Vertebral-specific activation of the CX3CL1/ICAM-1 signaling network mediates non-small-cell lung cancer spinal metastasis by engaging tumor cell-vertebral bone marrow endothelial cell interactions

Rationale: The spine is one of the most common metastatic sites of non-small cell lung cancer (NSCLC), and NSCLC spinal metastasis results in serious consequences. Metastatic extravasation of disseminated cancer cells including increased invasiveness, adhesion and transendothelial migration is crucial for tumor metastasis. This study aimed to investigate the mechanisms underlying NSCLC spinal metastasis based on the C-X3-C motif chemokine ligand 1- (CX3CL1) and intercellular adhesion molecule-1- (ICAM-1) mediated signaling network. Methods: Immunohistochemistry, western blotting, and reverse transcription-quantitative PCR were conducted to detect the distribution of CX3CL1/ICAM-1 in different organs. Transwell, adhesion, and transendothelial migration assays were performed to evaluate the regulatory effects of CX3CL1/ICAM-1 on NSCLC cell invasion, adhesion, and transendothelial migration in vitro. A spontaneous spinal metastasis mouse model was established via injection of NSCLC cells into the left cardiac ventricle of NOD/SCID mice. The effects of CX3CL1/ICAM-1 on NSCLC spinal metastasis in vivo were validated using bioluminescent, micro-computerized tomography, immunohistochemistry and histological analyses. Results: CX3CL1 expression was specifically higher in vertebral bone compared with limb bones and lung tissue, and was associated with NSCLC spinal metastasis. Mechanically, vertebral bone marrow endothelial cells (VBMECs) enhanced NSCLC cell invasion via CX3CL1 signaling-mediated activation of the PI3K/AKT pathway. Furthermore, we found that VBMECs effectively induced ICAM-1-dependent NSCLC cell adhesion in coordination with platelets through the CX3CL1/ICAM-1/LFA-1 pathway. Meanwhile, CX3CL1 enhanced NSCLC cell transendothelial migration by increasing permeability of VBMECs via ICAM-1-dependent activation of the Src/GEF-H1 pathway. Interestingly, NSCLC cells were indicated to promote CX3CL1 secretion of VBMECs through MAPK14/ADMA17-dependent CX3CL1 release and NF-κB-dependent CX3CL1 synthesis. Based on these findings, we revealed a novel feedback cycle between circulating NSCLC cells and VBMECs mediated by CX3CL1/ICAM-1 signaling. Further disengagement of the CX3CL1/ICAM-1-mediated feedback cycle in vivo significantly restricted metastasis and prolonged mouse survival. Conclusions: Our results indicated a unique feedback cycle between circulating NSCLC cells and VBMECs mediated by CX3CL1/ICAM-1 signaling, which is necessary for NSCLC spinal metastasis. This work provides a new perspective for underlying the mechanisms of NSCLC spinal metastasis and indicates potential novel targets for the prevention of NSCLC spinal metastasis.

The effect of ultrasound pulse length on microbubble cavitation induced antibody accumulation and distribution in a mouse model of breast cancer

In solid tumors, the limited diffusion of therapeutic molecules in the perivascular space is a known limitation impacting treatment efficacy. Ultrasound Targeted Microbubble Cavitation (UTMC) has been shown to increase vascular permeability and improve the delivery of therapeutic compounds including small molecules, antibodies (mAb), nanoparticles and even cells, notably across the blood-brain-barrier (BBB). In this study, we hypothesized that UTMC could improve the accumulation and biodistribution of mAb targeting the adenosinergic pathway (i.e. CD73) in mice bearing bilateral subcutaneous 4T1 mammary carcinoma. METHODS: A bolus of fluorescently labeled mAb was given intravenously, followed by a slow infusion of microbubbles. UTMC therapy (1 MHz, 850 kPa) was given under ultrasound image guidance for 5 minutes to the right side tumor only, using three different pulse lengths with identical ultrasound energy (5000cyc "long", 125x40cyc "mid" and 500x10cyc "short"), and leaving the left tumor as a paired control. Longitudinal accumulation at 0 h, 4 h and 24 h was measured using whole-body biofluorescence and confocal microscopy. RESULTS: Our data support an increase in antibody accumulation and extravasation (# extravasated vessels and extravasated signal intensity) at 0 h for all pulses and at 4 h for the mid and short pulses when compared to the control non treated side. However, this difference was not found at 24 h post UTMC, indicative of the transient nature of UTMC. Interestingly, confocal data supported that the highest extravasation range was obtained at 0 h with the long pulse and that the short pulse caused no increase in the extravasation range. Overall, the mid pulse was the only pulse to increase all our metrics (biofluorescence, fraction of extravasated vessels, amount of extravasated Ab, and extravasation range) at 0 h and 4 h time points. CONCLUSIONS: Our results support that UTMC can enhance antibody accumulation in solid tumors at the macroscopic and microscopic levels. This preferential accumulation was evident at early time points (0 h and 4 h) but had started to fade by 24 h, a time dependence that is consistent with the ultrasound blood brain barrier opening literature. Further development and optimization of this theranostic platform, such as repeated UTMC, could help improve antibody based therapies against solid cancer.