Using ficoll-gradient centrifugation, peripheral blood mononuclear cells were isolated from the blood of outpatient volunteers (University of Florida Institutional Review Board protocol 2015-01051). Characteristics of the volunteers are shown in Table 1 . Monocytes were purified using a monocyte enrichment kit (Stemcell Technology, Vancouver), following the manufacturer’s instruction. Monocytes were plated in 12-well plates at 300,000 cells per well in macrophage differentiation media (RPMI 1640 containing 10% heat-inactivated FBS, 20 Units/ml penicillin, 20 μg/ml streptomycin, 250 ng/ml Amphotericin B, recombinant human GM-CSF (0.5 ng/ml) and recombinant human M-CSF (5 ng/ml)) and differentiated for 7 days. Both growth factors, GM- and M-CSF, exist in the lung, and especially GM-CSF is important to induce AAT expression (24). Supplemental medium (50% of the volume in each well) was added every 3 days after removal of half of the old media, and cells were used on day 7 for any treatment. To induce inflammatory signaling in MDMs, the cells were stimulated with 10 ng/ml LPS from Escherichia coli O111:B4 (Sigma-Aldrich, St. Louis) overnight. Non-treated and LPS-treated MDMs were harvested for RNA extraction using the Qiagen RNeasy kit (Qiagen, Hilden). To inhibit LPS-mediated TNFα signaling, MDMs were pretreated with LPS for 1 h and incubated with TNFα neutralizing antibody (MAB210, R&D Systems, Minneapolis) or isotype control (MAB002, R&D Systems, Minneapolis) for 18 h.

To examine AAT distribution, MDMs were differentiated on glass slides and fixed in 4% paraformaldehyde for 20 min and permeabilized for 10 min in PBS containing 0.01% Triton X-100. The permeabilized cells were incubated with rabbit anti AAT polyclonal antibody (Abcam, Cambridge) at 1:400 dilution in PBS containing 0.1% Tween 20 for 1 h. After washing with PBS-Tween 20, cells were immunostained with Alexa Fluor488 goat anti-rabbit (Abcam, Cambridge) at 1:500 dilution at room temperature for 1 h. The immunostained cells were mounted on glass slides using VECTASHIELD mounting media with DAPI and examined using a fluorescence microscope (BZ-X700, Keyence, Osaka). For quantification of intracellular AAT, AAT fluorescent intensity and number of cells were measured with BZ software, and the fluorescent intensity was normalized to the cell number. To examine CD36 distribution, MDMs were incubated with mouse anti CD36 monoclonal antibody (ThermoFisher, Waltham) at 1:20 dilution in blocking solution (Invitrogen, Carlsbad) overnight at 4°C. After washing with PBS-Tween 20, cells were immunostained with Alexa Fluor647 goat anti-rabbit (Abcam, Cambridge) at 1:1000 dilution at room temperature for 1 h. For quantification of CD36, ~1000 MDMs were evaluated for each MDM group (n=4).

Total proteins were extracted from MDMs using RIPA lysis buffer (Cell Signaling, Danvers) plus protease and phosphatase inhibitors. The protein concentration of each sample was measured using a standard Bradford assay (BioRad, Hercules) and equal amounts of protein were loaded onto a 12% SDS polyacrylamide gel. After gel electrophoresis, the proteins were transferred onto a nitrocellulose membrane using a wet-transfer system, and the membrane was blocked in Tris-buffered saline with 0.1% Tween 20 (TBST) containing 5% nonfat dry milk. The membrane was immunoblotted overnight at 4°C with AAT rabbit polyclonal antibody (DAKO, Carpinteria) at a dilution of 1:5,000 in TBST. Horseradish peroxidase conjugated anti-rabbit antibody (BioRad, Hercules) was used for secondary labeling at 1:5,000 in TBST for 1 h at room temperature. The membrane was reprobed with GAPDH rabbit polyclonal antibody (Proteintech, Rosemont) at 1:5,000 in TBST. A horseradish peroxidase conjugated anti-rabbit (BioRad, Hercules) was used for secondary labeling. Protein bands were visualized by enhanced chemiluminescence (ECL, GE Healthcare, Chicago).

