Lean and obese bASCs reduce their differentiation capacity within the TME

The importance of the differentiation capacity of MSCs is well-known for proper tissue homeostasis and regeneration [30, 31]. We questioned how the TME would affect the differentiation capacity of ln-aT bASCs and ob-aT bASCs, compared to their counterpart ln-dT bASCs and ob-dT bASCs isolated from adipose tissue distant to the tumors. Clinical information of the patients is listed in Table S1. These bASCs were subjected to adipogenic, osteogenic and chondrogenic differentiation stimuli and their differentiation capacities were evaluated by lineage specific staining and gene analysis. The analysis showed that ob-bASCs had a highly reduced differentiation capacity in all three analyzed lineages compared to ln-bASCs (Fig. 1), supporting our previous reports of visceral and subcutaneous ASCs [9, 32, 33]. Specifically, both ln- and ob-aT bASCs displayed a significantly reduced adipogenic differentiation capacity compared to ln- and ob-dT bASCs (Fig. 1A-D). In accordance, ln-dT bASCs had the highest percentage of immature- (44.2%) and mature adipocytes (24.1%), compared to the other three subgroups (Fig. 1A-C). Although ob-dT bASCs showed no difference in the percentage of immature adipocytes relative to ob-aT bASCs (ob-dT bASCs: 28.5% vs. ob-aT bASCs: 27.7%) (Fig. 1A and B, 4th and 5th bars), the ob-dT bASCs displayed a significantly increased number of mature adipocytes (ob-dT bASCs: 17.0% vs. ob-aT bASCs: 8.6%) (Fig. 1A and C, 4th and 5th bars), suggesting a reduction of adipogenic differentiation ability of ob-aT bASCs. In line with this observation, the expression of adipogenic-related genes such as peroxisome proliferator-activated receptor γ (PPARγ), adiponectin (ADIPOQ) and LEPTIN were highly upregulated by 66–83% in ln-dT bASCs compared to ln-aT bASCs (Fig. 1D, 5th vs. 6th bar). Reduced adipogenic gene expression was also observed in ob-aT bASCs in comparison to ob-dT bASCs (Fig. 1D, 7th vs. 8th bar), yet not to the same extent as in ln bASCs, and a significant difference was observed only for PPARγ.

Fig. 1

Lean and obese bASCs adjacent to breast cancer display an impaired differentiation capacity. A-J bASCs (ln-aT, ln-dT, ob-aT and ob-dT) were induced to adipogenic (adipo. Diff.) (A-D) for 14 days, osteogenic (osteo. Diff.) (E-H) and chondrogenic differentiation (chondro. Diff.) (I and J) for 21 days and their differentiation rates were evaluated. ln-dT bASCs were used as control cells without differentiation medium. A bASCs were stained for α-tubulin (green), phalloidin (red) to visualize the cytoskeleton, and DNA (DAPI, blue). Example images are shown. Scale bar: 50 μm. B and C The percentage of immature adipocytes (lipid vacuoles < 5 nm) (B) or mature adipocytes (lipid vacuoles > 5 nm) (C) was quantified. The results of individual bASC subgroups are presented as mean ± SEM (n = 500 cells for each condition, pooled from three experiments). D The gene expression of PPARγ, ADIPOQ and LEPTIN is shown for undifferentiated (−) and differentiated (+) bASCs. The results are from three individual experiments and presented as mean ± SEM. E bASCs were stained with Alizarin Red S to visualize calcium deposition. Representative images are shown. Scale bar: 50 μm. F and G The percentage of bASCs showing calcium deposition (F) and the mean gray value (G) were evaluated. The results are presented as mean ± SEM (F: n = 500 cells for each condition, pooled from three experiments, G: n = 30 images for each condition, pooled from three experiments). H The gene expression of KLF4, PTCH1, c-MYC, OSTEOPONTIN and RUNX2 is shown for undifferentiated (−) and differentiated (+) bASCs. The results are from three individual experiments and presented as mean ± SEM. I and J bASCs were stained with Alcian blue to visualize acidic polysaccharides. Representative bright-field images are shown (J). Scale bar: 50 μm. The quantification of the mean gray value is presented (I). The results are shown as mean ± SEM (n = 30 images for each condition, pooled from three experiments). Unpaired Mann-Whitney U test was used in (B and C), (F and G) and (I). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. Student’s t test was used in (D) and (H). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001

