Interleukin-1 induces pro-inflammatory leukocyte infiltration and regulates fibroblast phenotype in the infarcted myocardium ^†

Animals

C57BL/6 and IL-1R1 −/− (16) mice in a C57BL/6 background were purchased from Jackson Laboratories (Bar Harbor, ME). All protocols were approved by the committee on animal research care at Albert Einstein College of Medicine.

Murine Model of Reperfused Myocardial Infarction

Female and male mice (2-3 months of age) were anesthetized by isoflurane inhalation (isoflurane 2-3% vol/vol). Myocardial infarction was induced using a closed-chest mouse model of reperfused myocardial infarction, as previously described (6). All animals underwent 1h coronary occlusion; after 24h or 7 day of reperfusion, the chest was opened and the heart was immediately excised and processed for either flow cytometric experiments, fibroblast culture experiments or for harvesting of fibroblasts and CD11b+ myeloid cells. Sham animals were prepared identically without undergoing coronary occlusion/reperfusion. Mice used to harvest CD11b+ cells and fibroblasts for RNA extraction underwent 24h, and 7 days of reperfusion (WT, n = 7; KO, n = 8 per group). Wild-type mice used for flow cytometric analysis underwent reperfusion at 24h (N=7) and day 7 (N=7) in addition to uninjured controls (N=10). Flow cytometric studies for the quantitatively assessment of the cellular infiltrate at Day 7 in WT and IL-1R1 null reperfused infracted hearts included (N=5) in each group.

RNA Extraction and Quantitative Polymerase Chain Reaction

Total RNA isolated from stimulated cardiac fibroblasts and from CD11b+ myeloid and CD11b-negative nonmyeloid cells was reverse transcribed to cDNA using iScript cDNA synthesis kit (Bio-Rad) following the manufacturer’s guidelines. Quantitative PCR was performed using the SYBR green (Bio-Rad) method on the iQ5 Real-Time PCR Detection System (Bio-Rad) for 40 cycles at an annealing temperature (TΔOpt °C) specific for a primer pair as generated Premier Biosoft. Primers were synthesized using Beacon designer from Premier Biosoft (Version 8.02, Palo Alto CA). Each sample was run in duplicate. The CT method using GAPDH or 18S as indicated as the reference gene was used for relative quantification of expression of various genes studied in this study. The following sets of primers were used in the study:

IL-1R1 (forward) CAGGATTCATCAGCAGAA, (reverse) GAGGCAGTAAGTTGAGTT; IL-1R2 (forward) GGCTATTACAGATGTGTTATGAC, (reverse) GTTCCGTGGTTGTTCCTT; Matrix metalloproteinase (MMP)-3 (forward) GGAAATCAGTTCTGGGCTATACGA, (reverse) TAGAAATGGCAGCATCGATCTTC; MMP-8 (forward) GAAGGCAGGAGAGGTTGT, (reverse) TGGAGGAAGATCAGTAATGGAA; Collagen 1 (forward) AAGAAGACATCCCTGAAG , (reverse) ATACAGATCAAGCATACCT; Collagen 3 (forward) TTGCGATGACATAATCTG, (reverse) GCACAACATTCTCCAAAT; α-smooth muscle actin (α-SMA) (forward) GAGTAATGGTTGGAATGG , (reverse) TGTTCTATCGGATACTTCA; MCP-1 (forward) AAGTTGACCCGTAAATCT, (reverse) CTAGTTCACTGTCACACT; IL-10 (forward) TTGCTATGGTGTCCTTTC, (reverse) ATCTCCCTGGTTTCTCTT; IL-6 (forward) GACCATCCAATTCATCTT, (reverse) CATTCATATTGTCAGTTCTT; IL-1β (forward) CAAAGAAGAAGATGGAAA, (reverse) ATGGTGAAGTCAATTATG; GAPDH (forward) AACGACCCCTTCATTGACCT , (reverse) CACCAGTAGACTCCACGACA; 18S (forward) GGCTCATTAAATCAGTTATG , (reverse) GCTCTAGAATTACCACAG; Activin receptor-like kinase (ALK)5 (forward) ACGAATGACATACTACCTA, (reverse) CCTGATTATAGCACAAGAG; BAMBI (forward) GTCTGAACTATAAATGGAACC, (reverse) TCTCCAAGATGACTAAGC; TNFα (forward) GTCTGTATCCTTCTAACTTA, (reverse) TCTTGTGTTTCTGAGTAG; Endoglin-L (forward) CGTAATGATGGAACTGAG, (reverse) ATGAACCAGGAGACATAT, Endoglin-S (forward) CGTAATGATGGAACTGAG, (reverse) ATGAACCAGGAGACATAT, periostin (forward) TGCCAACAGTTACTATGA, (reverse) CAGCATTCATATAGCACAG.

