MHC class I-specific antibody binding to nonhematopoietic cells drives complement activation to induce transfusion-related acute lung injury in mice

Abstract

Transfusion-related acute lung injury (TRALI), a form of noncardiogenic pulmonary edema that develops during or within 6 h after a blood transfusion, is the most frequent cause of transfusion-associated death in the United States. Because development of TRALI is associated with donor antibodies (Abs) reactive with recipient major histocompatibility complex (MHC), a mouse model has been studied in which TRALI-like disease is caused by injecting mice with anti-MHC class I monoclonal Ab (mAb). Previous publications with this model have concluded that disease is caused by FcR-dependent activation of neutrophils and platelets, with production of reactive oxygen species that damage pulmonary vascular endothelium. In this study, we confirm the role of reactive oxygen species in the pathogenesis of this mouse model of TRALI and show ultrastructural evidence of pulmonary vascular injury within 5 min of anti-MHC class I mAb injection. However, we demonstrate that disease induction in this model involves macrophages rather than neutrophils or platelets, activation of complement and production of C5a rather than activation of FcγRI, FcγRIII, or FcγRIV, and binding of anti-MHC class I mAb to non-BM-derived cells such as pulmonary vascular endothelium. These observations have important implications for the prevention and treatment of TRALI.

Figure 1.

mTRALI resembles human TRALI. (A) BALB/c male mice (four mice per group) were injected i.v. with 34-1-2s (mouse IgG2a anti–H-2D^d/K^d mAb) or mouse IgG2a isotype control mAb (δIA6.2) and evaluated for the next 26 min for changes in breathing pattern suggestive of dyspnea (shown as increases in Penh). Data are representative of more than five experiments. (B) BALB/c (H-2^d), B10.D2 (H-2^d), BALB/k (H-2^k), and C57BL/6 (H-2^b) background male mice (four mice per group) were injected i.v. with 34-1-2s. Penh was evaluated. Pooled graph is representative of two BALB/k, three C57B/6, three B10.D2, and three BALB/c experiments. (C) BALB/c male mice (10 mice per group) were injected i.v. with 34-1-2s or control mAb. Arterial blood pO₂ was measured 15 min later. Results are pooled from two experiments. (D) BALB/c male mice (five mice per group) were injected i.v. with 34-1-2s or control mAb; 30 min later, lungs were harvested, and percent water content was measured. Data are representative of three experiments. (E) BALB/c male mice (five mice per group) were injected i.v. with 34-1-2s or control mAb; 30 min later, BAL fluid was collected, and its protein content was measured. Data are representative of two experiments. (F) Representative H&E-stained sections of lung from BALB/c male mice injected i.v. with control mAb (1 and 2) or 34-1-2s (3–6). Lungs were harvested 0.5 (3 and 4) or 24 h (1, 2, 5, and 6) later. Arrows are pointing to neutrophils. (G) BALB/c male mice (five mice per group) were injected i.v. with 34-1-2s. Blood was drawn before or 30 or 240 min after challenge, and absolute neutrophil count (ANC) was measured. Data are representative of three experiments. (H) BALB/c male mice (10 mice per group) were injected i.v. with 34-1-2s or control mAb. Maximum Penh during the subsequent 26 min was determined. Lungs were harvested 30 min after injection and evaluated for percent water content. r² for percent water weight of lungs versus maximum Penh = 0.521. Data are representative of two experiments. (I) Representative electron micrographs of lungs harvested from BALB/c male mice before or 5 min after i.v. injection of 34-1-2s. Black boxes in top pictures outline the enlarged area below. White arrowheads point to areas of separation of vascular endothelium from basement membrane. Bars: (F) 70 µm; (I) 2 µm. (J) BALB/c male mice ages 8 and 18 wk (four mice per group) were injected i.v. with 34-1-2s or control mAb, and lungs were harvested for evaluation of percent water content 30 min later. Data are representative of two experiments. (K) BALB/c male mice (five to six mice per group) were injected i.v. with 34-1-2s or control mAb, and Penh was evaluated over a 1,440-min period. Data are representative of two experiments. (L) BALB/c male mice (five to six mice per group) were injected i.v. with 34-1-2s or control mAb, and lungs were harvested for evaluation of percent water content 0, 30, 120, and 1,440 min later. Data are representative of two experiments. *, P < 0.05 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 2.