AAT was measured in conditioned media of MDMs using a sandwich enzyme-linked immunosorbent assay (ELISA). Ninety-six well ELISA plates were coated with goat anti-human AAT antibody at 4°C overnight. The wells were blocked with PBS containing 0.05% Tween 20 and 0.5% BSA for 1 h at room temperature. Control and samples were added to the ELISA plate and incubated for 2 h at 37°C. After the ELISA plates were washed, bound AAT remained in each well. Rabbit anti-human AAT antibody (DAKO, Carpinteria) was added to the plate and incubated for 1 h at 37°C, followed by HRP-conjugated goat anti-mouse antibody (Bio-Rad, Hercules). After washing, QuantaBlu Fluorogenic Peroxidase Substrate (ThermoFisher, Waltham) was added to the wells and incubated for 5 min at room temperature. HRP activity was read at an excitation/emission maxima of 325/420 using a spectrophotometer (LS50B LuminSpectrometer, Perkin Elmer, Waltham). The concentrations of CXCL-8 (Abcam, Cambridge), LTB4 (Cayman Chemical, Ann Arbor), and TNFα (Abcam, Cambridge) were measured in conditioned media of MDMs by ELISA, following the manufacturer’s instruction.

Primary neutrophils were isolated from PiMM volunteers using EasySep direct human neutrophil isolation kit (Stemcell Technologies, Vancouver), following the manufacturer’s instruction. The isolated cells were incubated with fluorescent CD16 and CD66b antibodies (BioLegend, San Diego). Percentage of CD16 and CD66b-positive cells was calculated using a Gallios flow cytometer with Kaluza software (Beckman Coulter, Brea, Supplementary Figure 1 ). Neutrophil chemotaxis was assayed in a Transwell system using polycarbonate membranes with 3-μm pore size (Corning, Corning). MDMs were incubated with or without LPS and, supernatant from the cell culture was transferred into the bottom layer of the chemotaxis chamber. Freshly isolated neutrophils (1 x 10⁶) were added into the top layer of the chamber. The neutrophils migrated from the top to the bottom layer for 30 min. Migrated cells were counted using a hemocytometer and automated cell counter (Invitrogen, Carlsbad).

Lyophilized AAT (Prolastin-C) was reconstituted with deionized water, following the manufacturer’s instruction and stored at -80°C. To examine whether AAT is able to inhibit LPS-mediated CXCL-8 expression, MDMs were incubated with LPS and different concentrations of AAT for 18 h. AAT-treated MDMs were lysed for RNA extraction using the Qiagen RNeasy kit (Qiagen, Hilden), and CXCL-8 expression was examined using qRT-PCR.

To induce apoptosis in the cells, neutrophils were aged for 20 h and then incubated with 1 μM staurosporine for another 3 h. The apoptotic rate of the neutrophils was assessed by flow cytometry with Annexin V/propidium iodide staining (Invitrogen, Carlsbad). For efferocytosis assay, the apoptotic neutrophils were labeled with CellTracker Red CMTPX dye (Invitrogen, Carlsbad), following the manufacturer’s instruction.

Following 7 days of macrophage differentiation, MDMs were incubated with or without LPS for 18 h, and then the MDMs were incubated with CellTracker Red CMTPX dye-labeled apoptotic neutrophils suspended at 1 x 10⁶/ml at 37°C for 30 min, providing a phagocyte to target ratio of 1:4. After the incubation, the non-ingested neutrophils were removed by repeated washing with PBS. Removal of the non-ingested neutrophils was confirmed with light microscope (Olympus 1X70), and phagocytosis of the neutrophils by MDMs was confirmed by fluorescence microscopy (BZ-X700, Supplementary Figure 2 ). MDMs were incubated with Accutase (Stemcell Technologies, Vancouver) at room temperature for 20 min, followed by 15 min on ice. After the incubation with Accutase, MDMs were collected by a gentle scraping with a plastic scraper and analyzed by Gallios flow cytometer with Kaluza software (Beckman Coulter, Brea). A minimum of 10,000 events was acquired per sample. For each MDM sample, the efferocytosis rate of control MDM was set to 100%.

Total RNA (1 μg), extracted from MDMs, was reverse-transcribed using SuperScript^® VILO Master Mix (Invitrogen, Carlsbad), according to the manufacturer’s instruction. Quantification of PCR products was performed with 7500 Fast Real-time PCR (Applied Biosystems, Foster City). SensiFAST Real-Time PCR Kit (Bioline, London) was used to produce fluorescence-labeled PCR products and to monitor increasing fluorescence during repetitive cycling of the amplification reaction. Taqman probes/primers specific for the CXCL-1, CXCL-8/IL-8, TNFα, CD14, CD36, RARα genes, and for the GNB2L1 gene, as a housekeeping gene, were used in the real-time PCR reaction. Expression levels of the genes were obtained using the classical 2^(-ΔΔCt) method.

