Cutting edge: FasL(+) immune cells promote resolution of fibrosis.

Journal of Autoimmunity xxx (2015) 1e10

Contents lists available at ScienceDirect

Journal of Autoimmunity journal homepage: www.elsevier.com/locate/jautimm

Cutting edge: FasLþ immune cells promote resolution of fibrosis Shulamit B. Wallach-Dayan a, *, Liron Elkayam a, Regina Golan-Gerstl a, Jenya Konikov a, Philip Zisman a, Mark Richter Dayan b, Nissim Arish a, Raphael Breuer a, c a

Lung Cellular and Molecular Biology Laboratory, Institute of Pulmonary Medicine, Hadassah e Hebrew University Medical Center, Jerusalem, Israel Department of Emergency Medicine, Shaare Zedek Medical Center, Jerusalem, Israel c Department of Pathology, Boston University School of Medicine, Boston, MA, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 June 2014 Received in revised form 26 January 2015 Accepted 23 February 2015 Available online xxx

Immune cells, particularly those expressing the ligand of the Fas-death receptor (FasL), e.g. cytotoxic T cells, induce apoptosis in ‘undesirable’ self- and non-self-cells, including lung fibroblasts, thus providing a means of immune surveillance. We aimed to validate this mechanism in resolution of lung fibrosis. In particular, we elucidated whether FasLþ immune cells possess antifibrotic capabilities by induction of FasL-dependent myofibroblast apoptosis and whether antagonists of membrane (m) and soluble (s) FasL can inhibit these capabilities. Myofibroblast interaction with immune cells and its FasL-dependency, were investigated in vitro in coculture with T cells and in vivo, following transplantation into lungs of immune-deficient syngeneic Rag/ as well as allogeneic SCID mice, and into lungs and air pouches of FasL-deficient (gld) mice, before and after reconstitution of the mice with wild-type (wt), FasLþ immune cells. We found that myofibroblasts from lungs resolving fibrosis undergo FasL-dependent T cell-induced apoptosis in vitro and demonstrate susceptibility to in vivo immune surveillance in lungs of reconstituted, immune- and FasL-deficient, mice. However, immune-deficient Rag/ and SCID mice, and gld-mice with FasL-deficiency, endure the accumulation of transplanted myofibroblasts in their lungs with subsequent development of fibrosis. Concomitantly, gld mice, in contrast to chimeric FasL-deficient mice with wt immune cells, accumulated transplanted myofibroblasts in the air pouch model. In humans we found that myofibroblasts from fibrotic lungs secrete sFasL and resist T cell-induced apoptosis, whereas normal lung myofibroblasts are susceptible to apoptosis but acquire resistance upon addition of anti-s/ mFasL to the coculture. Immune surveillance, particularly functional FasLþ immune cells, may represent an important extrinsic component in myofibroblast apoptosis and serve as a barrier to fibrosis. Factors interfering with Fas/FasL-immune cellemyofibroblast interaction such as sFasL secreted by fibrotic-lung myofibroblasts, may abrogate immune surveillance during fibrosis. Annulling these factors may pave a new direction to control human lung fibrosis. © 2015 Elsevier Ltd. All rights reserved.

Keywords: FasL sFasL Immune cells Lung fibrosis resolution Myofibroblast apoptosis

1. Introduction Patients with idiopathic pulmonary fibrosis (IPF) manifest increased levels of soluble Fas-death receptor ligand (sFasL) in the circulation [1] and in fluids obtained via broncheoalveolar lavage

* Corresponding author. Institute of Pulmonary Medicine, HadassahdHebrew University Medical Center, POB 12000, Jerusalem 91120, Israel. Tel.: þ972 2 6776622; fax: þ972 2 6435897. E-mail addresses: [email protected] (S.B. Wallach-Dayan), [email protected] (L. Elkayam), [email protected] (R. Golan-Gerstl), [email protected] (J. Konikov), [email protected] (P. Zisman), [email protected] (M.R. Dayan), [email protected] (N. Arish), Raffi@hadassah.org.il (R. Breuer).

(BAL) [2], which correlate with disease activity [1]. It has been previously reported that normal myofibroblasts, but not those from fibrotic lungs with IPF, are susceptible to apoptosis induced by Fas agonists [3,4]. Membrane FasL can also induce cell apoptosis, and the soluble form of Fas (sFas) [4] or sFasL [5] contributes to resistance to Fas-induced apoptosis. Immune cells in general, and particularly cytotoxic T cells that express FasL [6], induce apoptosis in ‘undesirable’ self and nonself cells bearing the Fas-death receptor, including lung fibroblasts [6,7]. FasL-expressing immune cells may thus provide a means of immune-surveillance (e.g. in cancer) [8,9]. FasL is primarily expressed on T cells after their activation, whereas Fas, a 45-kDa type I cell surface receptor that belongs to the TNF receptor family, is constitutively expressed on most cell

http://dx.doi.org/10.1016/j.jaut.2015.02.006 0896-8411/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: S.B. Wallach-Dayan, et al., Cutting edge: FasLþ immune cells promote resolution of fibrosis, Journal of Autoimmunity (2015), http://dx.doi.org/10.1016/j.jaut.2015.02.006

2

S.B. Wallach-Dayan et al. / Journal of Autoimmunity xxx (2015) 1e10

surfaces [10]. FasL is a 37-kDa type II glycoprotein that acts through cellecell contact [11,12]. We have shown that FasL expression levels are increased following mouse lung bleomycin (BLM) injury and during evolution of fibrosis [13]. In parallel, myofibroblasts from lungs with active fibrosis acquired an ‘immune-privilege-like’ phenotype [14] with FLIP-mediated Fas signaling of proliferation rather than apoptosis [15], permitting their unremitting accumulation. Here we sought to determine the ability of immune cells to induce apoptosis and removal of myofibroblasts from coculture in vitro and from lungs in vivo, and to understand the role of FasL expressed by immune cells and of sFasL, which is increased in IPF, in interactions leading to the resolution vs. evolution of fibrosis. We show that FasLþ immune cells play a critical role in dissipating myofibroblasts from the lungs following BLM injury. Our data reveal a clear linear correlation between in vivo aSMA (myofibroblast) cell exclusion from the lungs following injury and resolution of fibrosis, with increased myofibroblast susceptibility to T cell-induced apoptosis. We further show that myofibroblasts from lungs resolving fibrosis, which we show to lose their capability of evading immune-surveillance, can freely accumulate in lungs of immune-and FasL-deficient mice and trigger fibrosis. In humans, IPF lung myofibroblasts release sFasL and resist Fasinduced apoptosis. In contrast, fibroblasts from normal human lungs can be killed by T cells with dependence on Fas/FasL interaction. Pretreatment of T cells with FasL antagonists before coculture rendered normal-human lung myofibroblasts resistant to T cell-induced apoptosis. We demonstrate the effects of immune surveillance and the role of FasL in the interplay of human, as well as mouse lung myofibroblast apoptosis and the impact on resolution of fibrosis. Taken together this implies that the immune system is capable of combating fibrosis by FasL-mediated immune surveillance, and that sFasL released by IPF-lung myofibroblasts may represent an additional mechanism of escape from immune surveillance.

approved by the University of Pittsburgh Institutional Review Board. Patients with IPF were confirmed to have histopathology characteristic of usual interstitial pneumonia (UIP), without evidence of other known causes. Normal lung tissue obtained from organ donors or from patients undergoing lung biopsy for tumor diagnosis, was cultured and some fibroblasts were grown in culture. After expansion in culture, fibroblasts were determined to be “normal-lung myofibroblasts.” (For details see the Online Data Supplement [ODS] Fig. S1). All human lung fibroblasts were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, penicillin, streptomycin, anti-mycotic agent, sodium pyruvate, L-glutamine, and MEM non-essential amino acids, all purchased from Biological Industries (Beit HaEmek, Israel). LL 97A (AlMy) (ATCC® CCL-191™)-IPF-lung, and LL 24 (ATCC® CCL-151™)-normal lung fibroblast cell lines were also used in all the experiments with human cells. Mouse cells were obtained as we previously detailed and as detailed in ODS Fig S1.