Similar results were obtained from microscopic analyses of the calcium deposition in differentiated osteocytes stained with Alizarin Red S (Fig. 1E). Compared to dT bASCs, aT bASCs demonstrated a reduced percentage of differentiated osteocytes and a lower mean gray value, a parameter used for the evaluation of the staining signal [34], of the Alizarin Red S staining (Fig. 1E-G, 2nd vs. 3rd bar and 4th vs. 5th bar). In particular, ob-dT bASCs had a decreased differentiation capacity by 11% (Fig. 1E-G, 3nd vs. 5th bar). Consistent with these results, all five genes KLF4, PTCH1, c-MYC, OSTEOPONTIN and RUNX2 associated with osteogenic differentiation were significantly higher in ln- and ob-dT bASCs compared to ln- and ob-aT bASCs (Fig. 1H). In further support, the chondrogenic differentiation, indicated with the specific staining of Alcian blue used to identify sulfated proteoglycans, showed a comparable trend (Fig. 1I and J). bASCs in the proximity of breast cancers exhibited drastically reduced chondrogenic differentiation capacity, as indicated by a weaker staining of polysaccharides quantified by the mean gray value (Fig. 1I, 2nd and 4th bars). Both distant bASCs subgroups displayed a moderate blue staining, whereas ln-dT bASCs showed a 26% increased gray value compared to ob-dT bASCs (Fig. 1I, 3rd vs. 5th bars). These results clearly demonstrate an impaired differentiation capability of obese- and cancer-near bASCs, suggesting a cancer-educated phenotype of bASCs in the TME of breast cancers.

To exclude the possibility that these differentiation alterations were a result of a changed cell cycle distribution or cell proliferation, cell cycle analyses and cell viability assays were performed. None of the analyzed bASC groups (ln/ob-aT/dT) showed significant differences in both assays (Fig. S1A and B). Moreover, all bASCs subgroups displayed a comparable CD (cluster of differentiation) marker profile specific for MSCs (Table S2 and S3) [35].

bASCs de-differentiate into distinct CAF-like phenotypes

Tumors are known to shape their TME to support their development, proliferation, metastasis and therapeutic resistance [36]. Especially, CAFs derived from various cell origins such as fibroblasts and MSCs [16, 37] are key components of the TME and important in TME remodeling [38]. To address if aT-bASCs de-differentiate into a myofibroblastic cancer-associated (myCAF) phenotype, we assessed their specific protein markers alpha-smooth muscle actin (αSMA), fibroblast-specific protein1 (FSP1) and integrin subunit beta 1 (CD29) by flow cytometry. As presented in Fig. 2A-C, ln-aT bASCs showed a moderately increased expression of αSMA and CD29 compared to ln-dT bASCs (Fig. 2A and C, 1st vs. 2nd bar). FSP1 did not show any differences in these lean bASCs subtypes (Fig. 2B, 1st vs. 2nd bar). Interestingly, ob-aT bASCs displayed a significantly enhanced expression of all three myCAF marker proteins in comparison to ob-dT bASCs (Fig. 2A-C, 3rd vs. 4th bar). CAFs, isolated from the same breast cancer as reported [39] and served as positive control, showed a remarkably high expression of αSMA and FSP1 (Fig. 2A and B, 5th and 6th bar), whereas CD29 was only upregulated in CAFs isolated from patients with obesity, similar to ob-aT bASCs (Fig. 2C, 3th and 6th bars).

