Almonertinib significantly inhibited the viability of NSCLC cells

To investigate the sensitivity of different EGFR genotypes to almonertinib, the normal cell line Beas-2B (lung epithelial cells), the lung adenocarcinoma cell line A549 (EGFR wild-type cells), and H1975 (EGFR T790M mutant cells) were treated with 0, 2, 4, 8, 16, or 32 µM almonertinib for 24, 48, or 72 h and subjected to a CCK8 assay. H1975 cells presented the fastest decline in cell viability, followed by A549 cells, whereas Beas-2B cells presented the slowest decline (Fig. 1A). The proliferation, migration, and invasion ability of cells treated with 2.4 µM allorertinib for 24 h were altered in colony formation and transwell invasion assays. When the concentration was increased to 4 µM, 2 µM allorertinib considerably decreased the proliferation, migration, and invasion ability of H1975 cells, and the viability of A549 cells decreased (Fig. 1B-F). Beas-2B cells were largely unaffected by almonertinib treatment. These results showed that almonertinib significantly inhibited the viability of NSCLC cells compared with normal cells, whereas EGFR T790M-mutant cells were more sensitive than were EGFR wild-type cells.

Fig. 1

Almonertinib significantly inhibited the viability of NSCLC cells. (A) Beas-2B, A549 and H1975 cells were treated with different concentrations (0, 2, 4, 8, 16, or 32 µM) of almonertinib for 24, 48, or 72 h. Cell viability was measured by a CCK8 assay. Then, Beas-2B, A549 and H1975 cells were treated with 0, 2, or 4 µM almonertinib, respectively. (B-C) Colony formation was detected via a colony formation assay. (D) Cell migration was evaluated via a wound-healing assay. (E-F) Cell invasion was determined by a Transwell invasion assay (magnification, ×100). The values are the means ± SEMs. Significance: ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001

CAFs were successfully induced by TGF-β1

We selected the classical cytokine TGF-β1 (transforming growth factor β1) for the activation of CAFs, which has been confirmed to be associated with the transformation of CAFs in previous studies [15, 16]. Human embryonic HFL1 lung fibroblasts were treated with 10 ng/mL TGF-β1 for 24–48 h for 2–3 passages, and their morphology was observed under a microscope. Normal fibroblasts (NFs) and fibroblasts stimulated with TGF-β1 (CAFs-24 h and CAFs-48 h) for 24 h and 48 h, respectively, did not differ in morphology. Both showed typical fibrocyte-like characteristics, which were flat, spindle shaped, and swirled at dense growth sites (Fig. 2A). α-SMA and FAP expression was evaluated by western blotting. The expression of both markers was greater in the TGF-β1-stimulated fibroblasts than in the NF-stimulated fibroblasts (Fig. 2B). The protein expression levels of α-SMA and FAP in the CAF-48 h group were significantly greater than those in the NF group. However, the differences between the CAF-24 h and NF groups were not statistically significant. The upregulation of CAF markers is an important feature of CAF activation. According to the western blotting results, TGF-β1 successfully induced the conversion of NFs into CAFs. All the CAFs used in the follow-up experiments were obtained from NFs after treatment with TGF-β1 for 48 h.

Fig. 2

CAFs were successfully induced by TGF-β1, and almonertinib increased the TGF-β1 secretion of H1975 cells. HFL1 cells were stimulated with 10 ng/mL TGF-β1 for 24–48 h, after which they were passaged 2–3 times. (A) Morphology was observed under a microscope (magnification, ×40). (B) The expression levels of α-SMA and FAP in the NF, CAF-24 h and CAF-48 h groups were detected by Western blotting. (C) The secretion of TGF-β1 in H1975 cells treated with 0, 2, or 4 µM almonertinib was determined by ELISA. The values are the means ± SEMs. Significance: ns, not significant; **p < 0.01; ***p < 0.001

Almonertinib increased the TGF-β1 secretion of H1975 cells

To detect the secretion of TGF-β1 in EGFR T790M-mutant NSCLC cells before and after almonertinib treatment, the supernatants of H1975 cells treated with 0, 2, or 4 µM almonertinib were collected. The concentration of TGF-β1 in the supernatants of the different groups was determined via an ELISA kit. The amount of TGF-β1 secreted by H1975 cells increased after stimulation with almonertinib (Fig. 2C). Furthermore, the secretion of TGF-β1 in H1975 cells also increased with increasing almonertinib concentration. In addition, H1975 cells themselves do not secrete a small amount of TGF-β1, which supports the use of TGF-β1 as a CAF inducer in this study.