mTRALI is partially FcγR independent. (A) BALB/c male mice (14–24 mice per group, pooled from four experiments) or male mice with a mixed BALB/c and C57BL/6 background (8–9 mice per group, pooled from two experiments) that were FcRγ sufficient (FcRγ^+/+) or deficient (FcRγ^−/−) were injected i.v. with 34-1-2s or control mAb. Penh was evaluated for this and all other experiments in this figure. Lungs harvested 30 min after challenge were evaluated for water content in this and all other experiments in this figure. (B) FcγRIII^+/+ and FcγRIII^−/− male mice (six mice per group) were injected i.v. with 34-1-2s or control mAb. Results are representative of at least two experiments for this and all subsequent panels in this figure. (C) FcγRI^+/+ and FcγRI^−/− male mice (six mice per group) were injected i.v. with 34-1-2s or control mAb. (D) FcγRI/III^+/+ and FcγRI/III^−/− male mice (four mice per group) were injected i.v. with 34-1-2s or control mAb. (E) FcγRI/III^−/− male mice (four mice per group) were injected i.v. with anti-FcγRIV mAb or isotype control mAb and then challenged the next day with 34-1-2s or isotype control mAb. *, P < 0.05; **, P < 0.005; and ***, P < 0.0005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 3.

Suppression of mTRALI by FcγRIIb ligation. (A) BALB/c male mice (four to eight mice per group) were pretreated s.c. with 2.4G2 (rat IgG2b anti-FcγRII/III mAb) or control mAb and then injected i.v. the next day with 34-1-2s or control mAb. Penh and percent water weight of lungs were measured for this and all other experiments in this figure. Results are representative of at least two experiments for this and all subsequent panels in this figure. (B) BALB/c FcγRIIb^+/+ and FcγRIIb^−/− male mice (four mice per group) were pretreated i.p. with mouse IgG2a anti-Ly17.2 mAb (anti-FcγRIIb mAb) or isotype control mAb and then injected i.v. the next day with 34-1-2s or isotype control mAb. (C) FcγRI/III^−/− male mice (four mice per group) were pretreated s.c. with anti-FcγRII/III mAb or control mAb and then injected i.v. the next day with 34-1-2s. (D) BALB/c FcγRIIb^+/+ or FcγRIIb^−/− male mice (six mice per group) were pretreated s.c. with anti-FcγRII/III mAb or control mAb and then injected i.v. the next day with 34-1-2s or control mAb. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 4.

mTRALI is C5 and C5aR dependent. (A) BALB/c male mice (four mice per group) were injected i.v. with 34-1-2s or control mAb, and lungs were harvested 5, 15, and 30 min later. Deposition of C3 (bright green) in lung sections was evaluated by immunofluorescence staining. Representative slides are shown. Bars, 70 µm. (B) Mixed genetic background H-2^d C3^+/+ and C3^−/− male mice (four mice per group) were injected i.v. with 34-1-2s or control mAb. Penh was evaluated in this and all subsequent panels in this figure. Lungs harvested 30 min after challenge were evaluated for percent water content in this and all subsequent panels in this figure. Results are representative of at least two experiments for all panels in this figure except as noted. (C) B10.D2 C5^+/+ and C5^−/− male mice (four to eight mice per group) were injected i.v. with 34-1-2s or control mAb. (D) BALB/c WT, C5aR^−/−, and C3aR^−/− male mice (four to eight mice per group) were injected i.v. with 34-1-2s or control mAb. Penh is only shown for 34-1-2s–injected mice in this and subsequent panels in this figure. (E) BALB/c male mice sufficient (WT) or deficient for the C5a receptor, C5L2 (C5L2^−/−; four mice per group) were injected i.v. with 34-1-2s or control mAb. This experiment was performed once. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 5.

Low C5 levels prevent development of mTRALI in female mice. (A) BALB/c male and female mice (six mice per group) were injected i.v. with 34-1-2s or control mAb. For this and all subsequent experiments in this figure, except B and C, Penh was evaluated, and lungs harvested 30 min after challenge were evaluated for percent water content. Results are representative of at least two experiments for all panels in this figure. (B) Relative plasma C5 content was determined by ELISA for B10.D2 C5^+/+ male and female and C5^−/− male mice (12 mice per group). (C) Relative plasma C5 content (12 mice per group) was determined by ELISA for 3.5-wk-old (prepubertal) or 8-wk-old (adult) WT male mice. (D) BALB/c female mice (four to eight mice per group) were pretreated i.v. with saline (none) or plasma from male or female WT mice, male C5^−/− mice, or male C3^−/− mice and then injected i.v. with 34-1-2s. (E) Adult and prepubertal male mice (four mice per group) were injected i.v. with 34-1-2s or control mAb. (F) BALB/c WT and FcRγ-deficient (FcRγ^−/−) and mixed background C3-deficient (C3^−/−) and B10.D2 C5-deficient (C5^−/−) male mice had their plasma C3 and C5 levels measured by ELISA (n = 5 mice per group for C3 and n = 12 mice per group for C5; ND = none detected). (G) BALB/c male and female and mixed background C3-deficient male mice had their plasma C3 levels measured by ELISA (five mice per group). (H) BALB/c male mice were pretreated with plasma from male FcRγ^−/− mice before i.v. challenge with 34-1-2s or control mAb. (I) BALB/c female mice (four mice per group) pretreated with normal or heat-inactivated plasma from WT male before i.v. challenge with 34-1-2s or control mAb. (J) B10.D2 C5-deficient male mice (four mice per group) were pretreated with plasma from C5-deficent or -sufficient male mice before i.v. challenge with 34-1-2s or control mAb. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 6.