2. Materials and methods

2.5. Isolation of lung myofibroblasts

2.1. Reagents

Myofibroblast isolation was described by us in detail elsewhere [13,14]. Details can be found in the ODS.

Monoclonal antibodies (mAbs) and other reagents included 1 mg/ml propidium iodide (PI) stock solution in PBS purchased from Calbiochem® (San-Diego, CA, USA); CFSE from Molecular Probes® (Eugene, Oregon, USA); rabbit anti-human/mouse activecaspase-3 mAb for Western blot purchased from Cell Signaling Technology, Inc. (Boston, MA, USA); Cy3-anti-aSMA, Cy2-anti-CD3, and Cy2-anti-caspase 3 mAbs for IHC purchased from Jackson Laboratories (Bar Harbor, ME, USA) anti-GFP mAb for IHC purchased from ClonTech Laboratories (Mountain View, CA, USA); antiGFP and anti-aSMA mAbs for FACS purchased from SigmaeAldrich (St. Louis, MO, USA); Annexin V-FITC-Cy5 purchased from BioVision, Inc.(Milpitas, CA, USA); as well as purified NA/LE FasL antagonist rat anti-mouse MFL3, CD178 mouse anti-human NOK-1 mAbs, purified NA/LE agonist Jo2 rat anti-mouse or mouse antihuman CD95 (Fas) DX2 mAb, purchased from BD Pharmingen™ (Oxford, UK). 2.2. Human primary and cell-line lung myofibroblasts 2.2.1. Human tissues and cells IPF fibroblasts, isolated from lung tissues, were obtained from Dr. Carol Feghali-Bostwick (Pittsburgh University Medical Center, Pittsburgh, PA, USA). Fibroblasts were cultured from the explanted lungs of patients with IPF who underwent lung transplantation, as previously described by Dr. Feghali-Bostwick [16], under a protocol

2.3. Animals C.B-17/IcrHsd-SCID, BALB/c, C57BL/6 Rag/ and C57BL/6 mice (5e6 or 11e12 weeks old) were from Harlan Sprague Dawley (Indianapolis, IN, USA); Tg(ACTBEGFP)10sb mice (11e12 weeks old) and gld C57BL/6-based mice (5e6 weeks old) from Jackson Laboratory. Mice were maintained under specific pathogen-free conditions in the Animal Unit of the Hebrew University-Hadassah School of Medicine with adherence to institutional guidelines for the care and use of laboratory animals. 2.4. Intratracheal (IT) instillation and induction of fibrosis in mouse lungs We have performed and detailed IT BLM previously [14].

2.6. Chimeric mice and adoptive transfer of lymphoid cells Generation of chimeric mice has been reported by us elsewhere [13,14]. Description in ODS. 2.7. Intratracheal transplantation (ITT) of GFPþ lung myofibroblasts ITT of d 28-BLM GFPþ myofibroblasts (5 106) previously isolated from EGFP-C57BL/6 mice was performed into gld/gld vs. gld/ wt, BALB/c vs. SCID, or into reconstituted Rag/ vs. Rag / host mice. The mice were killed 7e14 days after myofibroblast IT (Pentothal, 100 mg/mouse in 0.5 ml, CTS Chemical Industries, Kiryat Malachi, Israel). 2.8. Quantitative assessment of fibrotic lung injury As we described previously [14,17]. 2.9. Immunohistochemical staining of lung tissue sections Lung section immunohistochemical (IHC) techniques were as previously described [13,14].



3

2.10. Immunofluorescence staining of lung tissue sections

3. Results

As previously detailed [13,14], using anti-aSMA rabbit anti-CD3 mAb (Jackson Laboratories).

3.1. Resolution of fibrosis and myofibroblast clearance from the lungs of BLM-treated C57BL/6 mice correlates with their susceptibility to T cell-induced apoptosis

2.11. Ariol imaging

The detection of FasL was performed by Western blot as we have previously described [15]. Briefly, cells were lysed in NP-40 lysis buffer. For analysis of sFasL the conditioned medium was subjected to ultracentrifugation (100,000 g) for 2 h as described previously [20]. GAPDH was used as a loading control.

Following a single IT instillation of BLM (0.05 mU) into the lungs of C57BL/6 mice, a fibrotic pathology (Fig. 1A) with correlating massive accumulation of a-smooth muscle actin (aSMA)-myofibroblasts, characteristic of fibrosis, was detected by day 14 (Fig. 1B). However, at days 28 and 56, there was a gradual disappearance of myofibroblasts with concomitant resolution of fibrotic pathology (Fig. 1A and B, respectively). A semiquantitative morphological index (SMI) was used to assess lung tissue damage (Fig. 1C) and a computerized quantitative image analysis (Ariol SL-50, Genetix Ltd., New Milton, UK) to score the frequency of aSMAþ cells (Fig. 1D). The Spearman correlation coefficient showed a strong positive association between reduced aSMA staining (myofibroblast disappearance) and a lower SMI (resolution of fibrosis) (Fig. 1E). Primary lung myofibroblast cultures were obtained from BLMtreated mice 1, 14, 28, and 56 days following IT, after second and third passage, and maintained in RPMI medium (Fig. S1). When cocultured with activated syngeneic CD4þ T cells isolated from naïve wt mice at a ratio of 1:10 for 24e48 h, myofibroblasts from lungs resolving fibrosis (days 28, 56) reacted very similarly to prefibrosis lung myofibroblasts (day 1) and, as we showed previously, to normal myofibroblasts [14]. Their number (Fig. 1F) and survival (Fig. 1G1) markedly decreased, and their apoptosis increased (Fig. 1H1, Fig. S2), compared to myofibroblasts cultured alone (Ctrl) and to fibrotic-lung myofibroblasts (day 14). We found a positive linear correlation between in vivo myofibroblast accumulation (high aSMA) and fibrosis (high SMI) with increased myofibroblast survival in vitro (Fig. 1G2). There was a concomitant significant negative linear correlation between myofibroblast accumulation in vivo and apoptosis in vitro (Fig. 1H2). This suggests that myofibroblast accumulation and consequent fibrosis in the lungs of C57BL/6 mice following IT-BLM are parallel, but also reversible following downregulation of resistance to immune cell-induced apoptosis.

