A TLR4 Agonist Induces Osteosarcoma Regression by Inducing an Antitumor Immune Response and Reprogramming M2 Macrophages to M1 Macrophages

Abstract

Osteosarcoma (OsA) has limited treatment options and stagnant 5-year survival rates. Its immune microenvironment is characterized by a predominance of tumor-associated macrophages (TAMs), whose role in OsA progression remain unclear. Nevertheless, immunotherapies aiming to modulate macrophages activation and polarization could be of interest for OsA treatment. In this study, the antitumor effect of a liposome-encapsulated chemically detoxified lipopolysaccharide (Lipo-MP-LPS) was evaluated as a therapeutic approach for OsA. Lipo-MP-LPS is a toll-like receptor 4 (TLR4) agonist sufficiently safe and soluble to be IV administered at effective doses. Lipo-MP-LPS exhibited a significant antitumor response, with tumor regression in 50% of treated animals and delayed tumor progression in the remaining 50%. The agent inhibited tumor growth by 75%, surpassing the efficacy of other immunotherapies tested in OsA. Lipo-MP-LPS modulated OsA's immune microenvironment by favoring the transition of M2 macrophages to M1 phenotype, creating a proinflammatory milieu and facilitating T-cell recruitment and antitumor immune response. Overall, the study demonstrates the potent antitumor effect of Lipo-MP-LPS as monotherapy in an OsA immunocompetent model. Reprogramming macrophages and altering the immune microenvironment likely contribute to the observed tumor control. These findings support the concept of immunomodulatory approaches for the treatment of highly resistant tumors like OsA.

Keywords: growth inhibition; immunotherapy TLR4 agonist; macrophages reprograming; osteosarcoma.

Figure 1

Transcriptomic pro-inflammatory signature is a prognostic marker for OsA’s metastatic spreading. (A) Characteristics of OsA cohorts; (B) MCPs score according to patients 5-years metastatic status. A higher score of genes linked to cytotoxic T-cells, monocyte lineage and NK cells was found in the 5-years metastatic-free group; (C) Expression of proinflammatory cytokines according to 5-years metastatic status. The expression of IL1β, IL6 and IFNγ was higher in the 5-years metastasis-free group; (D) Expression of antigen presentation related genes and TLR genes according to 5 years metastatic status. Higher score of activation of antigen presenting pathway, TLR pathway were found in the 5-year metastasis-free group; (E) Prognostic value of TLR4 expression for OsA’s metastatic spreading. A higher expression of TLR4 was associated with a significantly better metastasis free survival (p = 0.0035). * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001.

Figure 2

LipoMP-LPS induced tumor regression in a rat OsA model. (A) Mean tumor evolution of the treatments’s groups: LipoMP-LPS slowed down tumor progression; (B) Tumor inhibition rate at day 14: LipoMP-LPS significantly inhibited tumor growth by 75% (p < 0.005); (C) Individual tumor progression according to treatment. Control group had a homogeneous tumor progression, vehicle induced 4 stabilization of tumor progression (stabilized response SR); LipoMP-LPS induced 4 SRs and 4 complete responses (CRs); (D) Histological analyses (HPS and Ki67 stainings) of tumors according to treatments. HPS staining (×10) showed histological changes caused by LipoMP-LPS in the stabilized responder (SR) tumors (decrease of proliferative area, increase of necrotic areas, fibrosis) and fibrosis induced in the complete responder (CR) tumors. Ki67 staining (×40) showed a decrease in proliferative tumor cells caused by LipoMP-LPS in the stabilized tumors (SR). ** p < 0.005 Cte: control group (PBS), SR: stabilized response, CR: complete response.

