Targeting of mannan-binding lectin-associated serine protease-2 confers protection from myocardial and gastrointestinal ischemia/reperfusion injury

Abstract

Complement research experienced a renaissance with the discovery of a third activation route, the lectin pathway. We developed a unique model of total lectin pathway deficiency, a mouse strain lacking mannan-binding lectin-associated serine protease-2 (MASP-2), and analyzed the role of MASP-2 in two models of postischemic reperfusion injury (IRI). In a model of transient myocardial IRI, MASP-2-deficient mice had significantly smaller infarct volumes than their wild-type littermates. Mice deficient in the downstream complement component C4 were not protected, suggesting the existence of a previously undescribed lectin pathway-dependent C4-bypass. Lectin pathway-mediated activation of C3 in the absence of C4 was demonstrated in vitro and shown to require MASP-2, C2, and MASP-1/3. MASP-2 deficiency also protects mice from gastrointestinal IRI, as do mAb-based inhibitors of MASP-2. The therapeutic effects of MASP-2 inhibition in this experimental model suggest the utility of anti-MASP-2 antibody therapy in reperfusion injury and other lectin pathway-mediated disorders.

Fig. 1.

MASP-2 is essential for lectin pathway functional activity. (A) Lectin pathway-dependent C4 activation: plasma from WT, heterozygous (M2^+/−), and homozygous (M2^−/−) MASP-2–deficient mice were assayed for C4 convertase activity on microtiter plates coated with mannan or zymosan. Serial dilutions of pooled WT plasma were included on each plate, standard curves plotted, and C4 deposition calculated as a proportion of that in the pooled WT serum. Results are means (± SD). (B) Reconstitution of plasma with recombinant MASP-2. Plasma (0.5%) pooled from four MASP-2^−/− mice was preincubated with recombinant murine MASP-2 (rMASP-2) or inactive recombinant MASP-2 (iMASP-2). C4 convertase activity was then assayed on microtiter plates coated with mannan. (C) The classic pathway makes no significant contribution to C3 deposition on mannan: diluted mouse sera were added to microtiter plates coated with mannan and C3b deposition assayed. (D) MASP-2^−/− mice retain a functional classical pathway: C3b deposition was assayed on microtiter plates coated with immune complexes (generated in situ by coating with BSA then adding goat anti-BSA IgG). (E) MASP-2–deficient mice retain a functional alternative pathway: C3b deposition was assayed on zymosan coated microtiter plates under conditions that permit only alternative pathway activation (buffer containing Mg²⁺ and EGTA). Results in B to E are means of duplicates and are typical of three independent experiments.

Fig. 2.

MASP-2 deficiency significantly reduces MIRI-induced tissue loss. (A) MIRI-induced tissue loss following LAD ligation and reperfusion in MASP-2^−/− mice and WT littermate controls. Hearts were subjected to 30 min of focal ischemia by occluding the LAD, followed by 120 min reperfusion. At the end of the experiments, the LAD was reoccluded and 5% Evans blue injected into the left ventricle (LV) to determine the risk zone (RZ). Hearts were excised, sectioned, and incubated at 37 °C for 30 min in 3% TTC solution, which is a metabolic stain. Nonperfused infarcted tissue remains colorless, nonperfused and noninfracted tissue (RZ) stains red, and normal perfused myocardium stains blue. Right panels are image-enhanced. (B) Risk volume as a percentage of total myocardial volume. The infarct size (INF), calculated as a percentage of the AAR, is significantly reduced in MASP-2^−/− mice compared with their WT littermates. (*P = 0.035, two-tailed Student t test.) (C) Plots INF against AAR (both as percentages of total LV volume). MASP-2^−/− mice show a highly significantly reduction in infarct size, regardless of the extent of the ischemic insult (AAR). (D and E) Hearts from MASP-2–deficient mice are not protected from reperfusion injury when myocardial ischemia is induced ex vivo in buffer-perfused hearts (Langendorff mode heart preparations), which demonstrates that the differences shown in B and C are caused by a plasma factor, not by a lower susceptibility of the myocardium of MASP-2^−/− mice to ischemic damage. LAD ligations and reperfusion were conducted as in B and C, but in absence of plasma (n = 10 per group in B to E). (F and G) C4^−/− mice are not protected from MIRI-induced tissue loss. (F) INF and AAR in C4^−/− mice (n = 6) and WT littermate controls (n = 7). (G) INF as a function of AAR, showing that C4^−/− mice are as susceptible to MIRI as their WT littermates. (Error bars in B, D, and F show means ± SD; C, E, and G analyzed using ANCOVA).

