Tumor Necrosis Factor-α Accelerates the Resolution of Established Pulmonary Fibrosis in Mice by Targeting Profibrotic Lung Macrophages (VSports在线直播)

Animals

Male C57BL/6J, TNF-α–deficient (B6. 129S-Tnftm1Gkl/J) and MAcrophage Fas-Induced Apoptosis (MAFIA) [C57BL/6-Tg(Csf1r-EGFP-NGFR/FKBP1A/TNFRSF6)2Bck/J] mice (6–8 wk of age) were purchased from the Jackson Laboratory (Bar Harbor, ME) and studied as approved by the National Jewish Health Institutional Animal Care and Use Committee. Fifteen mice were instilled for each experimental treatment condition VSports app下载.

VSports - Induction, Measurement, and Treatment of Pulmonary Fibrosis

Pulmonary fibrosis was initiated by the intratracheal instillation of 50 μl of bleomycin (3 U/kg) (TEVA Phamaceuticals, North Wales, PA) as previously described (22). Control mice were instilled with 50 μl of saline V体育官网. Fibrosis was assessed by lung measurements of static compliance, collagen (hydroxyproline) levels, and evaluation of Masson’s trichrome–stained sections as reported (22). As described in the Results, mice were intratracheally instilled with 50 μl of recombinant mouse TNF-α (500 ng) (PeproTech, Rocky Hill, NJ) at 3 and 4 weeks and harvested at 5 weeks. No mortality was associated with TNF-α administration.

Depletion of Macrophages in MAFIA Mice

Macrophages were conditionally depleted in MAFIA mice, which express a chimeric FK506 binding protein-Fas death receptor under the c-fms intronic enhancer, by the intravenous injection of the synthetic dimerizer AP20187 (10 mg/kg) (Ariad Pharmaceuticals, Inc., Cambridge, MA) reconstituted as previously described (23).

Enzymatic Dispersal of Pulmonary Macrophages and Flow Cytometry (VSports)

Mice were killed by lethal intraperitoneal injection of Nembutal (Ovation Pharmaceuticals, Paramus, NJ), and the lungs were lavaged (22). Single-cell suspensions were obtained from perfused, enzymatically dispersed lungs (24). Cells were incubated with Mouse BD Fc Block (BD Biosciences, San Diego, CA) for 30 minutes and stained with fluorescently tagged monoclonal antibodies against CD11c, MHCII, Ly6G (eBiosciences, San Diego, CA), CD11b, and F4/80 (BD Pharmigen, San Diego, CA) as described (25). CD206 (Serotec, Raleigh, NC) and NOS2 (BD Biosciences) were used to stain alternatively and classically programmed macrophages, respectively. The mean fluorescence intensities (MFIs) for CD206 and NOS2 were determined for the entire population of CD11c⁺CD11b^var and CD11c⁻CD11b⁺ macrophages. Dendritic cells identified as CD11c⁺CD11b^varMHCII^hiF4/80⁻ were eliminated from macrophages followed by confirmation of being SSC^lo. Data were acquired with a LSR II flow cytometer (BD Biosciences) and analyzed with FlowJo software (Tree Star, Ashland, OR).

Immunofluorescence Analysis of Macrophage Programming

Immunofluorescence staining of paraffin-embedded sections using antibodies against F4/80 (Serotec), NOS2 (BD Biosciences), and arginase I (Santa Cruz Biotechnology, Santa Cruz, CA) was conducted as previously described (26). Images were acquired on a Zeiss Axioplan 2 epifluorescence microscope and analyzed with Axiovision software (Zeiss, Thornwood, NY). Quantification of fluorescence intensity from 10 images per animal by pixel counting was accomplished using ImageJ software. The total pixel intensity for arginase I or NOS2 in each image was divided by the pixel intensity for F4/80 from the same image and multiplied by 100 to generate a score (% positive macrophages) for each marker. This approach provides a quantitative value for the macrophage population sampled but does not distinguish specific changes occurring in individual cells. Arginase activity was measured in the middle right lobe as described (27).

