Role of CCR5 in IFN-γ–induced and
cigarette smoke–induced emphysema (original) (raw)
Effect of IFN- γon CCR5 chemokine ligands. To further define the mechanisms that mediate IFN-γ–induced tissue alterations, we determined whether IFN-γ altered the expression and or production of CC and CXC chemokines that might be expected to contribute to these responses. In lungs from Tg– mice on normal or doxycycline (dox) water, the levels of mRNA encoding MIP-1α/CCL-3, MIP-1β/CCL-4, and RANTES/CCL-5 were at or below the limits of detection of our assays (Figure 1A). In contrast, IFN-γ was a potent stimulator of MIP-1α/CCL-3, MIP-1β/CCL-4, and RANTES/CCL-5 mRNA and protein in lungs from Tg+ mice. The mRNA effects were seen after as few as 3 days of dox administration and persisted throughout the 3-month study interval in RT-PCR evaluations (Figure 1A and data not shown). Similar results were seen with real-time RT-PCR mRNA quantification (Figure 1B). Increases in levels of cytokine protein were also seen after as few as 3–7 days of dox administration and remained significantly elevated after 1 month of dox administration (Figure 1C). These effects were not specific for these cytokine moieties, as prominent induction of monocyte chemoattractant protein-1/CCL-2 (MCP-1/CCL-2), MCP-2/CCL-8, MCP-5/CCL-12, MIP-2/CXCL2/3, KC/CXCL-1, eosinophil neutrophil-activating–78/CXCL-5 (ENA-78/CXCL-5), monokine induced by γ interferon/CXCL-9 (Mig/CXCL-9), interferon-inducible protein 10/CXCL-10 (IP-10/CXCL-10), interferon-inducible T cell α chemoattractant/CXCL-11 (I-TAC/CXCL-11), stromal cell–derived factor-1/CXCL-12 (SDF-1/CXCL-12), C10/CCL-6, macrophage-derived chemokine/CCL-22 (MDC/CCL-22), and thymus-expressed chemokine/CCL25 (TECK/CCL25) were also noted (Figure 1A). In contrast, eotaxin/CCL-11, thymus- and activation-regulated chemokine/CCL-17 (TARC/CCL-17), and lungkine/CXCL-15 were not similarly regulated (Figure 1A). These studies demonstrate that IFN-γ is a potent and selective stimulator of pulmonary chemokines.
IFN-γ regulation of pulmonary chemokines. Tg– and Tg+ mice were given normal or dox water for up to 1 month. The levels of mRNA encoding the noted chemokines in lungs from mice on dox for 1 month were evaluated by RT-PCR (A) and real-time RT-PCR (B), and the levels of BAL chemokine protein were evaluated by ELISA (C). Each evaluation in A is representative of a minimum of 4 similar experiments. The values in B and C represent the mean ± SEM of evaluations in a minimum of 4 animals. *P < 0.05 in B. In C, P < 0.05 at all time points for MIP-1α and MIP-1β and at 7 and 30 days for RANTES. MDC, macrophage-derived protein; TECK, thymus-expressed chemokine; TARC, thymus- and activation-regulated chemokine.
Effect of IFN- γon CCR5. Because prominent alterations in MIP-1α/CCL-3, MIP-1β/CCL-4, and RANTES/CCL-5 were noted, studies were next undertaken to define the effects of IFN-γ on their receptor, CCR5. Levels of CCR5 mRNA and protein that were near or below the limits of detection of our assays were noted in lungs from Tg– mice on normal water or dox water and Tg+ mice on normal water (Figure 2, A and B and data not shown). In contrast, the levels of CCR5 mRNA and protein were markedly increased in dox-treated IFN-γ–transgenic animals (Figure 2, A and B). To localize this protein, immunohistochemistry was also undertaken. These studies demonstrate that CCR5 is in an interstitial location in lungs from Tg+ mice (Figure 2, C–F). This staining was CCR5 specific, as it was not noted in IFN-γ–transgenic mice with null CCR5 loci, and the antibody that was used detected an appropriately sized molecule in Western blot evaluations of proteins from lungs from Tg+ mice (Figure 2B). In double labeling experiments, confocal analysis demonstrated that, in many cases, CCR5 colocalized with vimentin (see Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI24858DS1). In contrast, CCR5 did not colocalize with CD45, CD3, griffonia lectin, keratin, or S100A4 (Supplemental Figure 1 and data not shown). In keeping with recent reports in other experimental systems (30), these studies demonstrate that IFN-γ stimulates the expression of CCR5 on pulmonary interstitial stromal cells.
