IL-1β-dependent extravasation of preexisting lung-restricted autoantibodies during lung transplantation activates complement and mediates primary graft dysfunction

Abstract

Preexisting lung-restricted autoantibodies (LRAs) are associated with a higher incidence of primary graft dysfunction (PGD), although it remains unclear whether LRAs can drive its pathogenesis. In syngeneic murine left lung transplant recipients, preexisting LRAs worsened graft dysfunction, which was evident by impaired gas exchange, increased pulmonary edema, and activation of damage-associated pathways in lung epithelial cells. LRA-mediated injury was distinct from ischemia-reperfusion injury since deletion of donor nonclassical monocytes and host neutrophils could not prevent graft dysfunction in LRA-pretreated recipients. Whole LRA IgG molecules were necessary for lung injury, which was mediated by the classical and alternative complement pathways and reversed by complement inhibition. However, deletion of Fc receptors in donor macrophages or mannose-binding lectin in recipient mice failed to rescue lung function. LRA-mediated injury was localized to the transplanted lung and dependent on IL-1β-mediated permeabilization of pulmonary vascular endothelium, which allowed extravasation of antibodies. Genetic deletion or pharmacological inhibition of IL-1R in the donor lungs prevented LRA-induced graft injury. In humans, preexisting LRAs were an independent risk factor for severe PGD and could be treated with plasmapheresis and complement blockade. We conclude that preexisting LRAs can compound ischemia-reperfusion injury to worsen PGD for which complement inhibition may be effective.

Keywords: Innate immunity; Organ transplantation; Transplantation.

Figure 1. Preexisting whole lung-restricted antibodies (LRAs) against self-antigens induce primary graft dysfunction after syngeneic murine lung transplantation.

Recipient mice received 150 μg each (i.v.) of isotype control, LRAs (anti–collagen type V plus anti–K-α1 tubulin), or LRA F(ab′)₂ portion, 24 hours before and 1 hour after lung transplantation. (A) Twenty-four hours after transplantation, arterial blood oxygenation was analyzed after clamping the right hilum (n = 3–4). Lungs were also harvested for assessment of (B) pulmonary edema (n = 3–5) and (C) neutrophils (live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3). (D) Histology showing capillaritis and alveolar edema in mice treated with LRAs but not isotype control antibodies. Scale bar: 20 μm. (E) Inflammatory cells (both polymorpho- and mononuclear) were counted in 10 high-power fields (×40) and averaged for each group. (F) Pictures from 2-photon microscopy showing LRA (magenta) deposition. Scale bar: 50 μm. Data are presented as mean ± SD. PaO₂/FiO₂, arterial oxygen pressure. Graphs in A–C were analyzed by 1-way ANOVA followed by Tukey’s post hoc test. Graph in E was analyzed by unpaired, 2-tailed Student’s t test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 2. Differential expression analysis of single-cell RNA-seq data from isotype- or LRA-treated murine allograft after lung transplantation within epithelial cells.

(A) Functional enrichment analysis with GO Biological Processes was performed with the significantly upregulated genes in epithelial cells from LRA-treated compared with isotype-treated murine lung allograft. (B and C) Violin plots of expression for select genes significantly upregulated in epithelial cells from LRA-treated compared with isotype-treated murine lung allograft.

Figure 3. LRAs activate complement pathway to mediate lung graft injury.

(A) WT or Fcrγ^–/– recipients received 150 μg each (i.v.) of isotype control or LRAs (anti–collagen type V plus anti–K-α1 tubulin) 24 hours before and 1 hour after lung transplantation. Arterial blood oxygenation was analyzed after clamping the right hilum (n = 4). (B) Lungs treated as in A were also harvested for neutrophil quantification (live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3). (C) WT or Fcrγ^–/– recipient mice were treated as in A. Twenty-four hours after transplantation, bronchoalveolar lavage fluid (BALF) was obtained and C3 concentration measured by ELISA (n = 3–6). (D) WT or C3^–/– recipient mice were treated as in A. Arterial blood oxygenation was analyzed after clamping the right hilum (n = 3). (E) Quantification of neutrophil recruitment into lungs in mice treated as in A (neutrophil gating: live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3–5). Data are presented as mean ± SD. PaO₂/FiO₂, arterial oxygen pressure. Graphs were analyzed by 1-way ANOVA followed by Tukey’s post hoc test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; NS, not significant.

