Novel Method to Assess Resident Alveolar Macrophage Efferocytosis of Apoptotic Neutrophils by Flow Cytometry

Abstract

Macrophage efferocytosis of apoptotic neutrophils (PMNs) plays a key role in the resolution of inflammation. In these studies, we describe a novel flow cytometric method to assess efferocytosis of apoptotic PMNs. Resident alveolar macrophages and PMNs were collected from lungs of mice exposed to inhaled ozone (0.8 ppm, 3 hr) followed by lipopolysaccharide (3 mg/kg, i.v.) to induce acute lung injury. PMNs were labeled with PKH26 or DilC₁₈(5)-DS (D12730) cell membrane dye and then incubated with resident alveolar macrophages at a ratio of 5:1. After 90 min, macrophage efferocytosis was analyzed by flow cytometry and confirmed by confocal microscopy. Whereas alveolar macrophages incubated with D12730-labeled PMNs could readily be identified as efferocytotic or non-efferocytotic, this was not possible with PKH26 labeled PMNs due to confounding macrophage autofluorescence. Using D12730 labeled PMNs, subsets of resident alveolar macrophages were identified with varying capacities to perform efferocytosis, which may be linked to the activation state of these cells. Future applications of this method will be useful in assessing the role of efferocytosis in the resolution of inflammation in response to toxicant exposure.

Introduction

Efferocytosis is the process by which phagocytic cells clear apoptotic cells and bodies from tissues. Initiated by inflammatory mediators released from apoptotic cells, efferocytosis plays a key role in the resolution of inflammation (Doran et al., 2020). In the absence of efferocytosis, apoptotic cells and bodies can induce secondary necrosis and prevent activation of pro-resolution signaling (Doran et al., 2020). In this context, evidence suggests that impaired macrophage efferocytosis of apoptotic neutrophils (PMN)s results in persistent pulmonary inflammation in acute respiratory distress syndrome (ARDS), a severe form of acute lung injury (ALI) (Grégoire et al., 2018).

We recently established a model of ALI induced by exposure of mice to ozone followed by intravenous administration of lipopolysaccharide (LPS) (Radbel, 2020). ALI induced by ozone and LPS is associated with persistent neutrophilia and activation of resident alveolar macrophages (AM)s. Thus, this represented a useful experimental model of a form of ALI to assess resident AM efferocytosis of lung PMNs. For this analysis, we developed a novel methodology, which involved magnetic separation of leukocytes followed by cell membrane staining with DilC₁₈(5)-DS (D12730) (Johnson, 2010) and flow cytometric analysis. Using this method, we demonstrated that resident AMs from mice with ALI readily phagocytize infiltrating lung PMNs. Moreover, use of D12730 eliminated confounding macrophage autofluorescence associated with the use other cell membrane dyes (Erriah et al., 2019), markedly improving the ability to assess efferocytosis.

Materials and Methods

Animals and exposures

Specific pathogen-free C57BL6/J male mice (12-13 wk) were obtained from The Jackson Laboratories (Bar Harbor, Maine). Mice were housed in filter-top microisolation cages and were maintained on food and water ad libitum. All animals received humane care in compliance with Rutgers guidelines, as outlined in the Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health. Animal protocols were approved by the Rutgers University Institutional Animal Care and Use Committee. Ozone was generated from oxygen gas via an ultraviolet light generator and mixed with air as previously described (Francis et al., 2020). Ozone concentrations inside the chamber were monitored continuously during the exposure period using a Photometric ozone analyzer (2B Technologies, Boulder, CO). Ozone concentrations were maintained by adjusting both the intensity of the UV light and the flow rate of ozone into the chamber. Mice were exposed to ozone (0.8 ppm) or filtered air for 3 hr. After 24 hr, mice were treated i.v with 3 mg/kg LPS (E. coli 0128:B12, Sigma-Aldrich, St. Louis, MO) to induce endotoxemia or phosphate buffered saline (PBS) as a vehicle control. While ozone and LPS alone can induce a PMN response in vivo (Hollingsworth et al., 2007; Matute-Bello et al., 2008), we used the combined exposure to maximize this response for the development of our efferocytosis assay.

