Macrophage Polarization in Sarcoidosis: An Unexpected Accomplice?

Sarcoidosis is a complex immune disease characterized by accumulation of antigen-activated CD4⁺ T cells and macrophages, progressing to granuloma formation (1). Granulomas result from a dynamic interplay between macrophages and T lymphocytes. Granulomatous inflammation in pulmonary sarcoidosis has historically been linked to compartmentalized elevation of T-helper cell type 1 (Th1) cytokines, including but not limited to IL-2, IFN-γ, and TNF-α (1–3). This cytokine milieu is believed to drive classical (M1) macrophage activation. Emerging data, including the findings of Locke and colleagues (pp. 84–95) presented in this issue of the Journal, challenge the paradigm that granulomatous inflammation in sarcoidosis is simply a Th1-mediated process, and suggest that this tissue response is much more contextual and nuanced (4). In their paper, they describe an innovative in vitro model in which treatment of peripheral blood mononuclear cells (PBMCs) isolated from patients with sarcoidosis by means of purified protein derivative (PPD)-coated polystyrene beads results in the formation of granuloma-like multicellular aggregates and alternative (M2) macrophage polarization (4, 5). Using an unbiased systems biology approach, they identified an IL-13–regulated gene expression pathway. Similarly, IL-13–regulated molecular pathways were found to be enriched in sarcoid mediastinal nodes and lung tissue as compared with control tissue. The authors also found induction of inhibitory leukocyte immunoglobulin-like receptor (LILRB) genes in PPD-stimulated PBMCs from patients with sarcoidosis. This family of immune checkpoint molecules has been shown to regulate diverse cellular processes, including macrophage M2 polarization (6, 7).

Although these findings are provocative, further mechanistic clarity is required. The pathways analysis suggests that the activated pathways are regulated by IL-13. It is somewhat surprising that IL-13 is not one of the upregulated genes during in vitro granuloma formation or in sarcoid tissues. However, data are shown to indicate a reduction in the ratio of IFN-γ to IL-13 (and IFN-γ to IL-10) in conditioned media of PPD-stimulated PBMCs isolated from patients with sarcoidosis as compared with healthy control PBMCs. Consistent with this observation, increased expression of IL-13 mRNA and protein has been reported in BAL cells or PBMCs isolated from patients with sarcoidosis (8, 9). There are several strengths of the in vitro model described in this report. First, we lack ideal animal models of sarcoidosis that recapitulate human disease. Moreover, the human in vitro model is an desirable vehicle for precisely dissecting molecular pathways involved in antigen-driven cell aggregation/granuloma formation. To that end, it is unfortunate that experiments to more definitively establish a causal role for IL-13 in downstream responses, such as cell aggregation/granuloma formation, M2 polarization, and LILRB molecule expression, were not vigorously pursued. To address the possible upstream contribution of IL-13, the authors show that the STAT6 inhibitor leflunomide diminished granuloma formation in PPD-stimulated sarcoid PBMCs. Given that IL-13 promotes alternative macrophage activation in a STAT6-dependent fashion, these findings implicate IL-13 as a possible upstream mediator (10). The results described need to be interpreted with some caution, as leflunomide blocks de novo pyrimidine production, resulting not only in STAT6 inhibition but also rather broad regulation of lymphocyte responses, including clonal expansion, terminal differentiation, and apoptosis (11). Contributions from other STAT6 activators, such as IL-4, cannot be excluded, although enhanced expression of IL-4 in sarcoidosis tissue or cells has not been described. A particularly appealing aspect of these experiments is that leflunomide is used clinically in the treatment of refractory sarcoidosis, and these studies provide some insights into possible mechanisms of its beneficial effects (12).

According to the current dogma, macrophage activation states can be classified as classically activated (M1 or “killer” macrophages) or alternatively activated (M2 or “healer” macrophages) (13). M2 macrophages can be further subdivided based on functional states and inducers of alternative activation. These M2 phenotypes include M2a macrophages, induced by IL-4 or IL-13; M2b macrophages, induced by concurrent immune complex and Toll-like receptor stimulation; and M2c macrophages, induced in response to antiinflammatory factors such as corticosteroids, TGF-β, and IL-10. In addition to their important roles in parasitic immunity and allergic inflammation, alternatively activated macrophages have been increasingly recognized for their contribution to wound healing and tissue remodeling (13, 14). In contrast to T-cell polarization, there is considerable plasticity in macrophage polarization phenotypes, whereby classical or alternative macrophage activation phenotypes exist on a continuum, and can be transient and highly reversible depending on signals present within the microenvironment. Previous work has found an increased proportion of M1 macrophages (as defined by CD40 cell surface expression) in the airspace of patients with sarcoidosis, whereas the proportion of M2 macrophages (as defined by CD163 cell surface expression) tended to be higher in other types of interstitial lung disease, including nonspecific interstitial pneumonia, idiopathic pulmonary fibrosis, and hypersensitivity pneumonitis (14). In contrast, Shamaei and associates recently reported enhanced CD163 staining in granulomas of patients with sarcoidosis as compared with tuberculous granulomas (15). The findings in the present study are consistent with the latter observation, demonstrating increased macrophage CD163 staining in PPD-stimulated sarcoid PBMCs. The use of CD163 alone as a marker of the M2 phenotype is suboptimal, as this marker may capture a unique subset of alternatively activated macrophages. However, concomitant expression of other M2 macrophage genes, such as MMP2, MMP12, and CCL18, further support M2 polarization. Pathway analysis identified IL-13 as a possible driver of the alternative macrophage activation in this study. Based on the limited induction of IL-13 in the in vitro granuloma model and sarcoid tissues, this does raise the possibility of epigenetic regulation of M2 genes. For example, the H3K27 demethylase Jmjd3 has been shown to be induced in a STAT6-dependent fashion and to lead to M2 polarization during helminthic infection (16).

As with all unexpected findings, new questions arise. Most importantly, what is the functional significance of M2 polarization in sarcoidosis? Given the association of M2 macrophages with diseases characterized by exuberant tissue repair and fibrosis (12, 13), one wonders whether the transition to a dominant M2 macrophage phenotype might orchestrate progressive fibroproliferation observed in more advanced stages of pulmonary sarcoidosis. Unfortunately, the limited number of patients studied does not allow for this question to be addressed. Moreover, it would be of considerable interest to understand the specificity of the granulomatous response observed in sarcoid cells and tissues. For instance, how does this response compare with that observed using cells and tissues recovered from other types of granulomatous or interstitial lung diseases, and would the response be different if different antigens (e.g., Th1 antigens such as Candida, or Th2 antigens such as Schistosoma egg antigen) were employed? Finally, the durability and reversibility of macrophage skewing is unknown and requires further investigation. Regardless, this timely study sheds fresh new light on pathogenic mechanisms in sarcoidosis and identifies additional potential accomplices involved in this disease.