CCR5 in IFN-γ–transgenic mice. Tg– and Tg+ mice were given normal or dox water for 1 month. The levels of CCR5 mRNA and protein in lungs from these mice were evaluated by real-time RT-PCR (A) and Western blot analysis (B), respectively. The CCR5 in the pulmonary specimens is compared with that in a thymus-positive control. (C–F) Immunohistochemistry was used to compare the CCR5 in lungs from Tg– mice stained with anti-CCR5 (C), Tg+ mice stained with anti-CCR5 (D), Tg+ mice stained with control antibody (E), and Tg+ mice with a null mutation of CCR5 stained with anti-CCR5 (F). The values in A represent the mean ± SEM of evaluations in a minimum of 4 animals. Each evaluation in B–F is representative of a minimum of 4 similar experiments. Original magnification, ×40. *P < 0.05.
Role of CCR5 in IFN-γ–induced inflammation. To understand the roles of MIP-1α/CCL-3, MIP-1β/CCL-4, RANTES/CCL-5, and CCR5 in the generation of IFN-γ–induced tissue alterations, we compared the inflammation in CC10-rtTA-IFN-γ Tg+ mice that had been randomized to normal water or dox water at 1 month of age and treated with antiserum against CCR5 or control antiserum. We also bred Tg+ mice with CCR5–/– mice and compared the effects of transgenic IFN-γ in CCR5+/+ and CCR5–/– mice. As previously reported (19), transgenic IFN-γ caused significant increases in bronchoalveolar lavage (BAL) total cell, macrophage, and neutrophil recovery and induced a patchy macrophage-rich tissue inflammatory response (Figure 3 and data not shown). Treatment with anti-CCR5 did not alter the number or differential of the cells that were recovered in BAL fluids and did not alter the tissue histology of lungs from Tg– mice on normal or dox water (Figure 3, A and B and data not shown). In contrast, this intervention significantly decreased BAL total cell and macrophage, neutrophil, and lymphocyte recovery and the tissue inflammatory response in lungs from dox-treated Tg+ animals (Figure 3, A and B and data not shown). Similar alterations were seen in comparisons of BAL and tissues from CCR5+/+ and CCR5–/– Tg+ mice (Figure 3, C and D and data not shown). Interestingly, this effect was not specific for IFN-γ, as a deficiency of CCR5 also decreased LPS-induced pulmonary inflammation (Supplemental Figure 2). Thus, CCR5 is a crucial regulator of the intensity and nature of IFN-γ–induced pulmonary inflammation.
Effect of CCR5 neutralization/ablation on IFN-γ–induced inflammation. (A and B) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig for 4 weeks. (C and D) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. BAL cell recovery and differential were then evaluated. The values represent the mean ± SEM of evaluations in a minimum of 4 mice in each group; *P < 0.05. Eo, eosinophil; Neu, neutrophil; Lym, lymphocyte; Mac, macrophage.