Mouse AM and PMN isolation

Twenty-four hours after exposure to ozone + LPS or filtered air + PBS, mice were euthanized by i.p. injection of xylazine (30 mg/kg) and ketamine (135 mg/kg). The lung was perfused in situ via the right ventricle with 5 ml of ice-cold sterile PBS. BAL was collected by slowly instilling and withdrawing 1 ml of ice-cold sterile PBS, pH 7.4 into the lung six times through a 20-gauge cannula inserted into the trachea, while gently massaging the tissue as previously described (Sunil et al., 2019). Lavage fluid was centrifuged (300 x g, 8 min, 4°C) and cell pellets resuspended in buffer (PBS containing 2% v/v fetal bovine serum (FBS) and 1 μM EDTA. Red blood cells present in BAL from all mice were lysed with 1 ml Hybri-Max^™ Lysing Buffer (Sigma-Aldrich) at room temperature for 5 min. Ly6G⁺ PMNs were collected from ozone + LPS exposed mice by magnetic separation using a MidiMACS^® magnet (Miltenyi Biotec, Auburn, CA). CD11b⁻ AMs were magnetically separated from Ly6G cells by negative selection using an EasySep^™ Mouse CD11b Positive Selection Kit (Stemcell technologies, Vanouver, Canada). Viable cells were then enumerated on a hemocytometer using trypan blue dye exclusion and purity assessed visually in Hema 3^™ Stat Pack (ThermoFischer, Waltham MA) stained cytospin preparations (300 cells). Ly6G⁺ cells were 92.7 ± 4.0 % PMNs; CD11b⁻ cells were 95.3 ± 2.0 % macrophages (mean ± SE, n=3 experiments).

Measurement of apoptosis

PMNs were stained with Annexin V and propidium iodine (PI) using a BD Pharmigen^™ FITC Annexin V Apoptosis Detection Kit (BD, Franklin Lakes, NJ) and analyzed on a Gallios flow cytometer (Beckman Coulter, Indianapolis, IN). The following laser/filters/voltages were used: forward scatter/0 volts; side scatter/0 volts; 488/525 nm ± 20/290 volts (Annexin V); 488/660 nm±20/330 volts (PI). PMNs were analyzed immediately after magnetic separation. Early apoptotic cells were identified as Annexin V⁺PI⁻ and late apoptotic cells as Annexin V⁺PI⁺ relative to unstained controls as previously described (Erriah et al., 2019). The percentage apoptotic cells was calculated as Annexin V⁺PI⁻ events (early apoptosis) or Annexin V⁺PI⁺ events (late apoptosis) divided by total events.

Flow cytometric analysis of macrophage efferocytosis

An outline of the efferocytosis assay is shown in Figure 1. After magnetic separation of BAL cells, PMNs were labeled with 2 μM PKH26 (Sigma-Aldrich) (Erriah et al., 2019) or 5 μM D12730 (Molecular Probes, Eugene, OR) according to the manufacturer’s instructions. The PKH26 dye concentration was selected based on the manufacturer’s recommendations for optimal staining and minimal cytotoxicity. To evaluate labeling efficiency, fluorescence of labeled and unlabeled PMNs was compared at excitation/emission 488/576 nm or 640/671 nm using a Beckman Coulter Astrios EQ cell sorter with the following lasers/filters/voltages (V): 488 nm/side scatter/255 V; 488 nm/forward scatter/240 V; 488/576 ± 21 nm/350 V (PKH26); 640/671 ± 30 nm/250V (D12730), or a Beckman Coulter Gallios flow cytometer with the following lasers/filters/voltages: forward scatter/0 V; side scatter/0V; 488/575 ± 20 nm/338 V(PKH26); 633/660 ± 20 nm/320 V (D12730). Using the same settings, unlabeled PMNs were also compared to unlabeled AMs to assess autofluorescence of the different cell types.

Figure 1. Efferocytosis Protocol.

Lung cells collected by bronchoalveolar lavage from mice treated with air + LPS or ozone + LPS were pooled and magnetically separated into Ly6G⁺ PMNs and Ly6G⁻CD11b⁻ alveolar macrophages (AM)s. PMNs were labeled with PKH26 or D12730 and then incubated with AM at a ratio of 5:1 in 24 well dishes at 37°C. After 90 min, cells were analyzed by flow cytometry (FC) to assess efferocytosis. Created with BioRender.com

