Plasma-purified human AAT (99% purity, Zemaira, CSL Behring, Kankakee, IL, USA) was used for experiments after buffer exchange to the sterile Hank’s Balanced Salt Solution (HBSS) (Merck Millipore, Darmstadt, Germany) using 10K centrifugal filter columns (Sartorius, Göttingen, Germany). The protein concentration was determined using the BCA Protein Assay Kit (Pierce™, Rockford, IL, USA) according to the instructions of the supplier. The oxAAT was prepared from AAT (Zemaira®) by adding N-chlorosuccinimide (Sigma-Aldrich, Merck, Darmstadt, Germany) at a molar ratio of 1:20 (AAT: N-chlorosuccinimide) for 20 min at room temperature. Afterward, to remove the N-chlorosuccinimide, AAT preparations were washed with phosphate-buffered saline (PBS; Sigma-Aldrich, St. Louis, MO, USA) using Vivaspin 20 centrifugal filter devices with a cutoff of 10K. The oxAAT did not form complexes with elastase and showed a retarded electrophoretic mobility relative to a native AAT. A highly purified (90%), glycosylated form of recombinant AAT protein produced in CHO cells was a gift from ExcellGene, Monthey, Switzerland.

Ten- to twelve-week-old male C57BL/6N mice (20 to 25 g) were obtained from Charles River Laboratories (Sulzfeld, Germany). Mice were maintained on mouse chow and tap water ad libitum in a temperature-controlled chamber at 24°C with a 12:12-h light–dark cycle. All procedures were approved by the local committee for the care and use of laboratory animals (Lower Saxony Office for Consumer Protection and Food Safety, LAVES no. 21-3761) and were performed in accordance with international guidelines on animal experimentation.

Polymicrobial sepsis in mice was induced by CLP surgery. In brief, mice were anesthetized with isoflurane (induction of 3%, maintenance of 1.5%, and oxygen flow of 3 L/min), and a 1-cm ventral midline abdominal incision was made. The cecum was then exposed, ligated with 4-0 silk sutures, and punctured through using a 24-gauge needle. The punctured cecum was gently squeezed to expel a 1- to 2-mm droplet of fecal material and returned to the abdominal cavity. The incision was closed in layers using 4-0 surgical sutures. Mice were fluid-resuscitated with pre-warmed normal saline (500 μL) intraperitoneally (i.p.) immediately after the procedure. Sham animals underwent the same procedure except for CLP. All experiments were performed at the same time of day. For pre-operative treatment, 0.1 mg/kg buprenorphine and 100 mg/kg metamizole were administered s.c. For post-operative analgesia, animals were s.c. injected with 0.1 mg/kg buprenorphine twice daily for 3 days.

For survival analysis, mid-grade sepsis was induced by CLP surgery with ligation of 50% of cecum length (21). Mice were treated i.p. with nAAT, oxAAT, or recAAT (200 mg/kg body weight) or HBSS (vehicle treatment) immediately after CLP or sham surgery. The survival was monitored up to 14 days after surgery. Blood samples were collected 3 days prior to and at 24 h after surgery under light isoflurane anesthesia. EDTA plasma samples were generated, stored at −80°C, and used for the quantification of TNF, interleukin 6 (IL-6), and chemokine ligand 2 (CCL2).

Short-term high-grade sepsis was induced by CLP surgery with ligation of 75% of cecum length (21), and immediately after surgery, mice were treated i.p. with recAAT or vehicle (HBSS buffer). Twenty hours after surgery, mice were anesthetized with isoflurane and, after blood collection, were sacrificed. EDTA plasma samples were generated, and peritoneal lavage (PL) was performed using 3 mL of PBS and collected in tubes without anti-coagulant. The volume of collected PL fluid was measured in each sample, and the total cell number was determined using a hemocytometer (Neubauer Zählkammer, Gehrden, Germany). Plasma and PL fluids were stored at −80°C for further analyses.

In separate experiments, the Evans blue assay was performed to estimate microvascular permeability. Immediately after CLP surgery, 200 µL of 0.25% wt/vol Evans blue dye (Sigma-Aldrich) in PBS was injected intravenously. Blood sampling was performed under isoflurane anesthesia at 20 h after CLP surgery, the mice were sacrificed, and PL was performed. The concentration of Evans blue dye in appropriate dilutions of plasma and PL fluid samples was measured spectrophotometrically at 620 nm. The following formula was used to correct the optical densities for contamination with heme pigments: E620 (corrected) = E620 (raw) − (E405 (raw) × 0.014. Plasma exudation was quantified as the ratio of extinction in PL fluid to extinction in plasma.