Flow cytometry in cells harvested from control and infarcted hearts

Single cell suspensions were obtained from infarcted WT and IL-1R1 null hearts as previously described (17). Cells harvested from the infarcted heart were counted and reconstituted in staining buffer (BD Biosciences) to a concentration of 1×10⁶cells/ml. Subsequently cells were incubated with LIVE/DEAD® Fixable Dead Cell Stain single-color dyes (Invitrogen) for 30min at room temperature to evaluate the viability. After one rinse with washing buffer, cells were incubated with anti- FcγIII/II (clone 2.4G2) antibody (BD Pharmingen) for 15 minutes and labeled at 4°C for 30 minutes simultaneously with following Rat anti-mouse antibodies purchased from Biolegend: Percp-labeled anti-CD45, FITC-labeled anti-CD11b, PE-Cy7-labeled-anti-F4/80, , APC labeled-anti-IL-1R1 and Pacific blue-labeled anti-CD19. APC-Cy7-labeled anti-Ly6C was from (BD Pharmingen). For intracellular staining, cells were fixed and permeabilized for 20 minutes at 4°C with fixation/permabilization kit (eBioscience). Subsequently, cells were incubated with PE-labeled anti-IL1β (BD Pharmingen). Finally, cells were washed twice, resuspended in staining buffer, and immediately analyzed with a Becton Dickinson LSRII flow cytometer (BD biosciences). Cell analysis using flow cytometry included analysis of only singlet cells; all doublets were excluded from the analysis by monitoring the SSC/FSC pulse width channel.

Monocytes/macrophages were defined as Live/deadlowCD19negCD45highCD11bhigh cells. Within this population, subsets were identified as either F4/80+ or F4/80-. Further, each F4/80 subgroup was identified as either Ly-6Chi or Ly-6Clo. The absolute number of cells in each subset was calculated by multiplying cell number by percent of cells in the subset, which was calibrated by heart weight and expressed as cells/mg. Data analysis was performed using FlowJo (Tree Star, Inc.).

Isolation of CD11b+ myeloid cells and CD11b-negative non-myeloid cells from infarcted hearts

CD11b+ myeloid cells and CD11b-negative fibroblasts were isolated from control and infarcted hearts for RNA extraction. Briefly, infarct tissue (1h ischemia followed by 24h, or 7 days of reperfusion), or healthy hearts were minced, and placed into a cocktail of 0.25 mg/ml Liberase Blendzyme 3 (Roche Applied Science), 20 U/ml DNase I (Sigma Aldrich), 10 mmol/L HEPES (Invitrogen), 0.1% Sodium Azide in HBSS with Ca2+ and Mg2+ (Invitrogen) and shaken at 37°C for 40 minutes. Subsequently cells were passed through 40μm nylon mesh and centrifuged (10 min, 200 g, 4°C). Finally, cells were reconstituted with staining buffer (d-PBS without Ca2+ and Mg2+, 2% FBS, 0.1% sodium azide) and total cell numbers were determined with trypan blue (Mediatech, Inc.). The resulting single-cell suspensions were washed with HBSS supplemented with 0.2% (wt/vol) BSA and 1% (wt/vol) FCS and centrifuged. Single cells were resuspended in a special MACS buffer containing phosphate buffered saline (PBS), pH 7.2, 0.5% BSA and 2 mM EDTA, incubated with CD11b+ microbeads (Miltenyi Biotec) 10μl/10⁷ cells at 4°C for 15min, then washed once and centrifuged. Resuspended cells went through a MACS Column (Miltenyi Biotec) set in a MACS Seperator (Miltenyi Biotec). Unlabeled cells which passed through were collected and washed once with PBS as the CD11b-negative fraction. These CD11b-negative cells were harvested for further experiments. The magnetically labeled CD11b+ cells were retained on the column. Approximately 2 ml of MACS buffer was applied onto the column. Cells were flushed out by firmly pushing the plunger and collected into a tube.

Protein extraction and Western blotting

In order to assess activation of p-Smad2:Smad2 and pErk1/2, fibroblasts harvested from WT and IL-1R1 null hearts underwent stimulation with IL-1β (10 ng/ml) or LPS (25 ng/ml) for 15-60 min. At the end of the experiment cell lysates were used for protein extraction. Western blotting was performed as previously described using antibodies to p-Smad2, Smad2/3 (Cell Signaling) and GAPDH. The ratio of p-Smad2 : Smad2 expression was assessed.