To analyze efferocytosis, resident AMs (CD11b⁻) were co-incubated for 90 min with labeled PMNs at ratios of 5:1, 3:1 and 1:1 PMNs to AMs in sterile non-coated Nunc^™ 24 well petri dishes in DMEM/Nutrient Mixture F-12 (without phenol red) supplemented with 10% v/v FBS and 1% v/v penicillin/streptomycin at 37°C as previously described (Erriah et al., 2019). Cells were collected by washing the plates 5 times with 1 ml ice-cold saline. Light microscopic evaluation confirmed that less than 1% of the cells remained adhered to the dishes after this procedure (data not shown). Cells were washed in PBS, centrifuged (300 x g, 8 min, 4°C), and then washed once with PBS to remove non-phagocytized material; they were then fixed in 3% paraformaldehyde overnight, and analyzed on a Beckman Coulter Astrios EQ cell sorter using the following lasers/filters/voltages (V): 488 nm/side scatter/255 V; 488 nm/forward scatter/240 V; 488/576 ± 21 nm/350 V (PKH26); 640/671 ± 30 nm/250V (D12730). PMNs and AMs were sorted into D12730⁻ and D12730+ populations based on fluorescence at 640/671 nm and assessed microscopically on a Leica TCS SP8 confocal microscope. In pilot studies we compared 5:1, 3:1 and 1:1 ratios of PMNs to AMs. We found that the greatest efferocytotic response was observed at a 5:1 ratio and this ratio was used in all subsequent studies.

To further characterize efferocytosis in AM subpopulations, CD11b⁻ AMs and PMNs were co-incubated for 90 min at 37°C and at 4°C (efferocytosis negative controls). Cells were then incubated with anti-mouse CD16/32 (1:100; clone 93; Biolegend, San Diego, CA) for 10 min in 100 μl of staining buffer (PBS, 2% FBS and 0.02% sodium azide) at 4°C, to block nonspecific binding, followed by 30 min with FITC-CD11b (1:100; cloneM1/70; Biolegend), PE-Siglec F (1:10; clone ES22-10D8; Miltenyi), PE-Viol770-CD45 (1:50; clone 30F11; Miltenyi), Brilliant Violet 421-CD11c (1:100; clone N418; Biolegend) and ViolGreen-Gr-1 (Ly6G; 1:50; clone REA810; Miltenyi) antibodies. Subsequently cells were washed with staining buffer, fixed in 3% paraformaldehyde overnight, and analyzed on a Beckman Coulter Gallios flow cytometer using the following lasers/filters/voltages: forward scatter/0 V; side scatter/0V; 488/525 ± 20 nm/320 V (CD11b); 488/575 ± 20 nm/338 V (Siglec F); 488/755 nm/440 V (CD45); 633/660 ± 20 nm/320 V (D12730); 405/450 ± 50 nm/320 V(CD11c); 405/550 ± 40 nm/310 V (Ly6G). Efferocytotic macrophages were identified based on D12730 fluorescence.

Statistical Analyses

Data were analyzed using a Students’ t test and presented as mean ± SE. A p value ≤ 0.05 was considered statistically significant.

Results

Assessment of neutrophil apoptosis

Initially we assessed apoptosis of PMNs by flow cytometry (Erriah et al., 2019; Vermes et al., 1995). Figure 2 (upper panel) shows a representative scattergram of early (AnnexinV⁺PI⁻) and late (Annexin V⁺PI⁺) PMN apoptosis. A summary of early, late and total (early+late) apoptosis is presented in Fig. 2 (lower panel). The majority of PMNs were found to be in early apoptosis.

Figure 2. Flow cytometric analysis of PMN apoptosis.

PMNs were prepared as described in the Materials and Methods. Cells were unstained or stained with propidium iodide (PI) and/or Annexin V and analyzed by flow cytometry. Upper Panel. Representative scattergrams of cells in early (AnnexinV⁺PI⁻) and late (AnnexinV⁺PI⁺) apoptosis based on fluorescence of unstained, Annexin V-stained, and PI-stained control cells. Lower Panel. The percentage of total, early, and late apoptosis of PMNs. Data are mean ± SE (n = 3 separate experiments).