TGF-β1 is a peptide growth factor that has multiple functions. Abnormalities in the signal transduction pathway of TGF-β1 are important factors in cancer development, progression, infiltration, and metastasis. TGF-β1 promotes cell proliferation in progressive tumors, which, in turn, promotes tumor invasion and metastasis. The function of Smad3 is dependent on signal transduction and the regulation of Smad and non-Smad proteins. Notably, Smad3, a major activator of the TGF-β/Smad signaling pathway, regulates gene expression and promotes fibroblast differentiation and proliferation. In addition, TGF-β1, an important member of the TGF-β family, has been shown to be a major inducer of EMT and promotes the development of NSCLC. Our experiments revealed that TGF-β1 was induced to convert NFs to CAFs. The secretion of TGF-β1 also increased with increasing almonertinib concentration. Almonertinib may promote the accumulation of CAFs in NSCLC cells by increasing the secretion of TGF-β1. Therefore, we hypothesized that almonertinib may promote tumor development by promoting signaling via the TGF-β1/Smad pathway and inducing the development of epithelial‒mesenchymal transition to achieve CAF-induced drug resistance in NSCLC. This hypothesis adds to the possible mechanism underlying the induction of TGF-β1 expression by almonertinib. However, this pathway needs to be further supplemented with relevant experimental validation in subsequently derived subjects to enrich the research.

CAFs alleviated the inhibitory effects of almonertinib on NSCLC cells

CAFs usually promote cancer by secreting various soluble factors and exosomes [17,18,19]. Therefore, the supernatants of NFs and CAFs collected as conditioned medium (CM) after centrifugation were used to treat H1975 and A549 cells in advance for 48 h. On the basis of previous experimental results, drug concentrations of 2 µM and 4 µM were selected for use in pretreated H1975 and A549 cells, respectively. We used colony formation, Transwell migration, and invasion assays to detect changes in the proliferation, migration, and invasion ability of NSCLC cells treated early with the negative control (NC), NFs, or CAF-CM in the presence of almonertinib. Compared with those of the tumor cells in the NC-CM group, the number of colonies, migration, and invasion of the cells in the CAF-CM group increased at the same drug concentration (Fig. 3). Interestingly, the intervention of NFs with NSCLC cells had a synergistic effect with almonertinib in reducing cell viability. The results showed that after CAF intervention, the survival of NSCLC cells in the presence of almonertinib improved. CAFs alleviated the inhibitory effect of almonertinib on NSCLC cells to a certain extent, whereas NFs had the opposite effect.

Fig. 3

CAFs alleviated the inhibitory effects of almonertinib on NSCLC cells. The supernatants of NFs and CAFs were centrifuged at 1000xg for 15 min and collected as conditioned medium (CM). H1975 and A549 cells were treated with different CMs for 48 h. Then, they were exposed to 2 or 4 µM almonertinib. (A-B) Colony formation was evaluated via a colony formation assay. (C-D) Cell migration was determined by a Transwell migration assay (magnification, ×100). (E-F) Cell invasion was measured via a Transwell invasion assay (magnification, ×100). The values are the means ± SEMs. Significance: **p < 0.01, ***p < 0.001, ****p < 0.0001

The Hippo signaling pathway is correlated with third-generation EGFR-TKI resistance

We selected the GSE193258 dataset from the GEO database, focusing on third-generation EGFR-TKI resistance sequencing, to evaluate the DEGs. The acute group treated with osimertinib for 24 h was considered the osimertinib-sensitive group, and the DTP group treated with osimertinib for 21 days was regarded as the osimertinib-resistant group. The differential expression criteria were determined by│logFC│>1 and adjusted P value < 0.05 via the Benjamini‒Hochberg method.

The data from the acute and DTP groups in the PC9, NCI-H1975, HCC827, and HCC2935 cell lines were analyzed using cell lines as covariates. A total of 393 upregulated and 445 downregulated genes were identified (Figure S1). Functional enrichment analysis was performed for the downregulated genes; that is, their expression was upregulated in NSCLC cells after the acquisition of osimertinib resistance. GO analysis revealed the top five enrichment pathways of the three functional classifications (BP, CC, and MF), most of which were associated with the extracellular matrix (Figure S2A). KEGG analysis revealed the top 10 enriched pathways, indicating that osimertinib resistance was related to the PI3K/AKT and TGF-β signaling pathways (Figure S2B).

The genes in the acute and DTP groups of H1975 cells were subsequently analyzed for differential expression, and 875 upregulated genes and 538 downregulated genes were screened (Fig. 4A-B). Functional enrichment analysis was subsequently performed for the downregulated genes. GO analysis revealed that the enriched pathways were associated with the extracellular matrix and signaling pathway activation, such as external encapsulating structure organization, the collagen-containing extracellular matrix, and receptor ligand activity (Fig. 4C). KEGG analysis identified only three enriched pathways. Among these pathways, the Hippo signaling pathway was the most significant (Fig. 4D). The Hippo signaling pathway is linked to a variety of signaling pathways and has been found to be associated with cell adhesion, tumor growth, and metastasis [20, 21].

Fig. 4

The Hippo signaling pathway is correlated with third-generation EGFR-TKI resistance. The data from the acute group and DTP group of H1975 cells were analyzed for differential expression. The (A) heatmap and (B) volcano map are shown. Functional enrichment analysis was subsequently performed for the downregulated genes. The results of (C) GO analysis and (D) KEGG analysis revealed the top enrichment pathways

These results demonstrate that third-generation EGFR-TKI resistance is correlated with the extracellular matrix and the Hippo signaling pathway.

