Subpopulations of human lung alveolar macrophages: ultrastructural features

Identification of a distinct glucocorticosteroid-insensitive pulmonary macrophage phenotype in patients with chronic obstructive pulmonary disease

BACKGROUND

In patients with chronic obstructive pulmonary disease (COPD), pulmonary macrophages increase in number, release increased levels of inflammatory mediators, and respond poorly to glucocorticosteroids. Whether this is due to a change in macrophage phenotype or localized activation is unknown.

OBJECTIVE

We sought to investigate whether macrophages from patients with COPD are a distinct phenotype.

METHODS

Macrophage populations were isolated from human lung tissue from nonsmokers, smokers, and patients with COPD by using Percoll density gradients. Five macrophage populations were isolated on the basis of density (1.011-1.023, 1.023-1.036, 1.036-1.048, 1.048-1.061, and 1.061-1.073 g/mL), and cell-surface expression of CD14, CD16, CD163, CD40, and CD206 was assessed by using flow cytometry. Release of active matrix metalloproteinase 9, TNF-α, CXCL8, and IL-10 was measured by using ELISA.

RESULTS

The 2 least dense fractions were more than 90% apoptotic/necrotic, with the remaining fractions greater than 70% viable. Macrophages from nonsmokers and smokers were CD163(+), CD206(+), CD14(+), and CD40(-), whereas macrophages from patients with COPD were less defined, showing significantly lower expression of all receptors. There were no differences in receptor expression associated with density. Macrophages from patients with COPD of a density of 1.036 to 1.048 g/mL released higher levels of active matrix metalloproteinase 9 compared with cells from nonsmokers, with no difference between the remaining fractions. This population of macrophages from patients with COPD was less responsive to budesonide compared with those from nonsmokers and smokers when stimulated with LPS. Glucocorticosteroid insensitivity was selective for proinflammatory cytokines because budesonide inhibition of LPS-stimulated IL-10 release was similar for all macrophages.

CONCLUSIONS

This study identifies a specific macrophage phenotype in the lungs of patients with COPD who are glucocorticosteroid insensitive with a density of 1.036 to 1.048 g/mL but do not correspond to the current concept of macrophage phenotypes.

Cockade-Like Structures in Alveolar Macrophages in Extrinsic Allergic Alveolitis

BACKGROUND AND OBJECTIVES

In immunocytochemical preparations of bronchoalveolar lavage (BAL) cells from patients with extrinsic allergic alveolitis (EAA), we observed the presence of alveolar macrophages with cockade-like structures in their cytoplasm (cockade+ alveolar macrophages). These cockade+ alveolar macrophages may reflect a subpopulation of alveolar macrophages which may show a different predominance in various interstitial lung diseases. In this study we aimed to compare the frequency of cockade+ alveolar macrophages in patients with EAA (n = 14) with the results obtained in patients with sarcoidosis (n = 11), idiopathic interstitial pneumonia (IIP; n = 10) and control subjects (n = 8). We also investigated the expression of the transferrin receptor CD71 on cockade+ alveolar macrophages.

METHODS

In BAL fluid, the total number of cells and differential counts were determined, and immunocytologic examinations of macrophages and lymphocytes were done using monoclonal antibodies. The percentage of cockade+ alveolar macrophages was determined by counting 300 macrophages in the CD20 field of an immunocytochemical slide.

RESULTS

The percentage of cockade+ alveolar macrophages was significantly higher in the EAA group (36 +/- 9%) compared to patients with sarcoidosis (12 +/- 5%) or IIP (11 +/- 10%) and control subjects (3 +/- 1%; p < 0.001). The proportion of CD71+ alveolar macrophages was significantly lower in EAA than in the other groups (p < 0.01), and the CD71 antigen was expressed on a significantly lower proportion of cockade+ alveolar macrophages compared to cockade- alveolar macrophages in EAA (p < 0.001).

CONCLUSION

We conclude that cockade+ alveolar macrophages could play a role in the pathogenesis and differential diagnosis EAA.