Fig. 2

The tumor microenvironment induces the de-differentiation of bASCs. A-C bASCs (ln-aT, ln-dT, ob-aT and ob-dT) were stained for αSMA, FSP1 and CD29 for FACS analyses. CAFs (ln-CAF and ob-CAF) were isolated and stained as positive controls. Quantification of αSMA (A), FPS1 (B) and CD29 (C) are shown as bar graphs. The results are from three independent experiments (n = 3, 60.000 cells for each condition and in each group) and presented as mean ± SEM. D Representative images of bASCs and CAFs stained for αSMA (green), phalloidin (red) and DNA (DAPI, blue) are shown. Red boxes indicate measured areas. Scale: 25 μm. Inset scale: 12.5 μm. E The evaluation of the mean fluorescence intensity of αSMA is presented as scatter plots. The results are from three independent experiments (n = 3, 90 cells for each condition and in each group) and presented as mean ± SEM. F Cellular extracts from bASCs were prepared for WB analysis with antibodies against αSMA, COL1A1, caveolin-1, AKT, pAKT and pGSK3β. GAPDH and β-actin served as loading controls. G Quantification of the αSMA signal in WB is shown, relative to the corresponding amount of GAPDH. The results are from three independent experiments and presented as mean ± SEM. H and I Relative gene levels of ACTA2, TAGLN and CTGF, important myCAF marker genes (H), and relative gene expression of CSF3, CXCL10, IL1β and HAS1, iCAF marker genes (I), are shown for bASC subgroups (ln-dT, ln-aT, ob-dT and ob-aT). The results are from three independent experiments and presented as bar graphs with mean ± SEM. Student’s t test was used in (A-C) and (G-I). Unpaired Mann-Whitney U test was used in (E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. ∗p < 0.05, ∗∗∗p < 0.001

To corroborate these results, bASCs were stained for αSMA, vimentin (VIM) and collagen type I alpha 1 chain (COL1A1) for intensity quantification. The analyses revealed elevated signals for αSMA, VIM and COL1A1 in ln/ob-aT bASCs (Fig. 2D and E, Fig. S1C-F), whereas ln/ob-dT bASCs only displayed low to moderate staining signals (Fig. 2D and E, Fig. S1C-F). Of note, the expression of COL1A1 in bASCs could be clustered into three groups (Fig. S1C, E and F): the majority of lean and obese dT bASCs had a low expression of COL1A1 (Fig. S1C and F, 1st and 3rd bars), a high expression was observed in ln/ob-aT bASCs (Fig. S1C and F), and a high percentage of ob-aT bASCs also presented intermediate levels (Fig. S1C and F).

Furthermore, we analyzed the protein levels of αSMA, COL1A1, pAKT, AKT and pGSK3β via Western blot analysis. ob-aT bASCs presented an upregulated αSMA compared to all other bASC subgroups (Fig. 2F and G). By contrast, αSMA was just marginally increased in ln-aT bASCs (Fig. 2F and G), but COL1A1 was increased in ln-aT bASCs as indicated by the previous immunofluorescence staining (Fig. 2F and Fig. S1C, E and F). Moreover, ln-aT bASCs, like ob-aT bASCs, showed an activated PI3K pathway evidenced by increased pAKT and pGSK3β (Fig. 2F).

Unlike ob-aT bASCS, ln-aT bASCs displayed an inconclusive expression of specific myCAF markers. We hypothesized that ob-aT bASCs represented a myCAF-like type, whereas ln-aT bASCs might exhibit a different cancer-associated phenotype, namely iCAF, as described for fibroblasts [40, 41]. To clarify this issue, gene expression analysis of the myCAF associated genes ACTA2, connective tissue growth factor (CTGF) and transgelin (TAGLN), and iCAF classified genes colony stimulating factor 3 (CSF3), CXCL10, IL1β and hyaluronan synthase 1 (HAS1) [40, 42] were performed. Indeed, these analyses revealed that ob-aT bASCs had a predominant myCAF phenotype with highly increased gene levels of ACTA2, TAGLN and CTGF compared to ln-dT bASCs (Fig. 2H, 2nd vs. 3rd bar). Strikingly, though displaying slightly elevated gene profiles of ACTA2 and CTGF (Fig. 2H, 1st vs. 2nd bar), ln-aT bASCs had a significantly increased expression of iCAF genes including CSF3, CXCL10 and IL1β (Fig. 2I, 1st vs. 2nd bar). The gene levels of HAS1 were low and not significantly altered in bASCs, because of the low gene copy number (Fig. 2I, 4th graph). Ob-aT bASCs mostly had lower expression of these genes compared to other bASCs (Fig. 2I). These results suggest that bASCs in the TME are de-differentiated toward CAF-like phenotypes, which explains the greatly reduced differentiation capacity of both bASCs subgroups. Furthermore, these data highlight that obesity impacts the de-differentiation of bASCs isolated from adipose tissue near the breast tumor. Of importance, ln-aT bASCs reflected the gene expression pattern of an iCAF-like phenotype, whereas ob-aT bASCs demonstrated a myCAF gene and protein profile.

