RKIP overexpression reduces lung adenocarcinoma aggressiveness and sensitizes cells to EGFR-targeted therapies

Abstract

Lung adenocarcinoma (LUAD), the most common subtype of non-small-cell lung cancer (NSCLC), is often driven by mutations, particularly in epidermal growth factor receptor (EGFR), that guide targeted therapy choices. However, resistance to these treatments remains a major clinical challenge. Raf kinase inhibitory protein (RKIP), encoded by the PEBP1 gene, a metastasis suppressor, modulates key oncogenic pathways and may influence tumor aggressiveness and therapy response. Yet, its specific role in NSCLC remains unclear. This study investigates the influence of RKIP expression on NSCLC aggressiveness and explores its impact on therapy response, particularly to EGFR-targeted therapies. In silico analyses revealed that lower RKIP mRNA expression correlates with poorer survival outcomes in LUAD patients but not in other NSCLC subtypes. Genetic modulation of RKIP expression in LUAD cell lines demonstrated that its overexpression reduced migration, spheroid integrity, and suppressed tumor growth, whereas RKIP knockout had opposite effects, particularly in vivo. Expression profiling showed that RKIP overexpression impacts the activation of mitogen-activated protein kinase (MAPK), RAC serine/threonine-protein kinase (AKT), and signal transducer and activator of transcription 3 (STAT3) pathways, as well as processes related to extracellular matrix regulation and inflammatory responses. Importantly, in vitro and in vivo experiments demonstrated that RKIP overexpression sensitizes cells to anti-EGFR treatments, whereas RKIP knockout diminished their sensitivity. Overall, our findings indicate that RKIP modulates LUAD progression and response to EGFR-targeted therapies, although its clinical value as a biomarker requires further validation. These findings highlight RKIP's potential in overcoming therapeutic resistance and the need for further investigation into its regulatory mechanisms.

Keywords: EGFR‐targeted therapy; Raf kinase inhibitory protein; lung adenocarcinoma; non‐small‐cell lung cancer; therapeutic resistance.

Fig. 1

Correlation between PEBP1 (RKIP) mRNA expression levels in normal and NSCLC tissues and its clinical relevance. (A) Comparative analysis of the PEBP1 mRNA expression levels between LUAD tissues (517 samples) or LUSC tissues (501 samples) and normal lung tissues (59 or 51 samples), upper and lower panel respectively. Data were plotted on the LCE online platform from the TCGA_LUAD_2016 or TCGA_LUSC_2016 database. For the comparative analysis, the t‐test was used, and P < 0.05 was considered significant. Boxplots show the median and interquartile range (IQR); whiskers extend to data within 1.5 × IQR, and outliers are shown as individual points. (B) Kaplan–Meier distribution of LUAD (upper graph) and LUSC (lower graph) patient's overall survival (OS) in months, performed in cBioPortal. Patients were divided into two groups considering PEBP1 expression levels: High (PEBP1: EXP > 0 (blue and green)) and Low (PEBP1: EXP < 0 (purple and yellow)). In LUAD's analysis, 258 samples and 243 were considered Low and High PEBP1, respectively. In LUSC's analysis, 220 samples and 258 were deemed Low and High PEBP1, respectively. For survival analysis, the Logrank test was performed, and P < 0.05 was considered significant. LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; NSCLC, non‐small‐cell lung cancer; PEBP1, gene that encodes RKIP.

Fig. 2

Evaluation of RKIP expression in NSCLC cell line, modulation of RKIP expression, and in vitro characterization of HCC827 RKIP overexpressing cell line. (A) Western blot analysis is used to assess the expression levels of the proteins EGFR, AKT, ERK, and RKIP and their respective phosphorylated forms in lung cancer cell lines, with and without EGF stimulation. This experiment was performed three times. (B) Immunofluorescence analysis of RKIP basal expression levels in lung cancer cell lines. These are representative pictures of two independent assays. Pictures were taken at 200× magnification. Scale bar: 50 μm. (C) Western blot analysis to assess the transfection efficiency to overexpress (OE) RKIP protein on the HCC827 cell line. On top is presented a representative image of three independent experiments, and on the bottom is shown the graphical representation of the quantification of the western blots. (D) Graphical representation of the cell's viability that was assessed at 24, 48, and 72 h using the MTS assay (N = 4). (E) Clonogenicity assay to assess both proliferation capacity and the capacity to form multicellular colonies after 14 days. Colonies were counted manually, and the representative pictures were taken using a stereomicroscope (Olympus SZ) at 1× magnification (N = 3). Scale bar: 1 cm. (F) Wound healing migration assay, where the capacity of the cells to migrate and close the wound was evaluated over time (12, 24, 36, 48, 60, 72, and 84 h) (N = 3). On the right are representative images taken at 100× magnification at 0 and 72 h. Scale bar: 200 μm. (G) Graphical representation of the cell's capacity to form spheroids over time (4, 7, 9, and 11 days). The spheroids area is represented in percentage, considering the spheroid size at day 4 at 100% (N = 3). Representative images are shown on the right and were acquired using the Axio LabA1 microscope at 100× magnification. Scale bar: 1 mm. (H) Western blot analysis to assess the expression levels of relevant proteins related to EMT, NFĸB, and EGFR signaling pathways in HCC827 RKIP OE and CTR cells. Regarding all western blots in this figure, cells were stimulated with EGF for 15 min at 10 ng·mL⁻¹ when relevant and the experiments were performed three times. Also, α‐Tubulin was used as the loading control. In all graphics, error bars indicate the standard deviation (SD). Single comparisons between the different conditions studied were made using Student's t‐test, and differences between groups were evaluated using the two‐way ANOVA test. Values statistically different from the control group are represented with *P < 0.05, ***P < 0.001 and ****P < 0.0001. CTR, control; EGF, epidermal growth factor; NSCLC, non‐small‐cell lung cancer; RKIP OE, RKIP overexpression.

