Abstract
The presence of an intact, impermeable blood-brain barrier (BBB) during the early (micrometastatic) stages of tumour development limits significantly both early detection and effective treatment of brain metastasis. Systemic administration of tumour necrosis factor (TNF) in murine models has been shown to transiently permeabilise the BBB selectively at sites of micrometastasis in the brain, facilitating both the enhanced detection of metastatic colonies employing contrast-enhanced MRI and improved delivery of therapeutically relevant molecules to the tumour site. Histological evaluation suggested that the BBB disruption was elicited primarily by the activation of one of the two endogenous receptors of TNF; TNF type 1 receptor (TNFR1), expressed on the endothelium delimiting the tumour associated vasculature [1]. Based on these findings, the aim of the current study was to use a mutein of TNF (mutTNF) with increased bioactivity at TNFR1 specifically to decipher the molecular underpinnings of TNF-induced BBB permeabilisation. Initially, cytotoxicity of mutTNF on L929s (mouse fibrosarcoma) cells, which only express TNFR1, was assessed compared to either human TNF (huTNF) or mouse TNF (mTNF). TNF-induced cell death was increased by ca. 35% at 48h post-administration of 10ng/ml of mTNF. Similarly, 10ng/ml huTNF, increased cell death by ca. 36%. In contrast, mutTNF induced minimal cytotoxicity ca. 5% at 48h, suggesting that mutTNF is a less potent ligand for mTNFR1- in comparison to both mTNF and huTNF - a specific agonist for mouse TNFR1 alone. All agents showed a dose-dependent effect on cell viability (Fig. 1A). Subsequently, the permeabilising effect of mutTNF was assessed in vitro using human brain endothelial cells (hCMEC/D3). Treatment of unstimulated hCMEC/D3 cells with pro-inflammatory cytokines (huTNF, mutTNF) resulted in increased paracellular permeability (ca. 48% and ca. 64%, respectively; Fig. 1B). Additionally, stimulation with non-cytotoxic formulations of tumour conditioned medium (TCM), derived from brain-seeking metastatic breast cancer cells, elicited activation of the endothelial monolayer and upregulation (ca. 85%) of TNFR1 mRNA expression levels. Exposure of hCMEC/D3 cells to TCM alone or in combination with huTNF/mutTNF was followed by significant increases (83%, 95% and 110%, respectively) in BBB permeability (Fig. 1C). These findings indicate that although the TNFR1-selective mutTNF did not induce L929s cell cytotoxicity, it was able to potently induce endothelial monolayer permeabilisation. These data provide preliminary evidence in support of a TNFR1-mediated permeabilising effect of TNF on the BBB.
Reference
1. Connell JJ, Chatain G, Cornelissen B, Vallis KA, Hamilton A, Seymour L, et al. J Natl Cancer Inst. 2013;105(21):1634–43.
Figure 1 (A) MTT assay following conditioning of L929s cells with the inducing agent actD (1μg/ml) for 1h and concomitant treatment with huTNF, mTNF or mutTNF for 48h. (B) Permeability of hCMEC/D3 monolayer treated with 10ng/ml of IL-1β (positive control), huTNF or mutTNF. (C) Permeability of hCMEC/D3 monolayer stimulated with TCM (1h) with/ without following addition of huTNF or mutTNF for 24h. *P < 0.05, **P < 0.01, ***P <0.001 compared with actD (A) or vehicle- unstimulated cells (B, C). Data are representative of at least four independent experiments and are given as mean + S.D.