Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2026 Apr 4;16:16233. doi: 10.1038/s41598-026-46997-7

Colorectal cancer cells respond differentially to autophagy induced by tyrosine kinase inhibitors

Lila Louadj 1,✉, Grigorios Gerotziafas 1,2, Michèle Sabbah 1,3,#, Annette K Larsen 1,#
PMCID: PMC13201848  PMID: 41935207

Abstract

Autophagy plays an important role in the response of tumors to environmental stress including colorectal cancer (CRC). The present study aimed to investigate the relationship between the autophagic capacities of CRC cells and their sensitivities to Nintedanib and Regorafenib, two TKIs with clinical activity in metastatic CRC. Our results showed various effects of these two TKIs on cell viability across a panel consisting of 12 well-characterized CRC cell lines. We showed an opposite cytotoxicity profile between HT-29 and LoVo cell lines for both TKIs which is accompanied by an induction of autophagy in cell-type and drug-dependent manner. Interestingly, pharmacologic and genetic autophagy inhibition decreased the cytotoxic activity of Nintedanib but did not affect the activity of Regorafenib. In addition, signaling pathways analysis revealed opposing effects on the Akt-mTOR and Erk-signaling pathways by the two TKIs. Nintedanib inhibits the Akt- mTOR pathway which is compensated by activation of Erk signaling, whereas Regorafenib is a strong inhibitor of Erk-signaling and is accompanied by Akt- mTOR activation. Taken together, our results indicate that autophagy contributes to the cytotoxic activity of Nintedanib but not that of Regorafenib. These results highlighted that CRC with high autophagic flux may be selectively sensitive to Nintedanib.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-46997-7.

Keywords: Colorectal cancer (CRC), Tyrosine kinase inhibitors (TKIs), Autophagy, Nintedanib, Regorafenib, Cell viability

Subject terms: Cancer, Cell biology, Drug discovery, Oncology

Introduction

Despite significant advances in diagnosis and treatment, colorectal cancer (CRC) remains one of the leading causes of cancer-related death worldwide, affecting both men and women, with more than 1.9 million new cases and 0.9 million deaths in 20201,2. CRC etiology involves both genetic and environmental factors. Autophagy plays an important transformational role in the switch that occurs when a cell shifts from normal to malignant3 and is a fundamental mechanism for responding to these genetic and environmental stresses.

Autophagy is a regulated and evolutionarily conserved catabolic process that serves to degrade damaged proteins and organelles and recycle their biochemical components for use in energy production and other biosynthetic reactions4–7. It occurs at low basal levels in virtually all cells and is typically induced under cellular stress conditions such as starvation. It is initially considered to perform homeostatic functions or a survival strategy requirement6,8, however, dysfunctions in this process are associated with a wide range of human diseases, including cancer8–12. Indeed, in the occurrence and development of cancer, many studies indicated that autophagy has been broadly recognized as a double-edged sword, playing a dual role in tumor suppression and growth promotion13,14. Thus, whether autophagy acts as an anti-tumor or a pro-tumor mechanism remains controversial.

Interestingly, at early stages of tumor development, autophagy may serve to restrain tumor development, as indicated by the tumor suppressor function of beclin 1, a central regulator of autophagy15. In contrast, for advanced tumors, autophagy appears to play a cytoprotective role and seems to promote tumor progression by providing increased resistance to the hypoxic, acidic and nutrient-poor tumor microenvironment16,17. In addition, autophagy can promote escape from T-cell-mediated lysis18,19.

Autophagy may also influence the tumor response to anticancer therapies in many kinds of cancer cells, specifically in CRC. Indeed, it is well established that virtually every form of cellular stress including cancer therapy (chemotherapy; radiation; targeted therapy) promotes autophagy by various mechanisms20. The mechanistic link between drug exposure and autophagy induction is particularly strong in the case of anticancer receptor tyrosine kinase inhibitors (TKIs) as mTOR (mammalian mechanistic target of rapamycin) is believed to act as the intracellular nodal point for most, if not all, of these receptors. mTOR is a serine/threonine kinase and master regulator of autophagy21. Numerous studies have reported that the autophagy induced by TKIs is cytoprotective22–28 which has encouraged clinical trials of autophagy inhibitors. A notable exception concerns a study with the EGFR (epidermal growth factor receptor) inhibitor erlotinib that convincingly shows that autophagy contributes to the cytotoxic activity of this compound, both in vitro and in vivo29. Thus, it cannot automatically be taken for granted that all receptor tyrosine kinase inhibitors induce cytoprotective autophagy.

Therefore, this study aimed to evaluate the activity of two multikinase inhibitors, Nintedanib and Regorafenib, both of which with clinical efficacy for metastatic colorectal cancer (CRC)29,30 in inducing autophagy in colorectal cancer cells and elucidate the underlying mechanism. Finally, we sought to determine whether autophagy induced by these TKIs was cytoprotective cytostatic or cytotoxic.