Association of mTRALI with complement in untreated and LPS-pretreated mice. (A) BALB/c male and female mice (four mice per group) were challenged i.v. with 34-1-2s; lungs were harvested 0, 0.5, or 48 h later. In this and all subsequent panels, black boxes in left photographs outline the enlarged area immediately to the right, and black arrowheads point to the region of separation of vascular endothelium from its basement membrane. (B) BALB/c female mice (four mice per group) were injected i.v. with female or male plasma and then injected i.v. with 34-1-2s. Lungs were harvested 30 min later. (C) BALB/c C5aR-deficient mice (four mice per group) were challenged i.v. with 34-1-2s; lungs were harvested 30 min later. (A–C) Representative electron micrographs of lungs are shown. Bars, 2 µm. (D) Mixed genetic background H-2^d/d male C3^+/+ and C3^−/− mice (four mice per group) were inoculated i.t. with 3 µg LPS and challenged i.v. 3 d later with 34-1-2s. Penh was evaluated for the next 26 min. Lungs were harvested 30 min after injection and evaluated for percent water content. (E) BALB/c male mice (four mice per group) were initially left untreated or inoculated i.t. with 2 µg LPS and challenged i.v. 3 d later with 8 µg 34-1-2s or control mAb. Penh was evaluated for the next 24 h. (D and E) A repeat experiment gave similar results. *, P < 0.05 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 7.

Neither platelets nor neutrophils are required for mTRALI pathogenesis. (A) BALB/c male mice (four mice per group) were pretreated i.v. with saline or rabbit antiplatelet pAb (anti-plt) both 24 and 48 h before or only 4 h before injection i.v. with 34-1-2s or control mAb. In all panels in this figure except B, Penh was evaluated, and then lungs were harvested 30 min after 34-1-2s injection and evaluated for percent water content. Results are representative of at least two experiments for all panels in this figure except as noted. (B) BALB/c male mice (4–10 mice per group) were injected i.v. with rabbit antiplatelet pAb 4 h or both 24 and 48 h before determination of peripheral blood platelet count or with neuraminidase or saline 24 h before determination of peripheral blood platelet count. (C) BALB/c male mice (four to eight mice per group) were pretreated with neuraminidase or saline i.p. 1 d before i.v. injection with 34-1-2s or control mAb. (D) BALB/c male mice (four to eight mice per group) were pretreated with saline or hydroxyurea i.p. daily on day 0–6 and/or anti-Ly6G/C mAb i.v. at a dose of either 0.1 mg (Lo) on days 5 and 6 or 1 mg (Hi) on day 6. On day 7, mice were bled for determination of absolute neutrophil count (ANC; top left) or injected i.v. with 34-1-2s or control mAb (bottom). Penh values for control mAb-injected mice were all <1 (not depicted). In addition, some mice (10 mice per group; top right) were left untreated or were treated i.p. daily with hydroxyurea from day 0 to 6 and with low-dose anti-Ly6G/C mAb on days 5 and 6. On day 7, all mice were challenged i.v. with 34-1-2s, and lungs were harvested before or 30 s or 30 min after i.v. 34-1-2s injection. The number of neutrophils per high power field was determined for 10 fields of H&E-stained lung sections from each mouse. The quantitation of lung neutrophils part of the experiment was performed once. (E) BALB/c male mice (four to five mice per group) were pretreated i.p. with TER119 (rat IgG2b anti-RBC mAb) or isotype control mAb and injected i.v. 24 h later with 34-1-2s or control mAb. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 8.