Figure 4. LRAs activate classical and alternative complement pathways to mediate lung graft injury.

(A) Diagram depicting experiments shown in B and C. WT, C1q^–/–, or Mbl^–/– recipient mice received 150 μg each (i.v.) of isotype control or LRAs (anti–collagen type V plus anti–K-α1 tubulin) 24 hours before and 1 hour after lung transplantation (Ltx). For some experiments, WT recipient mice treated with antibodies as described received C1INH, LNP023 (Factor B inhibitor, FBI), or anti-C5 neutralizing antibody. (B) Arterial blood oxygenation from mice described in A was analyzed after clamping the right hilum (n = 3). (C) Quantification of neutrophil recruitment into lungs in mice treated as described in A (neutrophil gating: live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3–5). (D and E) Immunocytochemistry for C4d (D) and C5b-9 (E) in allografts of LRA-treated mice (right) compared with isotype-treated mice (left). Data are presented as mean ± SD. PaO₂/FiO₂, arterial oxygen pressure. Graphs were analyzed by 1-way ANOVA followed by Tukey’s post hoc test. *P < 0.05; ***P < 0.001; ****P < 0.0001; NS, not significant. Scale bars: 20 μm and 10 μm (insets).

Figure 5. LRA-induced lung dysfunction is complementary to ischemia-reperfusion injury.

(A) Diagram depicting experiment shown in B and C. WT recipient mice received 150 μg each (i.v.) of isotype control or LRAs (anti–collagen type V plus anti–K-α1 tubulin) 24 hours before and 1 hour after lung transplantation (Ltx) using donor Nr4a1^–/– mice. (B) Arterial blood oxygenation from mice described in A was analyzed after clamping the right hilum (n = 3). (C) Quantification of neutrophil recruitment into lungs in mice treated as described in A (neutrophil gating: live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3–4). (D) WT recipient mice treated as in A were injected (i.p.) with isotype or anti-Ly6G antibody 24 hours before lung transplantation and harvested for determination of pulmonary edema 24 hours after transplantation (n = 3). (E) WT recipient mice treated as described in D. Arterial blood oxygenation was analyzed after clamping the right hilum (n = 3–4). Data are presented as mean ± SD. PaO₂/FiO₂, arterial oxygen pressure. Graphs were analyzed by 1-way ANOVA followed by Tukey’s post hoc test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; NS, not significant.

Figure 6. IL-1β is necessary to increase vascular endothelial permeability and extravasation of LRAs.

(A) Mouse primary lung microvascular endothelial cells were seeded at 200,000 cells per insert and cultured until confluent. Monolayers were incubated for 24 hours in the absence (control, CT) or presence of 50 ng/mL IL-1β in growth medium. FITC-antibody permeability testing was performed as described in the Methods. (B) Diagram depicting experiment shown in C and D. WT recipient mice received 150 μg each (i.v.) of isotype control or LRAs (anti–collagen type V plus anti–K-α1 tubulin) 24 hours before and 1 hour after lung transplantation (Ltx). Donor lungs were from Il1r^–/– mice or WT mice treated with IL-1RA 24 hours before and 1 hour after lung transplant. (C) Arterial blood oxygenation from mice described in B was analyzed after clamping the right hilum (n = 3). (D) Quantification of neutrophil recruitment into lungs in mice treated as described in B (neutrophil gating: live CD45⁺SiglecF^–CD11b⁺Ly6G⁺) (n = 3–5). (E–G) Immunohistochemistry for C4d (E and G) and LRAs (F) in allografts of LRA-treated wild-type mice (upper) compared with LRA-treated Il1r^–/– mice (lower) (E) or in lungs from PBS- or IL-1β–treated mice (F and G). Data are presented as mean ± SD. PaO₂/FiO₂, arterial oxygen pressure. Graphs were analyzed by 1-way ANOVA followed by Tukey’s post hoc test. *P < 0.05; **P < 0.01; ***P < 0.001. Scale bars: 20 μm.