Assessment of contraction of collagen pads populated with cardiac fibroblasts

To assess the contractile properties of cardiac myofibroblasts collagen pads populated with cardiac fibroblasts were used as previously described (18). Following polymerization, fibroblast-populated collagen pads were incubated in triplicates (WT; n=9) or duplicates (IL-1R1−/−; n=6) of 3 individual pads in each 6-well culture plate (BD Falcon, San Jose, CA) with the following conditions – 0% FCS DMEM (1x), 15 ng/ml IL-1β, and TGF-β (10 ng/ml) in the presence or absence of 15 ng/ml IL-1β. After 24 h the pictures of the plates were taken in flatbed scanner and the area of each pad was measured using Image Pro software. To study expression of MMPs and TGF-β receptors in fibroblasts populating the collagen pads, RNA extraction was performed at the end of the experiment.

Assessment of myofibroblast transdifferentiation and matrix protein synthesis by isolated cardiac fibroblasts

Cardiac fibroblasts from WT and IL-1R1 −/− hearts were isolated by enzymatic digestion with a collagenase buffer as previously described (18), (19). Cells at passage 3 were seeded in chambers of Culture Slides (BD Falcon) and allowed to attach overnight. Cells were serum-starved for 24h and subsequently stimulated with 25 ng/ml of TGF-β1 (R&D Systems) in the presence or absence of IL-1β (15ng/ml) for 24 or 72 hours. Cells were harvested and total RNA extracted using TRIzol (Invitrogen). Additional cells were fixed for 10 min in a 3.7% solution of paraformaldehyde (Sigma), then washed twice with PBS and stained for α-SMA. qPCR was performed to measure expression of collagen I, periostin and α-SMA. Immunofluorescence was performed after permeabilization with 0.1% Triton-X (Sigma) in PBS. Subsequently, slides were stained with FITC-labeled anti-α-SMA antibody (Sigma). Finally to visualize the polymerized actin cytoskeleton, slides were incubated in 0.5 μM solution of Alexa Fluor 594-conjugated phalloidin (Molecular Probes) in PBS for 20 min. Nuclei were stained with DAPI (Invitrogen).

Statistical analysis

Data are expressed as mean ± SEM. Statistical analysis was performed using GraphPad version 5 (GraphPad Software Inc., LaJolla, CA, USA) using unpaired, 2-tailed Student’s t test using Welch’s correction for unequal variances and 1-way ANOVA with Tukey’s multiple comparison test. Results were considered statistically significant at P<0.05.

1. Sequential infiltration of the infarcted myocardium with monocyte subpopulations

In order to study the time course of infiltration of the infarct with monocyte subpopulations we performed flow cytometric analysis of cells harvested from the infarcted mouse myocardium. Reperfused myocardial infarction had intense infiltration with CD45+ hematopoietic cells (Figure 1A) that peaked after 24h of reperfusion (Figure 1A-B). The number of CD45+/CD11b+ myeloid cells (Figure 1C) and of CD45+/F4/80+ macrophages (Figure 1D) also peaked after 24h of reperfusion showing a 20-fold increase in comparison to control myocardium. Ly6C identifies a “pro-inflammatory” monocyte phenotype (Ly6C(hi)) that infiltrates the inflamed tissue during the early stages after injury; in contrast, Ly6C(lo) cells have reduced inflammatory activity and exhibit reparative properties (20). In the reperfused mouse infarct abundant CD11b+/Ly6C(hi) pro-inflammatory monocytes were recruited in the infarcted myocardium during the peak of the inflammatory phase (24h of reperfusion - Figure 1A-E). Subsequently, the number of Ly6C(hi) cells decreased significantly after 3-7 days of reperfusion, reflecting resolution of the inflammatory infiltrate. In contrast, the number of CD11b+/Ly6C(lo) cells increased markedly after 7 days of reperfusion (Figure 1F), reflecting late infiltration of the infarct with reparative monocytes.

2. IL-1 receptor expression profile in cells infiltrating the infarcted myocardium

Changes in relative expression of the signaling IL-1 receptor, IL-1R1, and the decoy receptor, IL-1R2, may determine responsiveness to the cytokine. Thus, transition from the inflammatory to the resolution phase of the post-infarction reparative response may be characterized with dynamic alterations in IL-1R1 and IL-1R2 expression by inflammatory leukocytes. We used flow cytometry to study the IL-1R expression profile in cells harvested from reperfused infarcted hearts (Figure 2A). The number of IL-1R1+ cells in the infarcted myocardium peaked after 24h of reperfusion. In comparison to control hearts, the infarcted myocardium showed marked, but transient, increases in the number of IL-1R1-expressing CD45+ hematopoietic cells (130-fold), IL-1R1+ CD11b+ myeloid cells (42-fold) and IL-1R1+ F4/80+ macrophages (158-fold) (Figure 2B-E) after 24h of reperfusion. After 7 days of reperfusion the number of IL-1R1+ hematopoietic cells decreased significantly suggesting either clearance of IL-1R1-expressing leukocytes or suppressed expression of the signaling receptor (Figure 2B-E).