Transcriptional reprogramming of bASCs in the TME

Since ln/ob-aT bASCs displayed a functional decline and a transition to cancer-educated bASC phenotypes, a transcriptomic analysis should reveal the impact of the TME on the overall gene expression of aT bASCs in comparison to dT bASCs. Total RNAs were extracted from the four bASC subgroups for whole-genome mRNA-sequencing (RNA-seq) [43]. Strikingly, the transcriptome analysis revealed 967 deregulated genes with a significantly adjusted p-value in ln-aT bASCs versus ln-dT bASCs visualized by a heatmap (Fig. 3A) and a volcano plot (Fig. 3C). The transcriptome comparison between ob-aT bASCs and ob-dT bASCs revealed 824 significantly deregulated genes by using the p-value without adjustment (Fig. 3B and Fig. 3D), possibly due to distinct gene alterations in each obese individual. The “Kyoto Encyclopedia of Genes and Genomes” (KEGG) pathway analysis highlighted the most significant changes in their gene expression of ln-aT bASCs in the cell-cell receptor interaction (33 genes), pathways in cancer (40 genes), TNF signaling pathway (20 genes), PI3K-Akt signaling pathway (30 genes), and the chemokine signaling pathway (23 genes), compared to ln-dT bASCs (Fig. 3E). Interestingly, a further pathway analysis by the “Gene Ontology project” (GO) revealed that pathways involved in immune system processes (187 genes), response to stress (256 genes), signal transduction (337 genes) and cell differentiation (219 genes) were altered in ln-aT bASCs (Fig. 3F). The altered genes in the cell differentiation pathway included stemness- and proliferation-associated genes such as Erb-B2 receptor tyrosine kinase 4 (ERBB4), roundabout guidance receptor 2 (ROBO2), slit guidance ligand 2 (SLIT2), forkhead box C2 (FOXC2), SRY-box transcription factor 9 (SOX9), hyaluronan synthase 2 (HAS2), retinoblastoma-associated protein 1 (E2F1), forkhead box L1 (FOXL1) and nuclear factor kappa b subunit 2 (NFĸB2), all of which were upregulated in ln-aT bASCs (Fig. S2A, 1st to 9th plots). Their downregulated genes included platelet derived growth factor receptor beta (PDGFRβ), androgen receptor (AR) and homeobox A2 (HOXA2) (Fig. S2A, 10th to 12th plots).

Fig. 3

Transcriptomic profiles of lean and obese bASCs adjacent to breast cancers. A-G Total RNAs were extracted from each sample of bASC subgroups (ln-dT, ln-aT, ob-dT and ob-aT, 5 samples for each subgroup) for transcriptome analysis. A and B Heatmap of significantly differentially expressed genes of ln-dT vs. ln-aT bASCs (A) and ob-dT vs. ob-aT (B). Gene expression was analyzed using DESeq2 R package. Genes with a p-value < 0.01, and a fold change greater than 2 (red color code) and below − 2 (blue color code), respectively, were included. C and D Volcano plots showing the adjusted p-value (adj. p > 0.05) for genes differentially expressed between bASCs close to and distant to the breast cancers in lean (C) and obese (p > 0.05) (D) bASCs. Upregulated genes are depicted in red color, downregulated in green, and non-changed genes in blue (adjusted p-value > 0.05). E and F Significantly enriched KEGG pathways (E) and GO pathways (F) are presented for ln-aT bASCs compared to ln-dT bASCs. For each KEGG or GO pathway, the bar shows the adjusted p-value. The numbers (n) behind the pathway names indicate deregulated genes. G Violin plots present selected myCAF/iCAF genes differentially expressed. Values reflect the log expression levels of genes from the RNA-seq data. H Heatmap depicts significantly differentially expressed cytokine/chemokine genes in four bASCs subgroups. These genes have a fold change greater than 2 (red color code) or below − 2 (blue color code). Genes in (A-F) have an adjusted p-value of ≤0.05 and genes shown in (G) have an adjusted p-value of ≤0.05, at least for one condition (ln-aT or ob-aT)