Levels of the pro-inflammatory cytokines TNF, IL-6, and CCL2 were quantified in plasma and PL fluid by bead-based flow cytometry assay (CBA Kit; BD Biosciences, Heidelberg, Germany) in accordance with the instructions of the manufacturer.

The inflammatory cell populations in the PL fluid were analyzed by flow cytometry using a fluorescence-activated cell sorting (FACS) Canto cytometer (BD Biosciences, Franklin Lakes, NJ, USA). The following commercial monoclonal antibodies (BioLegend, San Diego, CA, USA) were used for the detection of leukocytes, macrophages, polymorphonuclear neutrophils (PMNs), B cells, and T cells, respectively: anti-CD11b, anti-F4/80, anti-Gr1, anti-CD115, anti-TCR β, and anti-CD19. Further analyses were performed using FlowJo software (Tree Star, Ashland, OR, USA).

Levels of endogenous AAT were estimated in plasma and PL fluid samples obtained at 20 h post-surgery from untreated septic and sham mice by Western blotting. In brief, 1 µL of plasma samples and 5 µL of PL fluid samples were separated on 10% sodium dodecyl sulfate–polyacrylamide gels (SDS–PAGE). From gels, proteins were transferred onto polyvinylidene fluoride membrane by semidry Western blotting. For specific detection of AAT, primary rabbit polyclonal anti-AAT antibody was used at a dilution of 1:800 (DAKO, Glostrup, Denmark). The immune complexes were visualized using appropriate secondary horseradish peroxidase-conjugated antibodies (DAKO, Denmark) at a dilution of 1:10,000 and enhanced chemiluminescence (ECL) Western blotting substrate (Thermo Fisher Scientific, Grand Island, NY, USA). The density of the specific bands was quantified using Image Lab v5.2.1 software (Bio-Rad, Hercules, CA, USA).

Measurement of labile heme concentration in PL fluid samples was performed in 96-well plates as previously described (22). Briefly, 5 µL of an appropriately diluted PL fluid sample was added to 95 µL of HBSS buffer containing 0.75 µM Apo-Horseradish Peroxidase (Apo-HRP) (BBI Solutions, Gwent, UK) and incubated for 10 min at 4°C. Simultaneously, hemin standards ranging from 0.25 to 2.5 nM were prepared in HBSS buffer in a final volume of 100 μL from a stock solution of 25 nM hemin (Frontier Scientific, Logan, UT, USA) and incubated for 10 min at 4°C. Then, 5 μL of each sample and standard were added to a 96-well plate. To start the assay, 200 μL of 3,3′,5,5′-tetramethylbenzidine (TMB) substrate was added per well. To determine the concentration of labile heme, the absorbance was measured at 652 nm for 2–3 min. The time point at which the highest hemin standard induced an absorbance from 1.6 to 2 was used.

EDTA blood, PL fluid, and spleens were obtained from septic mice treated with vehicle, nAAT, or oxAAT at 20 h after induction of high-grade sepsis. The peritoneal lavage fluid was serially diluted 1:10 in PBS five times (4°C). The spleen was submerged for 1 min in 70% ethanol to remove potential contaminations of its surface before rinsing the organ briefly with PBS. Then, the organ was homogenized using 100-µm BD Falcon™ Cell Strainer and pistil, thereby rinsing the sieve with 2 mL of sterile PBS. The spleen homogenate at a volume of 500 µL was diluted 1:10 in PBS. Then, 100 µL of the undiluted lavage fluid, the spleen homogenate, and each dilution was transferred onto agar plates for culture at 35°C. Columbia agar containing 5% sheep blood plates (Becton Dickinson, Franklin Lakes, NJ, USA; 4354071) used for the growth of Gram-positive and Gram-negative bacteria and MacConkey agar plates (Mast Diagnostica GmbH, Reinfeld/Stormarn, Germany; 202010) used for growth of Gram-negative bacteria were incubated under aerobic conditions; Schaedler agar plates (Becton Dickinson, 4354084) incubated under anaerobic conditions in an air-tight plastic container (AnaeroGen 3.5L, Thermo Fisher Scientific) were used to permit growth of Gram-positive and Gram-negative anaerobic bacteria. EDTA blood was diluted 1:10 in PBS, and 100 µL of the undiluted and diluted samples was evenly spread on a Columbia agar and Schaedler agar for culture under aerobic or anaerobic conditions, respectively. In all cases, after approx. 48 h of culture, bacterial colonies on the plates were counted. The number of colonies forming units in the original sample was calculated considering the applied volume and dilution factor.