Fig. 3.

Lectin pathway-dependent activation of C3 in the absence of C4. (A) The lectin pathway activates C3 in the absence of C4. C3b deposition was assayed on mannan-coated microtiter plates using recalcified WT and C4^−/− plasma at concentrations prohibitive for alternative pathway activation (1.25% and below). Preincubating C4^−/− plasma with excess (1 mg/mL) fluid-phase mannose before the assay completely inhibits C3 deposition. Results are typical of three independent experiments. (B) Inhibition of C3 deposition using anti–MASP-2 mAb. Plasma (1%) was mixed with various concentrations of anti-rat MASP-2 mAb AbyD 04211 (abscissa) and C3b deposition assayed on mannan-coated plates. Results are means (± SD) of four assays. (C) A comparative analysis of C3 convertase activity in plasma from complement deficient mouse strains tested under lectin or classic activation pathway-specific conditions. WT mice (n = 6), MASP-2^−/− mice (n = 4), MASP-1/3^−/− mice (n = 2), C4^−/− mice (n = 8), C4/MASP-1/3^−/− mice (n = 8), Bf/C2^−/− (n = 2), and C1q^−/− mice (n = 2) were tested in parallel. Reconstitution of Bf/C2^−/− plasma with 2.5 μg/mL recombinant rat C2 (Bf/C2^−/− + C2) restored C3b deposition. Serial dilutions of pooled WT plasma were included on each plate, standard curves plotted, and C3 deposition calculated as a proportion of that seen in the pooled WT serum. Individual samples were assayed in duplicated at a concentration of 1% plasma. Results are means ± SD **P < 0.01 (two-tailed t test). (D) Kinetics of C3 convertase activity in plasma from various complement deficient mouse strains tested under lectin activation pathway-specific assay conditions (1% plasma, results are typical of three independent experiments).

Fig. 4.

The absence of MASP-2 protects mice from GIRI-mediated injury. (A) MASP-2^−/− mice show a significant degree of protection from severe GIRI damage following transient (40 min) occlusion of the mesenteric artery and reperfusion (3 h) of ischemic gut tissue. *P < 0.05 as determined by Student t test (two-tailed). (B) In vivo ablation of lectin pathway activity following a single intraperitoneal injection of recombinant antimurine MASP-2 antibody AbyD 04211 (0.6 mg/kg body weight). At each time point, three mice were killed and serum assayed for lectin pathway-dependent C4 activation. Lectin pathway functional activity was normalized to lectin pathway activity in pooled sera from naive mice measured either in the absence (100%) or in the presence of 100-nM blocking antibody (0%). Results are means (± SEM). (C) Effect of anti–MASP-2 mAb treatment on the severity of GIRI: mice were dosed with 1 mg/kg of AbyD 04211, or an irrelevant isotype control antibody, or injected with saline 18 h before being subjected to 40 min gastrointestinal ischemia and 3 h of reperfusion (n = 10 per group). (*P < 0.05 when comparing animals treated with either the MASP-2 inhibitory antibody AbyD 04211 or an isotype control). Sham animals (n = 5 per group) were treated in an identical fashion, except that no clamp was applied to the mesenteric artery. (D) Histological presentation of GIRI-mediated pathology in the small intestine of WT C57BL/6 mice pretreated with a single intraperitoneal injection of either isotonic saline, an isotype control antibody (1 mg/kg body weight), or recombinant antimurine MASP-2 antibody AbyD 04211 (1 mg/kg body weight) 12 h before the induction of GIRI, and their respective sham controls. Arrowheads indicate subepithelial spaces in the luminal part of the villi (characterized by the lack of cellular content beneath the continuous epithelial layer) as typical features of GIRI pathology. (Magnification, 100×.)