Impact of mesenchymal stem cells' secretome on glioblastoma pathophysiology

Abstract

Background: Glioblastoma (GBM) is a highly aggressive primary brain cancer, for which curative therapies are not available. An emerging therapeutic approach suggested to have potential to target malignant gliomas has been based on the use of multipotent mesenchymal stem cells (MSCs), either unmodified or engineered to deliver anticancer therapeutic agents, as these cells present an intrinsic capacity to migrate towards malignant tumors. Nevertheless, it is still controversial whether this innate tropism of MSCs towards the tumor area is associated with cancer promotion or suppression. Considering that one of the major mechanisms by which MSCs interact with and modulate tumor cells is via secreted factors, we studied how the secretome of MSCs modulates critical hallmark features of GBM cells.

Methods: The effect of conditioned media (CM) from human umbilical cord perivascular cells (HUCPVCs, a MSC population present in the Wharton's jelly of the umbilical cord) on GBM cell viability, migration, proliferation and sensitivity to temozolomide treatment of U251 and SNB-19 GBM cells was evaluated. The in vivo chicken chorioallantoic membrane (CAM) assay was used to evaluate the effect of HUCPVCs CM on tumor growth and angiogenesis. The secretome of HUCPVCs was characterized by proteomic analyses.

Results: We found that both tested GBM cell lines exposed to HUCPVCs CM presented significantly higher cellular viability, proliferation and migration. In contrast, resistance of GBM cells to temozolomide chemotherapy was not significantly affected by HUCPVCs CM. In the in vivo CAM assay, CM from HUCPVCs promoted U251 and SNB-19 tumor cells growth. Proteomic analysis to characterize the secretome of HUCPVCs identified several proteins involved in promotion of cell survival, proliferation and migration, revealing novel putative molecular mediators for the effects observed in GBM cells exposed to HUCPVCs CM.

Conclusions: These findings provide novel insights to better understand the interplay between GBM cells and MSCs, raising awareness to potential safety issues regarding the use of MSCs as stem-cell based therapies for GBM.

Keywords: Conditioned media; Glioblastoma; Human umbilical cord perivascular cells; Mesenchymal stem cells; Migration; Proliferation; Proteomics; Secretome; Viability.

Fig. 1

Effect of HUCPVCs conditioned media (CM) on GBM cell viability. Cell viability was measured by MTT (a, b) and trypan blue (c, d) assays on U251 (a, c) and SNB-19 (b, d) GBM cell lines, after exposure to control media or HUCPVCs CM. HUCPVCs CM led to a statistically significant increase in viability of GBM cells in both assays, in all tested time points. All experiments were done in triplicate, at least in 3 independent experiments. Data is presented as the mean ± SD (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 and ****p ≤ 0.0001)

Fig. 2

Effect of HUCPVCs conditioned media (CM) on GBM cell migration. a, c Representative pictures showing the migratory capacity of U251 (a) and SNB-19 (c) GBM cells exposed to control media or HUCPVCs CM. b, c Quantification of U251 (b) and SNB-19 (d) cell migration presented as % of wound closure. Treatment with HUCPVCs CM led to a statistically significant increase of GBM cell migration. Data is presented as the mean ± SD of at least 3 independent experiments, each in triplicate (*p ≤ 0.05 and **p ≤ 0.01)

Fig. 3

Effect of HUCPVCs conditioned media (CM) on GBM cell proliferation and sensitivity to temozolomide treatment. a, c Proliferation capacity of U251 (a) and SNB-19 (c) GBM cells was determined by BrdU assay after treatment with control media or HUCPVCs CM. Exposure to HUCPVCs CM increased the proliferation rate of both GBM cell lines (**p ≤ 0.01). b, d Determination of the half inhibitory concentration (IC₅₀) values of temozolomide (TMZ) treatment in U251 (b) and SNB-19 (d) cell lines. For both cell lines, no statistically significant differences in the TMZ IC₅₀ values were found between cells treated with HUCPVCs CM or control media (p = 0.6738 for U251, and p = 0.3115 in SNB-19). Results are expressed as the mean ± SD of 3 independent experiments, each in triplicate

Fig. 4

Effect of HUCPVCs conditioned media (CM) in GBM cells growth and angiogenesis, in vivo CAM model. Representative pictures of CAM assay after 8 days of tumor growth in ovo (a, b) and ex ovo (e, f) (× 16 magnification). c, d Tumor growth was measured in ovo. A higher tumor area was found in tumors originated from both U251 (c) and SNB-19 (d) cells exposed to HUCPVCs CM (p = 0.0260 for U251 and p = 0.0290 for SNB-19). g, h Number of blood vessels surrounding tumors derived from U251 (g) and SNB-19 (h) cells. Both U251 and SNB-19 tumors exposed to HUCPVCs CM presented an increase in the number of blood vessels comparing with control conditions, however only CM-exposed SNB-19 GBM tumors reached statistical significance (p = 0.1416 and p = 0.0069, respectively). Results are expressed as the mean ± SD (*p ≤ 0.05; **p ≤ 0.01). Dashed circle, tumor; Arrowheads, intra-tumoral blood vessel; Arrow, extra-tumoral blood vessel

Fig. 5

Functional analysis of proteins present in HUCPVCs conditioned media (CM). a–c DAVID was used to query the functional annotation of HUCPVCs secretome. The top 20 statistically significant enriched Gene Ontology (GO) terms in Biological Process (a), Molecular Component (b) and Molecular Function (c) are shown. d,  e All statistically significant enriched Reactome (d) and KEGG (e) pathways are represented. The -log values of p values are displayed