Ultrastructural heterogeneity of the alveolar macrophages from tobacco smokers with chronic bronchitis

Alveolar macrophages recovered by bronchoalveolar lavage from 14 heavy smokers with chronic bronchitis were assessed. Ultrastructural examination revealed marked cellular heterogeneity. Three subpopulations of alveolar macrophages were readily identifiable. These have been termed "young," "mature," and "degrading," reflecting their ultrastructural features. In addition, a majority of the cells were found to be positive by TUNEL staining, indicating DNA damage, but a very small percentage tested positive for Caspase-3, suggesting that apoptosis might not account for the DNA damage in at least some of these cells. A small percentage of proliferating cells were noted.

Heterogeneity of procoagulant activity and cytokine release in subpopulations of alveolar macrophages and monocytes

We have studied the expression of tissue factor (TF) and fibrinopeptide A (FPA) generation as well as the release capacity of TNF-alpha, IL-1beta, and IL-6 in density-defined subpopulations of alveolar macrophages (AM) and monocytes (Mo). TF was equally expressed on all AM subpopulations and Mo, while the FPA-forming capacity was at the same level in low density AM as in Mo and was significantly (P < 0.05) higher in low density AM than in high density AM. The lipopolysaccharide (LPS)-induced release of TNF-alpha was higher (P < 0.05) in high density AM than in low density AM and in Mo. IL-1beta release was undetectable in unstimulated AM and in LPS-stimulated low density AM, while the LPS-induced IL-1beta release in high density AM was low compared to the levels demonstrated in Mo. LPS-stimulated IL-6 release was not distinctively different in the AM subpopulations and Mo. The presented study showed that FPA generation and LPS-stimulated TNF-alpha release were dependent on the density (i.e., maturity) of AM. This implies that a skewed distribution of AM subpopulations induced by disease processes may profoundly influence the inflammatory reactions, including extravascular activation of coagulation.

CD14 expression and binding of lipopolysaccharide to alveolar macrophages and monocytes

We have studied the expression of the lipopolysaccharide (LPS) receptor CD14 on monocytes (Mo) and alveolar macrophages (AM), including density- and size-defined subpopulations. Bronchoalveolar lavage (BAL) was performed on eleven healthy non-smokers and blood sampled from 5 of them, and the levels of cell CD14 expression was investigated using flow cytometry. The influence of LPS stimulation on the CD14 expression of AM was studied at various intervals during prolonged incubation. Further, the relationship between CD14 expression and LPS binding to Mo and subpopulations of AM was studied by measuring fluorescein isothiocyanate (FITC)-LPS binding (flow cytometry) and binding of radioiodinated LPS (125I-LPS). The CD14 expression was 13-fold higher (P < 0.02) on Mo than on unfractionated and high density AM. The CD14 level on the latter was higher than on low density AM, and also higher (P < 0.05) on small AM compared to large (flow cytometrically defined) AM. LPS stimulation had a downregulating effect on AM CD14 level, but after several hours of continuing decreased expression, an increased (P < 0.05) CD14 expression was demonstrated, indicating de novo synthesis. The binding of LPS to subpopulations of AM and isolated Mo was not significantly different, but the binding of FITC-LPS to Mo in whole blood was higher than to AM (P < 0.02). The presented results indicate that AM of different size and maturity have different and variable (activation dependent) CD14 levels. The LPS binding capacity was, however, not proportional to the CD14 expression, indicating that LPS binding mechanisms unrelated to CD14 levels were also operable.

Altered expression and localization of 5-lipoxygenase accompany macrophage differentiation in the lung