Immortalized mouse peritoneal mesothelial cells (MPMC) were cultivated as described previously (23). Briefly, the cells were grown on 6-well or 24-well cell culture plates to 80% confluence in Roswell Park Memorial Institute (RPMI) medium GlutaMAX™ (Thermo Fisher Scientific, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS), 1% Penicillin/Streptomycin (Sigma-Aldrich, St. Louis, MO, USA), 0.4 mg/mL hydrocortisone (Sigma-Aldrich, St. Louis, MO, USA), 1% Insulin-Transferrin-Selenium sodium pyruvate (Thermo Fisher Scientific, Waltham, MA, USA), and 10 U/mL recombinant mouse interferon γ (IFN-γ; Cell Sciences, Canton, MA, USA) at 33°C (permissive conditions). The cells were differentiated for 3 days in the same medium at 37°C without IFN-γ (non-permissive conditions) and starved overnight in a serum-free medium. MPMC were stimulated for 24 h with different concentrations of lipopolysaccharide (LPS) (E. coli O111:B4, Sigma-Aldrich), free hemin (Sigma-Aldrich), or recAAT separately or in combination. The production of pro- and anti-inflammatory cytokines TNF, IL-6, IL-10, and CCL2 were measured in the conditioned medium as described above for PL fluid samples.

Resident peritoneal macrophages (MPMΦ) were obtained from healthy C57BL/6N mice by peritoneal lavage (2 × 5 mL PBS). After centrifugation at 300 g for 10 min at 4°C, the supernatants were decanted, and the remaining pellets were washed with RPMI GlutaMAX™ medium supplemented with 10% FBS, resuspended in the same medium, and distributed into 24-well culture plates at a concentration of 1.0 × 10⁶ cells/mL per well. Plates were incubated overnight at 37°C, 5% CO₂, and 95% humidity to allow macrophage adhesion.

Non-adherent cells were removed by vigorous washing with RPMI 1640 medium. MPMΦ were starved for 4 h in 1% FBS/RPMI 1640 medium and then stimulated with 5 ng/mL LPS, with increasing concentrations of recAAT (0.1, 1, 10, and 100 µg/mL) or with LPS/recAAT combinations for 24 h. The MPMΦ medium without stimuli served as a control. The release of pro- and anti-inflammatory cytokines, TNF, IL-6, IL-10, and CCL2, was examined in the conditioned medium as described above for PL fluid samples.

RNA was isolated using NucleoSpin RNA II Kit (Macherey-Nagel GmbH & Co. KG, Duren, Germany) and reversely transcribed with M-MLV-RT (Promega GmbH, Mannheim, Germany) according to the manufacturer’s instructions. Real-time polymerase chain reactions were performed in triplicates on a LightCycler 480 or LightCycler 96 using SYBR Green (Roche, Grenzach-Wyhlen, Germany). Primer sequences were selected using PrimerBank and are reported in the Supplementary Methods . HPRT1 was used as a reference gene. Primers for HPRT1 were obtained from Biomol GmbH (Hamburg, Germany).

Apoptosis was assessed using fluorescein isothiocyanate (FITC) Annexin V Apoptosis Detection Kit with Propidium Iodide solution (PI) (BioLegend). To distinguish early-stage apoptotic cells from late-stage apoptotic and necrotic cells, Annexin V and Propidium Iodide solution PI were used according to the manufacturer’s instructions.

The Kaplan–Meier curves were used to illustrate survival between treatment groups, and statistical assessment was performed using the log-rank test. For other parameters, the D’Agostino and Pearson omnibus normality test was used to test for normality. Multiple comparisons were analyzed using one-way analysis of variance (ANOVA) with Sidak’s post hoc correction or the non-parametric Kruskal–Wallis test with Dunn’s post hoc correction. Data are presented as mean + standard deviation (SD). A p-value of <0.05 indicated statistical significance. All statistical analyses and data visualizations were performed using GraphPad Prism v.9.0 (GraphPad Software, La Jolla, CA, USA).