MDMs were treated with LPS (10 ng/ml) for 30 min and then they were incubated with apoptotic neutrophils or Jurkat cells for 3 h. To induce apoptosis in Jurkat cells, the cells were exposed to UV (200 mJ/cm²) in PBS and the UV-treated cells were incubated in RPMI 1640 for 4 h before being incubated with MDMs. Non-ingested apoptotic cells were washed off with PBS five times and MDMs were lysed for RNA isolation. The expression level of CXCL-8 was examined in the RNA samples using qRT-PCR. The efferocytosis-mediated suppression of cytokine levels was compared between M- and Z-MDMs.

Results are expressed as the mean of number of independent experiments using MDMs from different donors. The assessment was evaluated by two-tailed Student’s t-test, one-way ANOVA test or two-way ANOVA test. Bonferroni test was used for multiple comparisons. P-values of 0.05 or less were considered to be statistically significant.

AAT mRNA expression level was compared between M- and Z-MDMs. As shown in Figure 1A , the gene expression level of AAT was similar between the two groups (p-value=0.17). The concentration of AAT was also compared in conditioned media of M- and Z-MDMs, and the result shows that the AAT concentration is significantly higher in the media of M-MDMs than that of Z-MDMs ( Figure 1B , p-value<0.0001), indicating a higher secretion rate of M-AAT than Z-AAT. MDMs were immunostained for AAT ( Figure 1C ), and the level of intracellular AAT was quantified in the cells. The level of intracellular AAT was significantly higher in Z-MDMs than M-MDMs ( Figure 1D , p-value=0.0006).

The innate immune response usually begins with the activation of alveolar macrophages to produce various cytokines, including neutrophil chemoattractants in the lung, and chemotaxis by neutrophils plays a critical role in the innate immune response. CXCL-8 and CXCL-1 are potent neutrophil chemoattractants on an equimolar basis (27–29). The concentration of AAT is about five-fold higher in conditioned media of M-MDM compared with Z-MDM (24). To examine the effect of AAT on the expression of CXCL-8 and CXCL-1, their expressions were compared between LPS-treated M- and Z-MDMs. LPS remarkably increased the expression of the two neutrophil chemoattractants in both M- and Z-MDMs ( Figures 2A, B ). The expression level of CXCL-8 was increased ~200-fold by LPS in Z-MDMs, and CXCL-1 expression was increased ~40-fold by LPS treatment in Z-MDMs. The expression level of CXCL-8 and CXCL-1 was significantly higher in LPS-treated Z-MDMs than M-MDMs (p-value < 0.05 for the comparison). It was previously reported that the level of CXCL-8 is significant higher in bronchoalveolar lavage fluid of AATD individuals than healthy controls (30). We measured the concentrations of CXCL-8 in conditioned media of LPS-treated MDMs using ELISA. The result shows that CXCL-8 concentration is significantly higher in conditioned media of LPS-treated Z-MDMs than M-MDMs ( Figures 3A , p-value=0.0004), consistent with the previous finding. The higher expression level of CXCL-8 in Z-MDMs could result from lower concentration of extracellular AAT, higher level of accumulated intracellular AAT, or both. Exogenous AAT was added to M- and Z-MDM cultures to give a similar concentration of extracellular AAT between the two MDM cultures to eliminate the effect caused by different extracellular AAT concentrations on CXCL-8 expression between the cells. The result shows that even when the concentration of extracellular AAT is very similar between the two MDM cultures, the expression level of CXCL-8 is still significantly higher in Z-MDMs compared with M-MDMs ( Figure 3B ). It indicates that intracellular Z-AAT led to the higher expression of CXCL-8 in LPS-treated Z-MDMs. As AAT is an acute phase protein, its expression level is increased by LPS treatment, leading to a higher level of intracellular Z-AAT in the cells ( Figure 3C ). The level of intracellular AAT was quantified and compared between controls and LPS-treated MDMs. As shown in Figure 3D , the intracellular AAT level was not statistically different between M-MDM controls and LPS-treated M-MDMs, but it was significantly higher in LPS-treated Z-MDMs than their controls (p-value<0.0003). To examine whether AAT expression level is correlated with the expression of CXCL-8 in Z-MDMs, we calculated a Pearson’s correlation coefficient and p-value between the expression levels of the two genes. The result shows that the expression of AAT is highly correlated with the expression of CXCL-8 in Z-MDMs ( Table 2 ).

We also examined the level of leukotriene B4 (LTB4), a well-known neutrophil chemoattractant, in conditioned media of M- and Z-MDMs using ELISA. The concentration of LTB4 was below the lower limit of detection level in the media of M-MDM samples while it was detected in Z-MDM samples. The concentration of LTB4 was on average 93.6 ± 53.7 pg/ml in Z-MDM samples. The release of LTB4 is stimulated when alveolar macrophages are exposed to neutrophil elastases (10). It could explain the low level of LTB4 in conditioned media of the MDM cultures.