The results revealed clear differences in the downstream signaling pathways of the two drugs as well as their capacity to induce autophagy. We also show that both genetic and pharmacological inhibition of autophagy was associated with reduced toxicity of Nintedanib, but had no effect on Regorafenib. These findings suggest that Nintedanib may be particularly active in cells with high autophagic flux while autophagy has little, if any, impact on the activity of Regorafenib.

Results

Cytotoxic activity of Nintedanib and Regorafenib toward CRC cells

The cytotoxic activities of Nintedanib and Regorafenib were determined in a panel of twelve well-characterized CRC cell lines. Cells were incubated with Nintedanib or Regorafenib for 120 h, and their viability was subsequently measured using a cell viability assay. Nintedanib and Regorafenib were active at similar concentrations with IC50 values ranging from 1.3 to 5.0 µM and from 0.7 to 4.1 µM respectively (Fig. 1A). Comparison of the IC50 values for Nintedanib and Regorafenib in the different cell lines revealed no correlation in the sensitivity of the two drugs, suggesting that both TKIs have different activity profiles (Fig. 1B). Cell exposure to the cytotoxic drugs is accompanied with different cellular responses including cell cycle arrest, autophagy and cell death. To determine the chronology and the magnitude of these responses, we selected two cellular models, HT-29 the most sensitive cell line to Nintedanib and LoVo the most sensitive to Regorafenib. All subsequent experiments were carried out at the IC50 dose.

Fig. 1.

Fig. 1

Open in a new tab

Growth inhibitory activity of Nintedanib and Regorafenib toward CRC cells. (A) The growth inhibitory activity of Nintedanib (up right) and Regorafenib (up left) toward CRC cells was determined by the MTT viability assay after 120 h continuous drug exposure. The corresponding IC50 values for each cell line for both drugs are shown in the tables directly below the viability curves. All values are averages of at least 3 independent experiments each done in duplicate. (B) The IC50 values for Nintedanib were plotted against the IC50 values for Regorafenib for each cell line. There was no statistically significant correlation between the 2 parameters (r2 = 0.29). (C, D) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib or Regorafenib for the indicated times and the cell cycle distribution was determined by flow cytometry analysis. The values indicate the average values for 3 experiments. (E, F) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib and Regorafenib for the indicated times, and the protein levels of the cell-cycle regulators p27Kip1 and p21Waf1/Cip1 were determined by Western blot analysis. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of the indicated cell cycle regulator in drug-exposed cells compared to the corresponding untreated control. (G, H) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib and Regorafenib for the indicated times, and the fraction of fragmented DNA present in the sub-G1 fraction was determined by flow cytometry analysis. All data are expressed as a mean ± SD (n = 3). Data were analyzed by two-way ANOVA test and were compared to the corresponding untreated control; ****p < 0.0001; ***p < 0.001; *p < 0.01. ns < 0.05 vs. non treated.

Exposure of HT-29 cells to Nintedanib (Fig. 1C) was accompanied by a strong and prolonged G1-arrest after 48 h with a corresponding decrease of cells in the S- and G2-phase of the cell cycle. G1-arrest was also observed for Regorafenib, although the cells seem to be able to divide once, as indicated by the shortened G1-fraction by 48 h. To investigate the possible mechanism of the cell cycle arrest, we determined the levels of p21 Waf1/Cip1 and p27/kip1 proteins, which negatively regulate cell cycle progression from the G0/G1 to S phase and the S to G2 phase. In HT-29 cells, the G1-arrest was associated with upregulation of p27Kip1, which was particularly prominent for Nintedanib. In contrast, p21Waf1/Cip was not affected (Fig. 1E). Exposure of LoVo cells to Nintedanib and Regorafenib was also accompanied by a strong and prolonged G1-arrest (Fig. 1D). The G1-arrest is principally associated with upregulation of p21Waf1/Cip following Nintedanib treatment and with p27Kip1 after exposure to Regorafenib (Fig. 1F). Exposure of HT-29 and LoVo cells to Nintedanib or Regorafenib was accompanied by increased cell death starting at 24 h as determined by the amount of fragmented cellular DNA in the sub-G1 fraction (Fig. 1G, H). Longer drug exposure led to increased levels of cell death for both cell lines reaching 20–29% of the total cell population by 72 h.

Influence of Nintedanib and Regorafenib on autophagy induction

To investigate the ability of Nintedanib and Regorafenib to trigger autophagy in vitro, HT-29 and LoVo cell lines were labeled with Autophagy Blue™, a widely used specific marker of autophagosomes. As shown in (Fig. 2A), Nintedanib is a strong inducer of autophagy in HT-29 cells with maximal activity observed after 24 h drug exposure during which the autophagy activity reached 64-times the level compared to the untreated control cells. Regorafenib exhibits similar autophagy kinetics; however, the overall levels of autophagy are approximately 4–6 times lower than those observed with Nintedanib.

Fig. 2.

