05 for DOC vs. CON P = 0.40 for DOC vs. CON ANOVA P = 0.07 for interaction P = 0.72 for interaction myofibroblast markers in CCL2null mice at 8 weeks. These data are Figure 4. Cardiac fibrosis and mechanism behind cardiac remodeling at 8 weeks. Treatment groups as per Figure 1. A, Cardiac fibrosis at 8 weeks. B, MMP2/MMP9 activity in whole-heart consistent with the hypothesis that tissue. C, Quantitation of cardiac fibroblasts, FSP1⫹ cells. D, Quantitation of cardiac recruited macrophages may promyofibroblasts, ␣SMA⫹ cells. All data are analyzed by two-way ANOVA with results displayed mote myofibroblast activation (24). below each figure. *, P ⬍ .05 by Tukey’s multiple comparison post hoc tests. Mean ⫾ SEM; n ⫽ 6 –10. Bone marrow-derived fibroblasts of myeloid lineage may underlie the reduced cardiac fibrosis (25). HowCCL2-null mice and thus may exhibit a minor strain-specific difference in baseline parameters. However, any mi- ever, full characterization of fibrocytes in the myocardium nor differences between the control groups do not signif- of DOC/salt-treated mice has not been performed. In conicantly affect the core results and the conclusion of the trast, expression of some profibrotic markers in whole study that the loss of CCL2 expression largely prevents the heart (CTGF, TGF␤1, and COL3) was not dependent on macrophage infiltration but, rather, MR-dependent actiresponse to DOC/salt administration. vation in the vasculature, as indicated by immunostaining. Macrophages regulate cellular mechanisms These findings support our previous work investigating underlying MR-dependent cardiac remodeling DOC/salt responses in mice null for cardiomyocyte MR There is good evidence that the alternatively activated (MyoMRKO), where cardiac macrophage recruitment macrophage phenotype is directly involved in the tissue did not occur in response to DOC/salt (16) but vascular remodeling and fibrosis seen in other disease models. The MR responses were retained. late adverse remodeling process seen predominantly at 8 weeks in our study is consistent with published literature MR-mediated cardiac hypertrophy is independent (2, 4) and has previously been demonstrated to be inde- of tissue macrophage infiltration and pendent of BP changes (21). In our study, the lack of re- inflammation Cardiac hypertrophy is independently induced by cruited macrophages in the heart likely mitigates the amount of cardiac fibrosis in the CCL2-null mice at 8 DOC/salt treatment, notably late at 8 weeks, supporting a weeks. One mechanism that underlies the MR-dependent trophic effect of MR activation on cardiomyocytes (26).

A

Fibrosis

B % activity

Cells per frame

% area a cardiac collagen

Cells pe er frame

MMP2/MMP9 activity

110

1.0

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

1064

Shen et al

Macrophage and Cardiovascular Disease

Endocrinology, March 2014, 155(3):1057–1066

Figure 5. SBP and markers of vascular inflammation in the aorta at 8 weeks. Treatment groups as per Figure 1. A, SBP at 8 weeks. B, Gene expression of macrophage cytokine IL-23. C, Gene expression of inflammatory marker osteopontin. mRNA levels are normalized to glyceraldehyde 3-phosphate dehydrogenase. All data are analyzed by two-way ANOVA with results displayed below each figure. *, P ⬍ .05 by Tukey’s multiple comparison post hoc tests. Mean ⫾ SEM; n ⫽ 6 –10.

This is consistent with our previous study, which had also indicated this to be independent of macrophage or DOC/ salt-induced tissue inflammation/remodeling pathways (9). That the SBP response to DOC/salt treatment was markedly suppressed in the CCL2-null further highlights the direct, BP-independent, DOC/salt-mediated cardiac hypertrophy response (27). The role of macrophages and CCL2 in regulation of MR-dependent SBP We observed a significant increase in SBP in WT but not CCL2-null mice given DOC/salt treatment for 8 weeks. An important role for macrophages in regulating Ang II-dependent vascular inflammation and BP had been described (28). We have previously shown an essential role for macrophage MR signaling in BP regulation (9). In contrast, BP elevation after nitric oxide depletion with NG-nitro-l-arginine methyl ester and/or Ang II was observed to be less dependent on macrophage MR status (8, 10). In this present study, the lack of vessel wall macrophages and T cells may have protected the CCL2-null mice from an increase in SBP compared with WT mice. The underlying mechanism may be attributed to the remarkable difference in the gene expression of vascular inflammatory markers such as IL-23 and osteopontin between WT and CCL2-null mice, and between DOC/salt-treated WT and control-treated WT mice. IL-23 is a proinflammatory macrophage-derived cytokine (29), known for its role in activating T helper 17 cells (30), which are responsible for IL-17-mediated vascular inflammation. Synergistic actions of IL-17 and TNF␣ have been implicated in the pathogenesis of Ang II and DOC/salt-induced vascular hypertension (31, 32). Although the vascular TNF␣ response to DOC/salt treatment in our study is only modest, in conjunction with