In contrast, much like Ly6C(lo) monocytes, IL-1R2+/CD45+ hematopoietic cells appeared late in the infarcted myocardium (Figure 2A, F-I). After 7 days of reperfusion infarcted hearts exhibited a 36-fold increase in the number of IL-1R2+/CD11b+ myeloid cells and a 45-fold increase in the number of IL-1R2-expressing F4/80 macrophages (Figure 2F-I).

3. IL-1R expression profile in Ly6Chi and Ly6Clo cells

Because the time course of infiltration with pro-inflammatory Ly6C(hi) monocytes resembled the time course of recruitment of IL-1R1+ cells, we asked whether the pro-inflammatory phenotype of Ly6C(hi) cells may be due to increased expression of IL-1R1, thus reflecting increased responsiveness to the inflammatory actions of IL-1. Ly6C(lo) cells, on the other hand may exhibit attenuated pro-inflammatory signaling because of expression of the IL-1R2 decoy receptor. Using flow cytometry, we examined the IL-1R expression profile in Ly6C(hi) and Ly6C(lo) cells harvested from the infarcted mouse heart. IL-1R1 and IL-1R2 positive cells were found in both Ly6C(hi) and Ly6C(lo) subpopulations. The number of IL-1R1+/Ly6C(hi) cells peaked during the early inflammatory phase of infarct healing (after 24h of reperfusion) and was significantly reduced during the proliferative phase (after 7 days of reperfusion) (Table I, Figure 3A). In contrast, the number of IL-1R1+/Ly6Clo cells was very low (Table I, Figure 3B). Cells expressing the decoy receptor IL-1R2 were more abundant during the reparative phase (after 7 days of reperfusion) (Table I, Figure 3C-D). At all timepoints examined, the majority of IL-1R1+ hematopoietic cells were Ly6C(hi) (control: 54.1%+12, 24h: 58.7%+6.7, 7 days: 60.7%+7.9); whereas a significantly lower percentage of IL-1R1+ cells were identified as Ly6C(lo) cells (control: 30.1%+7.1, 24h: 1.7%+0.5, 7 days: 32%+10). Relative expression of IL-1Rs by Ly6C(lo) and Ly6C(hi) subsets was dependent on the stage of healing as evidenced by dynamic changes in the ratio of IL-1R2+ to IL-1R1+ cells within the Ly6C(hi) and Ly6C(lo) subpopulations (Table I). During the inflammatory phase of infarct healing IL-1R1+/Ly6Chi cells outnumbered IL-1R2+/Ly6C(hi) cells by 15:1, whereas after 7 days of reperfusion comparable numbers of IL-1R1+ and IL-1R2+ Ly6C(hi) cells were identified. In the Ly6C(lo) subpopulation the ratio of IL-1R1+:IL-1R2+ cells was also significantly lower after 7 days of reperfusion reflecting increased expression of the decoy receptor by Ly6C(lo) cells in the late phases of infarct healing (Table I).

4. IL-1R1 signaling mediates inflammatory leukocyte recruitment in the infarcted myocardium

We have previously demonstrated that IL-1R1 null mice exhibit attenuated adverse remodeling following myocardial infarction, associated with reduced myocardial expression of inflammatory cytokines and chemokines (7). We hypothesized that attenuated inflammatory activity in IL-1R1 KO animals may be due to decreased recruitment of pro-inflammatory leukocytes. Accordingly, we quantitatively assessed the cellular infiltrate in WT and IL-1R1 null hearts using flow cytometry. When compared with WT animals, IL-1R1 null mice exhibited a marked reduction in the total number of infarct cells after 7 days of reperfusion (Table II). Decreased cellular infiltration in the infarcted myocardium was due to reduced recruitment of CD45+ hematopoietic cells (Table II). In contrast, the number of CD45-negative cells was comparable between WT and IL-1R1 KO infarcted hearts (Table II). IL-1R1 loss was associated with markedly reduced infiltration of the infarct with CD11b+, Ly6Chi and Ly6Clo cells suggesting that IL-1 signaling plays an essential role in mediating the post-infarction inflammatory response (Table II).