2.16. Data analysis and statistics

3.2. FasL is essential to the cytotoxicity exerted by immune cells against myofibroblasts from lungs resolving fibrosis

IHC-slides stained with aSMA mAb were digitized (Ariol, Genetix, New Milton, UK) and semiquantitatively analyzed (Ariol system) [18]. Details in ODS. 2.12. Murine skin air pouch model and track of injected GFPþ lung myofibroblasts Dorsal air pouches were raised in host mice as we described [14]. See ODS. 2.13. CD4þ T cell isolation In vivo activated lymphoid organ-T cells were separated by specific columns (Biotech Laboratories, Houston, TX, USA) from wt or syngeneic gld mice, and anti-CD44 CD62L Abs were added, as described [14]. Activated CD4þ T cells showed increased FasL.

2.14. In vitro detection of apoptotic cells by Annexin V affinity labeling As described elsewhere [13,14,19].

2.15. Detection of mFasL and sFasL

Each experiment was repeated 2e3 times for each set of fibroblasts, which included 4e5 mice/human fibroblasts (n ¼ 4e5). Data were then pooled for statistical analysis. Western blot analyses and proliferation experiments for each separate experiment were repeated 3e4 times, depending on the experiment. Data are expressed as means ± SD. Statistical analyses of the Western blots data analysis was performed using the Student's t test with ManneWhitney modification or analysis of variance as applicable. The nonparametric KruskalleWallis test was applied to compare variables measured at different time intervals or following different treatments in vivo. Multiple pairwise comparisons were performed using the ManneWhitney nonparametric test with the Bonferroni correction for significance. The Spearman nonparametric correlation coefficient was calculated to assess associations between pairs of variables. Twoway ANOVA was used to assess time and treatment effects and the interaction between them. Using this statistical model, the Scheffe post-hoc procedure was applied for multiple pairwise comparisons. All statistical tests were two-tailed, and a p-value of 5% or less was considered significant.

CD4þ T cells from FasL-deficient FasL (gld) mice (Fig. 2A) and WT FasLþCD4þ T cells treated with MFL3 anti-FasL mAb-antagonist (Fig. 2B) showed diminished capability to deplete from coculture myofibroblasts isolated from lungs during resolution of fibrosis (day 28), in comparison with control wt- FasLþCD4þ T cells. The myofibroblast culture alone group (Ctrl) is presented in (Fig. S3a). This difference was consistent with myofibroblast survival (Fig. 2C and D, respectively) and apoptosis (Fig. 2E and F, respectively). In contrast, as we have previously shown with wt CD4þ T cells, [14,15] gld CD4þ T cells failed to induce apoptosis in day 14 Fas-resistant and ‘immune-privileged’ myofibroblasts from lungs with active fibrosis (Supplemental Fig. S3b). 3.3. Myofibroblast accumulation is restricted and evolution of fibrosis is impaired in the lungs of BLM-treated FasL-deficient gld mice reconstituted with FasLþ immune cells Lung sections from BLM-treated FasL (gld) chimeric mice that were reconstituted with FasLþ immune cells from WT mice (wt/gld) (Fig. S4) [13,14] were compared to sections from BLM-treated control chimeric FasL mice reconstituted with gld mice FasL


4


Fig. 1. Correlation between the disappearance of myofibroblasts from fibrotic lungs, their susceptibility to T cell-induced apoptosis, and resolution of fibrosis Lung sections from 4 to 5 C57BL/6 mice at days 1, 14, 28, and 56 following intratracheal (IT) instillation of bleomycin (BLM). (A) Hematoxylin-eosin (H&E) staining. (B) Immunohistochemistry aSMA staining (arrows). Representative micrographs of 12e15 sections at each time point. (C) Semi-quantitative morphological index (SMI) of tissue damage, scored 0e5, *p ¼ 0.008, and (D) morphometric analysis of aSMA frequency in 10 random fields per mouse using Ariol imaging, *p ¼ 0.02. (E) Spearman correlation coefficient plotting the percentage of aSMAþ stained cells and lung tissue damage, as assessed by SMI, in lung tissue sections; r ¼ 0.904, *p < 0.001. (F) Light microscopy images and trypan blue cell counts (±SD) of myofibroblasts (inserted numbers) following 24 h coculture with CD4þ T cells. Myofibroblasts were isolated from mouse lungs before development of fibrosis (day 1), during active fibrosis (day 14), and during resolution (days 28 and 56). Representative results of three independent experiments showing constant differences between aSMAþ cells at different time points of fibrosis, ex vivo. (G1) Percentage of surviving myofibroblasts after 24 h of coculture from each time point of IT-BLM, as assessed by trypan blue exclusion; *p ¼ 0.021. Survival of myofibroblasts at 24 h of coculture without T cells served as the control (100%). (G2) Spearman correlation coefficient for myofibroblast survival in vitro and aSMA staining in vivo, in lung tissue sections; r ¼ 0.811, p < 0.001. (H1) Percentage of Annexin Vpositive and propidium iodide (PI)-negative lung myofibroblasts following coculture, as assessed by flow cytometry, *p ¼ 0.029. (H2) Spearman correlation coefficient for myofibroblast apoptosis in vitro and aSMA staining in vivo in lung tissue sections; r ¼ 0.527, p < 0.05.

immune cells (gld/gld). Lung sections from mice reconstituted with FasLþ immune cells had no pathological evidence of fibrosis and stained negative for aSMA, indicating that FasLþ immune cells introduced into FasL mice abrogate the accumulation of lung myofibroblasts, despite BLM injury (Fig. 3A). Concomitantly,

Fig. 2. FasL is essential to the cytotoxicity exerted by CD4þ T cells against myofibroblasts from lungs resolving fibrosis. (AeB) Light-microscopy images and fibroblast cell counts ±SD (inserted numbers), (CeD) trypan blue exclusion (survival), and (EeF) percentage of Annexin V-positive staining of lung myofibroblasts isolated at different time points post IT-BLM following their coculture with T cells. Myofibroblasts were cocultured for 24 h with (A, C, E) CD4þ T cells from wild-type (wt) vs. FasL (gld) mice or, (B, D, F) CD4þ T cells (wt), with the addition of 10 mg/ml MFL3 anti-FasL mAb or control SHAM mAb. One of two experiments, n ¼ 4, *p < 0.03.

collagen accumulation in chimeric wt/gld mouse lungs was markedly decreased (Fig. 3B). To rule out the possibility that FasLþ immune cells affect the lymphocyte population, we performed broncheoalveolar lavage (BAL). We found no change in the extent of lung inflammation in FasLþ chimeric wt/gld compared with FasL gld/gld chimeric control BLM-treated mice, as assessed by total cell count and lymphocyte cell counts (Fig. 3C). We then directly assessed whether myofibroblast accumulation in gld/gld mice was due to impaired removal. To this end, ITT of exogenic GFPþ myofibroblasts from lungs resolving fibrosis (day 28) was performed into gld/gld- and chimeric wt/gld mice (Fig. S5). gld/gld mice with FasL immune cells exhibited a noticeable defect in the removal of GFPþ exogenous myofibroblasts from lung interstitium compared with chimeric wt/gld mice with FasLþ



