br inoculated with AKR B knockdown PC
inoculated with AKR1B10–knockdown PC9-BrM3 cells had a signif-icantly longer survival time compared to the control mice. While 60% of these mice survived >60 days after inoculation, all of the
control mice died after 50 days (Fig. 6D). Taken together, both our in vivo and in vitro experimental evidence supports that AKR1B10 knockdown markedly suppresses BM of lung cancer cells.
3.7. Down-regulation of MMP-2 and MMP-9 expression by AKR1B10 silencing is associated with deactivation of the MEK/ERK signaling pathway
Extracellular regulated kinase (ERK) signaling in cancer cells is a critical mediator of trans-endothelial migration . To further determine the potential mechanisms underlying the regulation of MMP-2 and MMP-9 expression by AKR1B10, we measured ERK activity in the PC9-BrM3 cells transfected with AKR1B10-targeted shRNA or treated with the ERK1/2 inhibitor SCH772984. Western blot analysis demonstrated that both silencing of AKR1B10 and pharmacological inhibition of ERK signaling significantly reduced MEK1/2 and ERK1/2 phosphorylation in the PC9-BrM3 cells (Fig. 7A). It is worth noting that SCH772984 treatment also down-regulated the expression of MMP-2 and MMP-9 in AKR1B10
control PC9-BrM3 cells. These data suggest that AKR1B10 regulates MMP-2 and MMP-9 protein expression in metastatic lung cancer cells, possibly through the MEK/ERK signaling pathway.
BM is the main cause of NSCLC-related mortality. It is a complex cascade of biological processes, which generally includes three key events: i) ‘‘primary tumor”: lung tumors develop in situ via epithelial-mesenchymal transition (EMT)  and invade the pul-monary vessels; ii) ‘‘circulating tumor cells (CTC)”: tumor cells sur-vive in the circulation and home to the Heparin ; iii) ‘‘brain metastasis” : CTCs adhere to and extravasate the BBB [8,50] to the parenchyma of brain and grow into a metastatic tumor mass
Fig. 7. AKR1B10 regulates MMP-2 and MMP-9 expression in a MEK/ERK signaling involved mechanism. (A) Representative western blot images showing the expression levels of proteins in the MEK/ERK signaling pathway (total MEK/p-MEK, total ERK/p-ERK) and the corresponding expression of MMP-9 and MMP-2. Ctrl, control PC9-BrM3 cells; NC, PC9-BrM3 cells transfected with negative control siRNA oligo; SiR-1, PC9-BrM3 cells transfected with AR1B10 siRNA-1 oligo; SiR-2, PC9-BrM3 cells transfected with AR1B10 siRNA-2 oligo; SCH772984, PC9-BrM3 cells treated with 1 lM ERK inhibitor SCH772984 for 48 h. **p < 0.01; ***p < 0.001. (B) Proposed working model of AKR1B10 in lung cancer brain metastasis. In metastatic lung cancer cells, elevated expression of AKR1B10 activates the MEK/ERK signaling cascade, and leads to increased expression of MMP-9 and MMP-2, which helps to break down the BBB by degrading the TJs to promote metastatic cells extravasation through the BBB.
by mesenchymal-epithelial transition (MET) . Among these steps, it is particularly unique that tumor cells penetrate the BBB.
To study BM pathology, especially the mechanisms underlying tumor cell extravasation through the BBB, we built a multi-organ microfluidic chip, which consists of two organ units - an upstream ‘‘lung” and a downstream ‘‘brain” - to mimic lung cancer BM. This chip platform was based on the upstream primary lung cancer bio-mimetic chip model established in previous work , while the downstream metastatic organ bionic chip ‘‘brain” was character-ized by its ‘‘BBB” key structure. We confirmed that the ‘‘BBB” we constructed mimics the physiological microenvironments for both structure integrity and barrier function. Using this chip platform, the whole BM process of lung cancer cells, from primary lung can-cer development in the upstream unit to the penetration of the BBB and colonization in the downstream unit, can be visually probed in real-time, which cannot be achieved by the Transwell system or in animal models. In addition, AKR1B10 expression was up-regulated in downstream brain metastases on the chip, compared with the upstream primary tumor mass, which is consistent with the up-regulation of AKR1B10 expression in PC9-BrM cells isolated from the brain metastases in animals. These results suggest that our chip recapitulate the in vivo changes of AKR1B10 expression, and data generated from the chips could be well correlated with that from in vivo animal studies. To further confirm the applicability of our chip platform, we subsequently investigated BM of lung cancer cells with differed metastatic abilities and explore the role of AKR1B10 in BM, and found that the chip platform was effective. Consistent results were obtained using both in vivo and in vitro models, suggesting the chip platform is an alternative model to further study the pathogenesis of cancer cell BM.