5. Disruption of IL-1 signaling in IL-1R1 null mice is associated with attenuated inflammatory activity in CD11b-negative non-myeloid cells

In addition to its effects on monocyte recruitment in the infarcted myocardium, IL-1R1 signaling may increase the level of pro-inflammatory activation of fibroblasts and leukocytes in the infarcted heart. In order to dissect the cellular targets of IL-1-mediated pro-inflammatory actions we harvested CD11b+ leukocytes and CD11b-negative non-myeloid cells from control and infarcted WT and IL-1R1 null hearts and we compared expression of pro-inflammatory cytokines. In comparison to cells harvested from normal hearts, CD11b+ myeloid cells and CD11b-negative non-myeloid cells harvested from infarcts after 24h of reperfusion showed markedly increased IL-1β and MCP-1 mRNA expression (Figure 4A-B, Figure 4D-E).

CD11b-negative cells isolated from IL-1R1 null infarcts after 24h of reperfusion had markedly lower IL-1β (Figure 4A) and MCP-1 (Figure 4B) expression than WT infarct CD11b-negative cells harvested at the same post-reperfusion timepoint. In contrast, in CD11b+ leukocytes, cytokine and chemokine expression levels were not affected by IL-1R1 loss (Figure 4D-E). IL-6 mRNA expression showed a distinct time course of expression in CD11b+ leukocytes and in CD11b-negative non-myeloid cells: CD11b-negative cells exhibited early upregulation after 24h of reperfusion (Figure 4C), whereas CD11b+ leukocytes had a late increase after 7 days (Figure 4C-F). IL-1R1 disruption did not significantly affect IL-6 expression by infarct leukocytes and nonmyeloid cells (Figure 4C, 4F).

6. IL-1 receptor expression profile in infarct fibroblasts

Because loss of IL-1 signaling selectively attenuated inflammatory activation of non-myeloid cells, we studied the IL-1R expression profile in CD11b-negative cells isolated from infarcted hearts. IL-1R1 mRNA expression was significantly increased in CD11b-negative cells harvested after 24h of reperfusion when compared with corresponding non-myeloid cells from control hearts (Figure 5A). In contrast, IL-1R2 mRNA expression was not detected in non-myeloid cells harvested from the infarcted heart (not shown). Because fibroblasts represent the majority of interstitial non-myeloid cells in control and infarcted hearts, we examined whether inflammatory cytokines regulate IL-1R1 mRNA expression in isolated cardiac fibroblasts. In cardiac fibroblasts harvested from normal WT mice, IL-1β significantly induced IL-1R1 mRNA synthesis after 4h of stimulation (Figure 5B). In contrast, IL-10 and TGF-β stimulation did not modulate fibroblast-derived IL-1R1 mRNA synthesis (Figure 5B).

7. IL-1β abrogates TGF-β-induced contraction of fibroblast-populated collagen pads in an IL-1R1-dependent manner

Although in the early stages following cardiac injury, fibroblasts may function as pro-inflammatory cells, the most dramatic alterations in their phenotype occur during the proliferative phase, when infarct fibroblasts undergo myofibroblast transdifferentiation. Acquisition of the myofibroblast phenotype is associated with increased expression of contractile proteins, such as α-SMA, and induces scar contraction. In order to study the effects of IL-1 on myofibroblast transdifferentiation we used a fibroblast-populated collagen pad assay. As previously reported by our laboratory (19), TGF-β, the key mediator in myofibroblast transdifferentiation, induced significant contraction of fibroblast-populated collagen pads (Figure 6A). IL-1β stimulation abolished TGF-β-induced gel contraction in pads populated with WT cells. The effects of IL-1β were dependent on IL-1R1 signaling; in collagen pads populated with IL-1R1 null cells IL-1 stimulation did not modulate TGF-β-induced gel contraction (Figure 6A). Moreover, unstimulated IL-1R1 null fibroblasts induced increased gel contraction when compared with WT cells suggesting that loss of IL-1 signaling may allow unopposed actions of autocrine or paracrine growth factors (Figure 6A).

