Generation of functionally active resident macrophages from adipose tissue by 3D cultures

Abstract

Introduction: Within adipose tissue (AT), different macrophage subsets have been described, which played pivotal and specific roles in upholding tissue homeostasis under both physiological and pathological conditions. Nonetheless, studying resident macrophages in-vitro poses challenges, as the isolation process and the culture for extended periods can alter their inherent properties.

Methods: Stroma-vascular cells isolated from murine subcutaneous AT were seeded on ultra-low adherent plates in the presence of macrophage colony-stimulating factor. After 4 days of culture, the cells spontaneously aggregate to form spheroids. A week later, macrophages begin to spread out of the spheroid and adhere to the culture plate.

Results: This innovative three-dimensional (3D) culture method enables the generation of functional mature macrophages that present distinct genic and phenotypic characteristics compared to bone marrow-derived macrophages. They also show specific metabolic activity and polarization in response to stimulation, but similar phagocytic capacity. Additionally, based on single-cell analysis, AT-macrophages generated in 3D culture mirror the phenotypic and functional traits of in-vivo AT resident macrophages.

Discussion: Our study describes a 3D in-vitro system for generating and culturing functional AT-resident macrophages, without the need for cell sorting. This system thus stands as a valuable resource for exploring the differentiation and function of AT-macrophages in vitro in diverse physiological and pathological contexts.

Keywords: 3D culture; adipose tissue; bone marrow; macrophage subpopulation; metabolism; phagocytosis; resident macrophage; unsupervised analysis.

Figure 1

Generation of AT-macrophages from 3D culture system. The seeding of the stromal vascular fraction (SVF) cells obtained from the murine sc–adipose tissue (AT) into plates with low adherence leads to the formation of spheroids. (A) Representative picture of these spheroids at 4, 7, and 13 days of culture. Acquisition was performed using an inverted Nikon Objective ×40 (scale bar = 500 µm). (B) Kinetic of the spheroid size. P-values were obtained by comparing the mean values of each time point with the mean value of day 4. (C) Spheroid cell number at days 7 and 13 of culture. (D, E) After dissociation, at days 7 and 13 spheroid composition was analyzed by flow cytometry. (D) Representative histograms of CD45 expression in dissociated cells from scAT SVF, spheroid day 7 and spheroid day 13, gated on singlets DAPI⁻ (4’,6-diamidino-2-phenylindole) cells. (E) Quantification of hematopoietic (CD45⁺) and non-hematopoietic (CD45⁻) cells in the scAT SVF and within the spheroid, expressed in % of DAPI⁻ cells. (F–K) Starting at day 10 of culture, dispersed cells migrate outside of the spheroids. (F) Number of scattered cells per spheroid obtained after 7 and 13 days in culture. (G) Representative picture of cells that have spread out of the spheroid after 13 days in culture (scale bar = 100 µm). Macrophages were stained with F4/80 (red) and their nuclei stained with DAPI (blue). (H). Representative dot plot of flow cytometry showing the expression of macrophage-specific markers on the surface of scattered cells out of the spheroid and selected among singlet DAPI⁻ cells. (I). Percentage of macrophages (CD45⁺/F4/80⁺/CD11b⁺) and other myeloid cells (CD45⁺/CD11b⁺/F480⁻) among CD45⁺ scattered cells surrounding spheroids and quantified by flow cytometry at day 13. (J) Representative histogram of flow cytometry showing the expression of Ki67 on AT cultured macrophages outside spheroids. (K) Quantification of Ki67 staining in the SVF and AT cultured macrophages outside spheroids. Results are expressed as mean ± SEM and compared using one-way ANOVA and t-test.

Figure 2

AT-macrophages are mature and phenotypically different from BM-macrophages. Macrophages generated from adipose tissue (AT) spheroids or from bone marrow (BM) monocyte differentiation were harvested and then seeded on adherent culture dishes, for 24h. Cells and supernatant were then collected and analyzed by flow cytometry. (A) Gating strategies for AT- and BM-macrophages. (B–D) Unbiased analysis of 5 independent samples of AT- and BM-macrophages realized using Uniform Manifold Approximation and Projection (UMAP) visualization with the signal strength of key phenotypic macrophages markers. (B) Cell clustering of both macrophage populations showing that AT- and BM-macrophages are distinct populations. (C) Protein expression patterns projected of CD206, CD16/32, MHC-II, MerTK, Dectin-1, CD36, F4/80, and CD11b MFI represented on a UMAP and illustrated using blue-green-yellow continuous color scale. (D) Heatmap visualization of surface markers expression on AT- and BM-macrophages. (E) Percentage of AT- and BM-macrophage (CD45⁺/CD11b⁺/F4/80⁺) expressing extracellular markers: CD206, CD36, CD16/32, Dectin-1, MHC-II, MerTK, TIM-4, Lyve-1, CCR2, (n = 3 to 14). Results are expressed as percentage of macrophage population and compared using paired t-test. (F). MFI of macrophages markers analyzed by flow cytometry in AT- and BM-macrophages (n = 3 to 14). Results are expressed as ratio of values obtained in BM-macrophages and compared using paired t-test. (G) Heatmap performed on the production of seven cytokines (pg/µl) quantified by LEGENDplex in supernatant of AT- and BM-macrophages cultures in unstimulated conditions. The dendrogram performed using complete linkage method was able to cluster BM- and AT-macrophages on the basis of their cytokines production. (n = 5). (H) Principal components analysis (PCA) performed on macrophages cytokine production. BM- and AT-macrophages were colored in orange and blue, respectively. (I) Arginase-1 and iNOS mRNA relative expression evaluated by RT-qPCR (n = 3–5). Results are express as a ratio of housekeeper gene expression (GAPDH). (J) Nitric Oxyde (NO) production was quantified in the culture medium using Griess Reagent (n = 5). Results are expressed in pg/µl. (K) ROS production was followed during 90 min using Luminol and Chemiluminescence was measured (n = 10). Results were compared using paired t-test.