IL-23 and osteopontin, it suggests a supportive role for macrophage-induced inflammatory cytokines in T cell and MR-mediated vascular inflammation and hypertension. Therefore, we postulate that macrophages play a synergistic role with T cells in MR-dependent BP regulation as reviewed recently (33). Increased endothelial cell activation and vascular remodeling by vascular smooth muscle cells are other mechanisms whereby DOC/salt treatment can regulate total peripheral resistance and thus BP. MR-mediated endothelial activation with induction of CCL2 has been recently shown in vivo, and a direct role for vascular smooth muscle cell MR signaling in hypertension has also recently been demonstrated (34, 35). Our findings demonstrate that vascular macrophage and T cell infiltration is required and may be synergistic with endothelial cell and vascular smooth muscle cell MR signaling in BP regulation. An important consideration is the limitation associated with the tail-cuff BP recordings in terms of intrinsic error. Although in our hands BP recordings are very reproducible with minimal variability, BP recordings by telemetry will provide definitive measurements as well as parameters that will offer valuable insights into the regulation of BP. CCR2 antagonist treatment has recently been shown to reduce both SBP in DOC/salt-treated mice as well as aortic macrophage numbers (36). Our study supports a specific role for CCL2, one of several ligands for CCR2, in effective recruitment of tissue macrophages and hence the control of BP in DOC/salt-mediated hypertension. The kidneys play a central role in the regulation of BP through salt and water retention. In this model, renal counterregulation is compromised by the uninephrectomy. Although the renal inflammatory profile could be a factor in the differ-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

doi: 10.1210/en.2013-1772

ential regulation of BP, any conclusion of cause and effect is limited by the uninephrectomy, which is fundamental to the model. Perspectives Although current MR antagonists have been shown to confer significant reductions in mortality and morbidity in large-scale clinical trials involving patients with heart failure, a significant incidence of hyperkalemia from the use of these MR antagonists has limited their widespread use (37). This has stimulated a search to understand the cellular and molecular basis of the cardiovascular benefits seen with MR antagonist therapy. In conjunction with our previous studies involving selective deletion in macrophage of the MR, this present study demonstrates an essential role for CCL2-dependent macrophage recruitment in promoting MR-mediated cardiovascular inflammation, remodeling, and hypertension. These data further suggest a role for targeting macrophage function as a therapeutic strategy in heart failure patients, particularly those with elevated plasma aldosterone levels. Therapeutic immunomodulation by a selective macrophage MR antagonist may confer a safer and more realistic approach than macrophage depletion therapy in patients with chronic heart failure. The role for T cells in MR-mediated cardiac pathology and the exact nature of interaction between macrophage, T cells, and other cell types, such as fibroblasts and cardiomyocytes in the heart and vascular smooth muscle cells in vessels, remains to be clearly defined. Further interrogation of each key player in a systematic manner is required and holds promise of valuable insights and the potential for identification of novel cardioprotective therapeutic targets.

Acknowledgments Address all correspondence and requests for reprints to: Dr Morag J. Young, Prince Henry’s Institute of Medical Research. P.O. Box 5152, Clayton 3168, Australia. E-mail: [email protected]. This work was supported by National Health and Medical Research Council of Australia Project Grant 1010034 and Fellowship 1002559 (to P.J.F.). P.J.F. is also supported by the Victorian Government’s Operational Infrastructure Support Program. J.Z.S. is supported by an Royal Australasian College of Physicians Clinical Research award and an Australian Postgraduate Award. Disclosure Summary: The authors have nothing to disclose.

endo.endojournals.org

2.