Fig. 2

Open in a new tab

Influence of Nintedanib and Regorafenib on autophagy in CRC cells. (A, B) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib (light grey columns) or Regorafenib (dark grey columns) for the indicated times, and the induction of autophagy was determined with a fluorescence imaging kit. The values reflect the degree of autophagy in drug-exposed cells compared to the corresponding untreated control. The values represent the average values for 3 experiments, each done in duplicate. Bars, SD (n = 3), data were analyzed by two-way ANOVA test and were compared to the corresponding untreated control; ****p < 0.0001; ***p < 0.001; *p < 0.01. ns < 0.05 vs. non treated. (C, D) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib or Regorafenib for the indicated times followed by cytochemistry to reveal cells undergoing autophagy. Cells were marked with antibodies directed against LC3 (red) or p62/Sequestosome1 (green). The nuclei were counterstained with DRACQ (blue). Representative images from one out of 2 experiments. (E, F) HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib or Regorafenib for the indicated times and the protein levels of the autophagy-associated proteins LC3, p62/Sequestosome1 and Beclin1 were determined by Western blot analysis. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of the indicated autophagy-associated protein in drug-exposed cells compared to the corresponding untreated control. Representative Western blots from one out of 2 experiments.

A similar pattern was observed in LoVo cells, with Nintedanib inducing higher levels of autophagy than Regorafenib (Fig. 2B). However, the overall autophagy levels in LoVo cells were substantially lower compared to those in HT-29 cells.

Taken together, these findings indicate that Nintedanib was a potent inducer of autophagy compared to Regorafenib both in HT-29 cells and in the LoVo cells.

To monitor the effects of Nintedanib and Regorafenib in individual cells over time, HT29 and LoVo cells were labeled with the autophagosome marker LC3 (red) and the autosomal flux marker p62/Sequestosome1 (green). The nuclei were counterstained with DRAQ5 (blue). The results (Fig. 2C) show a prominent cytoplasmic p62 signal in the untreated HT-29 cells which decreased over time in the presence of Nintedanib. Notably, the LC3 signal is strong after 24 and 48 h Nintedanib exposure, with a prominent co-localization between p62 and LC3 at 48 h. In comparison, Regorafenib-treated cells do not significantly alter LC3 or P62 levels.

A similar pattern was observed inLoVo cells (Fig. 2D). Indeed, cells displayed cytoplasmic p62 staining which is attenuated in the presence of Nintedanib. In contrast, the LC3 signal is prominent for 24 and 48 h. For Regorafenib, both the LC3 and p62 levels remained unchanged.

To further explore these findings, Western blot analysis was performed with antibodies againts LC3, p62 and Beclin1 proteins which are involved in the early autophagosome formation. The results show that exposure of HT-29 cells to Nintedanib is associated with a strong increase signal in the membrane-bound form of LC3 (LC3II), in clear contrast to what is observed with Regorafenib.

In contrast, the expression levels of both p62 and Beclin1 gradually decrease over time following Nintedanib exposure, consistent with their roles in the autophagic process (Fig. 2E). By comparison, the expression levels of these markers remain largely unchanged after Regorafenib treatment, suggesting limited activation of the autophagic pathway.

Similar findings are observed after LoVo cells exposure to Nintedanib or Regorafenib (Fig. 2F), although the LC3II signal for Nintedanib-treated cells was less prominent in LoVo cells as compared to the HT-29 cells.

Interestingly, similar results were obtained with three other cell lines with distinct genetic background suggesting that autophagic responses is observed regardless of their KRAS, BRAF or TP53 status (Figure S1 and S2).

Nintedanib and Regorafenib have differential effect on the Akt/mTOR-signaling pathway

mTOR is a serine/threonine kinase that serves as a central mediator of cellular growth, metabolism, and protein synthesis. In addition to promoting the anabolic process, mTOR also acts as a negative regulator of autophagy. It functions as a nodal point for multiple signaling pathways mediated by tyrosine kinase receptors expressed by CRC cells. Therefore, we speculated that Nintedanib/Regorafenib exposure might not only attenuate the activity of the angiokinases, but also inhibit their downstream targets, including Akt/mTOR signaling.

To investigate this hypothesis, the influence of drug exposure on the Akt/mTOR signaling pathway was assessed by Western blot in HT-29 and LoVo cells for the indicated times.

In HT-29 cells (Fig. 3A) Nintedanib exposure was accompanied by strong inhibition of Akt phosphorylation that was accompanied by reduced levels of phospho-mTOR and its downstream effector phospho-p70S6K.

Fig. 3.

Fig. 3

Open in a new tab

Influence of Nintedanib and Regorafenib on the Akt/mTOR-signaling pathway in CRC cells. HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib and Regorafenib for the indicated times, and the protein levels of proteins involved in the Akt/mTOR signaling pathway were determined by Western blot analysis. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of the indicated protein in drug-exposed cells compared to the corresponding untreated control. Representative Western blots from one out of at least 2 experiments.

A comparable response was observed for LoVo (Fig. 3B) cells, although the inhibition of Akt phosphorylation occurred later and the overall response was less marked.