AATD individuals have a higher number of alveolar neutrophils than non-AATD individuals (9, 31, 32). We suspected that the higher expression of the neutrophil chemoattractant factors in LPS-stimulated Z-MDMs could lead to higher neutrophil migration. We freshly isolated neutrophils from human blood and immediately used them for a neutrophil chemotaxis assay because their half-life is short, generally 6–8 h (33). We compared neutrophil migration rates between control and LPS-treated MDMs. Neutrophil migration rate was higher in the conditioned media of LPS-treated MDMs than that of non-treated controls, and the neutrophil migration rate of non-treated controls are shown in the Figure 4A . We calculated the neutrophil migration rate increased by LPS treatment in each MDM group. As shown in Figure 4B , LPS-mediated increase in the neutrophil migration rate was significantly higher in conditioned media of Z-MDMs than M-MDMs (p-value=0.039). This result correlates with our previous observation that the expression level of neutrophil chemoattractant factors was significantly higher in Z-MDMs than M-MDMs. To confirm that the neutrophils were migrated by chemotaxis other than chemokinesis, three different solutions of PBS, RPMI 1640, and macrophage-differentiated media were used as negative controls for the neutrophil migration assay. The result showed that random neutrophil migration rates in the three solutions were less than 3%, supporting that the observed neutrophil migration in conditioned media of MDMs was mediated by neutrophil chemoattractant factors produced by macrophages.

Neutrophils recruited to the site of infection phagocytize bacteria and the process typically accelerates neutrophil apoptosis. Alveolar macrophages clear the apoptotic neutrophils, which ultimately promotes resolution of the bacterial infection (34). Thus the ability of alveolar macrophages to phagocytize apoptotic cells in a timely manner is critical to orchestrate the resolution of inflammation. To determine the ability of M- and Z-MDMs to efferocytose apoptotic neutrophils, we performed efferocytosis of apoptotic neutrophils by MDMs in the presence or absence of LPS. The efferocytosis rate was on average similar between the two MDM groups in the absence of LPS ( Table 3 and Supplementary Figure 2D ). It was previously reported that LPS inhibits efferocytosis of apoptotic neutrophils by MDMs (26). We compared the reduced efferocytosis rate caused by LPS treatment between M- and Z-MDMs. The result showed that the inhibitory effect of LPS was significantly higher in Z-MDMs than M-MDMs ( Figures 5A–D (p-value=0.01)). We compared the expression of macrophage efferocytosis-related genes in LPS-stimulated M- and Z-MDMs. The examined molecules were PPARα, PPARγ, CEBPβ, ADAM17, and TNFα. These genes have previously been reported to be involved in efferocytosis. However, unlike the expression of TNFα, the expression levels of all other genes were not statistically different between M- and Z-MDMs (data not shown). LPS stimulation increased the expression of TNFα in both M-and Z-MDMs, but TNFα expression level was significantly higher in Z-MDMs ( Figure 6A , p-value=0.027). The protein level of TNFα was also significantly higher in conditioned media of LPS-treated Z-MDMs than that of M-MDMs ( Figure 6B , p-value=0.032). To investigate if extracellular AAT is able to inhibit LPS-induced TNFα expression in Z-MDMs, we incubated LPS-stimulated MDMs with two different concentrations, 1 and 10 μM, of M-AAT, and examined TNFα expression in the cells. The result shows that only a higher concentration of AAT, which is similar to the AAT concentration found in circulating blood of non-AATD individuals, could suppress TNFα expression in LPS-stimulated Z-MDMs, as shown in Figure 6C (p-value=0.018). It supports that AAT could inhibit LPS-mediated TNFα expression in macrophages. The ability of AAT to inhibit TNFα expression in human neutrophils was previously reported (35). We suspected that different TNFα expression levels between M-and Z-MDMs could be responsible for the different efferocytosis rates observed between the cells. In order for TNFα to exert its biological function, it has to bind to its specific receptors. To prevent TNFα from binding to its receptors, we incubated Z-MDMs with TNFα neutralizing antibodies. We then examined whether blocking TNFα signaling could abolish the inhibitory effect of LPS on the efferocytosis of apoptotic neutrophils by MDMs. The inhibitory effect of LPS on macrophage efferocytosis was significantly reduced when MDMs were incubated with the neutralizing antibody to TNFα ( Figure 7A , p-value=0.011). LPS reduced efferocytosis rate by 53% in Z-MDMs but the efferocytosis rate was recovered to 94% by TNFα neutralizing antibody in the cells. The result proposes that LPS inhibits macrophage efferocytosis through TNFα in the cells. The efferocytosis rate of the MDMs that were incubated with a combination of LPS and non-specific isotype antibody was similar to that of the MDMs incubated with LPS alone.