Role of CCR5 in IFN-γ–induced alveolar remodeling and destruction. To define the role of CCR5 in the pathogenesis of IFN-γ–induced alveolar remodeling, we compared the alterations in lung volume, alveolar size, and lung compliance in Tg+ mice on dox water that had been treated with anti-CCR5 or control Ig and CCR5+/+ and CCR5–/– Tg+ mice. In accordance with previous observations (19), dox induction of IFN-γ caused an impressive increase in all of these parameters (Figure 4 and data not shown). Treatment with anti-CCR5 did not alter these parameters in lungs from wild-type mice on normal or dox water (Figure 4 and data not shown). In contrast, lungs from dox-treated Tg+ mice treated with anti-CCR5 were significantly smaller and less compliant than lungs from Tg+ mice treated with control Ig (Figure 4A and data not shown). Alveolar size was similarly decreased when assessed with light microscopic or morphometric approaches (Figure 4, B and C). Importantly, similar decreases in alveolar remodeling were noted in comparisons of lungs from dox-treated CCR5+/+ and CCR5–/– Tg+ mice (Figure 4, D–F). Overall, treatment with anti-CCR5 or a null mutation of CCR5 caused 52.4% and 51.3% decreases in alveolar chord length enlargement, respectively, in dox-treated IFN-γ–transgenic mice (P < 0.01 compared with Tg+ dox-treated controls). When viewed in combination, these studies demonstrate that CCR5 plays a critical role in the pathogenesis of IFN-γ–induced alveolar remodeling and destruction in the murine lung.
Role of CCR5 in IFN-γ–induced remodeling. (A–C) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig (Ctrl Ig) for 4 weeks. (D–F) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. Lung volume (A and D), histology (B and E; ×10), and chord length (C and F) were evaluated. B and E are representative of a minimum of 5 similar evaluations. The values in the rest of the panels represent the mean ± SEM of evaluations in a minimum of 5 mice in each group. *P < 0.05.
Role of CCR5 in IFN- γelaboration. A deficiency of CCR5 could modify IFN-γ–induced tissue responses by altering the production of transgenic IFN-γ or modulating IFN-γ effector responses. To determine whether alterations in CCR5 regulated the production of IFN-γ, we compared the levels of BAL IFN-γ in Tg+ and Tg– mice treated with anti-CCR5 or control Ig and CCR5+/+ and CCR5–/– Tg+ mice. IFN-γ was not readily apparent in BAL fluids from Tg– mice on normal or dox water that had been treated with anti-CCR5 or control Ig (Figure 5). In contrast, significant levels of BAL IFN-γ were observed in Tg+ mice on dox water. These levels were similar in mice treated with anti-CCR5 or the control Ig (Figure 5A). Comparisons of BAL fluids from CCR5+/+ and CCR5–/– Tg+ mice also did not reveal differences in IFN-γ content (Figure 5B). Thus, treatment with anti-CCR5 and null mutations of CCR5 altered IFN-γ–induced tissue responses by modifying IFN-γ effector pathway activation.
Role of CCR5 in transgenic IFN-γ production. (A) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig for 4 weeks. (B) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. BAL IFN-γ was evaluated by ELISA. The values represent the mean ± SEM of evaluations in a minimum of 5 mice in each group.
Role of CCR5 in IFN-γ–induced chemokine elaboration. To investigate the mechanism(s) by which a deficiency of CCR5 ameliorated IFN-γ–induced inflammation, we compared the expression of selected chemokines in Tg+ mice treated with anti-CCR5 or control Ig. As noted above, the levels of MIP-1α/CCL-3, MIP-1β/CCL-4, RANTES/CCL-5, MCP-1/CCL-2, MCP-2/CCL-8, MCP-5/CCL-12, MIP-2/CXCL-2/3, KC/CXCL-1, ENA 78/CXCL-5, Mig/CXCL-9, IP-10/CXCL-10, I-TAC/CXCL-11, and SDF-1/CXCL-12 mRNA and protein were near or below the limits of detection of our assays in Tg– mice and were significantly increased by transgenic IFN-γ (Figure 6 and data not shown). Anti-CCR5 significantly diminished the ability of IFN-γ to stimulate the accumulation of _MIP-1_α/CCL-3, _MIP-1_β/CCL-4, RANTES/CCL-5, MCP-1/CCL-2, MIP-2/CXCL-2/3, KC/CXCL-1, IP-10/CXCL-10, and SDF-1/CXCL-12 mRNA (Figure 6A). In all cases, comparable decreases in the levels of BAL chemokines were also noted (Figure 6, B–D and data not shown). In contrast, anti-CCR5 did not alter the ability of IFN-γ to stimulate the expression of MCP-2/CCL-8, MCP-5/CCL-12, Mig/CXCL-9, or I-TAC/CXCL-11 (Figure 6A). Importantly, similar alterations in the levels of expression and production of these chemokines were noted in comparisons of dox-treated CCR5+/+ and CCR5–/– Tg+ mice (Figure 6, E–I and data not shown). Thus, IFN-γ stimulates pulmonary chemokines via CCR5-dependent and -independent pathways, with the CCR5 ligands MIP-1α/CCL-3, MIP-1β/CCL-4, and RANTES/CCL-3 and MCP-1/CCL-2, MIP-2/CXCL-2/3, KC/CXCL-1, IP-10/CXCL-10, and SDF-1/CXCL-12 being induced by the former and MCP-2/CCL-8, MCP-5/CCL-12, Mig/CXCL-9, and I-TAC/CXCL-11 being stimulated by the latter.
