Pharmacological Inhibition of Caspase-8 Suppresses Inflammation-Induced Lymphangiogenesis and Allograft Rejection in the Cornea
Wenru Su,Xiaoqing Chen,Yichen Xiao,Song Guo Zheng,Dan Liang
DOI: https://doi.org/10.1016/j.jaci.2018.02.005
2018-01-01
Abstract:Graft rejection (GR) remains a major hurdle in transplantation medicine, and lymphangiogenesis plays a critical role in this process. An improved understanding of the mechanisms involved in GR as well as new strategies to manage it are greatly needed. Caspases, which play critical roles in cell death, have received an increasing attention for their immune properties. Pharmacological inhibition of caspases has been proposed as a promising therapeutic strategy for the treatment of various diseases. However, the potential roles of caspase inhibitors and the molecular mechanisms involved in lymphangiogenesis and GR remain elusive. Cornea is accessible for experimental manipulation readily and the normal cornea lacks lymphatic vessels. Therefore, we used a mouse model of corneal transplantation (CT; Fig 1, A-D) to explore the clinical application of caspase inhibition and the molecular mechanisms involved in inflammation-induced lymphangiogenesis (ILA) and GR (see the Methods sections in this article's Online Repository at www.jacionline. org). Topical application of a pan-caspase inhibitor (Z-VAD-fmk) at 60 μM had a maximum inhibitory effect on ILA on day 9 (the experimental end point for lymphangiogenesis) after CT (Fig 1, E). We next tested different caspase-specific inhibitors to explore further the specific caspase responsible for ILA inhibition. We found that compared with the pan-caspase inhibitor, a specific caspase-8 inhibitor (Z-IETD-fmk) demonstrated similar inhibitory effects on ILA, whereas other caspase inhibitors only mimic partial (caspase-3 inhibitor and caspase-1 inhibitor) or no inhibitory effects on ILA (Fig 1, F). Therefore, the specific caspase-8 inhibition was chosen for subsequent experiments (for the choice of optimal doses of inhibitors, see the Results section and Fig E5 in this article's Online Repository at www.jacionline.org). CD11b+ macrophages are critically involved in ILA and primarily exist in the limbus of normal cornea (see Fig E1, A, in this article's Online Repository at www.jacionline.org).1Maruyama K. Ii M. Cursiefen C. Jackson D.G. Keino H. Tomita M. et al.Inflammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages.J Clin Invest. 2005; 115: 2363-2372Crossref PubMed Scopus (578) Google Scholar, 2Kataru R.P. Jung K. Jang C. Yang H. Schwendener R.A. Baik J.E. et al.Critical role of CD11b+ macrophages and VEGF in inflammatory lymphangiogenesis, antigen clearance, and inflammation resolution.Blood. 2009; 113: 5650-5659Crossref PubMed Scopus (295) Google Scholar, 3Ji R.C. Macrophages are important mediators of either tumor- or inflammation-induced lymphangiogenesis.Cell Mol Life Sci. 2012; 69: 897-914Crossref PubMed Scopus (93) Google Scholar Macrophage inflammatory protein-1 (MIP-1) and monocyte chemoattractant protein-1 (MCP-1), which could be produced by macrophages, corneal epithelial cells, and corneal fibroblasts, play critical roles in a further macrophage recruitment.4Gillitzer R. Goebeler M. Chemokines in cutaneous wound healing.J Leukoc Biol. 2001; 69: 513-521Crossref PubMed Google Scholar We found that caspase-8 inhibition decreased CD11b+ macrophage numbers, as well as MIP-1 and MCP-1 expression, in the grafted corneas on day 9 (Fig 1, G-I). During lymphangiogenesis, macrophages play critical roles via their structural contribution and secretion of prolymphangiogenic factors.1Maruyama K. Ii M. Cursiefen C. Jackson D.G. Keino H. Tomita M. et al.Inflammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages.J Clin Invest. 2005; 115: 2363-2372Crossref PubMed Scopus (578) Google Scholar, 2Kataru R.P. Jung K. Jang C. Yang H. Schwendener R.A. Baik J.E. et al.Critical role of CD11b+ macrophages and VEGF in inflammatory lymphangiogenesis, antigen clearance, and inflammation resolution.Blood. 