Transcriptome analysis corroborated the gene expression of myCAFs (Fig. 3G, 1st to 3rd plots). Important myCAF genes, including ACTA2, TAGLN and CTGF, were upregulated in ob-aT bASCs (Fig. 3G, 1st to 3rd plots), whereas ln-aT bASCs showed an increased expression of iCAF associated genes such as CCL2, leukemia inhibitory factor (LIF), IL1β, CSF3 and HAS1 (Fig. 3G, 4th to 8th plots). In addition, ln-aT bASCs displayed various upregulated cytokine genes including IL8, CXCL1–3 and CXCL10 reported for iCAFs (Fig. S2B). Moreover, presented with a heatmap, ln-aT bASCs displayed a prominent iCAF phenotype by showing an upregulated gene expression of chemokines, including CCL, colony stimulating factor (CSF), C-C motif chemokine receptor 5 (CCR5), CXCL, cytokines such as IL1A, IL1B, IL18BP, the IL1 receptor accessory protein (IL1RAP), LIF, and growth factors including vascular endothelial growth factor C (VEGFC), PDGFB and its receptor PDGFRB (Fig. 3H). In sum, the transcriptome analysis supports the notion that the TME greatly impacts the function and biology of bASCs. In particular, the TME educates ln-aT bASCs into an iCAF-like phenotype displaying a high gene expression of cytokines, chemokines and growth factors.

dT bASCs alter their cytokine secretion upon co-culture with triple negative breast cancer cells

Transcriptome analysis revealed a network of deregulated cytokines in bASCs isolated adjacent to breast cancer cells (Fig. 3H). To look at the secretion profiles of these cytokines, the conditioned media from bASCs were collected for evaluation using a human cytokine array containing 120 different targets. ln-aT as well as ob-aT bASCs secreted a higher amount of important migration and invasion stimulatory cytokines such as CXCL5, CXCL6, CXCL11, CCL27, IL8, the tissue inhibitor of metalloproteinase 1 (TIMP-1), the macrophage migration inhibiting factor (MIF), and the interleukin 6 signal transducer (IL6ST, also known as gp130 and IL6R-β) (Fig. S3A and B). CXCL11 was the only cytokine, which was significantly upregulated in ob-aT bASCs compared to ob-dT bASCs (Fig. S3B, 2ndgraph).

To investigate the impact of breast cancer cells on the cytokine secretion of bASCs, all subgroups of bASCs were indirectly co-cultured (Fig. S2C) with the triple negative breast cancer cell line MDA-MB-231 for 7 days. The cells were then cultured in serum-free medium for additional 3 days for collecting conditioned media for cytokine evaluation. Surprisingly, 7-day indirect co-culture with MDA-MB-231 cells was sufficient to induce an altered secretion pattern in dT bASC control cells (Fig. S3A, 2nd vs. 6th, and 4th vs. 8th row). By contrast, this indirect co-culture did not influence aT bASCs to the same extent as dT ASCs (Fig. S3A, 1st vs. 5th and 3rd vs. 7th row). Intriguingly, both ln/ob dT-bASCs significantly increased their secretion of stem cell factor (SCF), CCL5, VEGFA, oncostatin M (OSM), MIF, and osteoprotegerin (OPG) (Fig. S3C). In addition, the secretion of CXCL1 and CXCL5, two cytokines involved in breast cancer metastasis [44, 45], was moderately increased in all subgroups after indirect co-culture with MDA-MB-231 cells (Fig. S3D). These results highlight the importance of the interaction between breast cancer cells, the TME and bASCs, even an indirect short co-culture in this experimental setup, already partially educate dT bASCs by altering their secretion pattern of various cytokines.