Fig. 3

Molecular characterization of HCC827 cells after RKIP overexpression through NanoString analysis. (A) Heatmap of genes altered in HCC827 CTR (dark blue) and HCC827 RKIP OE cells (black). Red represents the overexpressed genes and in blue the downregulated genes. P adjusted < 0.01; fold‐change (FC) ≥ 1.5. (B) Functional protein association network associated with RKIP overexpression done in STRING, using the list of differentially expressed genes from NanoString analysis. The network formed had a significant interaction among proteins (PP1 enrichment P‐value = 1.11e⁻¹⁶). In this representation, each circle represents a protein (node), and each connection represents a direct or indirect connection (edge). Node color indicates the pathway that these proteins are related to green—necroptosis, red—NFKB pathway, purple—tumor necrosis factor (TNF) pathway, pink—apoptosis, and yellow—MAPK pathway. Line color indicated the type of interaction evidence: purple—experimental evidence, light blue—curate database, black—co‐expression, pink—experimentally determined, yellow—text mining, dark blue—gene co‐occurrence. (C–F) The differential expressed genes between HCC827 RKIP OE and CTR cells were also explored on ShinyGO platform through a functional enrichment analysis. Gene Ontology (GO) (C, D), KEGG (E), and Reactome (F) databases were used, and the top 20 pathways enriched considering our group of genes were compiled and represented. Enriched pathways were sorted considering the −log₁₀ (FDR), the size of the circles is proportional to the number of genes, and the color of the bars corresponds to the fold enrichment.

Fig. 4

In vivo assessment of the influence of RKIP modulation in tumor growth. (A) Effect of RKIP overexpression in tumor growth rate upon injection of HCC827 RKIP OE and CTR cells in immunocompromised mice. Tumors were manually measured with a caliper over time for 37 days (N = 8). (B) Representative immunohistochemistry pictures of the expression of RKIP, the proliferation marker Ki67, and H&E staining of RKIP CTR and RKIP OE tumors (N = 3). (C) Effect of RKIP knockout in tumor growth rate upon injection of PC9 RKIP KO and CTR cells in immunocompromised mice. Tumors were manually measured with a caliper overtime for 21 days (N = 8). (D) Representative immunohistochemistry pictures of the expression of RKIP, the proliferation marker Ki67 and H&E staining of RKIP CTR and RKIP KO tumors (N = 3). In all graphics, error bars indicate the standard deviation (SD) and the differences between groups were evaluated using the two‐way ANOVA test. Values statistically different from the control group are represented with ****P < 0.0001. CTR, control; H&E, hematoxylin and eosin staining; RKIP KO, RKIP knockout; RKIP OE, RKIP overexpression.

Fig. 5

Effect of RKIP modulation in cells response to EGFR‐targeted therapies, in vitro and in vivo. (A) Representative graphics of the cytotoxicity assays used for the anti‐EGFR drugs, Erlotinib, Afatinib, and AST1306, in HCC827 RKIP OE and CTR cells. Such was assessed for 72 h by MTS assay. The graphs represent the mean ± SD relative to DMSO alone (100%) viability. These are representative of four independent assays performed in triplicate. (B) Comparative IC₅₀ values analysis for HCC827 CTR and RKIP OE cells. IC₅₀ values are the mean of four independent assays performed in triplicate. (C) Representative graphics of the cytotoxicity assays used for the anti‐EGFR drugs, Erlotinib, Afatinib and AST1306, in PC9 RKIP KO and CTR cells. Such was assessed for 72 h by MTS assay. The graphs are represented as the mean ± SD, relative to DMSO alone (100%) viability. These are representative of four independent assays performed in triplicate. (D) Comparative analysis of IC₅₀ values for PC9 KO and CTR cells. IC₅₀ values are represented as the mean of four independent assays performed in triplicate. (E) Scheme of the in vivo experiment with Afatinib treatment upon subcutaneous injection of HCC837 RKIP OE or PC9 RKIP KO and respective control cells. (F) Relative tumor growth rate upon daily treatment with 10 mg·kg⁻¹ of Afatinib by oral gavage (N = 8). Tumor growth of all animals was normalized, taking into consideration the tumor size on the first day of treatment. (G) Representation and comparative analysis of the final tumor weight. The tumor weight of the nontreated animals was considered 1 to normalize the samples (N = 8). (H) Representation and comparative analysis of the final tumor weight upon daily treatment with 15 mg·kg⁻¹ of Afatinib by oral gavage. In this, the tumor weight of the non‐treated animals was considered as 1 to normalize the samples (N = 3). In all graphics, error bars indicate the standard deviation (SD). Single comparisons between the different conditions studied were made using Student's t‐test, and differences between groups were evaluated using the two‐way ANOVA test. Values statistically different from the control group are represented with *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001. CTR, control; RKIP KO, RKIP knockout; RKIP OE, RKIP overexpression.