2009; 113: 5650-5659Crossref PubMed Scopus (295) Google Scholar, 3Ji R.C. Macrophages are important mediators of either tumor- or inflammation-induced lymphangiogenesis.Cell Mol Life Sci. 2012; 69: 897-914Crossref PubMed Scopus (93) Google Scholar Our results show that LPS (a Toll-like receptor [TLR] 4 agonist, 100 ng/mL) increased active caspase-8 expression and activity (Fig 1, J, and Fig E1, B), and the expression of MIP-1, MCP-1, vascular endothelial growth factor (VEGF)-C, TNF-α, and IL-1β in a murine macrophage cell line (RAW264.7), whereas caspase-8 inhibition with Z-IETD-fmk (10 μM) blocked the LPS-induced increase (Fig 1, K, and Fig E1, C-F). In grafted corneas, the expression of VEGF-C, IL-1β, and TNF-α mRNA was also upregulated, whereas caspase-8 inhibition reduced their expression (Fig E1, G-I). These findings collectively indicate that the inhibition of macrophage recruitment and activation contributes to the suppression of ILA mediated by caspase-8 inhibition. Our in vitro experiments showed that a TLR4 agonist stimulated caspase-8 activity. In vivo, we found that TLR4 expression was increased in grafted corneas, and TLR4 knockout noticeably but only partially reduced caspase-8 activity (Fig 1, L-N, and Fig E1, H-K). Furthermore, ILA was reduced in TLR4-knockout mice in comparison with wild-type mice (Fig 1, O and P). In addition, LPS (5 μg/mL) increased caspase-8 activity and ILA, whereas caspase-8 inhibitor blocked LPS-mediated increase in grafted corneas (Fig 1, Q and R). Thus, we believe that TLR4 signaling plays an important role in caspase-8 activation in ILA. High-mobility group box-1 (HMGB1), a ubiquitous nuclear protein, is known as a damage-associated molecular-pattern molecule and an endogenous ligand for TLR4.5Tsung A. Tohme S. Billiar T.R. High-mobility group box-1 in sterile inflammation.J Intern Med. 2014; 276: 425-443Crossref PubMed Scopus (158) Google Scholar In the grafted corneas, extranuclear HMGB1 protein levels were upregulated, and HMGB1 inhibition with glycyrrhizic acid reduced TLR4 expression, caspase-8 activity, and ILA on day 9 (Fig 2, A-E). Treatment with recombinant HMGB1 increased TLR4 expression, caspase-8 activity, and ILA, whereas caspase-8 inhibition blocked HMGB1-mediated increase (see Fig E2, A-C, in this article's Online Repository at www.jacionline.org). The NLRP3 inflammasome comprises the NLRP3 scaffold, apoptosis-associated speck-like protein containing CARD (ASC), and caspase-1, and plays a significant role in various diseases.6Strowig T. Henao-Mejia J. Elinav E. Flavell R. Inflammasomes in health and disease.Nature. 2012; 481: 278-286Crossref PubMed Scopus (1642) Google Scholar, 7de Torre-Minguela C. Mesa D.C.P. Pelegrín P. The NLRP3 and Pyrin inflammasomes: implications in the pathophysiology of autoinflammatory diseases.Front Immunol. 2017; 8: 43Crossref PubMed Scopus (163) Google Scholar However, the potential roles of inflammasomes in lymphangiogenesis have never previously been explored. Our findings show that NLRP3 and ASC expression and caspase-1 activity were elevated 9 days after CT, whereas caspase-8 inhibition reduced this elevation (Fig 2, F and G; also see Fig E3, A-F). Notably, NLRP3 inhibition also reduced ILA (Fig 2, H). HMGB1 is passively released into the extracellular space by injured or necrotic cells or actively secreted into this space by activated immune cells.5Tsung A. Tohme S. Billiar T.R. High-mobility group box-1 in sterile inflammation.J Intern Med. 2014; 276: 425-443Crossref PubMed Scopus (158) Google Scholar, 8Harris H.E. Andersson U. Pisetsky D.S. HMGB1: a multifunctional alarmin driving autoimmune and inflammatory disease.Nat Rev Rheumatol. 