Defective efferocytosis by alveolar macrophages in IPF patients

Summary

Rationale

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive fibrosing interstitial pneumonia. The pathogenicity of IPF has been widely investigated but still remains to be clarified. Efferocytosis, the specialized recognition and ingestion of apoptotic cells by phagocytes, is essential for the resolution of inflammation in the lungs and repair of injured tissues. Impaired efferocytosis contributes to the pathogenesis of chronic lung diseases such as emphysema and cystic fibrosis. We hypothesized that efferocytosis would also be reduced in alveolar macrophages isolated from subjects with IPF.

Methods

Efferocytosis, was evaluated using Wright-Giemsa stained cell preparations isolated from the bronchoalveolar lavage (BAL) fluid of patients with IPF (n = 5), nonspecific interstitial pneumonitis (n = 6), cryptogenic organizing pneumonia (n = 4) and eosinophilic pneumonia (EP) (n = 5).

Results

Uningested apoptotic cells were significantly higher in BAL fluid from patients with IPF compared to other forms of interstitial lung disease. Macrophages isolated from patients with eosinophilic pneumonia had significantly fewer phagocytic ingestions than macrophages from the other three groups.

Conclusion

Efferocytosis by alveolar macrophages was significantly lower in subjects with IPF compared to subjects with other interstitial pneumonia. Dysregulated efferocytosis may contribute to the pathogenesis of IPF.

Introduction

Efferocytosis, the specialized recognition and ingestion of apoptotic cells by phagocytes, has been shown to down-regulate inflammation by decreasing inflammatory mediator production via production of TGF-β1, PGE2 and IL-10.¹ At the same time, efferocytosis drives tissue repair by upregulating cellular production of growth factors such as hepatocyte growth factor (HGF).² Impaired efferocytosis has been documented in a number of chronic respiratory diseases (e.g. COPD, cystic fibrosis and asthma) and contributes directly to their pathogenesis by enhancing inflammation and impairing epithelial repair.^3,4 Abnormal accumulation of inflammatory cells and dysregulated epithelial repair have been suggested to contribute to the development of fibrotic lung diseases. Accordingly, we sought to determine if efferocytosis was impaired in patients with idiopathic pulmonary fibrosis (IPF) and less-fibrotic idiopathic interstitial pneumonias (IIPs) such as nonspecific interstitial pneumonia (NSIP) and cryptogenic organizing pneumonia (COP).⁵

Methods

Efferocytosis, was evaluated using Dif-Quik™ stained cell preparations isolated from the bronchoalveolar lavage (BAL) fluid of patients with IPF (n = 5), NSIP (n = 6), COP (n = 4) and eosinophilic pneumonia (EP) (n = 5). We analyzed BAL from EP because it is rapidly reversible non-fibrotic lung disease. All BAL were performed at the time of diagnosis of IIP or EP. IPF was diagnosed according to ATS/ERS criteria(5) or UIPpattern by surgical biopsy. NSIP and COP were diagnosed by surgical biopsy ortrans-bronchial lung biopsy (TBLB). EP was diagnosed by clinical findings includingBAL and TBLB. We excluded cases in which collagen vascular diseases, infection orpneumoconiosis were present. Patient characteristics are available as online Supplementary data.Macrophages containing apoptotic bodies and apoptotic cells with condensednuclei³ were identified asshown in Fig. 1a. Efferocytosis was quantifiedusing the following formulas:

Figure 1.

a) Representative findings of Dif-Quik stained cell preparations isolated from the bronchoalveolar lavage (BAL) fluid. Top two panels show efferocytosis. Bottom panels demonstrate apoptotic cells. b – d) Phagocytic indexes (PI) (b), percentages of apoptotic cells (c) and defective index (DI) (d) in BAL cells from patients with IIPs and EP. Statistical analysis was performed by two-way ANOVA, followed by Student Newman Keuls test for multiple comparisons.

$$\text{Phagocytic index}(\text{PI})=\text{number of phagocytosed cells}/\text{number of AMs}\times 100$$

$$\text{Apoptotic cells}(\%)=\text{number of apoptotic cells}/(\text{total cells}-\text{number of AMs})\times 100$$

$$\text{Defective index}(\text{DI})=\text{apoptotic cells}(\%)/\text{PI}$$

Results

Phagocytic indexes were significantly lower in BAL obtained from patients with IIPs compared to EP (Fig. 1b). In addition, higher percentages of apoptotic cells were present in BAL from IPF patients (Fig. 1c) with corresponding elevation of the defective index (Fig. 1d), suggesting that clearance of apoptotic cells by AMs may be defective in IPF compared to other forms of IIP that have better clinical outcomes.