D12730 labeling of PMNs facilitates flow cytometric detection of efferocytosis

We next compared labeling of PMNs with PKH26 and D12730. PKH26 was found to label 81.7 ± 4.7% of PMNs; by comparison, a significantly greater percentage of PMNs (99.7 ± 0.1%) labeled with D12730 (Fig. 3). Importantly, PMNs labeled with D12730 were much more readily distinguished from unlabeled PMNs relative to PMNs labeled with PKH26. We also found CD11b⁻ AMs exhibited significant autofluorescence at 488/576 nm, the excitation/emission spectra for PHK26, when compared to unlabeled PMNs (Fig. 4). In contrast, autofluorescence of AMs at 640/671 nm, the excitation/emission spectra for D12730 was minimal (Fig. 4). No differences in autofluorescence were noted between CD11b⁻ AMs from control mice treated with air + PBS and mice treated with ozone + LPS.

Figure 3. Comparison of PMN labeling with PKH26 and D12730.

Unlabeled PMNs or PMNs labeled with PKH26 or D12730 were analyzed by flow cytometry at excitation/emission 488/576 nm or 640/671 nm as described in the Materials and Methods. The percentage positive PMNs (mean ± SE, n= 3-4) are shown. *Significantly different (p<0.05) from PKH26 labeled PMNs.

Figure 4. Autofluorescence of alveolar macrophages (AM)s and PMNs.

Unlabeled PMNs from ozone + LPS exposed mice and unlabeled AMs from mice exposed to air + PBS or ozone + LPS were incubated alone for 90 min at 37°C and then analyzed for autofluorescence by flow cytometry at excitation/emission 488/576 nm or 640/671 nm. The percentage positive cells are shown. Data are mean ± SE (n = 5-6). *Significantly different (p <0.05) from unlabeled PMNs. ND, not detectable.

In further studies we compared the ability to detect CD11b⁻ AM efferocytosis of PKH26-labeled PMNs and D12730-labeled PMNs. Flow cytometric analysis revealed two subpopulations of cells that varied in D12730 fluorescence; this was not evident in populations of AMs incubated with PKH26-labeled PMNs (Fig. 5). To identify the cells, the two subpopulations were sorted, stained with DAPI, and visualized microscopically. The D12730⁺ population was found to consist of AMs that had efferocytized PMNs and nonefferocytized PMNs (Figure 6). The D12730⁻ population consisted of non-fluorescent AMs.

Figure 5. Macrophage autofluorescence interferes with detection of efferocytosis of PKH26 labeled PMNs.

PKH26 or D12730 labeled PMNs and unlabeled AMs were co-incubated for 90 min at 37°C and then analyzed by flow cytometry at excitation/emission 488/576 nm or 640/671 nm. Negative and positive subsets of D12730 fluorescent cells were based on labeled and unlabeled PMN controls shown in Figure 3. Histograms are representative of at least 3 experiments.

Figure 6. Microscopic analysis of efferocytosis.

D12730 labeled PMNs and unlabeled AMs were incubated for 90 min at 37°C and then separated using a Beckman Coulter Astrios EQ cell sorter into D12730 positive (+) and D12730 negative (−) subsets, stained with DAPI, and visualized by confocal (left and middle panels) and brightfield (right panels) microscopy. Representative image of labeled PMNs with condensed nuclei and AM with intracytoplasmic fluorescence consistent with efferocytosis of labeled PMNs. Red, D12730; blue, DAPI.

To further characterize efferocytosis in resident AM subpopulations, following co-incubation of AMs with D12730 PMNs at 4°C or 37°C, the cells were stained with antibodies to CD45, CD11c, CD11b, Ly6G, and SiglecF and analyzed by flow cytometry. In these experiments, only singlets, identified by time of flight were included in the analysis (Fig 7). PMNs were identified as CD45⁺CD11c⁻CD11b⁺ Ly6G⁺ and resident AMs as CD45⁺CD11c⁺ SiglecF⁺ Ly6G⁻ cells. In cells incubated at 37°C, but not 4°C, two subpopulations of D12730⁺ resident AMs were identified based on high and low fluorescence. These subpopulations were sorted and examined microscopically. Greater intracellular D12730 fluorescence was noted in the D12730^hi population relative to the D12730^lo population, consistent with increased efferocytosis (Fig 8). We also identified a population of singlet cells that were Ly6G⁺SiglecF⁺. Since the identity of these cells is unknown, they were excluded from our analysis of the efferocytosis positive AM fraction.

Figure 7. Gating strategy for analysis of alveolar macrophage (AM) efferocytosis of PMNs.

