MISP Suppresses Ferroptosis via MST1/2 Kinases to Facilitate YAP Activation in Non-Small Cell Lung Cancer

Abstract

Despite advances in non-small cell lung cancer (NSCLC) therapies, resistance remains a major challenge. Ferroptosis, a form of regulated cell death, plays a key role in cancer progression and treatment response. However, the mechanisms governing ferroptosis in NSCLC are not fully understood. The Hippo pathway, which regulates cell proliferation, has recently been implicated in ferroptosis regulation. In this study, we identify Mitotic Spindle Positioning (MISP) as a critical inhibitor of ferroptosis in NSCLC. MISP is upregulated in NSCLC tissues, and its loss sensitizes cells to ferroptosis, reducing cell proliferation in vitro and in vivo. Mechanistically, MISP binds to the SARAH domain of MST1/2 kinases, inhibiting their homodimerization and autophosphorylation, leading to sustained activation of YAP, a transcriptional coactivator in the Hippo pathway. YAP activation increases SLC7A11 expression, which protects cells from ferroptosis. We also identify a mutant MISP-R390/391A that disrupts MISP-MST1/2 binding, further illustrating the MST1/2-dependent inhibition of Hippo signaling. Notably, MISP is a target of YAP, creating a feedback loop that amplifies YAP signaling. Our findings suggest a novel MISP-YAP axis regulating ferroptosis, positioning MISP as a potential therapeutic target for NSCLC, especially in cases with dysregulated YAP.

Keywords: MISP; MST kinases; SLC7A11; YAP; ferroptosis.

Figure 1

MISP is upregulated in NSCLC. A) Scatter plot showing MISP expression in NSCLC and adjacent tissues from the TCGA dataset. The figure was generated using GEPIA (http://gepia.cancer‐pku.cn/). n = 347 (adjacent) and n = 483 (NSCLC) in TCGA‐LUAD. B) Survival analysis of NSCLC patients from TCGA dataset stratified by top 25% quarter of low and high MISP expression. The figure was generated using GEPIA (http://gepia.cancer‐pku.cn/). C) Disease‐free survival analysis in NSCLC patients from the TCGA dataset stratified by top 25% quarter of low and high MISP expression. The figure was generated using GEPIA (http://gepia.cancer‐pku.cn/). D) Quantitative PCR analysis of MISP expression in NSCLC and adjacent tissues. n = 40 (adjacent) and n = 40 (NSCLC). E) Immunoblot analysis of MISP expression in NSCLC and adjacent tissues (cases 1 to 20). n = 20 (adjacent) and n = 20 (NSCLC). F) Protein quantification of MISP levels in E). Statistical significance was determined by a paired two‐tailed Student's t‐test as indicated. G) Kaplan‐Meier plots showing the overall survival of NSCLC patients with high or low MISP expression analyzed by qPCR. Data are presented as mean ± SEM. Unpaired t‐test was used in A and D to determine statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 2

MISP drives resistance to ferroptosis in NSCLC cells. A) Determination of cell proliferation in H1395 and HCC827 cells upon MISP overexpression using the cell counting kit 8 (CCK8). B) Determination of cell proliferation in control and MISP‐depleted H1395 and PC‐9 cells by CCK8. C,D) Demonstration and quantification of colony formation abilities in indicated cells upon MISP overexpression C) and MISP ablation D). Scale bar, 10 mm. E) Gross image of xenografts from nude mice implanted with wild‐type (WT) and MISP‐deficient H1395 cells. Scale bar, 10 mm. F) Analysis of tumor weight from nude mice implanted with wild‐type (WT) and MISP‐deficient H1395 cells. G) Representative images of H&E and IHC staining of Ki67 in indicated tumor sections. Scale bar, 50 µm. H) Trypan blue staining and quantification of cell death in WT and MISP‐depleted cells treated with DMSO, Lip‐1 (500 nm), Fer‐1 (10 µM), Z‐VAD‐FMK (10 µM), Nec‐1 (10 µM) and 3‐MA (1 mM) for 24 h. I–K) Relative levels of MDA I), 4‐HNE J), and ROS K) in WT and MISP‐depleted H1395 cells. L) Determination of GSH and GSSG levels in indicated cells. M) Trypan blue staining and quantification of cell death in vector and MISP‐overexpressing cells treated with Erastin (20 µM) for 24 h. N–P) Relative levels of MDA N), 4‐HNE O), and ROS P) in vector and MISP‐overexpressing cells treated with Erastin (20 µM) for 24 h. Q) Determination of GSH and GSSG levels in vector and MISP‐overexpressing cells treated with Erastin (20 µM) for 24 h. Data are presented as mean ± SEM. One‐way ANOVA was used in A,B, D, and H–L to determine statistical significance. Unpaired t‐test was used in C, F, M–Q to determine statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3