Discussion

Efferocytosis of dying cells is critical for the resolution of inflammation and repair of injured tissues. Efficient removal of apoptotic cells prevents the release of potentially toxic intracellular contents from the dying cells and enhances the production of growth factors and cytokines by the phagocyte, including TGF-β and HGF. The anti inflammatory effects of efferocytosis (i.e. through TGF-β production) may be considered of paramount importance in resolving inflammation in lung diseases where leukocyte infiltration plays a role, such as EP, COP and NSIP. On the other hand, sustained production of TGF-β may lead to fibrosis. In comparison, HGF enhances epithelial cell repair and may counteract the pro-fibrotic effects of TGF-β. Thus in non-inflammatory injury processes (such as IPF) the production of HGF may be of greater importance in driving epithelial repair. In other words, the normal resolution of inflammation and the initiation of lung repair almost certainly require a delicate balance between the production of anti-inflammatory factors (TGF-β) and molecules that stimulate repair (HGF). In this context, both the nature of the underlying disease and the timing of apoptotic cell clearance may be critical, such that impaired efferocytosis during early lung injury may delay the resolution of inflammation whereas during later phases, failure to clear apoptotic cells may result in disordered repair.

Alveolar macrophages are the most abundant leukocytes in the airspaces and are the major cell responsible for efferocytosis. Their programming (or activation) is highly dependent on the environment in which they reside. Accordingly, macrophages that exist in a Th1 environment are generally pro-inflammatory and “classically activated”. Conversely, a Th2 environment promotes the development of “alternatively activated” macrophages that produce molecules implicated in mesenchymal repair and fibrosis. It is therefore not surprising that AMs taken from patients with IPF exhibit features of alternative activation.⁶ As such, it has been suggested that they may contribute to fibrosis by producing TGF-β and CCL18.⁶ Indeed, intentional depletion of alveolar macrophages in animal models of IPF abrogates the development of fibrosis.⁷ Similar results have been achieved by inhibiting alternative macrophage activation.⁸ Interestingly, debris from dead or dying cells that have not been effectively cleared may promote alternative activation of macrophages.⁹ This raises the intriguing possibility that the presence of apoptotic cells drives alternative activation of macrophages, which in turn enhances fibroproliferation.

Nuclear condensation, cytoplasmic membrane blebbing, and reduced cell volume are morphologic hallmarks of apoptosis. Unfortunately, these changes make it difficult to distinguish the source (or origin) of apoptotic bodies. Therefore we cannot be sure whether the apoptotic cells identified in our study are effete leukocytes or sloughed epithelial cells. Since neutrophils and macrophages isolated from BAL of patients with IPF are resistant to apoptosis^10,11 we favor the latter, especially since lung tissue from patients with IPF contains high numbers of apoptotic epithelial cells.¹²

Increased apoptotic cell numbers may result from accelerated cell death or defective apoptotic cell removal. Since under most circumstances, clearance of apoptotic cells from injured tissues is incredibly efficient, we suggest that our results reflect the latter. For example, during bacterial pneumonia there are over 10⁶ neutrophils per mm³ of lung tissue, yet clearance of dying neutrophils is so efficient that less than 0.3% are apoptotic.¹³ On the other hand, impaired clearance of apoptotic cells has been suggested as a mechanism to explain the high levels of apoptotic cells noted in the parenchyma and airspaces in patients with COPD(4). While our study suggests that efferocytosis may be impaired in IPF, prospective research using primary cells taken from patients with IIPs is required to clarify this finding and determine the mechanisms that underlie this defect.