Resident versus recruited macrophage responses drive cardiac fibrosis and heart failure development

Xavier Revelo, Jop van Berlo, and co-authors investigated how resident versus recruited macrophages shape fibrotic responses in pressure-overload induced heart failure in mice.

Extracellular matrix remodelling and fibrosis is a hallmark of virtual any advanced heart disease causing irreversible scarring and organ dysfunction. Yet, there are few direct approaches targeting fibroblasts and collagen deposition and none in clinical practice.

Myeloid cells, mostly macrophages, are abundant in the heart (1). However, they are not a uniform cell population but heterogeneous and comprise resident tissue-homeostatic embryonically-derived macrophages and monocyte-derived macrophages that are recruited to the heart (2). After myocardial infarction, for example, these subsets exhibit distinct roles in the remodeling response (3, 4). Further, the interaction of adaptive immune cells such as T cells and myeloid cells in ECM remodelling has only recently gained broader interest (expertly reviewed in 5).

Ten years ago, we used sex-mismatched donor-recipient heart transplant patients to determine the contribution of resident versus recruited macrophages and fibroblasts to fibrosis and found that in transplanted hearts fibrosis appears mostly driven by tissue-resident donor-derived fibroblasts. In contrast, CD68+ macrophages in fibrotic areas were mostly of recipient origin(6). This was a descriptive study; meanwhile, the knowledge of immune-cell biology has expanded vastly as have technical capabilities.

In the current paper, the authors now take advantage of new experimental methods and apply them elegantly to elucidate immune-cell sub-populations following pressure-overload and mechanistically interrogate their function in fibrogenesis in preclinical mouse models.

Using flow cytometry and cytometry by time-of-flight (CyTOF) authors find a substantial accumulation of CD45+ leucocytes one week after transverse aortic constriction (TAC) mostly comprised of CD64+ F4/80+ macrophages that resides back towards normal levels after one month. Using scRNAseq, authors identified twelve monocyte/macrophage subtypes, eight of which were differently impacted after TAC. Ccr2 (and Timd4) were used to distinguish CCR2+ recruited vs TIMD4+CCR2- resident macrophages and gene expression and pathway analysis suggested distinct functional roles and upstream regulators of these subsets.

Next, to mechanistically probe the role of macrophage subsets in the early TAC response, the authors preferentially depleted resident macrophages using an anti-CD115 (blocking colony stimulating factor 1 binding) antibody to reduce tissue macrophage levels to ~1/4 to 1/3. Monocyte-derived macrophages were quickly replenished from the circulation. This approach did not alter cardiac hypertrophy or function (EF) one week after TAC, however, it amplified interstitial fibrosis and reduced angiogenesis. As isolated macrophages from hearts expressed pro-fibrotic cytokines such as TGFb but depletion of resident macrophages nonetheless increased collagen deposition authors speculated that recruited monocyte-derived macrophages play a critical role in fibrotic responses. They investigated this using CCR2 knockout mice, where recruited monocyte-derived macrophages are depleted. Indeed, fibrosis was attenuated in CCR2 knockouts (with and without anti-CD115 treatment).

At six weeks after TAC and anti-CD115 treatment, resident macrophage levels were replenished (prior work suggest this happens exclusively from resident pools, 2). As at the one-week time-point, fibrosis was augmented. In addition, these hearts exhibited worse EF and more cardiomyocyte hypertrophy later. Suggesting that early resident macrophages responses prime ensuing remodelling responses in an anti-fibrotic and protective fashion.

This study adds important evidence to our understanding of macrophage-driven remodelling responses in the heart and supports that tissue-resident macrophages oppose recruited macrophage effects in some aspects. Importantly, immune-cell to fibroblast crosstalk is a dynamic and injury-type specific response and may differ at early vs. late stages and ischemic vs. non-ischemic injuries. The current study suggests that it is important to target these cells/responses early (week 1 response). While depletion strategies are elegant and mechanistic approaches to interrogate the role of macrophage subsets there is some leakiness of these approaches and also extra-cardiac organs and myeloid cell niches are affected. The authors transcriptional data suggest that several (8 out of 12) macrophage subsets are impacted after injury. In this respect, it is important to emphasize that transcriptional clustering differs from established cell-surface marker-based characterization of cells and that the role of newly discovered transcription clusters often remains ill-defined. Future studies will investigate the role of these subsets and their balance and how they contribute to remodelling responses. Ultimately, this may lead to therapies that harness macrophage-induced stress responses to counter adverse remodelling.