Transcriptional repressors as guardians of tissue macrophage identity

Abstract

The association between specific transcription factors (TFs) and defined tissue‐specific macrophage phenotypes is far from being univocal. Many TFs that have been associated with tissue‐specific macrophages have relatively broad expression profiles suggesting the critical involvement of combinatorial regulation by multiple TFs in bringing about specific phenotypes. In the current issue of The EMBO Journal, Rauschmeier et al (2019) report the identification of the transcriptional repressors BHLHE40 and BHLHE41 as novel regulators of tissue‐specific macrophage properties that work both to promote alveolar macrophage (AM) identity and to suppress alternative tissue‐specific phenotypes.

Macrophages are key cells of the innate immune system that contribute a first line of immune defense in virtually all tissues of the body and critically contribute to organogenesis and tissue homeostasis. Tissue‐resident macrophage shares numerous biological functions, including the ability to assist the response to microbial threats via the production of soluble mediators (such as cytokines and chemokines) and to eliminate pathogens and dead cells by phagocytosis (Lavin et al, 2015). Despite these commonalities, macrophage properties and functions are strictly tissue‐specific (Lavin et al, 2015). Indeed, the hardwired core macrophage transcriptional network adjusts and diversifies in response to tissue‐specific niche signals (Gosselin et al, 2014; Lavin et al, 2014; Mass et al, 2016), which include cytokines, metabolites, and signals delivered by stromal and parenchymal cells via direct cell‐to‐cell contacts. The eventual generation of tissue‐specific macrophage subtypes can be considered a form of adaptation to the local environment, which can be rapidly reversed if macrophages are experimentally transferred to a different tissue (Natoli & Ostuni, 2019).

Niche signals act by regulating expression or function of specific TFs that work together with macrophage lineage‐determining TFs (notably PU.1 and IRF8) to control the transcriptional output characteristic of specific tissue macrophages (Fig 1). For example, the PU.1‐related TF SPI‐C is selectively required for the development of splenic red pulp macrophages, where it is specifically induced by the high levels of heme accumulated as a consequence of the destruction of senescent erythrocytes (Haldar et al, 2014). Along the same line, the liver X receptor alpha (LXRα) TF, which is activated by cholesterol metabolites (oxysterols), is essential for marginal zone macrophages in the spleen (A‐Gonzales et al, 2013); peritoneal macrophage is controlled by GATA6 induced in response to locally produced retinoic acid (Okabe & Medzhitov, 2014), and PPARγ orchestrates AM development downstream of GM‐CSF and TGFβ (Schneider et al, 2014).

Figure 1.

Schematic representation of the links between tissue‐specific niche signals, transcriptional regulators, and specification of distinct types of tissue macrophages.

The closely related helix–loop–helix transcriptional repressors BHLHE40 and BHLHE41 participate in a variety of molecular processes, from the regulation of circadian rhythms to the control of several immune cell types, including T and B cells. Rauschmeier and colleagues surveyed the relative expression of both transcription factors in macrophages of different origins and found that AMs are unique in that they express high levels of both proteins, although their expression is induced at distinct phases of development. While BHLHE40 expression is already expressed in fetal monocytes upon lung colonization, BHLHE41 expression correlates with the final stages of AM differentiation and mature phenotype. What are the exact cues leading to BHLHE40 and BHLHE41 expression remains to be addressed, but the contribution of local expression of TGFβ is likely. Consistent with expression data, the deletion of both BHLHE40 and BHLHE41 in AMs provoked functional abnormalities and a reduction in proliferative capacity that made double knockout cells disadvantaged at repopulating lungs in competitive transplantation experiments.

In keeping with previous observations, the authors could link the reduced proliferation to the de‐repression of two TFs, MAF and MAFB, that act in a redundant manner to enforce cell cycle exit and quiescence in macrophages (Aziz et al, 2009). In addition to the proliferative defects, BHLHE40/41‐deficient AMs displayed obvious functional defects. Specifically, a central and highly tissue‐specific function of AMs is to remove pulmonary surfactant accumulating in the alveolar cavity, thus promoting its turnover. The surfactant is a complex mixture of lipids and proteins that act to reduce surface tension in alveoli, thus preventing their collapse. In the absence of BHLHE40/41, a marked accumulation of intracellular lipids likely representing undigested components of pulmonary surfactant was observed and linked to changes in expression of sets of genes involved in lipid metabolism.

Another important aspect of the work by Rauschmeier et al (2019) is that it contributes to clarify the complex interplay among multiple players implicated in the control of tissue‐specific functions of macrophages. AMs are mostly if not uniquely originated from fetal liver monocytes that differentiate upon exposure to granulocyte‐macrophage colony‐stimulating factor (GM‐CSF) produced by epithelial cells of the alveoli. Exposure to GM‐CSF and TGFβ has been shown to trigger expression of PPARγ that in turn works with other TFs such as BACH1/BACH2 and CEBPβ to generate a transcriptional output required for AMs to exert their functions (Schneider et al, 2014). The current work shows that BHLHE40 and BHLHE41 work in concert with PPARγ to attenuate the expression of specific and shared sets of genes. Specifically, although PPARγ is mainly known to act as an activator and not a repressor of gene expression, genes repressed by BHLHE40/BHLHE41 were also targets of PPARγ‐mediated repression. A simple model that is partially supported by ChIP‐seq data is that co‐repression by BHLHE40/41 and PPARγ is mediated by combinatorial binding to the same genomic regulatory elements. From a functional point of view, it is of great interest the observation that BHLHE40 and BHLHE41 downregulate a wide spectrum of genes that are normally expressed by other tissue‐specific macrophages, implying that the acquisition of tissue‐ or niche‐specific gene expression programs must be accompanied by the extinction of alternative programs specific for distinct contexts.

The identification of BHLHE40 and BHLHE41 as novel regulators of AM phenotype acting in the context of both general macrophage core regulators and additional tissue‐ and niche‐specific TFs illustrates how complex is the transcriptional regulation of macrophages residing in distinct tissues and organs. In turn, this complexity is a reflection of the versatility of these cells in contributing to tissue homeostasis in contexts as different as those encountered in lungs, spleen, intestine, peritoneum, and so on. The deconvolution of the complex links between local signals, transcription factors, and transcriptional repressors in the generation and maintenance of different tissue‐specific macrophage populations will continue representing a major area of investigation with obvious therapeutic implications.

The EMBO Journal (2019) 38: e103271