8. IL-1β attenuates α-SMA expression in stimulated cardiac fibroblasts

Reduced gel contraction upon stimulation with IL-1β likely reflects impaired myofibroblast transdifferentiation. Accordingly, we examined whether IL-1β modulates α-SMA expression in stimulated fibroblasts. Unstimulated cardiac fibroblasts exhibited high expression of α-SMA after 24-72h of culture (Figure 6B-C); stimulation with IL-1β significantly suppressed α-SMA mRNA synthesis in both control and TGF-β-stimulated cells (Figure 6B-C). The effects of IL-1β were abrogated in IL-1RI KO cells (Figure 6B-C). Immunofluorescent staining of isolated fibroblasts for α-SMA supported the quantitative qPCR data showing that IL-1β reduced the incorporation of α-SMA in the cytoskeleton in an IL-1RI-dependent manner (Figure 6D). IL-1β also suppressed TGF-β-induced expression of the matricellular protein periostin (Figure 6G-H) , a key modulator of fibroblast phenotype and function in the infarcted and remodeling myocardium (21), (22). The effects of IL-1β on periostin synthesis were dependent on signaling through the type 1 receptor (Figure 6H). In contrast, expression of collagen type I mRNA by TGF-β-stimulated fibroblasts was not consistently modulated by IL-1β (Figure 6E-F).

9. IL-1β does not modulate TGF-βinduced Smad2/3 phosphorylation

Because TGF-β/Smad3 signaling is critically involved in cardiac myofibroblast transdifferentiation, we examined whether IL-1 directly modulates TGF-β-induced Smad2/3 activation. After 60 min of incubation, TGF-β stimulation significantly increased the levels of p-Smad2 in cardiac fibroblasts (*p<0.05); however, co-incubation with IL-1β did not alter Smad2 phosphorylation (Figure 6B-C).

10. IL-1β stimulation inhibits endoglin expression in cardiac fibroblasts and enhances TGF-β-induced synthesis of the decoy TGF-β receptor BAMBI

Next, we asked whether the inhibitory effects of IL-1β on TGF-β-induced fibroblast contractile activity are due to modulation of TGF-β receptor expression. TGF-β, but not IL-1β, stimulated mRNA expression of the TGF-β type I receptor ALK5; co-stimulation with IL-1β did not affect TGF-induced ALK5 synthesis (Figure 7A). IL-1β stimulation suppressed expression of the TGF-β co-receptor endoglin, inhibiting synthesis of both endoglin-L and endoglin-S transcripts, but did not modulate endoglin expression by TGF-β-stimulated cells (Figure 7B-C). The TGF-β decoy receptor BAMBI participates in negative regulation of TGF-β signaling; accordingly we examined whether IL-1β inhibits TGF-β responses by increasing BAMBI expression in cardiac fibroblasts. Although IL-1β stimulation did not alter BAMBI mRNA expression in cardiac fibroblasts, co-stimulation with TGF-beta and IL-1β significantly upregulated BAMBI mRNA (Figure 7D). Cardiac fibroblasts harvested from IL-1R1 KO mice had lower baseline expression of ALK5, endoglin-L and endoglin-S when compared with WT fibroblasts (Figure 7E-G). The inhibitory effects of IL-1β on endoglins were dependent on IL-1R1 signaling (Figure 7F-G). Moreover, the effects of IL-1β/TGF-β co-stimulation on BAMBI synthesis were also abrogated in IL-1R1 null cells (Figure 7H).

11. IL-1 enhances expression of the collagenases MMP-3 and MMP-8 in cardiac fibroblasts in an IL-1R1-dependent manner

In order to explore the effects of IL-1 on the matrix-degrading potential of cardiac fibroblasts we assessed expression of the collagenases MMP-3 and -8 by IL-1-stimulated fibroblasts cultured in collagen pads. After 24h of stimulation IL-1β markedly induced MMP-3 and MMP-8 mRNA synthesis; the effects of IL-1 were abrogated in IL-1R1 null fibroblasts.

Sequential infiltration of the infarcted heart with IL-1R1 and IL-1R2 cells may regulate the post-infarction inflammatory response

In the infarcted heart, sequential recruitment of leukocyte subpopulations with distinct properties may play an important role in regulation of inflammatory and reparative responses. Early recruitment of pro-inflammatory Ly6C(hi) monocytes through interactions that involve MCP-1/CCR2 signaling (23), (20) mediates phagocytotic clearance of dead cells and matrix debris from the infarct. These cells may be involved in the pathogenesis of dilative remodeling by producing and activating proteases, thus degrading the extracellular matrix. During the reparative phase of healing, the profile of infarct monocytes changes: reparative Ly6C(lo) cells with high expression of the fractalkine receptor CX3CR1 accumulate in the infarct and may mediate angiogenic and pro-fibrotic responses (20). Our findings provide further support for the presence of a biphasic monocytic response in reperfused myocardial infarcts (Table I). Flow cytometric analysis demonstrated that at the peak of the inflammatory phase (after 24h of reperfusion), the number of pro-inflammatory Ly6C(hi) cells increased 60-fold in the infarcted myocardium, whereas the number of reparative Ly6C(lo) cells was only modestly elevated (4-fold increase vs. control myocardium). In contrast, during the proliferative phase (after 7 days reperfusion), the number of Ly6C(hi) cells was markedly reduced, whereas Ly6C(lo) cells exhibited a 23-fold increase in comparison to WT hearts.