Role of CCR5 in IFN-γ–induced chemokine responses. (A–D) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig for 4 weeks. (E–I) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. Chemokine mRNA was evaluated by RT-PCR (A and E) and real-time RT-PCR (F). BAL chemokines (B–D and G–I) were evaluated by ELISA. Each evaluation in A and E is representative of a minimum of 4 similar experiments. The values in the remaining panels represent the mean ± SEM of evaluations in a minimum of 4 animals. *P < 0.05.
Role of CCR5 in IFN-γ–induced protease and antiprotease alterations. We reasoned that a deficiency of CCR5 could modulate the IFN-γ–induced inflammatory and alveolar phenotypes by regulating the production of respiratory proteases and or antiproteases. To test this hypothesis, we compared the levels of mRNA encoding lung-relevant MMPs, cathepsins, and antiproteases in Tg– and Tg+ mice treated with anti-CCR5 or control Ig and mice with wild-type and null CCR5 loci. Comparable levels of mRNA encoding MMP-2, MMP-9, MMP-12, MMP-14, cathepsin B, cathepsin H, cathepsin L, cathepsin S, α-1 antitrypsin (α1-AT), secretory leukocyte protease inhibitor (SLPI), and tissue inhibitor of metalloproteinases (TIMPs) 1–4 were noted in lungs from Tg– mice treated with anti-CCR5 or control Ig and Tg– mice with wild-type and null CCR5 loci (Figure 7, A–C). In these mice, the levels of mRNA encoding a number of these moieties were near or below the limits of detection of our assays. In accordance with previous studies from our laboratory (19), dox induction of IFN-γ increased the levels of expression of MMP-9 and -12 and cathepsins B, H, L, and S, while inhibiting the expression of SLPI and not altering the expression of α1-AT or the TIMPs (Figure 7, A–C). The levels of MMP-9 protein were similarly increased (Figure 7D). Interestingly, the neutralization or ablation of CCR5 decreased the ability of IFN-γ to stimulate the accumulation of MMP-9 mRNA and protein and inhibit the expression of SLPI (Figure 7, A–D). It did not, however, alter the effects of IFN-γ on the other MMPs, cathepsins, and TIMPs. Thus, IFN-γ selectively stimulates MMP-9 and inhibits SLPI via CCR5-dependent activation pathways.
Role of CCR5 in IFN-γ–induced protease and antiprotease responses. (A) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig (anti-CCR5–) for 4 weeks. (B–D) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. Protease and antiprotease mRNAs were evaluated by RT-PCR (A and B), and MMP-9 mRNA and protein were evaluated by real time RT-PCR (C) and Western blotting (D), respectively. Each evaluation is representative of a minimum of 4 similar experiments. α1-AT, α-1 antitrypsin.
