The ferroptosis sensitivity of NSCLC cell lines affected by cell density
NSCLC are susceptible to ferroptosis induced by erastin[17, 18]. We found that the sensitivity of erastin is determined by cell density. When the NSCLC cell line A549 was grown at low cell density (< 50% fusion), the round morphological changes of the cells were observed under light microscope and were highly sensitive to ferroptosis induced by erastin. In contrast, when cells grow at high cell density (> 80% fusion), they are less sensitive to erastin (fig 1A). Next, we used cell survival tests to compare the sensitivity to erastin between low density and high density. The ferroptosis sensitivity of cell density regulation was further verified by MTT assay (Fig. 1C) and crystal violet staining (Fig. 1B). In order to exclude the possibility that the differences in cell death at high / low densities were due to the different levels of available erastin per cell, we seeded the different number of cells in the same areas to recreate low or high cell densities. Crystal violet staining and quantitative analysis showed that A549 cells plated at high density were more resistant to erastin. Similarly, MTT assay showed high sensitivity when A549 cells were cultured at low density, and high resistance when cultured at higher densities (Fig. 1C). Therefore, these data indicated that cell density affects the sensitivity of ferroptosis in NSCLC.
Hippo pathway regulators YAP and TAZ are molecular sensors that regulate the proliferation of density-dependent cancer cells. There exists a compensatory mechanism between YAP and TAZ, such that the corresponding YAP expression decreases when the expression of TAZ is high, which led to our investigation on the role of YAP and TAZ in ferroptosis. First, we detected the expression of YAP and TAZ proteins in A549 cells. From the comparison with breast cancer cell lines MDA-MB-231, we found that the main coactivator in NSCLC is TAZ protein (Fig 1 D), not YAP.
The activities of YAP/TAZ are regulated by their phosphorylation and intracellular localization. YAP and TAZ are phosphorylated, cytosolically retained, and subjected to proteasome degradation in high density. In contrast, in low density, YAP and TAZ are dephosphorylated and transferred to the nucleus, where they bind to the TEAD protein and drive the expression of genes that regulates cell proliferation, differentiation, and migration[15]. In order to compare the expression levels of TAZ in A549 at different densities, we performed western blot analysis, and the results showed that the expression level of TAZ protein decreased with increasing cell density (Fig. 1E). To confirm that activated TAZ expression also occurs in vivo, we isolated the cytoplasm and nucleus of A549 at low or high cell densities. As shown in Figure 1F, the nuclear level of TAZ increases when growing at low cell density. On the contrary, when growing at high cell density, its nuclear level decreased. Therefore, these results suggest that TAZ is the major Hippo effector in lung cancer, and its subcellular localization is regulated by cell densities. Therefore, we focused on TAZ as a major Hippo effector in regulating ferroptosis in lung cancer cells.
TAZ sensitivity regulation of ferroptosis induced by erastin
In order to investigate the role of TAZ on regulating ferroptosis, we first knocked down the TAZ gene of A549 cells (Fig 2A). We found that the resistance of A549 cells to ferroptosis which induced by erastin (Figs. 2B and 2C) was enhanced after knocking down the TAZ gene. In the low-density A549 cells, the knockdown of TAZ gene also showed ferroptosis resistance in high-density cells (Fig. 2D), suggesting that the activation of TAZ contributes to the development of density-dependent ferroptosis sensitivity. In addition, TAZ in A549 cells was knocked down by different concentrations of small interfering RNA (SiRNA), which decreased the sensitivity to erastin. When the content of TAZ in the cell was knocked down to a certain extent, the change of erastin resistance caused by the concentration of siRNA became not obvious (Figs. 2D and 2E), which indicated that TAZ was related to ferroptosis. In contrast, the expression of the constitutively active form of TAZ, TAZS89A, increased the sensitivity to erastin (Figs. 2F-2H). Taken together, these data suggest that the activation of TAZ regulates the sensitivities to ferroptosis under different cell densities.
Higher expression levels of TAZ in NSCLC Tissues
To compare the difference of TAZ expression between NSCLC tissues and adjacent tissues, we collected tumor and adjacent tissue samples from 33 NSCLC patients, including 15 patients had lung adenocarcinoma (LUAD), 17 had squamous carcinoma (LUSC) and 1 had pleomorphic carcinoma (PC). Western blot, qPCR and IHC were used to detect the difference of TAZ expression between NSCLC tissues and adjacent tissues. Western blot analysis showed that the expression level of TAZ protein in NSCLC tumor tissues was higher than that in adjacent normal tissues (Fig 3). Quantitative PCR showed that TAZ mRNA in tumor tissues was higher than that in paracancerous tissues, and there was a positive correlation between TAZ mRNA and protein expression in tumor tissues (Figs. 3A and 3B). The expression of TAZ detected by IHC also showed that the expression of TAZ protein in NSCLC tumor tissues was higher than that in adjacent normal tissues (Fig. 3C). In general, these results demonstrate that TAZ expression is elevated in NSCLC, suggesting a possibility of activating TAZ to induce ferroptosis in the treatment of NSCLC.
SnoN regulates ferroptosis sensitivity by affecting TAZ
Past studies have found that SnoN was mainly distributed in the cytoplasm in non-malignant breast tissues, but it is also expressed in the nucleus in some malignant tissues and cell lines[19]. In the process of studying the differential localization of SnoN in cells, Zhu et al[20] found that the localization of SnoN was highly sensitive to cell density: nuclear localization at low density and cytoplasmic localization at high density. This change in the localization of cell density has an interesting correlation with the expression of TAZ. This suggests that SnoN may act as an upstream gene of TAZ and regulate the expression of TAZ gene. Therefore, we used SiRNA to knock down SnoN gene, and real-time quantitative PCR observed that the expression of TAZ gene was affected (Fig 4B), confirming that SnoN acts on the upstream of TAZ.
Since SnoN regulates the expression of TAZ gene, we speculated that SonN may regulate ferroptosis sensitivity at different cell densities through TAZ. Firstly, we knocked down the SnoN gene (Fig. 4A), and MTT assay and crystal violet staining experiments showed that the resistance of A549 cells to ferroptosis was enhanced (Fig. 4C and 4F). In low-density A549 cells, the same ferroptosis resistance phenomenon was found in high-density cells after knockdown of SnoN gene (fig 2E), indicating that the activation of SnoN promotes the sensitivity of cells to density-dependent ferroptosis. Secondly, after the knockdown of SnoN in A549 cells by different concentrations of SiRNA (with matching TAZ concentration), when the TAZ content in cells is knocked down to a certain extent, the erastin resistance changes caused by different concentrations of SiRNA become insignificant (Fig. 4D). Conversely, overexpression of SnoN sensitizes A549 to erastin (Figs 3G and 3H). Together, SnoN acts on the upstream of TAZ, and knocking down SnoN may be closely related to the ferroptosis resistance caused by down-regulation of TAZ gene.