5

Fig. 3. FasLþ immune cells increase myofibroblast apoptosis, impede their accumulation, and decrease collagen deposition in BLM-treated lungs of gld mice. (A) Immunohistochemistry using aSMA staining (brown), in lung sections of control chimeric gld (with FasL immune cells) vs. chimeric gld mice (with FasLþ immune cells), 14 days post BLM IT instillation. Representative of 15 fields (20) in each mouse; n ¼ 4. Arrows indicate positive staining. (B) Lung collagen in BLM-treated FasL chimeric gld vs. FasLþ chimeric gld mice; *p ¼ 0.026. (C) Bronchoalveolar lavage (BAL) total cell count and percentage of lymphocytes in BLM-treated FasL chimeric gld vs. FasLþ chimeric gld mice. Representative results of two experiments. (D) Immunohistochemistry using GFP staining performed as in A. (E) Flow cytometry analysis quantifying the percentage of GFPþ myofibroblasts in lungs of FasL chimeric gld vs. FasLþ chimeric-gld mice. (F) Confocal-microscopy of lung tissue double staining using Cy3-anti-aSMA (red) and Cy2-anti-CD3 (green) or Cy2-anticaspase 3 (yellowegreen) mAbs in gld mice with FasLþ vs. FasL immune cells. Images from a 63, 1.4 oil immersion, zoom 3 Zeiss lens are also presented. Representative staining from two independent experiments (n ¼ 5). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

immune cells (Fig. 3DeF). Lung sections stained with anti-GFP mAb (Fig. 3D) and quantified by FACS analysis (Fig. 3E) revealed a twofold increase in the percentage of accumulated GFPþ myofibroblasts calculated from the total myofibroblast population in lungs from chimeric gld/gld vs. chimeric wt/gld host-mice. Using confocal microscopy, we further determined in vivo in lung sections whether aSMAþ cells in BLM-treated lungs of mice with FasLþ immune cells undergo increased apoptosis compared to mice with FasL immune cells. To this end, lung sections from BLMtreated chimeric gld mice reconstituted with FasLþ immune cells (wt/gld) and control gld/gld mice with FasL immune cells were double-stained with anti-aSMA and either anti-CD3 or anti-caspase 3 Abs. Compared to CD3þ cells, aSMAþ cells were almost absent in lungs of gld mice reconstituted with FasLþ immune cells (Fig. 3F; aSMA and CD3 in FasLþ vs. FasL). Moreover, aSMAþ cells from lungs of gld mice reconstituted with FasLþ immune cells stained positive for caspase 3 activity, as determined by colocalization of aSMA and caspase 3 (Fig. 3F; aSMA and caspase 3 in FasLþ), indicating their apoptotic state. In contrast, there was almost no caspase 3 activity in aSMAþ cells, with resulting increased aSMAþ cell accumulation in lung sections of BLM-treated control gld/gld mice (Fig. 3F; FasL).

3.4. Absence of an adaptive immune response in allogeneic mouse lungs results in uncontrolled accumulation of exogenic GFPþ myofibroblasts, with subsequent fibrosis In order to be removed, myofibroblasts from lungs resolving fibrosis should lose the ability to escape in vivo immune surveillance, a feature that we have found to characterize fibrotic-lung myofibroblasts in air pouches raised in allogeneic BALB/c mice [14] (Fig. S6). To this end, we performed ITT of GFPþ myofibroblasts from the lungs of BLM-treated Tg(EGFP) C57BL/6 mice resolving fibrosis (d28) into air pouches of BALB/c mice. Using immunohistochemistry in host lung tissue sections (Fig. 4A; WT), and FACS-flow cytometry of isolated host-lung myofibroblasts (Fig. 4B; WT), we demonstrate that GFPþ myofibroblasts from lungs resolving fibrosis were totally removed from allogeneic host wt BALB/c mouse lungs. Similarly, we found no traces of GFPþ myofibroblasts in the air pouches, as assessed by fluorescent microscopy (Fig. 4C; WT) and analyzed quantitatively by integrated optical density (IOD) arbitrary units using Image-Pro Plus software (Media Cybernetics, Rockville, MD, USA) (Fig. 4D; WT). To determine whether adaptive immunity is involved in myofibroblasts lung clearance, we performed ITT of myofibroblasts from


6


Fig. 4. Absence of adaptive immune response in allogeneic mice results in uncontrolled accumulation of exogenic GFPþ myofibroblasts in the air pouch and lungs, and in subsequent lung fibrosis (A) Immunohistochemistry of GFP staining (brown) in lung sections of allogeneic BALB/c WT-vs. SCID host mice following IT injection of GFPþ myofibroblasts of lungs resolving fibrosis (day 28). Arrows indicate positive GFP (myofibroblast) staining. Representative of 10e12 fields, 20. (B) Flow cytometry analysis quantifying the percentage of GFPþ myofibroblasts from lungs resolving fibrosis in WT vs. SCID mouse lungs; *p ¼ 0.05. Representative results of two independent experiments (n ¼ 4). (C) Carboxyfluorescein diacetate succinimidyl ester (CFSE) green fluorescence of CFSE-stained myofibroblasts from lungs resolving fibrosis in allogeneic WT vs. SCID mouse air pouches. Detection was performed using a fluorescent binocular microscope (Zeiss), and Nikon Coolpix digital camera. Representative results of three independent experiments (n ¼ 5); *p < 0.01. (D) Individual values of integrated optical density (IOD) are presented for day 28 myofibroblasts in WT and SCID mice. (E) IHC, hematoxylin and eosin (H&E) staining, in WT vs. SCID mice following GFPþ myofibroblast transplantation. (F) Semiquantitative morphological index (SMI) of tissue damage, scored 0e5; *p ¼ 0.03. (G) Lung collagen in WT vs. SCID mice following GFPþ myofibroblast-transplantation; *p ¼ 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

mice lungs resolving fibrosis (d28), into BALB/c-C.B-17 mice with severe combined immunodeficiency (SCID) and consequent impaired adaptive immune response (Fig. S7). In strong contrast to the complete loss of GFPþ detected in C.B-17 WT mice, a marked accumulation of exogenous GFPþ myofibroblasts was found in the lungs (Fig. 4AeB; SCID) and air pouches (Fig. 4CeD; SCID) of the C.B-17 SCID mice, indicating a regulatory role for immune cells (e.g. T cells) in the accumulation of lung myofibroblasts. With hematoxylin and eosin (H&E) staining (Fig. 4E) and SMI analysis (Fig. 4F), we further found pathological evidence of fibrosis with increased collagen deposition (Fig. 4G), in the lungs of C.B-17 SCID mice that had undergone IT instillation of GFPþ lung myofibroblasts, demonstrating that impaired immune surveillance of lung myofibroblasts results in the development of lung fibrosis. 3.5. Following reconstitution, syngeneic Rag/ mice dissipate exogenic GFPþ myofibroblasts from their lungs We then determined whether GFPþ myofibroblasts are directly dissipated in vivo by syngeneic, rather than allogeneic, immune cells. We found that reconstitution of C57BL/6-Rag/ mice that had undergone ITT of GFPþ C57BL/6 d28 myofibroblasts (Fig. S8) restored immune surveillance by suppressing GFPþ fibroblast

accumulation, as assessed by GFP staining in FACS analysis as well as immunohistochemistry (Fig. 5A and inserts, respectively). Moreover, using H&E we found that reconstitution of Rag/ mice abolished the development of fibrotic foci in these mice (Fig. 5B). We tracked growth rates of myofibroblasts in vivo [21] in syngeneic mouse air pouches and in vitro by assessment of carboxyfluorescein diacetate succinimidyl ester (CFSE) staining [22]. There were no differences in the in vitro or in vivo proliferation at day 28 vs. day 14 of GFPþ myofibroblasts, excluding effects of this parameter on myofibroblast accumulation (Fig. S9).