3.

4.

5.

6.

7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

References 21. 1. Milliez P, Girerd X, Plouin PF, Blacher J, Safar ME, Mourad JJ. Evidence for an increased rate of cardiovascular events in patients

1065

with primary aldosteronism. J Am Coll Cardiol. 2005;45:1243– 1248. Rocha R, Rudolph AE, Frierdich GE, et al. Aldosterone induces a vascular inflammatory phenotype in the rat heart. Am J Physiol Heart Circ Physiol. 2002;283:H1802–H1810. Wilson P, Morgan J, Funder JW, Fuller PJ, Young MJ. Mediators of mineralocorticoid receptor-induced profibrotic inflammatory responses in the heart. Clin Sci. 2009;116:731–739. Robert V, Silvestre JS, Charlemagne D, et al. Biological determinants of aldosterone-induced cardiac fibrosis in rats. Hypertension. 1995; 26:971–978. Sun Y, Zhang J, Lu L, Chen SS, Quinn MT, Weber KT. Aldosteroneinduced inflammation in the rat heart : role of oxidative stress. Am J Pathol. 2002;161:1773–1781. Young M, Funder JW. Eplerenone, but not steroid withdrawal, reverses cardiac fibrosis in deoxycorticosterone/salt-treated rats. Endocrinology. 2004;145:3153–3157. Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol. 2011;11:723–737. Usher MG, Duan SZ, Ivaschenko CY, et al. Myeloid mineralocorticoid receptor controls macrophage polarization and cardiovascular hypertrophy and remodeling in mice. J Clin Invest. 2010;120: 3350 –3364. Rickard AJ, Morgan J, Tesch G, Funder JW, Fuller PJ, Young MJ. Deletion of mineralocorticoid receptors from macrophages protects against deoxycorticosterone/salt-induced cardiac fibrosis and increased blood pressure. Hypertension. 2009;54:537–543. Bienvenu LA, Morgan J, Rickard AJ, et al. Macrophage mineralocorticoid receptor signaling plays a key role in aldosterone-independent cardiac fibrosis. Endocrinology. 2012;153:3416 –3425. Tesch GH, Schwarting A, Kinoshita K, Lan HY, Rollins BJ, Kelley VR. Monocyte chemoattractant protein-1 promotes macrophagemediated tubular injury, but not glomerular injury, in nephrotoxic serum nephritis. J Clin Invest. 1999;103:73– 80. Dewald O, Zymek P, Winkelmann K, et al. CCL2/monocyte chemoattractant protein-1 regulates inflammatory responses critical to healing myocardial infarcts. Circ Res. 2005;96:881– 889. Lu B, Rutledge BJ, Gu L, et al. Abnormalities in monocyte recruitment and cytokine expression in monocyte chemoattractant protein 1-deficient mice. J Exp Med. 1998;187:601– 608. Palframan RT, Jung S, Cheng G, et al. Inflammatory chemokine transport and presentation in HEV: a remote control mechanism for monocyte recruitment to lymph nodes in inflamed tissues. J Exp Med. 2001;194:1361–1373. Lumeng CN, Bodzin JL, Saltiel AR. Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest. 2007;117:175–184. Rickard AJ, Morgan J, Bienvenu LA, et al. Cardiomyocyte mineralocorticoid receptors are essential for deoxycorticosterone/salt-mediated inflammation and cardiac fibrosis. Hypertension. 2012;60: 1443–1450. Johar S, Cave AC, Narayanapanicker A, Grieve DJ, Shah AM. Aldosterone mediates angiotensin II-induced interstitial cardiac fibrosis via a Nox2-containing NADPH oxidase. FASEB J. 2006;20: 1546 –1548. Barron L, Wynn TA. Fibrosis is regulated by Th2 and Th17 responses and by dynamic interactions between fibroblasts and macrophages. Am J Physiol Gastrointest Liver Physiol. 2011;300: G723–G728. Deshmane SL, Kremlev S, Amini S, Sawaya BE. Monocyte chemoattractant protein-1 (MCP-1): an overview. J Interferon Cytokine Res. 2009;29:313–326. Mosser DM, Edwards JP. Exploring the full spectrum of macrophage activation. Nat Rev Immunol. 2008;8:958 –969. Young M, Head G, Funder J. Determinants of cardiac fibrosis in experimental hypermineralocorticoid states. Am J Physiol. 1995; 269:E657–E662.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

