Reg-ulating macrophage infiltration to alter wound healing following myocardial infarction

This editorial refers to ‘Reg proteins direct accumulation of functionally distinct macrophage subsets after myocardial infarction’ by H. Lörchner et al., pp. 1667–1679.

Macrophages are fundamental regulators for cardiac repair after myocardial infarction (MI).^1,2 While initially thought to exist in just two forms (inflammatory and anti-inflammatory), ample evidence now exists to show multi-dimensional phenotypes representing roles ranging from amplifying the pro-inflammatory status to initiating inflammation resolution to promoting repair through endothelial cell activation to stimulate neovascularization and fibroblast activation to produce extracellular matrix to form the infarct scar.^3–6

The Braun laboratory has previously identified regenerating islet-derived protein (Reg) 3β as a novel factor to signal macrophages to enter the infarct region.⁷ They now extend these findings to explore roles of Reg3β, as well as Reg3γ and Reg4, in the post-MI macrophage.⁸ From their evaluation, they classify three macrophage subpopulations: (i) pro-inflammatory major histocompatibility complex (MHC)-II^hi and Ly6C^lo cells; (ii) anti-inflammatory MHC-II^lo and Ly6C^lo cells; and (iii) neovascularization stimulating Ly6C^hi cells. The differences between these subtypes are summarized in Table 1.

This report establishes the what, where, and when of scientific investigation by showing that all three Reg proteins are robustly induced after MI, both in a mouse MI model and in human MI autopsy samples. Reg expression is majorly from the myocytes, as CD11b+ myeloid cells expressed these genes at very low levels. Using immunohistochemistry, they showed that macrophages and not fibroblasts bind Reg proteins. By time course evaluation, Reg3β peaks at Day 1, Reg3γ peaks at Day 2, and Reg 4 peaks at Day 4 after MI. What the significance of this divergence in peak expression times means remains to be evaluated. For all three Reg proteins, expression had returned to baseline values by Day 14, consistent with roles in wound healing. In vitro, all three proteins stimulated macrophage migration. Transcriptomic analysis of Day 4 post-MI subtypes revealed heterogeneity in profiles in the different subtypes. Whether individual subtypes also change in profiles over time was not assessed, and there may be temporal and spatial heterogeneity within the subpopulations.

Reg3β gene deletion prevented recruitment of macrophages (particularly the MHC-II^hi and Ly6C^lo cells), while local delivery of any of the three Reg proteins at the time of MI stimulated recruitment of the MHC-II^hi/lo Ly6C^lo subtypes and repressed MHC-II^hi/lo Ly6C^hi cells. The absence of macrophages in the MI resulted in a rupture and heart failure phenotype, highlighting the importance of the macrophage in wound healing.¹

Cardiomyocytes have long been known to stimulate the post-MI pro-inflammatory reaction by releasing damage-associated molecular patterns; in particular, complement activation is a key initiator of the inflammatory response.^9,10 We can now add Reg proteins to this list, as the cardiomyocyte was the major cell type generating the Reg proteins. Cardiomyocytes in the border zone (and likely in the infarct region) provided a strong signal to recruit macrophages into the damaged area. In vitro, oncostatin M (OSM) was a potent inducer of Reg3β. OSM signalling was sufficient for Reg protein induction in cardiomyocytes in vitro, but was not required for Reg3γ or Reg 4 expression after MI in vivo, as Reg3β was the only one dependent on presence of the OSM receptor. Of note, transforming growth factor β (TGFβ) suppressed the OSM-stimulated production of Reg3β and Reg3γ in cultured cardiomyocytes, bringing out the possibility that TGFβ has a role in tempering the pro-inflammatory component.

Previous dogma that there were only two macrophage subtypes (pro-inflammatory and anti-inflammatory) was based on an in vitro classification system that did not take into account multiple stimuli that occurs in the MI setting. MI is a complex and dynamically shifting environment, and the macrophage network reflects both the complexity and temporal components. The overwhelming evidence supports discontinuation of the M1/M2 nomenclature for the MI setting and this study by the Braun laboratory furthers this argument.³

While the study by Lörchner et al. extends our MI macrophage knowledge map, there are a number of future evaluations still needed.⁸ Mapping the signalling cascades invoked by the Reg family members will provide further information on how these proteins overlap and are distinct. While Reg3β deletion increased the cardiac rupture phenotype, more details on Reg roles in post-MI cardiac physiology and long-term remodelling outcomes are needed.

In summary, the Braun laboratory report intriguing findings on Reg modulation of macrophages. Selective modulation of macrophage subsets to generate predictable cardiac remodelling phenotypes is an immediate to long-term goal in the field, and these results provide information that will help us to achieve that goal.

Acknowledgements

The author acknowledges funding from the National Institutes of Health under Award Numbers GM104357, GM114833, GM115428, HL051971, HL075360, and HL129823, and from the Biomedical Laboratory Research and Development Service of the Veterans Affairs Office of Research and Development under Award Number 5I01BX000505. The content is solely the responsibility of the author and does not necessarily represent the official views of the National Institutes of Health or the Veterans Administration.

Conflict of interest: none declared.

The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.

References