The alveolar macrophage (AM) exhibits a greater capacity to synthesize bioactive leukotrienes from arachidonic acid than does its circulating precursor the peripheral blood monocyte. Macrophage differentiation in the lung entails cellular residence within both the pulmonary interstitial and alveolar compartments. In the present study, we sought to determine 1) whether this enhanced metabolic activity was acquired during maturation within the alveolar space and 2) the underlying mechanisms responsible for this upregulation. Rat AMs were separated by Percoll gradient centrifugation into four density-defined subpopulations thought to reflect their degree of maturation. On stimulation with a calcium ionophore, synthesis of leukotriene B4 increased with the degree of maturation, although it was diminished in the oldest subpopulation. This maturation-dependent upregulation was not explained by increases in arachidonic acid release but was associated with increased expression of 5-lipoxygenase (5-LO) protein as determined by immunoblot analysis. Whereas 5-LO is primarily cytosolic in monocytes, it is known to be primarily intranuclear in unfractionated AMs. Here, the localization of 5-LO was investigated by immunofluorescence microscopy and was found to be predominantly nuclear in all AM subpopulations; by contrast, the protein was cytosolic in interstitial macrophages isolated by mechanical and enzymatic lung digestion. These divergent localization patterns in AMs and interstitial macrophages were verified in situ by immunohistochemical staining of sections of normal rat lung. When unfractionated AMs were isolated and maintained in culture for 3 days, a shift in 5-LO distribution from nucleus to cytosol was observed. We conclude that 1) nuclear import of 5-LO occurs within the alveolar space and is reversible on removal from the alveolar milieu and 2) leukotriene synthetic capacity increases further during AM residence within the alveolar space as a result of a progressive increase in the amount of 5-LO protein.

Smoking alters the phenotype of macrophages in induced sputum

To investigate the effect of chronic smoke exposure on pulmonary macrophages (PM), the expression of seven different surface and intracellular molecules of PM was studied in induced sputum (IS) samples from healthy volunteers--nine smokers and seven non-smokers. Sputum was induced by inhalation of nebulized saline (3.5% NaCl). Cell viability and total cell counts (TCC) were performed immediately. Cell differentials were determined on May-Grunwald Giemsa-stained cytospin preparations. The PM were immunologically characterized by use of the following monoclonal antibodies: RFD1, RFD7, CD11b, CD54, CD68, CD71 and HLA-DR. The stainings were performed with a three-step, indirect immuno-alkaline phosphate method. Viability and TCC did not differ between the groups. Smokers had a higher percentage of macrophages (P < 0.05) and a lower proportion of neutrophils (P < 0.05). The percentage of macrophages expressing RFD1, HLA-DR, CD71 (P < 0.01 for all) and CD54 (P < 0.05) was significantly lower in smokers, whereas the remaining markers were expressed equally in the two groups. The results indicate that smoking induces a decrease in the expression by PM of surface molecules known to be associated with the antigen-presenting function.

An immunohistochemical study on the cytogenetic origin of pulmonary multinucleate giant cells in porcine dermatosis vegetans

The origin of pulmonary multinucleate giant cells (MGC) in porcine dermatosis vegetans was studied in six Norwegian Landrace pigs ages 4 (male), 5 (female), 6 (female), 10 (female), and 12 (one male, one female) weeks, using an avidin biotin peroxidase and alkaline phosphatase complex immunohistochemical method on sections of formalin- and ethanol-fixed and frozen tissue specimens. Well-characterized, commercially available antisera/monoclonal antibodies to keratin, vimentin, lysozyme, a monocytic antigen, and a myelomonocytic antigen were used. The immunoreactivity to intermediate-sized filaments in MGC was negative for keratins and positive for vimentin. In addition, a positive reaction was found in alveolar macrophages, chondrocytes, fibrocytes, alveolar lymphocytes, and granulocytes in ethanol-fixed tissue. Marked masking was observed in formalin-fixed tissue. Antilysozyme antiserum gave a positive cytoplasmic reaction in alveolar macrophages and MGC, although the reaction was variable in ethanol-fixed tissue. In trypsinized formalin-fixed tissue, a moderate and more consistent cytoplasmic reaction was observed in alveolar macrophages and MGC. Two monoclonal antibodies that identify human cells of the MMS, EMB 11 and Mac 387, were negative for EMB 11 in ethanol-fixed and frozen sections, whereas Mac 387 showed a positive and specific cytoplasmic immunoreaction in alveolar macrophages and small MGC in ethanol- and formalin-fixed tissue. Pulmonary MGC in dermatosis vegetans are derived from mesenchymal cells, and substantial evidence is provided in support of a monocyte/macrophage origin based on a positive reaction for lysozyme and a myelomonocytic antigen. The importance of adequate fixatives for immunohistochemical demonstration of cell-specific markers is clearly shown.