Role of CCR5 in IFN-γ–induced DNA injury and cell death. In keeping with the proposed role of apoptosis in the pathogenesis of the alveolar remodeling in emphysema (24–28), studies were undertaken to determine whether IFN-γ induced DNA injury and apoptosis and define the role(s) of CCR5 in these responses. In these experiments, TUNEL and dual propidium iodide/annexin V (PI/annexin V) evaluations were used to compare the DNA injury and cell death in lungs from Tg– and Tg+ mice that had been treated with either anti-CCR5 or control Ig and CCR5+/+ and CCR5–/– Tg+ mice. In Tg– mice on normal or dox water, 3% or less of cells were TUNEL+ regardless of they type of antibody treatment that was administered (Figure 8, A and B). In contrast, dox induction of IFN-γ caused a significant increase in TUNEL staining in lungs from Tg+ mice (Figure 8, A and B). This response was seen after as few as 2 days of dox administration. At its peak (after 2 weeks of dox), 15–25% of cells were TUNEL+. At this time point, TUNEL staining was seen predominantly in alveolar epithelial cells, with lesser numbers of endothelial cells and macrophages manifesting TUNEL reactions (data not shown). Double annexin-V and PI staining demonstrated that the cell death response in these mice was predominantly apoptosis that manifested as an increase in annexin V staining only (Figure 8C). Significant numbers of cells also manifested increases in annexin V and PI and were undergoing a response with features of apoptosis and necrosis (Figure 8C). Treatment with anti-CCR5 did not alter the levels of these responses in lungs from Tg– mice on normal or dox water (Figure 8, A and B and data not shown). Treatment with anti-CCR5 did, however, cause a significant decrease in DNA injury and cell death in dox treated Tg+ animals (Figure 8, A and B and data not shown). Similar decreases in DNA injury and cell death were seen with TUNEL and dual PI/annexin V evaluations of lungs from dox-treated CCR5+/+ and CCR5–/– Tg+ mice (Figure 8, C–E). When viewed in combination, these studies demonstrate that IFN-γ is a potent inducer of DNA injury, apoptosis, and a mixed apoptosis/necrosis cellular response. They also demonstrate that these responses are mediated by pathways that are, at least partially, CCR5-dependent.
Role of CCR5 in IFN-γ–induced DNA injury and cell death. (A and B) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig for 4 weeks. (C–E) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. TUNEL evaluations (A and D; ×40), TUNEL quantification (B and E), and dual PI/annexin V evaluations (C) were undertaken. Each evaluation in A, C, and D is representative of a minimum of 4 similar experiments. The values in the remaining panels represent the mean ± SEM of evaluations in a minimum of 4 animals. *P < 0.05.
Contributions of CCR5 to IFN-γ–induced apoptosis. To further understand the mechanisms by which CCR5 regulated DNA injury and cell death, the expression and activity of caspases and key cell death regulators were evaluated in mice that were treated with anti-CCR5 or control Ig and CCR5+/+ and CCR5–/– Tg+ mice. The levels of mRNA encoding Fas, FasL, TNF-α, caspase-3, -8, and -9, Bid, and Bax in Tg– mice were at or near the limits of detection in our assays and were not significantly altered by CCR5 neutralization or ablation (Figure 9, A–C). IFN-γ was a potent stimulator of these moieties that increased the levels of Fas, FasL, TNF-α, caspase-3, -8, -9, and Bid and Bax mRNA and the levels of TNF-α protein (Figure 9, A–E and data not shown). IFN-γ also activated caspase-3, -8, and -9 and Bid, causing readily detectable increases in caspase bioactivities and truncated Bid (tBid) accumulation (Figure 9, F and G). In all cases, these events were CCR5 dependent, as anti-CCR5 treatment or a null mutation of CCR5 decreased the levels of mRNA encoding Fas, FasL, TNF-α, caspase-3, -8, and -9, Bid, and Bax, the levels of TNF-α protein, and caspase and Bid activation (Figure 9). When viewed in combination, these studies demonstrate that IFN-γ is a potent activator of the extrinsic/death receptor and intrinsic/mitochondrial apoptosis pathways and that these activation events are, at least in part, CCR5 dependent.