Mouse plasma and PL fluid samples were collected 20 h after high-grade sepsis induced by CLP surgery. No differences in AAT plasma levels were observed between sham and septic mice ( Figure 1A ). However, significantly higher AAT levels were observed in PL fluid in septic mice compared to sham mice ( Figure 1B ).

In survival experiments, mid-grade sepsis was induced in mice by CLP surgery (n = 8/group) with ligation of 50% of the cecum length. This semi-lethal CLP mouse model with a survival rate of 40% is the most suitable CLP model to investigate the protective effects of therapeutic interventions on sepsis-related mortality (21). Sham-operated mice (n = 6) served as controls. Septic mice received 200 mg/kg recAAT, human nAAT, or oxAAT or were injected with vehicle (HBSS buffer) immediately after CLP surgery. As shown in Figure 2A , only 18% of vehicle-treated mice survived, compared to 87%, 71%, and 57% in the groups treated with recombinant, oxidized, and plasma-derived native AAT, respectively. Mice treated with all three forms of AAT showed significantly better outcomes compared to HBSS treatment (recAAT p < 0.01, nAAT p < 0.05, and oxAAT p < 0.01 vs. vehicle treatment).

The systemic inflammatory response was assessed by measuring the plasma concentration of the pro-inflammatory cytokines TNF and IL-6 and the chemokine CCL2. As expected, CLP-induced peritonitis was associated with strong systemic upregulation of all three pro-inflammatory mediators. Treatment with any of the three forms of AAT significantly reduced plasma levels of CCL2 and TNF but had no effect on IL-6 levels ( Figures 2B–D , respectively).

To investigate possible mechanisms underlying the beneficial effects of AAT on survival and the systemic inflammatory response, high-grade sepsis was induced by CLP surgery with ligation of 75% of the cecum length. The in vivo clearance of bacteria was examined in the PL fluid, blood, and spleen 20 h post-CLP operation. Colony-forming bacterial loads in blood, PL fluid, and spleen homogenates were slightly lower, but not significantly changed, in the septic mice treated with human nAAT or oxAAT compared to the vehicle-treated septic mice. These results were observed using Columbia blood agar plates ( Figure 3 ) as well as using Schaedler agar and MacConkey agar plates (data not shown).

Since recAAT showed the best effect on mouse survival, the following short-term experiments (follow-up 20 h) were performed using recAAT only. High-grade sepsis was induced in mice by CLP surgery with ligation of 75% of the cecum length (n = 10–12 mice per group). Sham-operated mice served as controls (n = 6–8 mice per group). Plasma levels of lactate dehydrogenase (LDH) reflecting the degree of sepsis-induced overall tissue damage were determined in the plasma of sham- and CLP-operated mice 20 h after surgery. As expected, induction of sepsis resulted in a significant increase in LDH levels compared to sham mice. Plasma LDH levels in septic mice treated with recAAT were significantly lower compared to those in vehicle-treated septic mice and were not significantly different from those in sham mice ( Figure 4A ).

Next, renal function was estimated by measuring plasma creatinine after sham or CLP surgery. As shown in Figure 4B , plasma creatinine levels were significantly higher in septic mice treated with a vehicle, whereas creatinine levels in septic mice treated with recAAT did not differ from sham controls. Finally, we determined plasma levels of alanine aminotransferase (ALT) and aspartate transaminase (AST) as markers of sepsis-induced hepatocellular injury. ALT and AST levels were significantly higher in the plasma of vehicle-treated septic mice compared to sham animals ( Figures 4C, D ). Compared to vehicle-treated septic mice, the increase in AST and ALT levels was less pronounced in recAAT-treated septic mice, but without statistical significance.

The levels of inflammatory mediators were analyzed in plasma and PL fluid 20 h after CLP or sham surgery. As expected, CLP-induced peritonitis was associated with strong systemic ( Figure 5A ) and local ( Figure 5B ) upregulation of pro-inflammatory cytokines IL-6, CCL2, and TNF as well as upregulation of anti-inflammatory IL-10 compared to sham animals. Levels of neutrophil chemoattractant CXCL2 were also increased in PL of septic mice ( Figure 5B ). Compared to vehicle-treated septic mice, recAAT-treated mice showed significantly lower plasma TNF levels and a strong trend toward lower CCL2, IL-6, and IL-10 levels ( Figure 5A ). In contrast, in PL fluid, the levels of IL-6, CCL2, CXCL2, and IL-10 were significantly lower in septic mice treated with recAAT, whereas TNF levels were not affected ( Figure 5B ).