Effect of Bacillus Calmette-Guérin inoculation on numbers of dendritic cells in bronchoalveolar lavages of rats

Dendritic cells (DC) are present in all lymphoid tissues and are widely distributed in the airway epithelium and lung parenchyma. In this study DC were morphologically and cytochemically identified in normal rat bronchoalveolar lavage (BAL), although in very low percentages. Furthermore, the total population as well as different Percoll density fractions demonstrated poor antigen-presenting capacity and even suppressed antigen-specific stimulation by rat splenic DC. In contrast, when an inflammatory response was induced by intratracheal inoculation of Bacillus Calmette-Guérin (BCG), an increase of Ia-positive cells, containing high percentages of monocytes and DC (MDC) was found. In BAL, DC increased about 25 times within 48 h after BCG inoculation. These BCG-induced BAL cells as well as the different density fractions showed a high antigen-presenting capacity at low concentrations. However, at higher concentrations they were suppressive, except for the highest density fraction which lacked alveolar macrophages (AM). These results indicate that the increased numbers of Ia-positive MDC during an inflammatory reaction are very likely responsible for antigen presentation in vitro. In contrast, AM suppress the antigen-specific T cell proliferation in a concentration dependent manner.

Characterization of immune inducer and suppressor macrophages from the normal human lung

Monoclonal antibodies (MoAbs) that are able to discriminate between dendritic cells (MoAb RFD1+) and mature macrophages (MoAb RFD7+) in normal tissues were used in combination with density separation techniques to isolate relatively homogeneous subpopulations of macrophages from human bronchoalveolar lavage (BAL). A characterization of surface antigen expression, and functional capacity was then carried out on each isolated alveolar macrophage (AM) subset. One population with the phenotype RFD1+RFD7- obtained from the non-adherent cell pool showed the characteristics of antigen-presenting cells having absent or poor expression of Fc and C3b receptors, a low content of lysozomal hydrolase and poor phagocytic capacity. This population strongly stimulated T lymphocytes in allogeneic mixed lymphocyte reactions (MLR). A second AM population, isolated by adherence and density centrifugation expressed the phenotype RFD1+RFD7+. These cells showed the same phenotypic characteristics of mature macrophages with strong expression of C3b and Fc receptors, and marked phagocytic capacity. Such AM were very poor stimulators of allogeneic MLR. Under certain circumstances the RFD1+RFD7+ cells were shown to actively repress the stimulatory capacity of the RFD1+RFD7- subpopulation. These results suggest that variations within the functional capacity of AM subsets may be capable of influencing the strength of acquired T cell immune responses of the lung.

Characterization of cytoadhesion molecules on human monocytes and tissue macrophages

The presence of proteins antigenically related to the GPIIb/IIIa complex expressed on platelets have been investigated on tissue macrophages recovered by bronchoalveolar lavage (lung alveolar macrophages, LAM) or peritoneal lavage (peritoneal macrophages, PM) as well as on monocytes. Polyclonal antibodies (pab) directed against human platelet GPIIb/IIIa and the vitronectin receptor (VnR), and mouse monoclonal antibodies (mab) against human GPIIb, GPIIIa or the GPIIb/IIIa-complex were used. Triton X-100 extracts of bronchoalveolar cells (BAC) (containing 94% LAM) and the ultrasedimentable fraction of cell-free bronchoalveolar lavage (US) reacted with the polyclonal antibodies against the GPIIb/IIIa-complex and the VnR, but only with one (P4) of the mabs. Cell microscopy after immunogold labelling and alkaline phosphatase immunostaining, as well as immunofluorescence using the P4 mab and the polyclonal anti-GPIIb/IIIa clearly demonstrated positive membrane staining of LAM, PM and monocytes. Both BAC and US extracts gave rise to immunoprecipitates in crossed and rocket immunoelectrophoresis using anti-GPIIb/IIIa and anti-VnR. Our data indicate that monocytes and their monocyte-derived tissue counterparts constitutively express GPIIb/IIIa-like antigen(s) on their membrane. The presence of such antigen(s) on tissue macrophages makes it unlikely that platelet contamination is responsible for these findings.