Role of CCR5 in IFN-γ regulation of cell death pathways. (A, D, and F) Tg– and Tg+ mice were placed on dox water and treated with anti-CCR5 or control Ig. (B, C, E, and G) CCR5+/+ and CCR5–/– Tg– and Tg+ mice were treated with dox water for 4 weeks. mRNA was evaluated by RT-PCR (A and B) and real-time RT-PCR (C). BAL TNF levels were evaluated by ELISA (D and E). (F) Caspase-3, -8, and -9, Bid, and truncated Bid (tBid) were evaluated by Western blot analysis. (G) Caspase bioactivity was evaluated. Each evaluation in A, B, and F is representative of a minimum of 4 similar experiments. The values in the remaining panels represent the mean ± SEM of evaluations in a minimum of 5 animals. *P < 0.05.
Role of CCR5 in cigarette smoke–induced inflammation and alveolar remodeling. The studies noted above demonstrate that CCR5 plays a key role in the pathogenesis of IFN-γ–induced inflammation and emphysema. Since human pulmonary emphysema is often caused by cigarette smoke exposure (20), studies were undertaken to define the role of CCR5 in the pathogenesis of cigarette smoke–induced emphysema. Studies were first undertaken to determine whether IFN-γ plays an important role in cigarette smoke–induced emphysema and stimulates the production of CCR5 ligand chemokines. In these experiments we first compared the emphysema induced by cigarette smoke in wild-type and IFN-γ–null mice. As can be seen in Figure 10A, cigarette smoke caused notable emphysema in wild-type mice. Importantly, this emphysematous response was significantly ameliorated in the IFN-γ–null animals (Figure 10A). In additional experiments, we measured the levels of CCR5 ligand chemokines in BAL fluids from cigarette smoke– and room air–exposed mice. These experiments demonstrated that cigarette smoke stimulates MIP-1α/CCL-3, MIP-1β/CCL-4, and RANTES/CCL-5 in the murine lung (Figure 10B).
Role of IFN-γ and CCR5 in cigarette smoke–induced responses. (A) IFN-γ+/+ and _IFN-γ_–/– mice were exposed to cigarette smoke (CS) for 6 months. (B–G) Tg– mice with wild-type and null CCR5 loci were exposed to cigarette smoke for 2 (B–E) or 6 (F and G) months. Alveolar chord length (A and G); BAL RANTES/CCL-5, MIP-1α/CCL-3, and MIP-1β/CCL-4 (B); total BAL cell recovery (C); the differential distribution of the recovered BAL cells (D); TUNEL staining (E); and alveolar histology (F; ×10) were assessed. Each evaluation in F is representative of a minimum of 4 similar experiments. The values in the remaining panels represent the mean ± SEM of evaluations in a minimum of 4 animals. *P < 0.05.
Because cigarette smoke induction of emphysema was IFN-γ dependent and associated with the induction of CCR5 ligands, we next evaluated the roles of CCR5 in these responses. This was done by comparing the responses in wild-type and CCR5-null mice that were chronically exposed to cigarette smoke or room air. In these studies, cigarette smoke caused a macrophage-rich BAL and tissue inflammatory response. This response was most prominent after 2 months of smoke exposure and persisted throughout the 6-month exposure interval (Figure 10, C and D and data not shown). It also caused DNA injury/apoptosis that could be observed with TUNEL evaluations (Figure 10E) and histologically and morphometrically apparent emphysema (Figure 10, F and G). In all cases, these responses were CCR5 dependent, since tissue and BAL inflammation, TUNEL staining, and emphysema were all decreased in cigarette smoke–exposed CCR5–/– and CCR5+/+ mice (Figure 10, C–G and data not shown). These studies demonstrate that CCR5 plays a critical role in cigarette smoke–induced chemokine induction, inflammation, DNA injury/apoptosis, and alveolar remodeling.