3.6. Human IPFe rather than normal lung myofibroblasts secrete sFasL and resist T cell-induced apoptosis. However, normal lung myofibroblasts acquire resistance to apoptosis following pretreatment of cocultured T cells with anti-FasL mAb Myofibroblasts from normal human lungs, in contrast to those isolated from the lungs of humans with IPF, that were cocultured for 48 h with previously activated (5 mg/ml Con-A [Sigma] for 2 h) human primary- or Jurkat-T cells, were cleared from the coculture and apoptosized, as determined by trypan-blue exclusion (Fig. 6A; Normal- and IPF-Fibs. þ T cells vs. Fibs. alone) and by assessment of



Fig. 5. Dissipation of syngeneic, transplanted GFPþ myofibroblasts in lungs of reconstituted Rag/ mice. (A) GFPþ cell staining in FACS, and in immunohistochemistry (arrows in inserts), demonstrating day 28 GFPþ myofibroblasts relative percentage in the lungs of Rag/ vs. reconstituted Rag/ mice. (B) IHC, H&E staining, in Rag/ vs. reconstituted Rag/ mice following GFPþ myofibroblast transplantation. Representative results of three independent experiments (n ¼ 5). In IHC, 7e8 fields were assessed (20).

the Annexin V marker of apoptosis with FACS (Fig. 6C2e3; Normal vs. IPF in inserts; Fibs. þ T cells vs. Fibs. alone), respectively. Since patients with IPF manifest increased sFasL in the circulation [1] and in BAL [2], which correlates with disease activity [1], we

7

then directly assessed whether myofibroblasts from IPF lungs can be one of the sources for sFasL. To this end we determined whether sFasL is released into culture medium from myofibroblasts isolated from lungs of five different patients with IPF compared to five normal subjects (Fig. 6B). Following ultracentrifugation of the culture medium, sFasL levels were determined in Western blot using specific NOK-1 anti s/mFasL mAb, and were found to significantly increase in IPF-lung myofibroblasts (Fig. 6B; sFasL in, Normal Fibs.vs. IPF Fibs. and Fig. S10A NL vs. IPF). Membrane, mFasL, was also determined in cultured cells and was found to have no change between IPF and normal lungs (see Fig. S10BeC). To equalize initial protein quantities GAPDH was detected and OD ratios of sFasL and mFasL to GAPDH were calculated (Fig. 6B; sFasL/GAPDH and Fig. S10B mFasL/GAPDH, respectively). Pretreatment of the human T cells with anti-s/mFasL NOK-1 antagonist mAb (1 mg, 30 min), before their addition to the coculture with human lung myofibroblasts, annulled their capability to induce normal lung myofibroblast apoptosis and clearance from coculture (Fig. 6C1e3; Fibs. þ T cells vs. Fibs. þ T cells þ anti-FasL) and had no effect on already resistant IPF-lung myofibroblasts (Fig. 6C1 IPF and inserts in Fig. 6C2e3 Fibs. þ T cells vs. Fibs. þ T cells þ anti-FasL), as determined by trypan-blue exclusion (Fig. 6C1) and Annexin V staining (Fig. 6C2). Differences in apoptosis fold ratios are presented (Fig. 6C3 for Fibs. þ T cells vs. Fibs. þ T cells þ anti-FasL). 4. Discussion Immune cells such as T lymphocytes were previously reported to be involved in both the attenuation and the promotion of fibrosis. These contradictory observations are most likely a reflection of the phenotypic heterogeneity of involved T cells, as

Fig. 6. Human IPFe as opposed to normal lung myofibroblasts secrete sFasL and resist T cell-induced apoptosis. However, normal-lung myofibroblasts acquire resistance to apoptosis following pretreatment of cocultured T cells with anti-FasL mAb. Normal- vs. IPF lung myofibroblasts cocultured with Jurkat cells (Fibs. þ T cells), vs. control culture containing only myofibroblasts (Fibs.) or cocultured with Jurkat cells previously treated with 10 mg/ml NOK-1 anti FasL mAb (Fibs. þ T cells þ anti FasL). (A and B1) Light microscope images and fibroblast cell counts ±SD (inserted numbers), and (B2) FACS Annexin V staining of lung fibroblasts isolated from normal human lungs or human fibrotic lungs (insert). (B3) Graphical presentation of Annexin V staining in normal vs. IPF (insert) lung myofibroblasts. One of two experiments, n ¼ 4, *p < 0.03. (C) Western blot analysis of sFasL released to the culture medium by normal-lung myofibroblasts (Normal Fibs.) vs. IPF-lung myofibroblasts (IPF Fibs.). Myofibroblasts were lysed in lysis buffer. For sFASL analysis the conditional medium was subjected to ultracenrifugation (100,000 g, 2 h). Optical density ratios of sFasL/GAPDH in five different patients with IPF and in five different normal subjects are presented. One of two experiments (n ¼ 5) with significant results (see Fig.S10A).


8


reviewed by Lo Re et al. [23] and Luzina et al. [24]. The significance of these issues and the extent of debate are reflected in the breadth of opinions regarding the importance and role of immune cells in scar resolution vs. evolution of pathologic-fibrosis [23e29]. The role of T cells in BLM-induced pulmonary fibrosis has been questioned [28]. We show here, in vitro and in vivo, their beneficial role. Immune cells, specifically human and murine T cells, can induce apoptosis in myofibroblasts from normal human lungs or from mouse lungs resolving fibrosis. Conditioned media of T cells that was added to myofibroblast culture in different concentrations failed to induce myofibroblast apoptosis (Fig. S10), indicating the need for a direct contact between the two cell types and excluding the possible effect of secreted factors in the induction of apoptosis. In addition, we show that with reconstitution of Rag/ mice the accumulation of previously transplanted exogenic GFPþ lung myofibroblasts is abated, and BALB/c, as opposed to congenic immune-deficient SCID mice, reject exogenic myofibroblast accumulation and inhibit fibrosis evolution. This cell death is dependent on Fas/FasL interaction. In this context, we further show here and elsewhere in vivo [14], that CD3þ cells are in close proximity to aSMAþ/caspase 3þ cells expressing the Fas death receptor [13], indicating the required contact and consequences of interaction between these cells. Moreover, we [13] and others [30] have shown in IPF patients that FasL is upregulated in inflammatory lung cells. We also show here that the in vivo accumulation of myofibroblasts and subsequent collagen deposition are markedly decreased in BLM-treated FasL gld mice following their reconstitution with WT (FasLþ) immune cells. Moreover, we found decreased fibrosis but similar lung inflammation in wt/gld vs. gld/gld mice, ruling out the possibility that FasLþ cells restrain other possible profibrotic immune cells as T cells, and excluding possible AICD and/or T celleT cell autoregulation. A delicate balance exists between FasLþ immune cells and FasLþ myofibroblasts [14]. In order to shift the balance towards an increase in FasLþ immune cells, we used chimeric gld mice, which harbor FasLþ immune cells and myofibroblasts with a ‘low FasL profile,’ rather than WT mice, which develop fibrosis even though they possess FasLþ immune cells [14]. It has been previously suggested that the presence of Fas/FasL interactions can indeed promote fibroblast apoptosis [31,32], and that mice with inactivating mutations in FasL (gld) show a decreased protective CD4þ Th1 response [33]. These mice spontaneously developed interstitial pneumonitis, lung hyperplasia with an increased number of fibroblasts, and lung pathology similar to that seen in patients with the connective tissue disease often associated with lung fibrosis [34]. Interactions of Fas/FasL in murine BLM-induced fibrosis are a multifaceted phenomenon [35,36]. Epithelial cell apoptosis has been hypothesized to play an important role in the development of pulmonary fibrosis, and the Fas/FasL pathway has been implicated as both a pro-apoptotic and profibrotic factor. However, we extensively discussed and ruled out the possibility that FasLþ immune cells promote apoptosis of Fasþ epithelial cells [13,14]. Of note, activation of lung fibroblasts from humans induces class II MHC antigen expression [37], promotes adhesion of fibroblasts to T cells [38], and even upregulates the display of costimulatory molecule CD40 [39]. Activation also results in Fas upregulation and sensitivity to Fas-mediated apoptosis upon ligation with CD40 ligand [40]. Why this possible antifibrotic mechanism fails in chronic diseases such as IPF is still under investigation. We have previously shown that IPF lung myofibroblasts overexpress FLIP, divert Fasinduced apoptosis towards proliferation, and escape immune surveillance [13,14]. In addition, IPF patients have increased levels of sFasL in their circulation [1] and broncheoalveolar lavage (BAL)