Figure 3

AT-macrophages are more metabolically active than BM-macrophages. Macrophages generated from adipose tissue (AT) spheroids or from bone marrow (BM) monocyte differentiation were seeded onto seahorse plates for 24h before to analyze mitochondrial respiration and glycolytic capacity. (A–D) Mitochondrial respiration and glycolytic capacity of AT- and BM-macrophages quantified using Seahorse Agilent Technology 96eXF. Results were normalized to the cell number using DAPI staining after the assay. (A, B) Extracellular acidification rate (ECAR) upon glycolytic stress (injection of glucose, oligomycin, and 2-deoxyglucose) was measured. (A) Representation of real-time measurement of ECAR. (B) Glycolysis, glycolytic capacity and glycolytic reserve were calculated (n = 5). (C, D) Oxygen consumption rate (OCR) upon mitochondrial stress (injection of oligomycin, FCCP, rotenone and antimycin A) was measured. (C) Representation of real-time measurement of OCR. (D) Basal respiration, maximal respiration and respiration-coupled ATP production were calculated (n = 5). (E) Energy map of maximal respiration versus glycolytic capacity after FCCP injection (n = 5). (F) Glyco- and mito-ATP production rates were calculated using the aforementioned ECAR and OCR data (n = 5). Results were compared using paired t-test.

Figure 4

AT- and BM-macrophages are differentially polarized by IL-4 or IFN-γ treatment. Macrophages generated from adipose tissue (AT) spheroids or from bone marrow (BM) monocyte differentiation were seeded onto adherent culture dishes for 24h, and treated with IFN-γ (A–C) or IL-4 (D–F) for another 24h. Flow cytometry was used to assess MFI (A, D) and % of positive cells (B, E) for each cell surface marker expressed by macrophages after treatment (n = 3–6). (C, F) Supernatants were collected and cytokine production quantified by LEGENDPlex (n = 5) in AT- and BM-macrophages. Results were compared using paired t-test. One Sample t-test analysis was used to compare stimulated and unstimulated conditions.

Figure 5

AT- and BM-macrophages show similar phagocytic activity. Phagocytic capacity of E. coli or C. Albicans yeast was assessed in adipose tissue (AT)– and bone marrow (BM)–macrophages generated from AT spheroids or from BM monocyte differentiation. (A–C) pH-Rodo E.Coli were added to macrophages and phagocytosis (apparition of red particles) was analyzed on IncuCyte during 48h. (A) Representatives pictures (IncuCyte) of phagocytose by BM- (up) and AT- (down) macrophages before (left) and after (right) E. coli addition. (B) A representative kinetic curve of bacteria phagocytosis by AT- and BM-macrophages. (C) Quantification of E. coli phagocytosis. Results are express in AUC of red particles reported to cell area. (n = 3). (D, E) C. Albicans yeast were added to macrophages for 30 min before assessing binding (D) and killing capacity (E). Macrophages were lysed and the supernatants were plated on Sabouraud Petri dishes. Twenty-four hours later, colonies were counted. Results are express as percent of bound among total colonies or killed colonies among total bound colonies. (n = 6) Results were compared using Paired t-test.

Figure 6

3D cultures to generate macrophages that mirror the phenotypic traits of in-vivo AT-resident macrophages. (A, B) A bulk RNA-seq analysis was performed on mouse-sorted sc– adipose tissue (AT) macrophages, AT- and bone marrow (BM)–cultured macrophages. (A) Hierarchical clustering of the 6,855 genes differentially expressed between AT- and BM-macrophages, according to their expression pattern across all samples, using multiClust R library. GO processes and adjusted P-values are mentioned. (B). Principal components analysis (PCA) on standardized gene expression of AT- (blue), BM- (orange), and sc-AT sorted (green) macrophages. (C, D) Unbiased analysis of SVF cells obtained from mouse sc-AT was realized by flow cytometry. (C). UMAP showing the clustering (Louvain method) of sc-AT macrophages in two populations and their percentages. (D) Heatmap of specific cell surface markers expression of sc-AT macrophages (clusters 0 and 1) (n = 7). (E–G) Single-cell RNA-seq analysis of sc-AT macrophages extracted from Emont et al. dataset (#GSE176171) (GSM5820690_Mm_ING_08–3) (29). (E) UMAP showing agglomerative clustering of macrophage population based on the gene expression. (F) Heatmap visualization of Mrc1, Fcgr3, Fcgr2b, Mertk, H2-Aa, H2-Eb1, H2-Ab1, and Cd36 gene expression (for clusters 0 and 1). (G). Heatmap visualization of Ccr2, Lyve1, Timd4 and Folr2 (for clusters 0 and 1).