In contrast, Regorafenib exposure was accompanied by increased levels of phospho-Akt in both HT-29 and LoVo cells, possibly indicating a feedback activation or compensatory response within the signaling network.

Interestingly, the expression of Deptor, a natural mTOR inhibitor, was notably upregulated under all conditions.

Nintedanib and Regorafenib have differential effect on the Erk/RSK-signaling pathway

MAPK (mitogen-activated protein kinase) is a family of serine/threonine protein kinases that play a pivotal role in the regulation of various cellular processes, including proliferation, differentiation, and stress responses. Among its subgroups, the ERK (extracellular signal-regulated kinase) pathway is particularly well-characterized and closely linked to oncogenic signaling in many cancers. In addition, MAPK/ERK signaling is closely associated with the autophagosomal pathway. Indeed, Erk can upregulate the cellular levels of LC3B and SQSTM1/p62 at the transcriptional level, thereby influencing the autophagic process.

Therefore, to investigate the influence of drug exposure on the Erk/RSK-signaling pathway, Western blot analysis were conducted after exposing HT-29 or LoVo cells to Nintedanib or Regorafenib for the indicated times.

Nintedanib treatment of both HT-29 (Fig. 4A) and LoVo (Fig. 4B) cells was accompanied by strong ERK activation as indicated by the induction of a strong increase of phospho-ERK levels. Surprisingly, the phosphorylation of classical Erk downstream targets was not affected. Specifically, phosphorylation of p90RSK (p90 ribosomal s6 kinases) was unaffected, while phosphorylation of rpS6 (ribosomal protein S6) a downstream target of p90RSK was significantly reduced.

Fig. 4.

Fig. 4

Open in a new tab

Influence of Nintedanib and Regorafenib on the Akt/mTOR-signaling pathway in CRC cells. HT-29 and LoVo cells were exposed to the IC50 dose of Nintedanib and Regorafenib for the indicated times, and the expression of proteins involved in the Erk/RSK pathway were determined by Western blot analysis. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of the indicated protein in drug-exposed cells compared to the corresponding untreated control. Representative Western blots from one out of at least 2 experiments.

In comparison, ERK phosphorylation was strongly reduced by Regorafenib in both HT-29 and LoVo cells, leading to reduced phosphorylation of all the classical downstream targets including p90RSK and rpS6.

Overall, these results reveal distinct effects of Nintedanib and Regorafenib on the MAPK/ERK axis. While Nintedanib paradoxically activates ERK but impairs downstream signaling, Regorafenib exerts a more classical inhibitory effect on the pathway.

Autophagy induction contributes to the cytotoxic activity of Nintedanib but not of Regorafenib

To determine if autophagy promotes or attenuates the cytotoxic activity of Nintedanib and Regorafenib, we first evaluated the influence of pharmacological autophagy inhibition on the cytotoxicity of these two drugs. For autophagy inhibition, we used 3-methyladenine (3-MA), which blocks autophagosome formation via inhibition of type III phosphatidylinositol-3-kinases. Thus, HT-29 and LoVo cells were treated with Nintedanib or Regorafenib in the absence or presence of 3-MA. Then, in one hand, the induction of autophagy was assessed by Western blot analysis of LC3 after 24 h, and on the other hand, the cytotoxic activity of both drugs was evaluated by MTT viability assay.

As expected, Nintedanib treatment is accompanied by strong induction of autophagy as indicated by the presence of high level of LC3II, which is greatly attenuated in the presence of 3-MA in both HT29 (Fig. 5A) and LoVo (Fig. 5B) cells. In contrast, Regorafenib exposure had modest influence on LC3II levels both in the absence or presence of 3-MA.

Fig. 5.

Fig. 5

Open in a new tab

Influence of autophagy modulation on the cytotoxicity of Nintedanib and Regorafenib. Autophagy induction was monitored by expression of LC3II in HT-29 (A) and LoVo cells (B) following 24 h exposure to Nintedanib or Regorafenib in the absence or presence of 3-MA. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of total LC3 in drug-treated cells compared to the untreated HT-29 cells. HT-29 cells (C) and LoVo cells (D) were exposed to Nintedanib (left) or Regorafenib (right) for 120 h in the absence or presence of 3-MA (2.5 mM) and the viability was determined by the MTT viability assay. The curves represent the average of 3 independent experiments, each done in triplicate. Bars, SD (n = 3). (E) HT-29 cells were transfected with shRNA toward Beclin1 or, as control, empty vector. Expression of BECN1 mRNA in parental, control or Beclin1-transfected cells. (F) Expression of Beclin1 protein in parental, control or shBeclin1-transfected cells. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of Beclin1 in transfected cells compared to the parental cells. (G, H) Autophagy in parental, control and shBeclin1 cells following 24 h exposure to Nintedanib or Regorafenib. Autophagy induction was monitored by expression of LC3II. The protein levels of β-actin are shown as the loading control. The numbers reflect the expression of total LC3 in drug-treated cells compared to the corresponding untreated cells. (I, J) The cytotoxicity of Nintedanib or Regorafenib was determined by colony formation of parental, control and shBeclin1 cells after 14 days continued drug exposure. The curves represent the average of 3 independent experiments, each done in triplicate. Bars, SD (n = 3). Representative Western blots from one out of 2 experiments.