fluid [2], which was correlated with disease activity [1]. Concomitantly, we have found that myofibroblasts isolated from lungs of humans with IPF secrete increased levels of sFasL. It was previously reported that shedding sFasL acts to regulate membrane FasL (mFasL) cytotoxic activity [41e43]. In addition, sFas has been shown to downregulate the apoptotic activity of mFasL [4,44]. It is thus conceivable that sFas and/or the increased levels of sFasL detected in the sera and lungs of IPF patients govern the activity of mFasL, implying downregulation of immune surveillance exerted upon Fasþ myofibroblasts. In accordance, we show here that addition of recombinant sFas or of NOK-1 anti FasL mAb inhibited apoptosis of human lung myofibroblasts upon coculture with activated FasLþ Jurkat or primary T cells. Similarly, in mice, interference with FasL interactions via either MFL3 antibody or the gld mutation significantly affected the cytotoxicity of T cells in vitro, allowing myofibroblast survival. To conclude, we found an inverse correlation between resistance to immune cell-induced apoptosis and regression of fibrosis. We show that myofibroblasts from lungs resolving fibrosis have lost the ‘immune privilege-like’ phenotype acquired during active fibrosis [14]. However, using immunodeficient allogeneic-BALB/c (SCID), as well as syngeneic-Rag/ mice we found that in the absence of immune cells such as T lymphocytes, myofibroblasts from lungs resolving fibrosis can accumulate and induce fibrosis rather than dissipate. Interestingly, IPF patients with a higher percentage of lymphocytes may have a more benign course of disease [26] and Elicker et al. found that despite prolonged high-dose immunosuppression, fibrotic disease within the native lung progresses rapidly in single lung transplant [45]. In addition, pulmonary fibrosis has been linked to several immune deficiency-related diseases, including HIV [46] and adenosine deaminase (ADA) deficiency, which typically cause SCID in infants [47], as well as ataxia telangiectasia, and common variable immune deficiency [48]. In addition, a new syndrome of fatal lung fibrosis was reported to be clearly associated with immunodeficiency [49]. It may therefore be possible that in IPF, the immune system deviates from protective autoimmunity [50]. In support of this possibility, immune-suppressing drugs have been shown to have no capability to limit the extent of fibrosis [51]. Alternatively or additionally, it may be possible that following tissue damage, excessive myofibroblast proliferation [52] leads to cell senescence, which further enhances immune surveillance [53] and limits fibrosis, as reported in the liver [52]. However, our results show that both myofibroblasts from lungs with active fibrosis (d14) and from lungs resolving fibrosis (d28) possess comparable in vitro and in vivo proliferation capacities based on cell fluorescence patterns detected in culture and in air pouches of syngeneic C57BL/6 mice. Nevertheless, this very interesting issue of myofibroblast cell senescence in the lung during the progression of fibrosis remains open for further investigation. To date there is no known cure for interstitial lung diseases such as IPF, although Pirfenidone was recently shown to increase survival [54].

5. Conclusions We propose a new concept in the treatment of fibrosis, namely, restoration of the immune system's intrinsic anti-fibrotic capability (see scheme in Fig. 7). This tactic entails enhancing immune surveillance by FasLþ immune cells and blocking the evasion of fibrotic tissue fibroblasts from immune cell-induced apoptosis, thus preventing their uncontrolled accumulation. To date it has not been recognized that the immune system can perform anti-fibrotic



9

David Nareznoy for his help preparing high quality figures, and Shifra Fraifeld for her editorial assistance in preparing this manuscript.

Appendix A. Supplementary material Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jaut.2015.02.006.

References

Fig. 7. Simplified scheme of the potential significance of immune surveillance in the resolution of fibrosis Lung injury is followed by myofibroblast accumulation and recruitment of immune inflammatory cells. FasLþ immune cells regulate myofibroblast accumulation during repair and resolution of fibrosis. Factors increasing myofibroblast resistance to immune cell-induced apoptosis, such as FLIP, and/or affecting immune surveillance, such as soluble Fas and/or FasL or immunodeficiency, prevent resolution and promote unremitting fibrosis.

functions by FasL-induced apoptosis of fibroblasts. Our findings elucidate this capability. Author contributions (1) Conception definition and hypothesis e SBWD, RGG, RB; study design e SBWD, LE, RGG, JK, PZ, NA, MRD, (2) Data acquisition e SBWD, LE, RGG, JK, PZ, NA; data interpretation e SBWD, LE, RGG, PZ, NA, MRD, RB. (3) Manuscript preparation e SBWD, LE, RGG, JK, PZ, NA; substantial revision e SBWD, MRD, RB. Funding and disclosures This work was supported by the Israel Science Foundation (855/ 10), the David Sheinberg Fund, and the Israel Ministry of Health (633333-6). The authors have no conflicts of interest. Major findings Physiological fibrosis is a normal process whereby damaged tissue undergoes repair by healthy scar formation, followed by scar resolution and return to normal organ function. Our studies demonstrate that this process is mediated by the immune system, and in particular by T cells expressing the Fas-Ligand (FasL) molecule. We have further found that antagonists of Fas/FasL interaction play a key role in modulating T cell capability to induce fibroblast apoptosis. In patients with idiopathic pulmonary fibrosis (IPF), soluble FasL, a FasL antagonist, is upregulated and secreted by lung myofibroblasts and scar tissue accumulates, suggesting that sFasL may confer protection, enabling myofibroblasts to resist immunecell induced apoptosis. Equilibrium FasL expression may appropriately modulate fibroblast response during fibrosis evolution and resolution. Acknowledgments We thank Prof. David Naor for helpful discussions and for reviewing the manuscript, Anita Kol for the experimental work, Tali Bdolach-Abram for her assistance with the statistical analysis,