Figure 3

LipoMP-LPS antitumoral response was associated with increased CD8+ T-cells peritumoral cells. (A) Mean number CD8+ Tcells/FOV according to treatments; (B) Representative CD8+ staining fr each group of treatment. The mean density of CD8+ T-cells/FOV varies with tumors’s response to treatments (SR: stabilized response; CR: complete response). Magnification ×200. The induction of tumor response (i.e., tumor growth arrest (SR)) is associated with an increase in CD8+ T-cells, while the density of CD8+ T-cells in tumors that regressed (CR) is similar to the one of the control tumors. * p < 0.05; ** p < 0.005. FOV: field of view.

Figure 4

Lipo-MP-LPS modify TAMs infiltrate in OsA’s immune environment. (A) Mean density of CD68+, CD163+ and TLR4+ macrophages/FOV according to treatments. Total CD68+ macrophages were significantly increased in LipoMP-LPS High, but not in LipoMP-LPS Low group (both in CR and SR; n = 7) compared with control group (saline, n = 7) while immunosuppressive CD163+ macrophages were only slightly increased. Density of TLR4+ macrophages was also significantly increased in LipoMP-LPS High groups. ** p < 0.005; (B) Representative staining of CD68 total macrophages, CD163 M2 macrophages and TLR4+ macrophages. Magnification ×200.

Figure 5

LipoMP-LPS modifies immune cell populations and cytokine environment in tumors at the onset of tumor response. (A) Flow cytometry analyses of immune cell populations in the tumors of animal treated with high or low doses of LipoMP-LPS or PBS as control. Data are presented as mean ratio ± SEM of positive cells in each group. ns: not significant * p < 0.05, ** p < 0.005. High dose of LipoMP-LPS increases tumor infiltration by CD4 T-cells and increases the levels of M1 macrophages but not total macrophages into the tumors; (B) Cytokine expression profile evaluated by LegendplexTM in the same tumors. Data are mean ± SEM of the concentration (pg/mL) of each cytokine. High and low doses of LipoMP-LPS induce a moderate proinflammatory environment in tumors, characterized by a slight increase in levels of IL-6, IL-1b, IL-1a, IL-10 and GM-CSF and of the chemokines CCL2 and CXCL1 in tumors compared with control groups. ns: not significant * p < 0.05.

Figure 6

LipoMP-LPS induced systemic adaptive immune responses. (A) Flow cytometry analysis of the immune cell populations in the spleen of OsA tumor bearing rats injected with high- or low-dose LipoMP-LPS, or with PBS as controls. For each group, the modification in cells density is expressed as fold change in comparison to control group (set up arbitrary at 1). Data are presented as mean ± SEM ratio of mean percentages of positive cells among CD45+ cells, normalized to mean spleen weights. ** p < 0.005; *** p < 0.001; (B) Mean spleen weights in groups. Data are presented as mean ratio ± SEM of spleen weights. *** p < 0.001; (C) Analysis of IFN-γ secretion by total splenocytes in each groups using ELISpot assays. Data are mean ± SEM of IFN-γ spots (4 × 10⁴ cells/wells in triplicate). IFN-g production (i.e., number of spots per 40,000 cell/well) was increased in LipoMP-LPS treated group compared with control; (D) Plasmatic levels of IFN-γ in each group. Data are mean ± SEM of plasma concentrations of IFN-γ evaluated for 4 animals/treatment group. * p < 0.05.

Figure 7

Phenotypes of bone marrow derived macrophages incubated with Lipo-MP-LPS. Rat bone marrow-derived macrophages polarized toward an M1 or M2 phenotype were incubated with LipoMP-LPS, or with PBS or Lipo as controls. (A) Flow cytometry analysis of the expression of the surface markers CD86 and CD163 on M1 and M2 polarized macrophages after 24 h of stimulation. Data are presented as mean fold change in MFI/Control ± SD; * p < 0.05; (B) Cytokines profile of stimulated M1- and M2-polarized macrophages evaluated by LEGENDplex multiplex bead assays. Data are presented as mean concentrations (pg/mL) of each cytokine ± SD. * p < 0.05; (C) Phagocytosis function of stimulated M1- and M2-polarized macrophages. Data are presented as mean fold change in % phagocytosis/Control ± SD.