"V体育2025版" Statistics

Data are presented as the mean ± SEM. Differences between conditions at specific time points were examined using Student’s unpaired t test. One-way ANOVA with Newman-Keuls post hoc analysis was used to compare results from more than two groups, with P < 0.05 considered to be significant.

In Vitro Analysis of Bone Marrow Macrophage Programming

Bone marrow macrophages from wild-type mice were prepared and cultured as previously described (13). After 7 days in culture, macrophages were pretreated for 24 hours with rmIL-4 and rmIL-13 (each at 20 ng/ml) (PeproTech) to induce alterative programming. TNF-α (20 ng/ml) (PeproTech) was added for 24 hours and macrophage programming was assessed by flow cytometry for CD206 and by arginase activity in cell lysates as described (27).

Quantification of TNF-α by ELISA

TNF-α was measured in lung homogenates by ELISA following the manufacturer’s protocol (R&D Systems, Minneapolis, MN).

TNF-α Accelerates the Resolution of Established Pulmonary Fibrosis in Mice

To test the hypothesis that therapeutic TNF-α administration accelerates the resolution of established pulmonary fibrosis, we used the single-dose model of intratracheal bleomycin instillation in mice (28). This model leads to a predicable course of alveolar protein leak (Figure 1A) and airspace inflammation (Figure 1B) at 1 to 2 weeks, followed by the development of parenchymal fibrosis between Weeks 2 to 5 as reflected by increased lung hydroxyproline (Figure 1C), decreased static compliance (Figure 1D), and increased lung collagen content as assessed by Masson’s trichrome staining (Figure 1E). Between Weeks 5 and 8 the fibrotic response gradually resolved until all fibrotic parameters approach that of saline-instilled control mice by Weeks 6 to 8. TNF-α was measured by ELISA and was undetectable in bronchoalveolar lavage fluid (BALF) at any time point but was detected in lung homogenates for up to 6 days, peaking 12 hours after bleomycin instillation (see Figure E1 in the online supplement).

To determine if therapeutically administered TNF-α accelerates the resolution of established pulmonary fibrosis, we intratracheally instilled mice with recombinant murine TNF-α (500 ng) (29) or saline as a control at Weeks 3 and 4 after bleomycin treatment (i.e., after fibrosis was established) and assessed inflammatory and fibrotic parameters at Week 5 (Figure 2A). Intratracheal delivery of TNF-α reduced lung collagen levels (P < 0.001) (Figure 2B) and improved compliance in fibrotic mice (P < 0.001) (Figure 2C). TNF-α instillation also reduced BALF protein levels (P < 0.001) (Figure 2D) and inflammatory cell numbers (P < 0.01) (Figure 2E). Examination of trichrome-stained lung sections confirmed reduced lung collagen in TNF-α–treated mice and an improvement of lung architecture with areas of relatively normal lung tissue juxtaposed to patches of residual fibrosis (Figure 2F). Instillation of TNF-α into the lungs of saline-instilled mice had no effect on lung histology or on the fibrotic and inflammatory endpoints (Figure 2).

Given the therapeutic effect of TNF-α in resolving established fibrosis, we next determined if endogenously produced TNF-α contributed to the resolution of pulmonary fibrosis. Wild-type and TNF-α^−/− mice were instilled with bleomycin, and fibrosis was assessed at 3, 6, and 8 weeks. Bleomycin-instilled wild-type and TNF-α^−/− mice developed similar changes in lung collagen levels (Figure 3A) and compliance (Figure 3B) at 3 weeks. However, although lung collagen levels and compliance in wild-type mice had returned to control levels by Weeks 6 and 8, collagen levels remained elevated (P < 0.05) (Figure 3A), and compliance continued to be reduced (P < 0.05) (Figure 3B) in TNF-α^−/− mice. TNF-α^−/− mice also exhibited elevated inflammatory cell numbers in BAL (P < 0.001) compared with wild-type mice at 3, 6, and 8 weeks (Figure 3C). Examination of trichrome-stained sections confirmed persistent fibrosis, collagen deposition (stained blue), septal thickening, and inflammatory cell infiltration (stained pink) for up to 8 weeks in the lungs of TNF-α^−/− mice, whereas in wild-type mice, inflammation and lung architecture were considerably improved by Weeks 6 and 8 (Figure 3D). Collectively, these findings indicate that therapeutic delivery of TNF-α accelerated the resolution of established pulmonary fibrosis, whereas genetic deficiency of TNF-α prolonged the fibrotic response.