CD11b⁻ resident AMs were incubated in the absence or presence of D12730-labeled PMNs for 90 min at 37°C or 4°C (negative control) and then analyzed by flow cytometry. Upper Panel. BAL cells were separated into singlets based on forward scatter time of flight. Cells were then separated into resident AMs (CD45⁺ CD11c⁺ Siglec F⁺Ly6G⁻ cells) and PMNs (CD45⁺ CD11c⁻ CD11b⁺Ly6G⁺ cells). Lower Panel. Resident AMs from the 37°C or 4°C co-culture temperatures were analyzed for D12730 fluorescence indicative of efferocytosis of labeled PMNs

Figure 8. Microscopic analysis D12730^lo and D12730^hi alveolar macrophages (AM)s.

D12730 labeled PMNs and unlabeled AMs were co-incubated for 90 min and then analyzed using a Beckman Coulter Astrios EQ cell sorter. AMs were identified by their autofluorescence at 488/576 nm and then separated into D12730 low (lo) and D12730 high (hi) subsets. Cells were then stained with DAPI and visualized by confocal (left and middle) and brightfield (right panels) microscopy. Red, D12730; blue, DAPI.

Discussion

The present studies describe a new method to effectively assess resident AM efferocytosis of apoptotic PMNs. The advantage of this method is that it is rapid, automated, and accurate with high throughput. The efficacy of our method was verified by direct visualization of cytoplasmic fluorescent bodies within macrophages using mouse lung PMNs from our experimental model of ozone + LPS induced ALI. Based on our studies we predict that this model will be useful for assessment of macrophage efferocytosis in diverse biological models.

Previous studies have shown that monocyte-derived macrophage efferocytosis of circulating PMNs can be measured by flow cytometry (Erriah et al., 2019). The present studies demonstrate that in order to effectively assess resident AM efferocytosis, use of D12730 membrane dye is needed to overcome the confounding effects of endogenous autofluorescence due to flavins and cellular lipids (Njoroge et al., 2001). D12730 is excited outside of the autofluorescence range of resident AMs and, as such, permits a more accurate estimate of the extent of efferocytosis by these cells (Viksman et al., 1994).

Using D12730-labeled PMNs and flow cytometry, we identified three distinct resident AM subsets exhibiting non-detectable, low or high levels of efferocytosis. To our knowledge, this has not been reported previously using a flow cytometric approach. Our findings that PMNs that have infiltrated into the lungs of mice after acute injury are phagocytized by resident AMs demonstrate that efferocytosis can be assayed in relevant biological systems. It remains to be determined if subpopulation variability within resident AMs with respect to their ability to phagocytize apoptotic PMNs is due to different activation states of these cells or early AM apoptosis. In this context, it is now well established that efferocytosis is linked to M2 macrophage proresolution activity (Zhong et al., 2018). It may be that low efferocytosis reflects the response of M2 macrophages that are incompletely activated and/or a subset of proinflammatory M1 macrophages.

It should be noted that there are some limitations of our assay that will require further refinement. In preliminary experiments, we found a distinct effect of neutrophil to macrophage ratio on the percentage AM efferocytosis. Thus, it is important to consider the biological system being modeled when choosing the co-incubation ratio as our ozone + LPS model does not represent all forms of ALI. Another potential limitation of our studies is the use of primary cells isolated from lungs of injured mice, which may be contaminated with other cell types. However, our magnetic gating strategy prior to cell membrane staining resulted in >90% purity in the populations of separated PMNs and resident AMs, suggesting robust and specific efferocytosis measurements. While there may be some residual macrophages adhering to cell culture plates and adherent AMs might be disproportionally efferocytotic or non-efferocytotic cells, the majority (~99%) were removed by using non-coated plates and washing with ice-cold saline. It should also be noted that the effect of D12730 cell membrane staining itself on macrophage efferocytosis in this model is unknown.

In summary, the present studies describe a novel method to measure resident AM efferocytosis of apoptotic PMNs using D12730 cell membrane dye and flow cytometry. This method provides information on subsets of resident AMs that perform efferocytosis at varying levels, which may be valuable for further investigation of the activation state and/or phenotype of these cells. Future applications of this method will be useful in determining the relative effects of toxicant exposure(s), disease pathogenesis and/or therapeutics to promote macrophage efferocytosis and the resolution of inflammation.

Highlights.

• Alveolar macrophage efferocytosis can be measured with a new flow cytometry method

• DilC₁₈(5)-DS labeling reduces confounding macrophage autofluorescence

• Varying degrees of macrophage efferocytosis can be identified using this new method