The viability of HT-29 cells to Nintedanib in the absence or presence of 3-MA was determined by the MTT assay after 120 h of continuous exposure. The results show that autophagy inhibition by 3-MA was accompanied by decreased Nintedanib cytotoxicity effect with IC50 values of 1.3 µM for Nintedanib alone and 3.8 µM for Nintedanib in the presence of 3-MA (Fig. 5C, left). In comparison, the presence of 3-MA had no effect on the IC50 of Regorafenib which was around 3 µM under both conditions (Fig. 5C, right).

Autophagy inhibition of LoVo cells resulted in comparable findings although the influence of 3-MA was more modest, with IC50 of 2.5 µM for Nintedanib alone and 4.8 µM in the presence of Nintedanib and 3-MA (Fig. 5D, left). In contrast, 3-MA had no influence on the cytotoxicity of Regorafenib which an IC50 around 0.7 µM under both conditions (Fig. 5D, right).

Next, we constructed autophagy-deficient HT-29 cells by genetic downregulation of Beclin1 using small hairpin RNA (shRNA). Beclin1 mRNA (Fig. 5E) and protein levels (Fig. 5F) revealed that the expression of Beclin1 was reduced by 90% in clones 1 and 2 while the control cells expressed comparable levels of Beclin1, compared to the parental cells. To verify the attenuation of drug-induced autophagy, cells were incubated in the absence or presence of Nintedanib or Regorafenib for 24 h, and the expression of LC3 was assessed by Western blot analysis. Nintedanib exposure was accompanied by a strong increase in the expression of LC3II in parental and control cells but not for clones 1 and 2 (Fig. 5G). In comparison, no apparent increase of LC3II was observed for Regorafenib in any of the cell lines (Fig. 5H).

Next, the cytotoxic activity of Nintedanib and Regorafenib was determined by colony formation assay after 14 days of continuous drug exposure. We observed decreased sensitivity to Nintedanib when beclin 1 was knocked down. The IC50 values were 0.80 and 0.65 µM for the parental and control cells, respectively, whereas the corresponding values were 1.8 and 1.7 µM for shBeclin1 clones 1 and 2 (Fig. 5I). In contrast, when cells were exposed to Regorafenib, the IC50 values were 0.55 and 0.50 µM for the parental and control cells, respectively, while the values for shBeclin1 clones 1 and 2 were 0.45 and 0.60 µM (Fig. 5J) indicating that knock down of beclin1 did not modify sensitivity to this drug.

Discussion

Autophagy is a physiological process that is involved in the progression of colorectal cancer and drug resistance. Here, we characterize the influence of two angiokinases, Nintedanib and Regorafenib on CRC cells with an emphasis on autophagy. In particular, we aimed to determine whether the two tyrosine kinase inhibitors induce autophagy and if this induction occurs through the same signaling pathways.

In the present study using multiple CRC cell lines, we demonstrated that HT-29 cell line is the most sensitive to Nintedanib and the most resistant to Regorafenib while the LoVo cell line exhibits the opposite profile. Based on these results, we selected two cellular models for further studies.

Both drugs induced a prominent G1 arrest that was accompanied by activation of the cell cycle regulators p27kip1 and p21Waf1/cip1, and an increase of the proportion of dead cells (sub G1) to reach maximal levels at 72 h in both cellular models. These findings agree with previous findings31. The cell cycle arrest was also associated with the induction of autophagy, although the extent of autophagy varied depending on the cell type and the drug used. Interestingly, Nintedanib proved to be a more potent autophagy inducer than Regorafenib in both cell lines. However, the overall autophagic response was markedly lower in LoVo cells as compared to HT-29 cells. This is coherent with previous findings reporting that the capacity to undergo autophagy varies widely between CRC cell lines32.

Exposure of HT-29 cells to Nintedanib, and to a lesser extent LoVo cells, was accompanied by a marked increase in the autophagic markers LC3II, along with reduced levels of p62/Sequestosome1 and Beclin1 starting at 24 h. Similar findings were observed for Regorafenib but to a lesser degree and later time points.

Although the PI3K-Akt-mTOR pathway is well established as a regulator of autophagy, we next investigated the signaling pathways activated by the two drugs with a particular focus on mTOR-related processes. Phosphorylation of Akt at Ser473 is a marker of mTORC2 activity33,34, while phosphorylation of mTOR at Ser2448 serves as a biomarker for mTORC1 activity, as do its direct downstream targets, p70S6 kinase and 4EBP1. There is ongoing controversy regarding the kinase responsible for Ser2448 phosphorylation of mTOR, with some studies pointing to Akt, while other studies suggest that this site is being phosphorylated by its downstream substrate p70S6 as part of a positive feed-back loop35.