MISP stimulates SLC7A11 expression via Hippo signaling. A) Immunofluorescent staining showing the subcellular localization of YAP in WT and MISP‐deficient H1395 cells. Nuclei were stained with DAPI. Scale bar, 20 µm. B) Cytoplasmic and nuclear fractions derived from WT and MISP‐deficient H1395 cells were subjected to immunoblotting as indicated. α‐Tubulin and Lamin A/C were used as cytoplasmic and nuclear markers, respectively. C) qPCR evaluation of the transcriptional levels of SLC7A11 and YAP target genes in WT and MISP‐deficient H1395 cells. D) Immunoblot analysis of SLC7A11, YAP, YAP phosphorylation, and its target expression in WT and MISP‐deficient H1395 cells. E) Examination of TEAD luciferase reporter activity in 293T cells upon MISP knockdown. F) Association of YAP with 14‐3‐3 by immunoblot analysis. 293T cells were co‐transfected with Myc‐14‐3‐3 and vector or Flag‐YAP in the presence of control or small interfering RNA targeting MISP (siMISP). Co‐immunoprecipitates were detected using the indicated antibodies. G) Immunofluorescent staining showing the subcellular localization of YAP in vector and MISP‐overexpressing H1395 cells. Nuclei were stained with DAPI. Scale bar, 20 µm. H) Cytoplasmic and nuclear fractions derived from vector and MISP‐overexpressing H1395 cells were subjected to immunoblotting as indicated. I) qPCR evaluation of the transcriptional levels of SLC7A11 and YAP target genes in vector and MISP‐overexpressing H1395 cells. J) Immunoblot analysis of SLC7A11, YAP, YAP phosphorylation and its targets expression in vector and MISP‐overexpressing H1395 cells. K) Examination of TEAD luciferase reporter activity in 293T cells upon MISP overexpression. L) Association of YAP with 14‐3‐3 by immunoblot analysis. 293T cells were co‐transfected with Myc‐14‐3‐3 and vector or Flag‐YAP in the presence of MISP. Co‐immunoprecipitates were detected using the indicated antibodies. M) Correlation analysis between MISP and the levels of CYR61 and CTGF in NSCLC patients (n = 40 (adjacent) and n = 40 (NSCLC)). Correlation was determined by the Pearson correlation test. Data are presented as mean ± SEM. One‐way ANOVA was used to compare differences between groups in C and E. Unpaired t‐test was used to determine statistical significance in I and K. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 4

MISP is a target gene of YAP/TEAD signaling. A) Examination of MISP and YAP target genes in H1395 cells with or without YAP overexpression. B) Immunoblot analysis of MISP and CYR61 expression in vector and YAP‐overexpressing H1395 cells. C) Determination of MISP and YAP target genes in H1395 cells upon YAP/TAZ or TEAD4 depletion. D) Immunoblot analysis of MISP and CYR61 expression in H1395 cells upon YAP/TAZ or TEAD4 depletion. E,F) Determination of the mRNA E) and protein F) levels of MISP and YAP target genes in H1395 cells upon serum starvation or FBS recovery after serum starvation. H1395 cells were cultured in serum‐free medium (SFM) for 12 h and treated with FBS‐containing medium for 1 h. G,H) Determination of MISP and YAP target genes in H1395 cells at low or high cell density. H1395 cells were plated at different cell densities for 12 h and subjected to qPCR G) and immunoblotting H) analysis. I) Schematic diagram of the MISP promoter with conserved TEAD‐binding motifs (termed as R1, R2, and R3). TSS, transcription starting site. J) ChIP‐qPCR analysis of TEAD4 enrichment at the MISP and CTGF promoters. K) Analysis of luciferase reporter activity driven by the MISP promoter or R2‐defective MISP promoter in HEK293T cells transfected with the indicated plasmids. L) Analysis of luciferase reporter activity driven by the MISP promoter or R2‐defective MISP promoter in HEK293T cells transfected with the indicated siRNAs. M,N) Relative levels of MDA M) and ROS N) in H1395 cells upon YAP/TAZ or TEAD4 depletion. Data are presented as mean ± SEM. Unpaired t‐test was used to determine statistical significance in A, E, G, and J. One‐way ANOVA was used to compare differences between groups in C and K–N. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 5

MISP interacts with MST1/2 kinases. A) Binding of MISP to MST1/2 kinases. PC‐9 cells were lysed and subjected to immunoprecipitation with IgG control or MISP antibody. Immunoprecipitates were detected using indicated antibodies. The IgG heavy chain is denoted by an asterisk. B) Interaction between MISP and MST1/2. Immunoprecipitation was performed with IgG control, MST1, or MST2 antibodies, and the interaction between MISP and MST1/2 was detected. C) MISP binds to MST1 in 293T cells. 293T cells were transfected with MISP‐Flag and HA‐MST1 plasmids alone or in combination for 24 h, lysed, and subjected to immunoprecipitation using the indicated antibodies. D) Immunofluorescent staining of MST1 and MISP using the indicated antibodies in H1395 cells. The boxed inset within the left‐hand panels highlights the specific area depicted in the enlarged image (right panel). Scale bar, 40 µm. E) The aa352‐524 region of MISP is required for MISP/MST1 interaction. Diagram of the domains and structural motifs of MISP (upper). Immunoblotting analysis of MISP fragments using the indicated antibody after Co‐IP with anti‐HA tag antibody (lower). F) The SARAH domain is required for MISP/MST1 interaction. Diagram of the domains and structural motifs of MST1 (upper). Immunoblot analysis of MST1 fragments using the indicated antibody after Co‐IP with anti‐Flag tag antibody (lower). G) Coomassie brilliant blue staining after GST or GST‐MISP pulldown. His‐MST1 (aa301‐487) or His‐MST2 (aa279‐491) recombinant proteins were incubated with GST or GST‐MISP (aa352‐524), and detected by Coomassie brilliant blue staining. Ten percent of each recombinant protein was used as input.