Ln- and Ob-aT bASCs upregulate the gene expression and secretion of tumor promoting cytokines

To further corroborate the transcriptome analysis data, we extracted RNA from four bASCs subgroups for quantitative gene analysis. Compared to ln-dT bASCs, ln-aT bASCs showed increased gene expression of IL6, IL8, CXCL1–3, CCL2, VEGFC, fibroblast growth factor 1 (FGF1) and FGF2, and LIF (Fig. 4A, 1st vs. 2nd bar) supporting the finding from the transcriptome analysis (Fig. 3H and Fig. S2B). The most significant hits were LIF, CXCL1, CXCL2 and CXCL3, which were all highly upregulated with RQ values of 6.5 to 48.4 in ln-aT bASCs compared to the ln-dT bASCs (Fig. 4A). Remarkably, ob-aT bASCs showed a similar trend with an increased gene expression of LIF (RQ: 2.4), CXCL1 (RQ: 3.0), CXCL3 (RQ: 2.8), CCL2 (RQ: 4.5), FGF1 (RQ: 1.8), FGF2 (RQ: 3.5) and VEGFC (RQ: 2.1), but not to the extent of ln-aT bASCs (Fig. 4A). Strikingly, ob-dT bASCs already displayed an enhanced gene expression of CXCL2, CXCL1, CXCL3, FGF2 and CCL2, compared to ln-dT bASCs (Fig. 4A), suggesting a crucial impact of obesity on bASCs.

Fig. 4

bASCs adjacent to breast cancer cells have increased gene expression and protein secretion of cancer promoting cytokines. A Relative gene levels of IL6, IL8, CXCL1, CXCL2, CXCL3, CCL2, VEGFC, FGF1, FGF2, and LIF are shown for bASC subgroups (ln-dT, ln-aT, ob-dT and ob-aT). The results are from three independent experiments and presented as bar graphs with mean ± SEM. B-E ELISA assays were performed with conditioned media from bASCs subgroups and the levels of cytokines IL6 and IL8 (B), chemokines CXCL1–3 and CCL2 (C), growth factors VEGFC and FGF1/2 (D), and inflammatory cancer-associated cytokines CSF3, CXCL10, IL1β and LIF (E) were analyzed. The results are from three independent experiments and presented as scatter bar graphs with mean ± SEM. Student’s t test was used. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001

To examine cytokine secretion, ELISA assays were performed with conditioned media from individual bASC subgroups. IL6 was upregulated in ob bASCs, whereas IL8 was slightly downregulated (Fig. 4B). Both ln-aT bASCs and ob-aT bASCs secreted high levels of cytokines (Fig. 4B). Additionally, ln-aT bASCs, compared to ln-dT bASCs, displayed significantly increased levels of CCL2, CXCL1–3, FGF1, FGF2 and VEGFC (Fig. 4C and D, 1st vs. 2nd bars). As observed by gene analysis (Fig. 4A), ob-aT bASCs also secreted increased levels of CCL2, CXCL1–3 and VEGFC, in comparison to ob-dT bASCs (Fig. 4C and D, 3rd vs. 4th bars), but not to the same extent as ln-aT bASCs did (Fig. 4C and D, 1st vs. 3rd bar).

To corroborate the gene expression data from transcriptomic analysis that suggest an iCAF phenotype for ln-aT bASCs, we analyzed the secretion of inflammatory cytokines and chemokines. Indeed, ln-aT bASCs had significantly increased secretion of CSF3, CXCL10 and IL1β, compared with all three other subgroups (Fig. 4E, 1st to 3rd graphs). Of importance, ob-dT bASCs were highly capable of releasing LIF with a value of 25.4 pg/ml relative to ln-dT bASCs with values of 9.7 pg/ml (Fig. 4E, 4th graph, 2nd and 4th bars), whereas ln- and ob-aT bASCs secreted it in a significantly high amount with 2.4 ng/ml in ln-aT bASCS and 2.2 ng/ml in ob-aT bASCs (Fig. 4E, 4th graph, 1st and 3rd bars). These data underscore the findings from the transcriptome analysis that ln-aT bASCs are de-differentiated into an iCAF-like phenotype. Importantly, most of these cytokines are tumor and metastasis promoting [16, 46, 47], suggesting that one of the main molecular mechanisms how ln/ob-aT bASCs support tumor development is via paracrine signaling.