Nintedanib treatment of HT-29 cells resulted in a strong inhibition of phospho-Akt, accompanied by reduced phosphorylation of mTOR and its downstream target p70S6K. The inhibition of Akt phosphorylation agrees with what has been reported for Nintedanib-treated endothelial cells and pericytes36. It is also coherent with a role for mTORC2 as a downstream target of VEGF-signaling since Nintedanib is a potent inhibitor of the tyrosine kinase activity of VEGFR while mTORC2 is considered as the major Ser473 Akt kinase. The inhibition of p70S6K is in line with in vivo observations where Nintedanib treatment was accompanied by reduced levels of Ser240/Ser244-phosphorylated S6, a robust biomarker of p70S6K activity in CRC xenografts37,38.

In LoVo cells, Nintedanib treatment to initially induced by Akt activation, which was followed by a delayed inhibition at 48 h. The reduced levels of phospho-Akt at this time were accompanied by decreased phosphorylation of mTOR and its dowstreal target p70S6K.

Regorafenib exposure of HT-29 cells was accompanied by Akt activation in both HT-29 and LoVo cells which, unexpectedly, was accompanied by decreased levels of phospho-mTOR in both cellular models. However, the decreased levels of phospho-mTOR by 48 h is coherent with the late induction of autophagy in these cells.

A surprising finding was the universal upregulation of Deptor, a natural mTOR inhibitor that is part of both the mTORC1 and the mTORC2 complex. This upregulation was most pronounced in cells treated with Regorafenib. It is unclear if the upregulation of Deptor is a cause or an effect of the decreased mTOR activity. Deptor is negatively regulated by both mTORC1 and mTORC2 through phosphorylation which leads to ubiquitination and subsequent proteasomal degradation39. Therfore, decreased mTOR activity would be expected to result in increased Deptor protein levels. Conversely, Deptor expression can also be upregulated in response to several stress conditions leading to increased levels of Deptor and further decreased mTOR activity. An intriguing question is if the upregulation of Deptor is somehow linked to inhibition of VEGF-signaling, since both Nintedanib and Regorafenib are strong VEGFR inhibitors. Specifically, Deptor has been described as a negative regulator of endothelial cell migration40 while mTORC2 is a known downstream effector of VEGFR needed for endothelial cell assembly and angiogenesis41.

MAPK/Erk signaling is closely associated with the autophagosomal pathway. Erk can upregulate the cellular levels of LC3B and SQSTM1/p62 at the transcriptional level42. Furthermore, recent results suggest that ATG proteins, and LC3 in particular, can serve as scaffold for the organization of proteins involved in MAPK signaling thereby promoting the formation of active phosphorylated Erk43. Our results show that Nintedanib treatment of both HT-29 and LoVo cells was accompanied by activation of pERK. However, this was not accompanied by increased phosphorylation of any of its conventional down-stream targets, Thr359/Ser363 on p90RSK and Ser235/236 on the ribosomal protein S6. Nintedanib is known to inhibit the VEGFR, FGFR and PDGFR tyrosine kinase receptors thereby preventing their downstream signaling. At the same time, p90RSK is a protein that is phosphorylated on multiple sites by different kinases, most likely in a specific order. Therefore, if one of the kinases needed to phosphorylate p90RSK before the phosphorylation of the Thr359/Ser363 residues can take place is affected by Nintedanib, Erk would not be able to phosphorylate this site even when activated.

Phosphorylation of ribosomal protein S6 on Ser235/236 is generally considered to be mediated by p90RSK44. However, previous work with B-Raf inhibitors has revealed that this site can also be phosphorylated by p70S6K45, which is in agreement with the decreased phosphorylation observed here for both p70S6K and rpS6. Regorafenib was a strong inhibitor of Erk activation in both HT-29 and LoVo cells which was accompanied by attenuated phosphorylation of all the downstream targets including RSK, and rpS6.

Taken together, these results show that Nintedanib and Regorafenib have opposite effects on the early part of the Akt- and Erk-signaling pathways. Nintedanib inhibits Akt-signaling which is compensated by activation of Erk whereas Regorafenib is a strong inhibitor of Erk-signaling which is accompanied by Akt activation. These effects may be directly linked to the activity spectra of the two compounds since Nintedanib principally inhibits VEGFR, FGFR and FGFR36 whereas Regorafenib is a more wide-spectrum angiokinase inhibitor which also show strong activity toward both wt and mutant b-Raf as well as c-Raf which are all upstream of the Erk/MAPK cascade. However, whatever the upstream signaling, the phosphorylation of mTOR becomes attenuated at some point in both cell lines, which is most pronounced for HT-29 cells. One possible explanation for this would be because release of the inhibitory effect of the TSC complex on mTOR requires phosphorylation by both Akt- and Erk-signaling46. Another, not necessarily mutually exclusive, possibility is the drug-induced upregulation of Deptor. These observations are consistent with previously described feedback mechanisms linking PI3K/Akt and MAPK/ERK signaling pathways, which often operate in an opposing manner, as well as the well-documented feedback loop between both pathways47–50.