TNF-α Reduces Profibrotic Alternative Macrophage Programming in Fibrotic Lungs

We hypothesized that the mechanism of TNF-α resolution of fibrosis involved a reduction in profibrotic alternative macrophage programming. To address this hypothesis, pulmonary macrophages were evaluated by flow cytometry, as described (30, 31). Macrophages were identified as CD11c⁺CD11b^varF4/80⁺Ly6G⁻, a subset that includes resident alveolar macrophages and recruited alveolar and interstitial inflammatory macrophages, or as CD11c⁻CD11b⁺F4/80⁺Ly6G⁻, a subset that primarily contains naive interstitial macrophages (Figure 4A). Alternative and classical macrophage programming markers were evaluated by quantifying the MFI of CD206 (mannose receptor) and intracellular NOS2 staining, respectively, in these two distinct subsets. Compared with saline-instilled mice, the number of CD11c⁺CD11b^varF4/80⁺ macrophages was increased at the peak of fibrosis in bleomycin-instilled mice at Week 3 but returned to baseline levels by Week 6 (Figure 4B, black bars). CD11c⁻CD11b⁺F4/80⁺ macrophage numbers were relatively unaffected by bleomycin instillation (Figure 4B, white bars). Of these subsets, CD11c⁺CD11b^varF4/80⁺ macrophages alone exhibited alternative programming markers at the peak of fibrosis at 3 weeks, as revealed by increased CD206 expression, which decreased during resolution (P < 0.05) (Figure 4C, black bars). NOS2 expression was low in both macrophage subsets and was largely insensitive to bleomycin instillation (Figure 4C). We confirmed these findings by confocal microscopy of lung sections stained with fluorescently tagged antibodies against arginase I (another marker of alternative programming), NOS2, and F4/80 (Figure 4D). Figures 4D and 4E shows that the majority of F4/80⁺ macrophages in lung sections from bleomycin-instilled mice expressed arginase I and low levels of NOS2. Arginase catalytic activity similarly increased in lung homogenates from fibrotic bleomycin-instilled mice at 3 weeks and decreased toward control levels as fibrosis resolved (Figure 4F). Taken together, these data indicate that alternative programming was increased only in CD11c⁺CD11b^varF4/80⁺ lung macrophages during the development of pulmonary fibrosis, whereas NOS2 was basally expressed at low levels by both macrophage subsets.

On the basis of these results, we questioned whether the accelerated resolution of fibrosis in TNF-α–treated mice was associated with alterations in macrophage programming. Fibrotic, bleomycin-instilled mice were given TNF-α at Weeks 3 and 4 and assessed for macrophage programming at Week 5 (Figure 5B). Compared with saline-instilled fibrotic mice, TNF-α treatment reduced the number of CD11c⁺CD11b^varF4/80⁺ macrophages (P < 0.01) (Figure 5B, black bars) and the expression of CD206 in the remaining cells (P < 0.01), whereas NOS2 expression was modestly increased (P < 0.01) in this subset (Figure 5C). TNF-α treatment had no effect on the number of CD11c⁻CD11b⁺F4/80⁺ macrophages or their programming status (Figures 5B and 5C). The changes in CD206 and NOS2 expression (Figure 5B) in the CD11c⁺CD11b^varF4/80⁺ subset occurred in a single population of dual-positive cells that expressed variable levels of both markers. Immunostaining of lung sections for F4/80, arginase I, and NOS2 confirmed the down-regulation in macrophage arginase I expression and the reciprocal increase in NOS2 expression in TNF-α–treated fibrotic mice compared with saline-instilled fibrotic mice (Figure 5C). Analysis of pixel density also revealed an increase in NOS2 expressing F4/80⁺ macrophages after treatment with TNF-α, consistent with the flow cytometry data (P < 0.001) (Figure 5D). Measurements of lung arginase activity also confirmed that TNF-α instillation reduced arginase levels (P < 0.05) (Figure 5E).