Figure 6

R390 and R391 of MISP are essential for interaction with MST1/2 kinases. A) Identification of the aa382‐396 region required for MISP/MST1 interaction. Indicated deletion mutants were co‐transfected with MST1 and subjected to immunoprecipitation. B) Determination of the binding affinity between WT MISP and the R2A mutant (R390/391A) for MST1. C) Immunoblot analysis of YAP phosphorylation, YAP, and CYR61 expression in H1395 cells in the presence of vector, WT MISP, or R2A mutant. D) Determination of YAP target genes expression in H1395 cells in the presence of vector, WT MISP, or R2A mutant. E) Examination of TEAD luciferase reporter activity in 293T cells upon forced expression of vector, WT MISP, or R2A mutant. F) Effects of WT MISP and R2A mutant on the association of YAP with 14‐3‐3 by Co‐IP and immunoblot analysis. G) Determination of MST1 homodimerization by Co‐IP and immunoblot analysis in 293T cells upon forced expression of vector, or WT MISP. H) Determination of MST1/2 heterodimerization by Co‐IP and immunoblot analysis in 293T cells upon forced expression of vector, or WT MISP. I) Immunoblot analysis of MST1 phosphorylation in H1395 cells upon MISP expression. J) Immunoblot analysis of MST1 phosphorylation in WT and MISP‐deficient H1395 cells. (K‐L) Effects of WT MISP and R2A mutant on MST1 homodimerization K) and heterodimerization L) by Co‐IP and immunoblot analysis in 293T cells. M) Immunoblot analysis of MST1 phosphorylation in H1395 cells upon WT MISP and R2A mutant expression. N,O) Relative levels of MDA N) and ROS O) in H1395 cells upon WT MISP and R2A mutant expression treated with Erastin (20 µM) for 24 h. Data are presented as mean ± SEM. One‐way ANOVA was used to compare differences between groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 7

MST1/2 kinases are required for MISP to trigger YAP activation. A) Immunoblot analysis of YAP phosphorylation, YAP, and CYR61 expression in H1395 cells upon transfection with the indicated siRNA. B) Determination of YAP target genes by qPCR in H1395 cells upon transfection with the indicated siRNA. C) Evaluation of TEAD luciferase reporter activity in 293T cells upon transfection with the indicated siRNA. D) Determination of cell death by trypan blue staining in cells with indicated treatment. E–H) Relative levels of 4‐HNE E), MDA F), ROS G), and GSH/GSSG H) in H1395 cells upon transfection with the indicated siRNA. I) Immunoblot analysis of YAP phosphorylation, YAP, and CYR61 expression in H1395 cells upon transfection with the indicated siRNA with or without XMU‐MP‐1 treatment (2.5 µM for 12 h). J) Determination of YAP target genes by qPCR in H1395 cells following transfection with the indicated siRNA with or without XMU‐MP‐1 treatment (2.5 µM for 12 h). K) Determination of TEAD luciferase reporter activity in 293T cells upon transfection with the indicated siRNA with or without XMU‐MP‐1 treatment (2.5 µM for 12 h). L) Determination of cell death by trypan blue staining in H1395 cells with indicated treatment. M–P) Relative levels of 4‐HNE M), MDA N), ROS O), and GSH/GSSG P) in H1395 cells upon transfection with the indicated siRNA with or without XMU‐MP‐1 treatment (2.5 µM for 12 h). Data are presented as mean ± SEM. One‐way ANOVA was used to compare differences between groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 8

MISP promotes tumor growth in vivo and correlates with YAP and SLC7A11 in NSCLC samples. A) Gross images of xenografts from the indicated mice with or without XMU‐MP‐1 treatment (5 mg kg⁻¹). Scale bar, 10 mm. B) Analysis of tumor weight from the indicated groups. C) Kaplan‐Meier plots showing the overall survival of the indicated nude mice with or without XMU‐MP‐1 treatment (5 mg kg⁻¹). D) Representative images of H&E and IHC staining of Ki67 in the indicated tumor sections. Scale bar, 50 µm. E) Representative images of IHC staining of MISP, YAP, SLC7A11, and MDA in NSCLC specimens. Scale bars, 150 µm. F,G) Correlation analysis between MISP and the levels of YAP (F) and SLC7A11 G) in NSCLC samples. Samples with grades 0 (no staining) and 1 (weak staining) were grouped as “low” expressions and samples with grades 2 (strong staining) were grouped as “high” expressions. H) Proposed working model describing the MISP‐MST1/2‐SLC7A11 signaling axis dictating ferroptosis and lung cancer growth. Data are presented as mean ± SEM. One‐way ANOVA was used to compare differences between groups. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.