Involvement of ROIs, monocytes, and non-BM–derived cells in mTRALI pathogenesis. (A) BALB/c male mice (four mice per group) were pretreated i.v. with N-acetylcysteine (NAC) or saline and injected i.v. 5 min later with 34-1-2s or control mAb. In all panels, except for D and F, Penh was evaluated and then lungs were harvested 30 min after 34-1-2s injection and evaluated for percent water content. Results are representative of at least two experiments for this and all subsequent experiments in this figure. (B) BALB/c male mice (five to eight mice per group) were pretreated i.v. with gadolinium or saline and injected i.v. the next day with 34-1-2s. (C) BALB/c male mice (four mice per group) were pretreated i.p. and i.t. with liposomes containing clodronate or PBS and injected i.v. 2 d later with 34-1-2s or control mAb. (D) BALB/c male mice (four mice per group) were injected with liposome-encapsulated clodronate (Clod) or PBS i.p. or i.t. Absolute monocyte count was determined for blood obtained 2 d later. Another group was injected i.v. with gadolinium (Gd). Blood was drawn the next day, and absolute monocyte count was determined. (E) BALB/c male mice (four mice per group) were injected with liposome-encapsulated clodronate or PBS either i.p. or i.t. and challenged i.v. 2 d later with 34-1-2s. (F) BALB/c male mice (four mice per group) were injected i.v. with biotinylated 34-1-2s or control mAb. Lungs were obtained 60 s later and stained for biotin (brown). Representative sections are shown. Bars, 70 µm. (G) BALB/c (H-2^d) and C57BL/6 (H-2^b) RAG2-deficient male mice (four mice per group) were lethally irradiated and injected i.v. the next day with BM cells from RAG2-deficient H-2^d or H-2^b mice. 6 wk later, H-2^d+ blood neutrophil counts were determined. 1 wk after that, mice were injected i.v. with 34-1-2s. An experiment in which BM chimeras were produced by reconstituting irradiated BALB/c or C57BL/6 mice with BM from (BALB/c × C57BL/6) F1 mice and challenged with 34-1-2s gave nearly identical results. Horizontal bars indicate the mean. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 9.

mAb properties required to induce mTRALI. (A) BALB/c male mice (four mice per group) were challenged i.v. with eight different mouse IgG2a anti–H-2^d mAbs. Mean maximum Penh during the next 26 min is shown. Results are representative of at least two experiments for all panels in this figure. (B) C57BL/10, B10.D2, and B10A-H2^a male mice (four mice per group) were injected i.v. with 34-1-2s or control mAb. For all panels in this figure except C and D, Penh was evaluated, and lungs harvested 30 min after injection were evaluated for percent water content. In this panel, Penh values for control mAb-injected mice were all <1 (not depicted). (C) Spleen cells harvested from BALB/c male mice were evaluated in vitro by flow cytometry for neutrophil and CD8⁺ T cell binding of six different mouse IgG2a anti–mouse H-2^d mAbs to neutrophils (MFI, mean fluorescence intensity). (D) BALB/c male mice (four mice per group) were challenged i.v. with 34-1-2s, 34-5-8s, SF1-1.1.10, or control mAb. Blood drawn 3 min later was evaluated by flow cytometry for C3 binding to nucleated cells. (E) BALB/c male mice (four mice per group) were pretreated i.v. with SF1-1.1.10, 34-5-8s, or control mAb and injected i.v. 2 h later with 34-1-2s or control mAb. Blood drawn 3 min later from some mice was evaluated for 34-1-2s binding to nucleated cells. Penh and percent water content of lungs were determined for the other mice. (F) BALB/c male mice (four mice per group) were injected i.v. with 34-1-2s, SF1-1.1.10, 34-1-2s + SF1-1.1.10, or control mAb. Some mice were evaluated for abnormal breathing (Penh; middle) and percent lung water content (right). Mononuclear cells in blood drawn from other mice 3 min after mAb injection were evaluated by flow cytometry for C3 binding (left). (G) BALB/c male mice (four mice per group) were injected i.v. with control mAbs, 34-1-2s, SF1-1.1.10, 34-5-8s, or 34-4-20s, or with the combinations of these mAbs shown. *, P < 0.05; and **, P < 0.005 using a nonparametric two-tailed t test. Mean ± SEM is shown.

Figure 10.

Induction of mTRALI by anti–MHC class I mAb. The simplest mechanism of mTRALI induction consistent with our data is shown. Anti–H-2^d mAb (1) binds to H-2^d on vascular endothelial cells, activating complement (2) with production of C5a (3), which binds to the C5aR on monocyte/macrophages (4), attracting them to lung vasculature (5) and inducing them to secrete ROIs (6), which disrupts pulmonary vascular endothelial basement membranes (7), allowing fluid and protein to leak into the lung parenchyma (8), causing noncardiogenic pulmonary edema.