[1] K. Kuwano, T. Maeyama, I. Inoshima, K. Ninomiya, N. Hagimoto, M. Yoshimi, et al., Increased circulating levels of soluble Fas ligand are correlated with disease activity in patients with fibrosing lung diseases, Respirology 7 (2002) 15e21. [2] K. Kuwano, M. Kawasaki, T. Maeyama, N. Hagimoto, N. Nakamura, K. Shirakawa, et al., Soluble form of fas and fas ligand in BAL fluid from patients with pulmonary fibrosis and bronchiolitis obliterans organizing pneumonia, Chest 118 (2000) 451e458. [3] T. Tanaka, M. Yoshimi, T. Maeyama, N. Hagimoto, K. Kuwano, N. Hara, Resistance to Fas-mediated apoptosis in human lung fibroblast, Eur. Respir. J. 20 (2002) 359e368. [4] F. Buhling, A. Wille, C. Rocken, O. Wiesner, A. Baier, I. Meinecke, et al., Altered expression of membrane-bound and soluble CD95/Fas contributes to the resistance of fibrotic lung fibroblasts to FasL induced apoptosis, Respir. Res. 6 (2005) 37. [5] T. Suda, H. Hashimoto, M. Tanaka, T. Ochi, S. Nagata, Membrane Fas ligand kills human peripheral blood T lymphocytes, and soluble Fas ligand blocks the killing, J. Exp. Med. 186 (1997) 2045e2050. [6] S. Hanabuchi, M. Koyanagi, A. Kawasaki, N. Shinohara, A. Matsuzawa, Y. Nishimura, et al., Fas and its ligand in a general mechanism of T-cellmediated cytotoxicity, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 4930e4934. [7] T. Stalder, S. Hahn, P. Erb, Fas antigen is the major target molecule for CD4þ T cell-mediated cytotoxicity, J. Immunol. 152 (1994) 1127e1133. [8] S. Hahn, R. Gehri, P. Erb, Mechanism and biological significance of CD4mediated cytotoxicity, Immunol. Rev. 146 (1995) 57e79. [9] C.J. Melief, Tumor eradication by adoptive transfer of cytotoxic T lymphocytes, Adv. Cancer Res. 58 (1992) 143e175. [10] M.E. Peter, P.H. Krammer, Mechanisms of CD95 (APO-1/Fas)-mediated apoptosis, Curr. Opin. Immunol. 10 (1998) 545e551. [11] S. Nagata, P. Golstein, The Fas death factor, Science 267 (1995) 1449e1456. [12] T. Suda, T. Takahashi, P. Golstein, S. Nagata, Molecular cloning and expression of the Fas ligand, a novel member of the tumor necrosis factor family, Cell 75 (1993) 1169e1178. [13] R. Golan-Gerstl, S.B. Wallach-Dayan, G. Amir, R. Breuer, Epithelial cell apoptosis by fas ligand-positive myofibroblasts in lung fibrosis, Am. J. Respir. Cell Mol. Biol. 36 (2007) 270e275. [14] S.B. Wallach-Dayan, R. Golan-Gerstl, R. Breuer, Evasion of myofibroblasts from immune surveillance: a mechanism for tissue fibrosis, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 20460e20465. [15] R. Golan-Gerstl, S.B. Wallach-Dayan, P. Zisman, W.V. Cardoso, R.H. Goldstein, R. Breuer, Cellular FLICE-like inhibitory protein deviates myofibroblast fasinduced apoptosis toward proliferation during lung fibrosis, Am. J. Respir. Cell Mol. Biol. 47 (2012) 271e279. [16] J.M. Pilewski, L. Liu, A.C. Henry, A.V. Knauer, C.A. Feghali-Bostwick, Insulin-like growth factor binding proteins 3 and 5 are overexpressed in idiopathic pulmonary fibrosis and contribute to extracellular matrix deposition, Am. J. Pathol. 166 (2005) 399e407. [17] S. Kremer, R. Breuer, I.S. Lossos, N. Berkman, T.G. Christensen, M.W. Connor, et al., Effect of immunomodulators on bleomycin-induced lung injury, Respiration 66 (1999) 455e462. [18] S. Gokhale, D. Rosen, N. Sneige, L.K. Diaz, E. Resetkova, A. Sahin, et al., Assessment of two automated imaging systems in evaluating estrogen receptor status in breast carcinoma, Appl. Immunohistochem. Mol. Morphol. 15 (2007) 451e455. [19] S.B. Wallach-Dayan, G. Izbicki, P.Y. Cohen, R. Gerstl-Golan, A. Fine, R. Breuer, Bleomycin initiates apoptosis of lung epithelial cells by ROS but not by Fas/ FasL pathway, Am. J. Physiol. Lung Cell. Mol. Physiol. 290 (2006) L790eL796. [20] M. Schulte, K. Reiss, M. Lettau, T. Maretzky, A. Ludwig, D. Hartmann, et al., ADAM10 regulates FasL cell surface expression and modulates FasL-induced cytotoxicity and activation-induced cell death, Cell Death Differ. 14 (2007) 1040e1049. [21] H. Fujioka, P.J. Hunt, J. Rozga, G.D. Wu, D.V. Cramer, A.A. Demetriou, et al., Carboxyfluorescein (CFSE) labelling of hepatocytes for short-term localization following intraportal transplantation, Cell Transplant. 3 (1994) 397e408. [22] X. Li, H. Dancausse, I. Grijalva, M. Oliveira, A.D. Levi, Labeling Schwann cells with CFSE e an in vitro and in vivo study, J. Neurosci. Methods 125 (2003) 83e91. [23] S. Lo Re, D. Lison, F. Huaux, CD4þ T lymphocytes in lung fibrosis: diverse subsets, diverse functions, J. Leukoc. Biol. 93 (2013) 499e510.


10


[24] I.G. Luzina, N.W. Todd, A.T. Iacono, S.P. Atamas, Roles of T lymphocytes in pulmonary fibrosis, J. Leukoc. Biol. 83 (2008) 237e244. [25] G.P. Downey, Resolving the scar of pulmonary fibrosis, N. Engl. J. Med. 365 (2011) 1140e1141. [26] E. Fireman, N. Vardinon, M. Burke, S. Spizer, S. Levin, A. Endler, et al., Predictive value of response to treatment of T-lymphocyte subpopulations in idiopathic pulmonary fibrosis, Eur. Respir. J. 11 (1998) 706e711. [27] E. Corsini, M.I. Luster, J. Mahler, W.A. Craig, M.E. Blazka, G.J. Rosenthal, A protective role for T lymphocytes in asbestos-induced pulmonary inflammation and collagen deposition, Am. J. Respir. Cell Mol. Biol. 11 (1994) 531e539. [28] M. Helene, V. Lake-Bullock, J. Zhu, H. Hao, D.A. Cohen, A.M. Kaplan, T cell independence of bleomycin-induced pulmonary fibrosis, J. Leukoc. Biol. 65 (1999) 187e195. [29] S.L. Collins, Y. Chan-Li, R.W. Hallowell, J.D. Powell, M.R. Horton, Pulmonary vaccination as a novel treatment for lung fibrosis, PLoS ONE 7 (2012) e31299. [30] K. Kuwano, H. Miyazaki, N. Hagimoto, M. Kawasaki, M. Fujita, R. Kunitake, et al., The involvement of FaseFas ligand pathway in fibrosing lung diseases, Am. J. Respir. Cell Mol. Biol. 20 (1999) 53e60. [31] N. Khalil, R. O'Connor, Idiopathic pulmonary fibrosis: current understanding of the pathogenesis and the status of treatment, CMAJ 171 (2004) 153e160. [32] Y.P. Moodley, P. Caterina, A.K. Scaffidi, N.L. Misso, J.M. Papadimitriou, R.J. McAnulty, et al., Comparison of the morphological and biochemical changes in normal human lung fibroblasts and fibroblasts derived from lungs of patients with idiopathic pulmonary fibrosis during FasL-induced apoptosis, J. Pathol. 202 (2004) 486e495. [33] F. Conceicao-Silva, M. Hahne, M. Schroter, J. Louis, J. Tschopp, The resolution of lesions induced by Leishmania major in mice requires a functional Fas (APO-1, CD95) pathway of cytotoxicity, Eur. J. Immunol. 28 (1998) 237e245. [34] P.L. Cohen, R.A. Eisenberg, The lpr and gld genes in systemic autoimmunity: life and death in the Fas lane, Immunol. Today 13 (1992) 427e428. [35] K. Kuwano, N. Hagimoto, M. Kawasaki, T. Yatomi, N. Nakamura, S. Nagata, et al., Essential roles of the FaseFas ligand pathway in the development of pulmonary fibrosis, J. Clin. Invest. 104 (1999) 13e19. [36] K. Aoshiba, S. Yasui, J. Tamaoki, A. Nagai, The Fas/Fas-ligand system is not required for bleomycin-induced pulmonary fibrosis in mice, Am. J. Respir. Crit. Care Med. 162 (2000) 695e700. [37] N.A. Saunders, R.J. Smith, A.M. Jetten, Differential responsiveness of human bronchial epithelial cells, lung carcinoma cells, and bronchial fibroblasts to interferon-gamma in vitro, Am. J. Respir. Cell Mol. Biol. 11 (1994) 147e152. [38] T.H. Piela, J.H. Korn, Lymphocyteefibroblast adhesion induced by interferongamma, Cell. Immunol. 114 (1988) 149e160. [39] G.D. Sempowski, P.R. Chess, R.P. Phipps, CD40 is a functional activation antigen and B7-independent T cell costimulatory molecule on normal human lung fibroblasts, J. Immunol. 158 (1997) 4670e4677.