Because TNF-α accelerates the resolution of established fibrosis in conjunction with reduced macrophage numbers and alternative programming status, genetic TNF-α deficiency would be predicted to prolong alternative programming in the persistently fibrotic lungs of TNF-α^−/− mice. Immunostaining of lung sections from bleomycin-instilled wild-type and TNF-α^−/− mice confirmed the predominant alternative macrophage programming response at 3 weeks in both strains (Figure 6A). However, unlike wild-type mice that exhibited a decline in arginase I staining to basal levels by Week 8, greater that 75% of TNF-α–deficient, F4/80⁺ lung macrophages continued to express arginase I at 6 and 8 weeks (Figure 6B, blue bars). Levels of NOS2 expression (Figure 6A) and the percentage of NOS2-positive cells (P = 0.912) (Figure 6B, green bars) were unaffected by the absence of TNF-α. Elevated lung arginase activity confirmed the persistence of this alternative macrophage programming-associated enzyme in the lungs of TNF-α^−/− mice (Figure 6C).

To determine if TNF-α directly affected macrophage programming, wild-type bone marrow–derived macrophages were alternatively programmed in vitro with recombinant IL-4 and IL-13. The alternatively programmed macrophages were then incubated with recombinant murine TNF-α and analyzed for changes in CD206 expression and arginase activity. Figure E2 shows that TNF-α significantly reduced the expression of CD206 (P < 0.001) and arginase activity (P < 0.01), suggesting that TNF-α qualitatively reduces alternative macrophage programming.

Taken together, our findings indicate that the accelerated resolution of established fibrosis occurring in the presence of TNF-α is accompanied by a specific reduction in CD11c⁺CD11b^varF4/80⁺ macrophages that have decreased expression of alternative macrophage programming markers and increased expression of NOS2. In contrast, the impaired resolution of fibrosis seen in TNF-α–deficient mice was associated with the persistence of alternatively programmed macrophages and a failure to up-regulate NOS2. Furthermore, TNF-α was able to directly reduce the expression of the alternative macrophage programming marker CD206 in bone marrow–derived macrophages in a manner similar to what was observed in vivo.

Loss of Inflammatory Lung Macrophages Promotes the Resolution of Established Pulmonary Fibrosis

We next addressed the importance of the loss of alternatively programmed lung macrophages in the resolution of fibrosis by conditionally depleting macrophages after fibrosis had been established. MAFIA mice were intratracheally instilled with bleomycin and monocytes, and macrophages were depleted by intravenous delivery of AP20187 between Weeks 2 to 3.5. Fibrosis and inflammatory indices were then assessed at Week 4 (Figure 7A). Fibrotic, bleomycin-instilled MAFIA mice treated with vehicle exhibited the expected increase in CD11c⁺CD11b^varF4/80⁺ lung macrophages (Figure 7B). Injection of AP20187 into fibrotic MAFIA mice reduced the number of CD11c⁺CD11b^varF4/80⁺ macrophages (P < 0.05) (Figure 7B, black bars) but had no effect on the number of alveolar macrophages in BAL (Figure 7B, red bars). Macrophage depletion in these mice reduced lung collagen (P < 0.001) (Figure 7C), restored compliance (P < 0.01) (Figure 7D), resulted in reduced BALF protein levels (P < 0.01) (Figure 7E), and led to an improvement in lung architecture (Figure 7F). Furthermore, macrophage depletion resulted in an almost complete loss of lung arginase activity, confirming that we had depleted alternatively programmed macrophages (Figure 7G). Akin to the lungs of TNF-α–treated fibrotic mice, 60% of the F4/80⁺ macrophages remaining in the lungs of AP20187-treated fibrotic MAFIA mice expressed an intermediate programming state reflected by the presence of arginase I and NOS2 (Figures 7H and 7I). Together, our findings show that the resolution of established fibrosis after conditional macrophage depletion reduces the number of alternatively programmed macrophages and phenocopies pulmonary delivery of TNF-α.