Inflammatory activity of leukocyte subsets may be regulated by alterations in the cell surface expression of the signaling and decoy IL-1 receptors (IL-1R1 and IL-1R2 respectively); these changes may modulate cellular responsiveness to IL-1. In vitro studies have demonstrated that in monocytes and macrophages, IL-1R1 and IL-1R2 expression is dynamically regulated by cytokines, lipopolysaccharide and other bioactive mediators (24), (25), (26). Thus, we examined whether transition from the inflammatory to the proliferative phase of infarct healing is associated with changes in IL-1 receptor expression profile. Flow cytometric analysis of cells harvested from the infarcted myocardium showed that the infarct sequentially recruits IL-1R1+ and IL-1R2+ cells. At the peak of the inflammatory phase (24h of reperfusion) cells expressing IL-1R1 predominate (Figure 2), however, their number is significantly reduced after 7 days of reperfusion, suggesting a loss of responsiveness to the effects of IL-1. Reduced numbers of IL-1R1+ cells during the proliferative phase of infarct healing may reflect suppression of IL-1R1 expression, recruitment of IL-1R1-negative leukocyte subpopulations, or clearance of apoptotic IL-1R1+ cells from the infarct. As the number of cells expressing the signaling receptor decreases, cells expressing the decoy type 2 receptor infiltrate the healing infarct. Increased numbers of IL-1R2+ cells may reflect upregulation of the decoy receptor in “anti-inflammatory” macrophages, or new recruitment of IL-1R2+ monocyte subpopulations. Late recruitment, or differentiation, of cells expressing the decoy receptor may serve as a mechanism for termination of IL-1 signaling in the infarcted heart.

IL-1R1 signaling mediates leukocyte infiltration in the infarcted heart

IL-1α and IL-1β signal by binding to IL-1R1, their only signaling receptor. IL-1R1 engagement leads to formation of a complex with the IL-1R accessory protein (IL-1RAcP); this signaling platform sequentially recruits MyD88, Interleukin Receptor Associated Kinase (IRAK)-1 and IRAK-4 initiating the cascade that ultimately leads to Nuclear Factor (NF)-κB activation and induction of a wide range of pro-inflammatory cytokines, chemokines and adhesion molecules (27). In vivo, several investigations have demonstrated the essential role of IL-1R1 signaling in inflammatory processes involving many different tissues. In a model of diet-induced obesity IL-1 signaling was found to be an essential mediator in inflammatory activation of adipose tissue macrophages (28). In bleomycin-induced lung injury, pulmonary inflammation required IL-1R1/MyD88 signaling (29); moreover, in a model of spinal cord injury IL-1R1 or MyD88 loss resulted in markedly attenuated infiltration with pro-inflammatory monocytes (30). In some studies the effects of IL-1 signaling appear to be dependent on the site of injury: IL-1R1 disruption reduced inflammatory leukocyte recruitment in a model of oral wounding, but had no effects on dermal injury (31).

Does IL-1 signaling selectively mediate recruitment of pro-inflammatory and reparative monocytes? IL-1R1 loss markedly reduced the cellular content of reperfused myocardial infarcts; decreased cellular infiltration in animals with impaired IL-1 signaling was due to attenuated recruitment of hematopoietic cells (Table II). IL-1R1 null mice had a global reduction in leukocyte recruitment affecting both pro-inflammatory Ly6C(hi) and reparative Ly6C(lo) cells. These findings may suggest that IL-1 is not selectively implicated in recruitment of pro-inflammatory leukocytes, but has global effects on leukocyte trafficking that may be mediated through induction of adhesion molecules, or through upregulation of a broad range of chemokines.