However, this approach has some caveats when applied to the study of protein kinase inhibitors in tumor cells. First, abnormalities in cell signaling pathways are ubiquitous in tumor cells and may become dominant in certain tumor types like alterations of the Wnt pathway in colorectal cancer and B-raf mutations in cutaneous melanoma51. These genetic alterations are likely to alter the signaling network compared to what has been described for normal cells under physiological conditions. Second, many biomarker proteins have multiple phosphorylation sites that are phosphorylated in tandem by different protein kinases. Therefore, even if the kinase responsible for a given phosphoepitope has been identified correctly, and the kinase of interest is active, this may not necessarily be apparent if some of the other kinases acting upstream on the same protein substrate are inhibited by the kinase inhibitor under study. Furthermore, studies of different kinase inhibitors have revealed the presence of alternative pathways or feed-back mechanisms that are not necessarily apparent under physiological conditions.

Most studies of receptor tyrosine kinase inhibitors report that these compounds induce cytoprotective autophagy. This was puzzling to us, since the most Nintedanib-sensitive cell line, HT-29, also displayed strong autophagy induction suggesting that the drug-induced autophagy might increase, rather than decrease, the cytotoxic effects. To clarify the impact of drug-induced autophagy, we inhibited autophagy by genetic (reduced expression of beclin1) and pharmacologically (3-MA) approaches. In agreement with the autophagy impairment, the Beclin1 knock down clones showed reduced formation of LC3II following exposure to both Nintedanib and Regorafenib. Long-time colony formation assays revealed that Beclin1 downregulation was accompanied by 2-3-fold increased resistance to Nintedanib whereas the sensitivity to Regorafenib remained stable. Pharmacological autophagy inhibition by 3-MA reduced the formation of LC3II following Nintedanib exposure in both cell lines. Similar results were obtained for LoVo cells although the effect of 3-MA on Nintedanib activity was less and only observed over a narrow dose range. In contrast, 3-MA had no influence on the activity of Regorafenib in both cell lines.

Taken together, we here characterize two angiokinases inhibitors, Nintedanib and Regorafenib, which exhibit clinical activity in the treatment of patients with metastatic colorectal cancer (CRC). The results revealed clear differences in the downstream pathways of the two agents as well as their capacity to induce autophagy, with strong induction of the mTOR inhibitor Deptor as the only common factor. Both genetic and pharmacological autophagy inhibition were accompanied by decreased toxicity of Nintedanib but did not affect Regorafenib. Therefore, Nintedanib-induced autophagy contributed to the cytotoxic effect while Regorafenib-induced autophagy had no apparent influence on cellular viability. These findings suggest that Nintedanib might be particularly active in cells/tumors with high autophagic flux while autophagy has little or no impact on the activity of Regorafenib.

Interestingly, recent organoid-based work52, also employed HT-29 and LoVo but reported an opposite role for autophagy in regorafenib sensitivity. It demonstrated that inhibition of autophagy enhanced sensitivity to Regorafenib, suggesting a pro-survival role of autophagy in this more physiologically relevant system. In contrast, our results obtained in 2D cell cultures indicate that Regorafenib induces a limited autophagic response that is insufficient to prevent cell cycle arrest and cell death. This apparent discrepancy may reflect fundamental differences between these experimental models, as organoids preserve tissue architecture, cellular heterogeneity, and microenvironmental cues such as oxygen and nutrient gradients, extracellular matrix interactions, and paracrine signaling, that can significantly impact the stress-adaptive pathways including autophagy. Furthermore, the functional role of autophagy is not only variable but also highly dependent on the strength and type of cellular stress. Thus, whereas in simplified 2D systems autophagy might mainly be a reaction to intrinsic signaling or metabolic stress, in organoid models the autophagy process is, to a certain extent, influenced by the limitations of the microenvironment and the heterogeneity of the phenotype, which may lead to autophagy playing a survival ‍‌role53–55.

Materials and methods

Drugs and chemicals

Nintedanib (BIBF 1120) was provided by Boehringer-Ingelheim and Regorafenib was purchased from Adooq Bioscience (# Bay73-4506). The stock solutions were prepared at 10 mM in DMSO. 3 methyladenine (3-MA) was purchased from Selleckchem (# S27672). DRAQ-5 (# 4084 S) was purchased from Cell Signaling.

Cell lines and culture conditions

LS513, FET, LS174T, SW48, DLD1, LIM1215 and LoVo cells were kindly provided by Richard Hamelin (Saint-Antoine Research Center, Paris, France). SW480 colon carcinoma cells were purchased from American Type Culture Collection (Rockville, MD). HT-29, and SW620 cells were generously provided by Richard Camalier, (National Cancer Institute, Bethesda, MD) while HCT-116 cells were a kind gift from Bert Vogelstein (John Hopkins, Baltimore, MD).