2012; 8: 195-202Crossref PubMed Scopus (530) Google Scholar In grafted corneas, caspase-8 or NLRP3 inhibition (glybenclamidethe, 100 μM) reduced extranuclear HMGB1 expression (Fig 2, I and J). In vitro, LPS induced NLRP3 and ASC expression, caspase-1 activity, and HMGB1 secretion, whereas caspase-8 inhibition blocked LPS-induced effects in macrophages (Fig 2, K; see Fig E4, A-D, in this article's Online Repository at www.jacionline.org). In addition, NLRP3 inhibition reduced HMGB1 secretion in macrophages (Fig 2, L). These findings reveal that the caspase-8/NLRP3 inflammasome pathway mediates HMGB1 secretion. We then explored the therapeutic roles of caspase-8 inhibition in corneal GR. In corneal isografts, ILA developed similarly, but not as rapidly as ILA in corneal allografts (Fig 2, M). Furthermore, topical caspase-8 inhibition reduced ILA in corneal isografts (Fig 2, N). However, there were significant differences in GR between corneal isografts and allografts (Fig 2, O). As expected, topical caspase-8 inhibitor, which inhibited ILA, obviously promoted corneal allografts survival (Fig 2, O). These results suggest that the CT procedure is sufficient to induce ILA, whereas both ILA and allograft antigens are necessary for GR in CT. The new lymphatic vessels transport antigen-presenting cells (APCs) carrying allograft antigens to provoke effector T-cell activation to initiate and promote GR. We found that caspase-8 blockade reduced the number and frequency of CD11c+CD80+ and CD11c+CD86+ dendritic cells (the most potent APCs) on day 12 after CT (Fig 2, P-U). CD4+ T cells, especially TH17 cells and regulatory T cells, are crucial for the development of GR.9Liu Z. Fan H. Jiang S. CD4(+) T-cell subsets in transplantation.Immunol Rev. 2013; 252: 183-191Crossref PubMed Scopus (88) Google Scholar We observed a reduction in the number and frequency of TH17 cells and an increase in the number and frequency of regulatory T cells in the draining lymph nodes after caspase-8 blockade (Fig 2, V-Y). These results collectively indicate that ILA inhibition by topical caspase-8 blockade interrupts the transportation of APCs to reduce TH17, which subsequently suppresses corneal GR (for the role of HMGB1 and TLR4 in corneal allograft rejection, see the Results sections and Fig E6 in this article's Online Repository at www.jacionline.org). In summary, this study provides the first evidence that pharmacological inhibition of caspase-8 suppressed lymphangiogenesis and GR; it also identifies an innovative mechanism whereby the caspase-8/NLRP3 inflammasome pathway promotes lymphangiogenesis and GR (see Fig E7 in this article's Online Repository at www.jacionline.org). The CT procedure initially leads to corneal cell damage, resulting in the release of HMGB1. The released HMGB1 bonds with TLR4 to activate caspase-8 signaling in the cornea. Then, the activated caspase-8 signaling mediates macrophage activation and recruitment to induce lymphangiogenesis. Furthermore, caspase-8 regulates NLRP3 inflammasome-mediated HMGB1 secretion in the cornea and macrophage. The secreted HMGB1 may, in turn, interact with TLR4, forming a positive feedback loop to further promote lymphangiogenesis. Finally, new lymphatic vessels transport APCs carrying allograft antigens to activate T cells, which initiate and promote GR. Therefore, the caspase-8/NLRP3 inflammasome pathway functions as an amplification system to promote lymphangiogenesis and GR, whereas the blockage of caspase-8 interrupts this process. These findings provide new insightful views into the pathogenesis of inflammatory lymphangiogenesis and GR. They also support the idea that the pharmacological inhibition of caspase-8 may serve as a new strategy for treating GR and other lymphatic disorders. Adult male C57BL/6 mice and BALB/c mice were obtained from the Model Animal Research Center of Nanjing University and the Guangzhou Animal Testing Center. C57BL/6-based homozygous Tlr4-d (B6.B10ScN-Tlr4lps-del/JthJ) mice and BALB/c-based homozygous Tlr4-d (C.C3-Tlr4Lps-d/J) mice were originally obtained from the Jackson Laboratory and were produced by the Model Animal Research Center of Nanjing University. This study was approved by the Institutional Animal Care and Use Committee of the Zhongshan Ophthalmic Center at Sun Yat-Sen University. All procedures involving animal eyes were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The animals were maintained in specific pathogen-free conditions under a 12-hour light-dark cycle with controlled temperature (23°C ± 2°C) and humidity (55% ± 10%). A mouse model of CT (Fig 1, A-C) was used as previously