TNFα is known to self-regulate its own expression by activating NF-kB signaling (36–38). Regarding that NF-kB signaling regulates the expression of CXCL-8 and CXCL-1 (39, 40), we suspected that TNFα might be responsible for the increased expression of CXCL-8 and CXCL-1 in LPS-stimulated Z-MDMs. We examined whether their expression is reduced when TNFα signaling is inhibited in Z-MDMs. The expression level of CXCL-8 was significantly reduced when MDMs were incubated with TNFα neutralizing antibody ( Figure 7B , p-value=0.019). The expression level of CXCL-1 was also reduced in the cells but the degree of the reduction was not significant (data not shown). We then performed neutrophil migration in conditioned media of the Z-MDMs in which TNFα signaling was inhibited. The result confirmed that neutrophil migration rate is reduced when MDMs were incubated with TNFα neutralizing antibody in Z-MDMs ( Figure 7C , p-value=0.024).

Efferocytosis plays a critical role in the resolution of inflammation by preventing the secondary necrosis of dead cells and triggering several anti-inflammatory signalings. Emerging evidence suggests that the expression of inflammatory cytokines is suppressed in post-efferocytotic macrophages (41, 42). We induced inflammatory signaling in MDMs using LPS and examined whether the efferocytosis of apoptotic neutrophils is able to suppress LPS-induced CXCL-8 expression. As shown in Figure 8A , macrophage efferocytosis of apoptotic neutrophils suppressed the expression level of CXCL-8 by ~20% in M-MDMs but the cytokine suppression was not observed in Z-MDMs (p-value=0.039). We repeated the assay using apoptotic Jurkat cells ( Figure 8B ). The efferocytosis-mediated suppression in the expression level of CXCL-8 was observed in both M- and Z-MDMs, and the suppressed level was significantly higher in M-MDMs (p-value=0.032). The results imply that AAT plays an important role in macrophage efferocytosis-medicated cytokine suppression and probably in resolving inflammation.

Our results showed that LPS-induced TNFα expression inhibits efferocytosis of apoptotic neutrophils by MDMs. It was previously reported that macrophages in a TNFα-rich inflammatory environment are less able to phagocytose apoptotic Jurkat cells (43). We wanted to identify the target molecules that TNFα modulates to inhibit macrophage efferocytosis. A number of genes are involved in the process of efferocytosis. We examined whether TNFα is able to reduce the expression of these genes. MDMs were incubated with TNFα for 1 or 18 h and the expression levels of the genes were examined in the TNFα-treated cells using qRT-PCR. We found that the expressions of CD14, CD36, and RARα are significantly down regulated by TNFα ( Figures 9A p-value=0.005, 9B p-value=0.0008, and 9C p-value=0.0001). These three genes play an important role in macrophage efferocytosis of apoptotic cells; CD14 functions as a bridging molecule that tethers apoptotic cells to macrophages and associates with other molecules within the phagocytic synapse (44). CD36 is essential for macrophage recognition of phosphatidylserine on the surface of apoptotic cells (45). RARα increases apoptotic cell phagocytosis by inducing the expression of phagocytosis-related genes (46). Therefore, TNFα could mediate reduction in efferocytosis of apoptotic cells by inhibiting the expression of CD14, CD36 and RARα in macrophages. We compared the expression level of the genes between LPS-treated M-MDMs and Z-MDMs. LPS treatment reduced the expression of CD36 and RARα in both M- and Z-MDMs, but the reduced level was statistically significant only in Z-MDMs ( Figures 9D p-value=0.0043 and 9E p-value=0.0018). The cell surface distribution of CD36 was examined in MDMs. LPS reduced the cell surface level of CD36 in both M- and Z-MDMs, but the reduced level was statistically significant only in Z-MDMs ( Supplementary Figure 3 , p-value=0.014). The expression of CD14 was also reduced by LPS in both M- and Z-MDMs, and it was, on average, reduced more in Z-MDMs than M-MDMs. However, there was no statistical difference in CD14 expression between LPS treated M-MDMs and Z-MDMs (data not shown). We suspect that the combined effect of the reduced expression of CD14, CD36, and RARα lead to the statistically different efferocytosis rate of apoptotic neutrophils between M- and Z-MDMs. However, we could not rule out that there are other molecules, playing a role in reducing the efferocytosis rate of LPS-treated MDMs.