1066

Shen et al

Macrophage and Cardiovascular Disease

22. Li YY, McTiernan CF, Feldman AM. Proinflammatory cytokines regulate tissue inhibitors of metalloproteinases and disintegrin metalloproteinase in cardiac cells. Cardiovasc Res. 1999;42:162–172. 23. Siwik DA, Chang DL, Colucci WS. Interleukin-1␤ and tumor necrosis factor-␣ decrease collagen synthesis and increase matrix metalloproteinase activity in cardiac fibroblasts in vitro. Circ Res. 2000; 86:1259 –1265. 24. Porter KE, Turner NA. Cardiac fibroblasts: at the heart of myocardial remodeling. Pharmacol Ther. 2009;123:255–278. 25. Xu J, Lin SC, Chen J, et al. CCR2 mediates the uptake of bone marrow-derived fibroblast precursors in angiotensin II-induced cardiac fibrosis. Am J Physiol Heart Circ Physiol. 2011;301:H538 – H547. 26. Yamamuro M, Yoshimura M, Nakayama M, et al. Direct effects of aldosterone on cardiomyocytes in the presence of normal and elevated extracellular sodium. Endocrinology. 2006;147:1314 –1321. 27. Young MJ, Moussa L, Dilley R, Funder JW. Early inflammatory responses in experimental cardiac hypertrophy and fibrosis: effects of 11 ␤-hydroxysteroid dehydrogenase inactivation. Endocrinology. 2003;144:1121–1125. 28. Wenzel P, Knorr M, Kossmann S, et al. Lysozyme M-positive monocytes mediate angiotensin II-induced arterial hypertension and vascular dysfunction. Circulation. 2011;124:1370 –1381. 29. Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest. 2012;122:787–795.

Endocrinology, March 2014, 155(3):1057–1066

30. Zhou L, Ivanov II, Spolski R, et al. IL-6 programs T(H)-17 cell differentiation by promoting sequential engagement of the IL-21 and IL-23 pathways. Nat Immunol. 2007;8:967–974. 31. Guzik TJ, Hoch NE, Brown KA, et al. Role of the T cell in the genesis of angiotensin II induced hypertension and vascular dysfunction. J Exp Med. 2007;204:2449 –2460. 32. Madhur MS, Lob HE, McCann LA, et al. Interleukin 17 promotes angiotensin II-induced hypertension and vascular dysfunction. Hypertension. 2010;55:500 –507. 33. Shen JZ, Young MJ. Corticosteroids, heart failure, and hypertension: a role for immune cells? Endocrinology. 2012;153:5692– 5700. 34. Hashikabe Y, Suzuki K, Jojima T, Uchida K, Hattori Y. Aldosterone impairs vascular endothelial cell function. J Cardiovasc Pharmacol. 2006;47:609 – 613. 35. McCurley A, Pires PW, Bender SB, et al. Direct regulation of blood pressure by smooth muscle cell mineralocorticoid receptors. Nat Med. 2012;18:1429 –1433. 36. Chan CT, Moore JP, Budzyn K, et al. Reversal of vascular macrophage accumulation and hypertension by a CCR2 antagonist in deoxycorticosterone/salt-treated mice. Hypertension. 2012;60: 1207–1212. 37. Juurlink DN, Mamdani MM, Lee DS, Kopp A, Austin PC, Laupacis A, Redelmeier DA. Rates of hyperkalemia after publication of the Randomized Aldactone Evaluation Study. N Engl J Med. 2004;351:543–551.

Need Part IVMOC Points? Learn more about the Endocrine Society’s Practice Improvement Modules (PIMs). www.endoselfassessment.org

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2014. at 13:32 For personal use only. No other uses without permission. . All rights reserved.

CCL2-dependent macrophage recruitment is critical for mineralocorticoid receptor-mediated cardiac fibrosis, inflammation, and blood pressure responses in male mice.

Recent studies show that mice with selective deletion of the mineralocorticoid receptor (MR) in macrophages are protected from mineralocorticoid-induc...
385KB Sizes 0 Downloads 0 Views