described (detailed in Online Repository text).E1Yamagami S. Dana M.R. The critical role of lymph nodes in corneal alloimmunization and graft rejection.Invest Ophthalmol Vis Sci. 2001; 42: 1293-1298PubMed Google Scholar, E2Liu Y. Hamrah P. Zhang Q. Taylor A.W. Dana M.R. Draining lymph nodes of corneal transplant hosts exhibit evidence for donor major histocompatibility complex (MHC) class II-positive dendritic cells derived from MHC class II-negative grafts.J Exp Med. 2002; 195: 259-268Crossref PubMed Scopus (179) Google Scholar To establish the mouse model of CT, BALB/c mice were used as corneal donors and C57BL/6 mice served as the corneal transplant recipients. For the mouse model of corneal isograft, C57BL/6 mice were used as both the donors and the recipients. Briefly, a donor mice corneal graft was prepared with a corneal trephine (2-mm diameter). The recipient graft bed was made in the central cornea by excising a circular 1.5-mm area. The donor graft was placed onto the recipient graft bed and secured with 8 interrupted 11-0 nylon sutures (Jiahe, Inc, Taibei, Taiwan, China) followed by application of antibiotic ointment. In some experiments, recombinant HMGB1 (80 μg/mL, Biolegend, San Diego, Calif), the HMGB1 inhibitor glycyrrhizic acid (80 μM, Sigma, St Louis, Mo), the NLRP3 inhibitor glybenclamidethe (100 μM, Sigma), caspase-8 inhibitor Z-IETD-fmk (BioVision, Milpitas, Calif), the caspase-1 inhibitor Z-YVAD-fmk (60 μM, BioVision), the caspase-3 inhibitor (60 μM, Z-DQMD-FMK, BioVision), the caspase-4 inhibitor (60 μM, Z-LEVD-FMK, BioVision), and the caspase-9 inhibitor (60 μM, Z-LEHD-FMK, BioVision), and all of which were suspended in sterile saline solution, were used. Five microliters of the preparation or vehicle was applied topically to the CT-treated eye 3 times a day for 9 days. Whole-mount corneal lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1) staining was performed as previously described.E3Cao R. Lim S. Ji H. Zhang Y. Yang Y. Honek J. et al.Mouse corneal lymphangiogenesis model.Nat Protoc. 2011; 6: 817-826Crossref PubMed Scopus (65) Google Scholar Briefly, the eyes were excised from sacrificed mice and fixed with 4% (wt/vol) paraformaldehyde at 4°C overnight. The corneas with an intact limbus were removed under a biomicroscope, washed 3 times in PBS, blocked with 3% BSA in PBST (0.3% (vol/vol) Triton X-100 in PBS) for 2 hours, and incubated overnight at 4°C with anti–LYVE-1 (Abcam, Cambridge, Mass). Then, the corneas were stained with Alexa Fluor 555–coupled goat antirabbit antibody (Cell Signaling Technology, Inc, Danvers, Mass) for 2 hours. Next, the samples were flat-mounted on microscope slides with Vectashield mounting medium, and images were obtained with a 5× objective with a fluorescent microscope. The lymphatic vessels was quantified as previously described.E4Han L. Su W. Huang J. Zhou J. Qiu S. Liang D. Doxycycline inhibits inflammation-induced lymphangiogenesis in mouse cornea by multiple mechanisms.PLoS One. 2014; 9: e108931Crossref PubMed Scopus (25) Google Scholar In brief, the merged images of the whole-mount corneas were imported into the Adobe Photoshop CC program. First, the “Polygonal Lasso Tool” was used to define the total area of the cornea. Then, the “Brightness/Contrast” and “Levels” were adjusted to clarify the lymphatic vessels. Next, the “Calculations” tool was applied to select the lymphatic vessel–covered areas. For CD11b staining in the frozen corneal sections, the eyes were removed from mice sacrificed at the end of the 9-day experiment and embedded in optimal cutting temperature compound. The eyeballs were cut into 6-μm frozen sections, fixed with 4% (wt/vol) paraformaldehyde for 10 minutes, and treated with anti-CD11b (Abcam) and subsequently with an Alexa Fluor 488–coupled goat antirabbit secondary antibody (Cell Signaling Technology). The CD11b-positive fluorescent signals in the frozen corneal sections (40× objective, corneal area ∼600 μm from the limbus) were imported into Adobe Photoshop CC to evaluate the cell density. The RAW264.7 murine macrophage cell line (Shanghai Cell Bank of Academia