Recruitment of inflammatory leukocytes such as neutrophils and monocytes into the abdominal cavity is a hallmark of abdominal sepsis (24). Therefore, we analyzed the total white blood cell counts and quantified the number of lymphocytes, monocytes, and granulocytes in the PL fluid 20 h after surgery. The total number of inflammatory cells in the PL fluid was greatly increased in septic mice compared to sham controls ( Figure 6A ), mainly due to an increase in granulocytes and monocytes ( Figures 6B, C ). The total number of lymphocytes did not change ( Figure 6D ). Treatment of septic mice with recAAT reduced granulocyte numbers ( Figure 6B ) but increased monocyte/macrophage content in the peritoneal cavity ( Figure 6C ).

Vascular leakage caused by endotoxemia is a devastating feature of the disproportionate host immune response (25). Compared to sham surgery, a significant increase in vascular leakage was observed 20 h after the CLP procedure, as reflected by increased extravasation of Evans blue dye into the abdominal cavity. This sepsis-induced vascular hyperpermeability was significantly reduced by treatment with recAAT ( Figure 7A ).

Both vascular leakage and cell damage caused by endotoxemia can contribute to the local accumulation of free heme in the abdominal cavity (8). Indeed, we observed significantly higher levels of free heme in the PL fluid of septic mice compared to sham controls. Treatment with recAAT significantly reduced the amount of labile heme in the PL fluid of septic mice ( Figure 7B ).

Mesothelial epithelium covering the internal body cavities and organs, and resident peritoneal macrophages pose the first line of defense in abdominal bacterial sepsis. In the early stages of infection, bacterial endotoxins activate Toll-like receptors on the surface of these cells, leading to a fulminant release of pro-inflammatory and chemotactic mediators, representing a crucial event in the pathogenesis of sepsis (26). To test whether AAT impairs the LPS-induced release of pro-inflammatory mediators, immortalized MPMC and primary naïve MPMΦ were used for in vitro experiments. As shown in Figure 8A , exposure of MPMC to increasing concentrations of LPS ranging from 1 to 100 ng/mL for 24 h resulted in an LPS dose-dependent increased release of CCL2. In the next series of experiments, MPMC were stimulated with a constant dose of LPS (100 ng/mL) for 4 h and analyzed for the expression of the IL-6, TNF, CCL2, CXCL1, and CX3CL1 genes. As expected, LPS upregulated the expression of all genes analyzed ( Figure 8B ). This LPS-induced inflammatory gene expression was significantly reduced in the presence of recAAT, whereas the expression of TNF was completely abolished ( Figure 8B ).

Similar results were obtained for primary peritoneal macrophages. recAAT reduced the cellular release of IL-6 and TNF in response to LPS in a concentration-dependent manner compared to LPS treatment alone ( Figure 8C ).

Recently, a synergistic effect of free heme on LPS-induced cytokine secretion in murine macrophages was demonstrated (27). Since we observed the accumulation of free heme in PL fluid from septic mice, in the following experiments, MPMΦ were stimulated with 5 ng/mL LPS alone or in the presence of 5 µM heme. While heme alone had no effect on TNF release into a conditioned medium, LPS-induced TNF release was significantly higher in the presence of heme. Notably, the addition of recAAT (100 µg/mL) to MPMΦ not only lowered LPS-induced TNF release but also completely blocked the synergistic effect of heme ( Figure 9A ). Similar results were obtained for MPMC stimulated with 5 ng/mL LPS alone or in the presence of 0.5 μM free heme. While free heme alone did not increase CCL2 levels in a conditioned medium, LPS-induced CCL2 release was significantly increased in the presence of heme ( Figure 9B ). Treatment of MPMC with recAAT not only reduced CCL2 release after LPS stimulation but also completely blocked the synergistic effect of free heme ( Figure 9B ).

Cells were stimulated with 5 ng/mL LPS, 10 μM free heme, or a combination of both. LPS alone did not increase the percentage of apoptotic (AV+/PI−) or necrotic (AV+/PI+) cells. In contrast, cytotoxicity was observed after 24 h of cell culture with 10 μM heme and with the LPS/heme combination. The addition of recAAT (10 μg/mL) to the cell culture medium slightly but not significantly reduced the percentage of apoptotic cells, whereas it reduced the percentage of necrotic cells significantly ( Figure 10 ).