Distinct responses of newly identified monocyte subsets to advanced gastrointestinal cancer and COVID-19

Abstract

Monocytes are critical cells of the immune system but their role as effectors is relatively poorly understood, as they have long been considered only as precursors of tissue macrophages or dendritic cells. Moreover, it is known that this cell type is heterogeneous, but our understanding of this aspect is limited to the broad classification in classical/intermediate/non-classical monocytes, commonly based on their expression of only two markers, i.e. CD14 and CD16. We deeply dissected the heterogeneity of human circulating monocytes in healthy donors by transcriptomic analysis at single-cell level and identified 9 distinct monocyte populations characterized each by a profile suggestive of specialized functions. The classical monocyte subset in fact included five distinct populations, each enriched for transcriptomic gene sets related to either inflammatory, neutrophil-like, interferon-related, and platelet-related pathways. Non-classical monocytes included two distinct populations, one of which marked specifically by elevated expression levels of complement components. Intermediate monocytes were not further divided in our analysis and were characterized by high levels of human leukocyte antigen (HLA) genes. Finally, we identified one cluster included in both classical and non-classical monocytes, characterized by a strong cytotoxic signature. These findings provided the rationale to exploit the relevance of newly identified monocyte populations in disease evolution. A machine learning approach was developed and applied to two single-cell transcriptome public datasets, from gastrointestinal cancer and Coronavirus disease 2019 (COVID-19) patients. The dissection of these datasets through our classification revealed that patients with advanced cancers showed a selective increase in monocytes enriched in platelet-related pathways. Of note, the signature associated with this population correlated with worse prognosis in gastric cancer patients. Conversely, after immunotherapy, the most activated population was composed of interferon-related monocytes, consistent with an upregulation in interferon-related genes in responder patients compared to non-responders. In COVID-19 patients we confirmed a global activated phenotype of the entire monocyte compartment, but our classification revealed that only cytotoxic monocytes are expanded during the disease progression. Collectively, this study unravels an unexpected complexity among human circulating monocytes and highlights the existence of specialized populations differently engaged depending on the pathological context.

Integrating AI-Powered Digital Pathology and Imaging Mass Cytometry Identifies Key Classifiers of Tumor Cells, Stroma, and Immune Cells in Non–Small Cell Lung Cancer

“…While some clusters occupied definite positions on the UMAP, the Mϕ populations expressing high levels of lineage markers (Mϕ, CD68 hi Mϕ, CD14 hi /CD16 hi Mϕ and CD163 hi Mϕ) were closer and overlapping, suggesting that they might represent the same population at different maturation states ( Fig. 3F ), confirming previous evidence obtained by trajectory analyses of transcriptomic data ( 36, 37 ). By UMAP visualization of the cluster arrangement with respect to their tissue location ( Fig.…”

“…The regional specialization of macrophage subtypes in tumor tissues is an important feature that can control their functional profile and correlate with survival benefit, as shown in melanoma ( 47 ) and in human colo-rectal liver metastasis ( 37 ). In line with previous studies, a considerable fraction of the Mϕ granularity was accounted by very similar populations (occupying the same position on a UMAP representation), possibly representing the same cell in different maturation states ( 36, 37 ). Another interesting aspect emerging from the integrative analysis was that tumor cells were significantly more frequent in poorly fibrotic tumors, while the presence of fibrosis did not correlate with the K score (i.e., the distribution pattern of tumor cells).…”

The slan antigen identifies the prototypical non-classical CD16+-monocytes in human blood

“…Totally, we sequenced 20890 cells, which decreased to 19085 after quality control. Dimensionality reduction by UMAP ( 26 ) clearly demonstrated a net segregation of the NC/slan + /slan − - and CL-monocyte groups, with INT-monocytes placing exactly in between them ( Figures 4A, B ), as expected ( 8 , 12 , 27 , 28 ). Moreover, in line with the bulk transcriptomic ( Figure 2A ) and proteomic ( Figure 3A ) results, the same UMAP confirmed that slan − /NC-monocytes stand closer to CL-and INT-monocytes than to slan + /NC-monocytes ( Figures 4A, B ).…”

Behind the monocyte’s mystique: uncovering their developmental trajectories and fates

“…Ym1 + Mos were deemed crucial for inflammation resolution, simultaneously producing higher levels of anti-inflammatory IL-10 and lower pro-inflammatory cytokines (IL-6, IL-12) than Ym1 − monocytes. NeuMos were also discovered to be present in the steady-state BM and blood [ 30 , 31 ], perhaps suggesting the presence of a background signal that maintains a basal level differentiation of GMPs into NeuMos in the BM. On the other hand, a deviation of GMP differentiation into cMoPs into a unique progenitor state could also occur in particular diseases such as bleomycin-induced fibrosis, whereby cells termed segregated-nucleus-containing atypical monocytes (SatMs) were discovered [ 32 ].…”

“…However, many questions still remain: How do these circulating phenotypes differ from monocytes that arise through the priming of BM progenitors such as NeuMos and DCMos? In a recent study, Rigamonti et al have attempted to address this question by demonstrating nine distinct functional populations of human circulating monocytes in healthy donors [ 31 ]. Specifically, there were five populations related either to inflammatory, neutrophil-like, interferon-related, or platelet-related pathways among the classical subset of monocytes.…”