Sinica, Shanghai, China) was cultured in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated FBS and 2 mM l-glutamine. For LPS activation, the cells were pretreated with caspase-8 inhibitor Z-IETD-fmk (10 μM) or the NLRP3 inhibitor glybenclamidethe (200 μM) for 1 hour, followed by incubation with or without LPS (100 ng/mL) for 16 hours. Next, the supernatants from the culture media and cells were collected for further analysis. The total RNA from corneal or cell lysates was obtained using an RNeasy Mini Kit (Qiagen, Valencia, Calif). The RNA was quantified with a NanoDrop spectrophotometer (ND-1000, NanoDrop Technologies), and cDNA was then generated using an Omniscript RT Kit (Qiagen). Next, the mRNA level was detected using ABsolute SYBR Green ROX Mix (Thermo, Waltham, Mass). The relative expression was calculated by normalizing the expression of each target to β-actin using the 2−ΔΔCt method. The total proteins of the cornea and cell homogenates were extracted with the KeyGen whole-cell lysis assay (KeyGen Biotech, Nanjing, China). Nuclear and cytoplasmic extracts were then prepared using the NE-PERTM nuclear and cytoplasmic extraction reagents (Thermo Fisher Scientific, Waltham, Mass), according to the manufacturer's instructions. Proteins (50-100 μg) were separated on polyacrylamide-SDS gels and electroblotted onto nitrocellulose membranes (Bio-Rad, Hercules, Calif). After blocking with Tris-buffered saline/5% nonfat dry milk, the membranes were treated with antibodies against HMGB1 (Abcam), TLR4 (Abcam), active caspase-8 (Cell Signaling Technology, Beverly, Mass), NLRP3 (Abcam), ASC (Millipore, Temecula, Calif), or β-actin (MultiSciences Biotech, Hangzhou, China) and then incubated with a horseradish peroxidase–conjugated secondary antibody. The signals were visualized by enhanced chemiluminescence (Pierce, Rockford, Ill). The concentrations of HMGB1 in the supernatants of cultured cells were detected by ELISA kit (Wuhan Mosak Biotechnology, Co, Wuhan, China). The activities of caspase-8, caspase-3, and caspase-1 in corneal or cell lysates were measured using colorimetric assay kits (BioVision). Cell surface markers and intracellular cytokine were analyzed using a FACS Calibur (BD Biosciences, San Jose, Calif) following standard protocols. The statistical analyses were performed with Student t test or appropriate 1-way ANOVA followed by Bonferroni post hoc tests using SPSS software (17.0; SPSS, Chicago, Ill). A P value of less than .05 was considered significant. When CT was performed on wild-type mice, we found that lymphangiogenesis appeared as early as 2 days and rapidly expanded to the interface (arrow) between the donor and host as early as 7 days after CT (Fig E5, A and B). Our results show that caspase-8 activity was significantly increased in the grafted corneas and remained elevated at 9 days after CT (Fig E5, C). CT also significantly increased the active caspase-8 protein levels in the grafted corneas after CT (Fig E5, D). Topical administration of Z-IETD-fmk led to the dose-dependent inhibition of caspase-8 activity, with an almost complete inhibitory effect at 120 μM in the grafted corneas at 9 days after CT (Fig E5, E). Of note, topically applied Z-IETD-fmk at 60 μM, not 120 μM, had a maximum inhibitory effect on ILA (Fig E5, F and G). We next explored the role of HMGB1 and TLR4 in corneal allograft rejection. As shown in Fig E6, A, the CT procedure induced the extracellular HMGB1 protein releasing. Next, we found that an HMGB1 inhibitor prolonged corneal allograft survival, whereas recombinant HMGB1 promoted corneal allograft rejection (Fig E6, B). The selective knockout of TLR4 also significantly prolonged corneal allograft survival, whereas TLR4 activation by LPS promoted allograft rejection (Fig E6, C). Taken together, these results imply that HMGB1 and TLR4 signaling promote caspase-8 activation and ILA and also play important roles in allograft rejection after CT. Fig E2Recombinant HMGB1 increases TLR4 signaling and caspase-8 activity in ILA. A and B, The topical instillation of recombinant HMGB1 protein increased TLR4 mRNA expression and caspase-8 activity 9 days after CT (n = 3). C, Recombinant HMGB1 promoted ILA 9 days after CT, whereas caspase-8 inhibitor blocked a recombinant HMGB1-mediated increase in ILA (n = 4). The data are presented as the means ± SDs. casp-8i, Caspase-8 inhibitor; rHMGB1, recombinant HMGB1. #P < .05 and ##P < .01 (compared with normal corneas); *P < .05 and **P < .01 (between the indicated groups).