CAFs downregulated the expression of YAP/TAZ in H1975 cells

YAP/TAZ are the principal molecules of the Hippo pathway and have been demonstrated to be related to CAFs and various physiological functions in tumors [22, 23]. GEPIA2 (http://gepia2.cancer-pku.cn/#index) was used to examine the correlation between YAP1, TAZ, and the CAF marker α-SMA (ACTA2). Both YAP1 and TAZ were weakly and significantly correlated with ACTA2 expression, respectively (Figure S3). YAP/TAZ were further determined to be related to CAFs.

To verify whether the effect of CAFs on NSCLC cells is related to YAP/TAZ, qPCR was used to determine the mRNA expression levels of YAP/TAZ. The YAP/TAZ expression level in EGFR wild-type A549 cells was lower than that in EGFR T790M mutant H1975 cells, which are more sensitive to almonertinib (Fig. 5A). The expression of YAP/TAZ was also downregulated in H1975 cells treated with CAF-CM (Fig. 5B). H1975 cells treated with the negative control or CAF-CM were treated with 2 µM almonertinib for 24 h, and YAP/TAZ expression in the CAF-CM group also decreased (Fig. 5C). Furthermore, we compared the expression of YAP/TAZ in H1975 cells after treatment with CAF-CM untreated with almonertinib and after treatment with 2 µM almonertinib in CAF-CM. Similarly, we discovered that in the presence of almonertinib, CAFs downregulated YAP/TAZ expression levels (Fig. 5D). Combined with the results of previous performance experiments (Figs. 1 and 3), these findings indicate that CAF intervention can downregulate the expression of YAP/TAZ in H1975 cells and that the change in expression may be related to CAF-mediated osimertinib resistance.

Fig. 5

CAFs downregulated the expression of YAP/TAZ in H1975 cells, and YAP1 was associated with poor prognosis in lung cancer patients. qPCR was used to evaluate the mRNA expression levels of YAP1 and TAZ. Statistical results of YAP/TAZ expression (A) in H1975 and A549 cells, (B) in H1975 cells treated with NC-CM or CAF-CM and (C) then exposed to 2 µM almonertinib for 24 h are represented by bar charts. (D) Statistical analysis of YAP/TAZ expression in H1975 cells treated with CAF-CM and H1975 cells exposed to 2 µM almonertinib for 24 h. (E) Survival analysis of YAP1 in 1,925 lung cancer patients from the TCGA database was performed. Significance: *p < 0.05, **p < 0.01, ***p < 0.001

YAP1 was associated with poor prognosis in lung cancer patients

Survival analysis of 1,925 lung cancer patients from the TCGA database was performed via the Kaplan‒Meier plotter (https://kmplot.com/analysis/) website, with the median YAP1 expression level used as the grouping criterion. The results indicated that the 5-year overall survival rate of patients with low YAP1 expression was lower than that of individuals with high YAP1 expression. Lower YAP1 expression was correlated with unfavorable outcomes in patients with lung cancer (Fig. 5E). This trend was consistent with the results of the functional experiments and qPCR.

YAP/TAZ is necessary for the ability of CAFs to protect cancer cells from damage induced by almonertinib

Multiple studies have suggested that the core transduction molecules YAP and TAZ in the Hippo pathway are associated with multiple physiological functions in tumors. On the basis of our previous experimental results, we hypothesized that CAFs promote almonertinib resistance by downregulating YAP/TAZ in lung cancer cells. To validate our hypothesis, we assessed the level of YAP/TAZ expression in cancer cells treated with CAF-CM and observed a significant reduction in YAP/TAZ protein levels compared with those in cells treated with NC-CM (Fig. 6A). This result was consistent with the results of previous qPCR and survival analyses (Fig. 5).

Fig. 6

CAFs influence the YAP/TAZ pathway to facilitate protection of cancer cells from almonertinib drug damage. (A) Statistical analysis of YAP/TAZ protein expression in H1975 cells treated with NC-CM or CAF-CM. (B-C) Transfection efficiency was verified by qPCR and western blotting after knockdown of YAP/TAZ. (D-G) Results of clones and invasion phenotypes of H1975 cells treated with NC-CM, NC-CM + siRNA or CAF-CM. Significance: *p < 0.05, **p < 0.01, ***p < 0.001

We subsequently knocked down YAP/TAZ expression to investigate whether CAFs contribute to almonertinib resistance in lung cancer. qPCR and western blotting demonstrated that siRNA-1 transfection was more efficient (Fig. 6B-C); therefore, we chose siRNA-1 to confirm that the YAP/TAZ molecular pathway was associated with CAFs to promote NSCLC drug resistance.

Furthermore, we cultured lung cancer cells with CM collected from NC, CAF, or H1975 cells after transfection with siRNA. We found that at the same drug concentration, the number of colonized and invasive cells was greater in both the NC-CM + siRNA group and the CAF-CM group than in the NC-CM-treated group (Fig. 6D-G). The results revealed that the knockdown of YAP/TAZ corresponded to the phenotypic trends observed after intervention with CAFs. These data provide evidence that CAFs protect cancer cells from almonertinib drug damage by downregulating YAP/TAZ in lung cancer cells.