Cardiac fibroblasts as cellular targets of IL-1 in the infarcted myocardium

Fibroblasts are the predominant non-cardiomyocyte population in the adult human myocardium (32). In addition to their role as matrix-synthetic cells, fibroblasts may also serve as sentinel cells that sense microenvironmental changes and play an important role in initiation of inflammatory responses (33), (34). Following myocardial ischemia, cardiac fibroblasts are capable of activating the inflammasome, triggering IL-1-driven inflammation (35) and may be important sources of cytokines and chemokines. Our experiments demonstrate that the pro-inflammatory actions of IL-1R1 signaling are not limited to leukocytes, but also involve fibroblasts. CD11b-negative cells harvested from the infarcted heart exhibited significant upregulation of pro-inflammatory genes; loss of IL-1R1 attenuated inflammatory gene expression by non-myeloid cells (Figure 4). Myeloid and CD11b-negative cells were harvested from whole hearts; thus, our experiments do not provide information on the potentially distinct cellular phenotypes in the infarcted and non-infarcted remodeling myocardium. Because during the inflammatory phase of infarct healing, cytokine upregulation is primarily localized in the infarcted area, the effects of IL-1R1 loss may primarily reflect IL-1 actions on cells infiltrating the infarct. In addition to its effects in promoting pro-inflammatory signaling, IL-1 also induced MMP expression in cardiac fibroblasts, thus stimulating matrix-degrading activity (Figure 8).

Myofibroblast transdifferentiation is the hallmark of the proliferative phase of cardiac repair (36); differentiated myofibroblasts are the main matrix-producing cells in the infarcted heart (37). Activation of TGF-β signaling provides a key signal for acquisition of myofibroblast phenotype; matrix alterations such as deposition of the ED-A isoform of cellular fibronectin and induction of matricellular proteins also contribute to the phenotypic transition of fibroblasts into myofibroblasts. However, despite the early presence of bioactive TGF-β in the infarcted myocardium within the first 3 hours of reperfusion (38), myofibroblast transdifferentiation in healing infarcts is delayed until pro-inflammatory signaling has been suppressed and the local environment can support a large population of matrix-secreting cells (39). Our findings demonstrate for the first time that IL-1 inhibits TGF-β-induced contraction of fibroblast-populated collagen pads, suppressing α-SMA and periostin synthesis, attenuating functional and molecular responses that directly reflect myofibroblast transdifferentiation (Figure 6). Thus, in addition to its pro-inflammatory actions, IL-1 signaling may also serve to prevent premature conversion of fibroblasts into myofibroblasts during the inflammatory phase of cardiac repair, at a timepoint when the infarct environment may be hostile to reparative and contractile cells. Suppression of IL-1 expression and termination of IL-1 signaling during the proliferative phase of infarct healing may allow unopposed actions of TGF-β on cardiac fibroblasts mediating myofibroblast transdifferentiation and scar contraction.

What is the mechanism of IL-1-mediated inhibition of the TGF-β response?

IL-1-mediated suppression of TGF-β signaling in cardiac fibroblasts is not due to direct effects on TGF-induced Smad2 phosphorylation (Figure 6B-C). The effects of IL-1β may reflect alterations on the TGF-β receptor expression profile in cardiac fibroblasts (Figure 7). IL-1-induced downregulation of the TGF-β co-receptor endoglin, a critical component of pro-fibrotic TGF-β signaling may be an important mechanism in mediating the inhibitory effects of IL-1 on TGF-β signaling. Expression of the TGF-β co-receptor endoglin is increased in failing hearts and has been linked with the development of cardiac fibrosis (40). IL-1 may also stimulate molecular signals that negatively regulate the TGF-β response. When added to TGF-β, IL-1β increased fibroblast expression of the BAMBI, a transmembrane molecule that decreases formation of the functional TβRI/TβR II complex, thus negatively regulating the TGF-β response (41). BAMBI overexpression in fibroblasts has been reported to enhance TGF-β-mediated effects (42).

Implications for the use of anti-IL-1 approaches in patients with myocardial infarction

Because loss of IL-1 signaling attenuates adverse post-infarction remodeling (7) and IL-1 exerts proapoptotic effects on cardiomyocytes (8), anti-IL-1 strategies have been proposed as therapy for patients with acute myocardial infarction. In a small pilot study, administration of anakinra was safe and attenuated dilative remodeling following myocardial infarction. These promising early results generate optimism regarding the potential effectiveness of IL-1 inhibition in preventing development of post-infarction heart failure. Our findings suggest that the protective effects of IL-1 inhibition may be due to decreased recruitment of pro-inflammatory leukocytes and to reduced activation of inflammatory and matrix-degrading properties in cardiac fibroblasts. However, our experiments also suggest caution regarding effects of disturbed IL-1 signaling on orchestration of the reparative response. Anti-IL-1 treatment may result in impaired recruitment of reparative monocyte subsets and in premature acquisition of the myofibroblast phenotype, thus causing perturbations in the reparative process. Moreover, disruption of IL-1 signaling may cause loss of unrecognized cytoprotective signals on ischemic cardiomyocytes. Dissection of the cell biological actions of IL-1 signaling is needed to optimally guide design of therapeutic approaches in the infarcted and remodeling heart.