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E3NLRP 3 inflammasome is actively involved in ILA. A-D, CT increased the expression of NLRP3 and ASC and activity of caspase-1; these changes were sustained throughout a 9-day observation period after CT compared with normal corneas (n = 3). E and F, Following caspase-8 inhibition, NLRP3 and ASC expressions were decreased compared with the control group 9 days after CT (n = 3). casp-8i, Caspase-8 inhibitor. The data are presented as the means ± SDs. ##P < .01 (compared with normal corneas) and **P < .01 (between the indicated groups).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E4Caspase-8 inhibition restrains NLRP3 inflammasome activation in macrophages. A-C, The inhibition of caspase-8 in macrophages reduced LPS-induced expression of NLRP3 and ASC. D, The inhibition of caspase-8 in macrophages decreased LPS-stimulated caspase-1 activity. casp-8i, Caspase-8 inhibitor. The data are presented as the means ± SDs. ##P < .01 (compared with normal corneas) and **P < .01 (between the indicated groups).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E5Pharmacological inhibition of caspase-8 suppresses lymphangiogenesis after CT. A and B, Corneal whole-mount immunostaining of LYVE-1, which shows an increase in the area covered with lymphatic vessels, occurring from day 0 to day 9 after CT in WT mice (n = 3). C, Colorimetric assay revealed significantly increased caspase-8 activity in grafted corneas after CT compared with that in normal corneas (n = 3). D, CT increased active caspase-8 protein expression in grafted corneas (n = 3). E, The efficiency of caspase-8 inhibitor Z-IETD-fmk (30 μM, 60 μM, and 120 μM) on caspase-8 activity. F and G, ILA was significantly decreased by caspase-8 inhibition (60 μM) compared with normal mice (n = 4). Casp, Caspase; casp-8i, caspase-8 inhibition; WT, wild-type. The data are presented as the means ± SDs. #P < .05 and ##P < .01 (compared with normal corneas); *P < .05 and **P < .01 (between the indicated groups).View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E6The role of HMGB1 and TLR4 in corneal allograft rejection. A, Immunofluorescence staining showed that the CT procedure induced the extracellular HMGB1 protein release (white arrow: extracellular HMGB1). B, HMGB1 inhibitor prolonged corneal allograft survival, whereas recombinant HMGB1 promoted CGR. C, The selective knockout of TLR4 also prolonged corneal allograft survival, whereas TLR4 activation by LPS promoted CGR. DAPI, 4′-6-Diamidino-2-phenylindole, dihydrochloride; GA, glycyrrhizic acid; rHMGB1, recombinant HMGB1.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig E7The proposed pathway and the molecular mechanisms of caspase-8 inhibition in lymphangiogenesis and allograft rejection in the cornea. The CT procedure initially leads to corneal cell damage, resulting in the release of HMGB1. Released HMGB1 bonds with TLR4 to activate caspase-8 signaling in the cornea. The activated caspase-8 signaling mediates macrophage recruitment and activation to induce lymphangiogenesis. Furthermore, caspase-8 regulates NLRP3 inflammasome-mediated HMGB1 secretion in the cornea and macrophage. The secreted HMGB1 may, in turn, interact with TLR4, forming a positive feedback loop to further promote lymphangiogenesis. Finally, new lymphatic vessels transport APCs carrying allograft antigens to activate T cells, thus initiating and promoting allograft rejection. Therefore, the caspase-8/NLRP3 inflammasome pathway functions as an amplification system to promote lymphangiogenesis and allograft rejection, whereas caspase-8 inhibition interrupts this process.View Large Image Figure ViewerDownload Hi-res image Download (PPT)