Ob- and ln-aT bASCs promote the growth of breast cancer spheroids via cell-cell contact

To study the effect of bASCs and their secreted cytokines, spheroids were generated with two luminal B breast cancer cell lines BT474(ER+, PR+, HER2+) and MDA-MB-361(ER+, PR+/−, HER2+) [27] and seeded on a bASC feeder layer as indicated in Fig. S3E. Area and diameter of spheroids were measured up to 96 h. The spheroids were then fixed, stained for phospho-histone H3 (pHH3, Ser10), a mitotic marker, α-tubulin and DNA (DAPI), and the mitotic index and cell number were quantified. Low malignant BT474 spheroids, which were grown in a direct co-culture setup on a feeder layer of ln-aT bASCs, showed a growth advantage at 24, 48, 72 and 96 h, compared to a feeder layer formed by ln-dT bASCs (Fig. S4A). The area and diameter were significantly increased in BT474 spheroids, especially at early time points 24 and 48 h (Fig. S4A, C and E). Interestingly, BT474 spheroids co-cultured with ob-aT bASCs displayed an even greater growth advantage compared to ln/ob-dT bASCs (Fig. S4A-F). The direct comparison of lean and obese aT bASCs showed that ob-aT bASCs led to an increased growth rate of BT474 spheroids by 29.97% at 72 h and 27.83% at 96 h compared to ln-aT bASCs (Fig. S5A, black scatter plots), whereas there was no significant difference between lean and obese dT bASCs (Fig. S5A, gray scatter plots). The similar effect was observed for BT474 spheroid diameter, which increased by 18 to 21% over the 72- to 96 h period after co-culture with ob-aT bASCs, compared to ln-aT bASCs (Fig. S5B, black scatter plots). In addition, the diameter of BT474 organoids co-cultured with ob-dT bASCs increased significantly by 13%, compared with ln-dT bASCs (Fig. S5B, gray scatter plots). In line with this, spheroids co-cultured with ob-aT bASCs had a higher cell number and mitotic index than with ln-aT bASCs (Fig. S4G-J, black bars). Only moderate differences were observed on breast cancer cell spheroids co-cultured with lean or obese dT bASCs (Fig. S4G-J, gray bars).

An increased diameter was also observed in MDA-MB-361 spheroids directly co-cultured with aT bASCs (Fig. S5C-J). The growth benefit of spheroids has been further significantly enhanced by co-culture with ob-aT bASCs, compared to ln-aT bASCs (Fig. S5G and H). Like BT474 spheroids, the mitotic index and cell number were also increased in MDA-MB-361 spheroids co-cultured with ob-aT bASCs (Fig. S6A-D, black bars). A moderate promoting effect was observed in the spheroids co-cultured with lean and obese dT bASCs (Fig. S5C-J and S6A-D). These data support previous reports that visceral ASCs had a stimulatory effect on cancer cell proliferation in a direct co-culture setup [20, 48]. Collectively, aT bASCs have a highly increased capacity to fuel the growth of low to intermediate malignant breast cancer spheroids, which is further exacerbated by the obese state.

The supernatant of aT bASCs stimulates breast cancer cell motility

A number of chemokines such as CXCL1–3, CXCL10 and CCL2 released from aT-bASCs are known to promote invasion and metastasis [49]. To examine this issue, epithelial MCF7(ER+, PR+, HER2+) and mesenchymal MDA-MB-231(ER-, PR-, HER2-) breast cancer cells incubated with supernatants of bASCs were tracked using time-lapse microscopy. The accumulated distance and the velocity of single-tracked cells are commonly used to assess the migratory capacity of cancer cells [50]. Supernatants of both ln- and ob-aT bASCs significantly increased the accumulated distance (ln-aT bASCs: 606 μm; ob-aT bASCs: 615 μm) and the velocity (ln-aT bASCs: 0.90 μm/s; ob-aT bASCs: 0.92 μm/s) of MCF7 cells, compared with control MCF7 cells, which had an accumulated distance of 363 μm and a velocity of 0.54 μm/s (Fig. S6E-G, 1st, 2nd and 4th scatter plots). Consistent with the proliferation results (Fig. S4 and S5), the supernatants of ln/ob-dT bASCs displayed moderate effects on MCF-7 cells with increased accumulated distances of 490 μm and 556 μm (Fig. S6F, 3rd and 5th scatter plots) and increased velocities of 27 to 46% (Fig. S