[40] E.J. Schattner, K.B. Elkon, D.H. Yoo, J. Tumang, P.H. Krammer, M.K. Crow, et al., CD40 ligation induces Apo-1/Fas expression on human B lymphocytes and facilitates apoptosis through the Apo-1/Fas pathway, J. Exp. Med. 182 (1995) 1557e1565. [41] A.M. Hohlbaum, S. Moe, A. Marshak-Rothstein, Opposing effects of transmembrane and soluble Fas ligand expression on inflammation and tumor cell survival, J. Exp. Med. 191 (2000) 1209e1220. [42] P. Schneider, N. Holler, J.L. Bodmer, M. Hahne, K. Frei, A. Fontana, et al., Conversion of membrane-bound Fas(CD95) ligand to its soluble form is associated with downregulation of its proapoptotic activity and loss of liver toxicity, J. Exp. Med. 187 (1998) 1205e1213. [43] M. Tanaka, T. Itai, M. Adachi, S. Nagata, Downregulation of Fas ligand by shedding, Nat. Med. 4 (1998) 31e36. [44] M. Mogi, K. Fukuo, J. Yang, T. Suhara, T. Ogihara, Hypoxia stimulates release of the soluble form of fas ligand that inhibits endothelial cell apoptosis, Lab. Invest. 81 (2001) 177e184. [45] B.M. Elicker, J.A. Golden, K.G. Ordovas, L. Leard, T.R. Golden, S.R. Hays, Progression of native lung fibrosis in lung transplant recipients with idiopathic pulmonary fibrosis, Respir. Med. 104 (2010) 426e433. [46] K. Crothers, L. Huang, J.L. Goulet, M.B. Goetz, S.T. Brown, M.C. RodriguezBarradas, et al., HIV infection and risk for incident pulmonary diseases in the combination antiretroviral therapy era, Am. J. Respir. Crit. Care Med. 183 (2011) 388e395. [47] H. Ozsahin, F.X. Arredondo-Vega, I. Santisteban, H. Fuhrer, P. Tuchschmid, W. Jochum, et al., Adenosine deaminase deficiency in adults, Blood 89 (1997) 2849e2855. [48] S.A. Schroeder, M. Swift, C. Sandoval, C. Langston, Interstitial lung disease in patients with ataxiaetelangiectasia, Pediatr. Pulmonol. 39 (2005) 537e543. [49] R. Somech, G.R. Somers, D. Chitayat, E. Grunebaum, A. Atkinson, E. Kolomietz, et al., Fatal lung fibrosis associated with immunodeficiency and gonadal dysgenesis in 46XX sisters e a new syndrome, Am. J. Med. Genet. A 146A (2008) 8e14. [50] E. Toubi, Y. Shoenfeld, Protective autoimmunity in cancer (review), Oncol. Rep. 17 (2007) 245e251. [51] A. Scriabine, D.U. Rabin, New developments in the therapy of pulmonary fibrosis, Adv. Pharmacol. 57 (2009) 419e464. [52] V. Krizhanovsky, M. Yon, R.A. Dickins, S. Hearn, J. Simon, C. Miething, et al., Senescence of activated stellate cells limits liver fibrosis, Cell 134 (2008) 657e667. [53] T.W. Kang, T. Yevsa, N. Woller, L. Hoenicke, T. Wuestefeld, D. Dauch, et al., Senescence surveillance of pre-malignant hepatocytes limits liver cancer development, Nature 479 (2011) 547e551. [54] T.E. King Jr., W.Z. Bradford, S. Castro-Bernardini, E.A. Fagan, I. Glaspole, M.K. Glassberg, et al., A phase 3 trial of Pirfenidone in patients with idiopathic pulmonary fibrosis, N. Engl. J. Med. 370 (2014) 2083e2092.


Cutting edge: maresin-1 engages regulatory T cells to limit type 2 innate lymphoid cell activation and promote resolution of lung inflammation.

Cutting edge: a further response.

Cutting edge: teaching future surgeons.

Confusion on the cutting edge.

Cutting Edge: NFAT Transcription Factors Promote the Generation of Follicular Helper T Cells in Response to Acute Viral Infection.

Immunotherapy and tolerance—cutting edge.

Rationing: at the cutting edge.

Cutting Edge: Regulatory T Cells Facilitate Cutaneous Wound Healing.

The Cutting Edge of Affinity Electrophoresis Technology.

Editorial: Cutting edge in systemic lupus erythematosus.

Cutting-edge Perspectives in Genomic Maintenance. Preface.

Gairdner presents cutting-edge cancer research.

Reviews on cutting-edge topics in toxicology.

François Jacob at the cutting edge.

Contemporary theme. Rationing: at the cutting edge.

Cutting edge: Central memory CD8 T cells in aged mice are virtual memory cells.

On the cutting edge of organ renewal: Identification, regulation, and evolution of incisor stem cells.

Cutting edge: identification and characterization of human intrahepatic CD49a+ NK cells.

Cutting Edge: Notch Signaling Promotes the Plasticity of Group-2 Innate Lymphoid Cells.

Cutting edge of endoscopic full-thickness resection for gastric tumor.

Combination of high spatial resolution and low minimum detection limit using thinned specimens in cutting-edge electron probe microanalysis.

Non-coding RNAs, the cutting edge of histone messages.

On the right side of cutting-edge education.

Cutting Edge: AhR Is a Molecular Target of Calcitriol in Human T Cells.