HCT-116 and SW620 cells were maintained in McCoy’s A, HT-29, LS513, FET, LS174T, SW48, LoVo, SW480, DLD1, RKO DMEM (Dulbecco modified Eagle’s medium) and LIM 1215 in RPMI 1640. All media were supplemented with 5% fetal calf serum and 1% penicillin/streptomycin. Upon defreezing, cells were monitored for mycoplasma contamination using the “MycoAlert™ mycoplasma detection kit (Lonza) and were replaced after 6 weeks in culture.

Antibodies

The following antibodies were obtained from Cell signaling and are directed against: LC3B (# 3868 S), Beclin1 (# 3738), SQSTM1/p62 (# 88588), Phospho-Ser235/236 S6 Ribosomal protein (# 2211 S), S6 Ribosomal protein (# 2217 S), Phospho-Ser371 p70S6 Kinase (# 9208), p70S6 Kinase (# 2708), Phospho-Thr359/Ser363 p90RSK (# 9344), p90RSK1/2/3 (# 9355), Phospho-Ser2448 mTOR (# 5536), mTOR (# 2983), Phospho-Ser209 elF4E (# 9741 S), elF4E (# 2069), phospho-Thr37/46 4E-BP1 (# 2855), 4E-BP1 (# 9644), phospho-Ser473 Akt (# 4060 S), Akt (# 9272 S), phospho-Thr202/Tyr204 Erk1/2 (# 4370 S), Erk (# 4695 S), p27Kip1 (# 3686), p21Waf1/Cip1 (# 2947) and β-Actin (# 5125 S). The secondary antibodies used for immunofluorescence were anti-rabbit IgG conjugated with Cy3, or anti-mouse IgG conjugated with Alexa fluor 448, both from Jackson ImmunoResearch.

MTT viability assay

Cellular viability was determined by the MTT (methylthiazolyldiphenyl-tetrazolium bromide) assay as described previously56 with minor modifications. The cells were seeded at a density of 5,000 per well on a 24-well plate. Cells were then exposed continuously (without media renewal) with different concentrations of Nintedanib (0, 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 10, 20, 50 µM) or Regorafenib (0, 0.1, 0.5, 1, 2, 3, 5, 6, 8, 10, 20, 40 µM) in the absence or presence of 3-methyladenine (2.5 mM) for 120 h followed by MTT determination. The IC50 value is defined as drug concentration inhibiting cell growth by 50% compared to untreated control cells. All values are averages of at least 3 independent experiments each done in duplicate.

Cell cycle analysis

Cells were plated at a density of 1 × 106 cells, and grown for 24 h prior to incubation with the IC50 dose of Nintedanib or Regorafenib for the indicated times. At the end of the treatment, cells were harvested and washed twice with PBS, re-suspended in 50 µl PBS and subsequently fixed in cold 70% ethanol added dropwise to the cell pellet under gentle vortexing, and then stored at -20 C° overnight. Cells were then rehydrated in PBS, and stained in PBS with propidium iodide (20 µg/ml) and ribonuclease A (100 µg/ml) for 30 min at room temperature in the dark followed by flow cytometry analysis. Cells were analyzed on a FACS Gallios (Beckman Coulter) with 20,000 events per sample and data analysis was performed using kaluza software.

Autophagy assay

HT-29 and LoVo cells were seeded at a density of 7 × 103 cells per well on 96-well black Plates, Clear Bottom, and were allowed to adhere overnight. Cells were then treated with the IC50 dose of Nintedanib or Regorafenib, and the plates were incubated at 37 °C with 10% CO2 for 24, 48–72 h. At the end of treatment, cell Meter™ Autophagy Assay Kit (AAT Bioquest # 23000-AAT) was prepared according to the manufacturer’s guidelines, and 100 µL of each reagent was added to each well followed by incubation for 1 h. The fluorescence was then determined on a microplate reader (Infinite pro-200, Tecan). The data were analysed, and expressed as percentage of the untreated cells (control). The experiments were repeated three times with duplicate samples.

Immunocytochemistry

HT-29 and LoVo cells (25 × 104) were plated on glass coverslip and incubated for 24 h followed by exposure to the IC50 dose of Nintedanib or Regorafenib for the indicated times (0–48 h). Cells were fixed for 15 min in 100% cold methanol on ice and the samples were saturated in PBS with 1% bovine serum albumin and 0.2% gelatine for 30 min at room temperature, followed by incubation with primary antibodies directed against LC3B or SQSTM1/p62 overnight at 4 C°. Incubation with secondary antibodies was carried out for 1 h at room temperature and the nuclei were then stained with DRAQ-5.

Confocal microscopy analyses were performed using a laser-scanning confocal microscope (model Leica SP2) at 60x magnification. All images were treated using Fiji software.

Western blot analysis

Western blot analysis was carried out as described previously (Escargueil et al., 2008; Mésange et al., 2014)37,57 with minor modifications. Cells (1.5 × 106) were plated in 60 mm dishes and cultured in the absence or presence of Nintedanib or Regorafenib at their respective IC50 dose. Cells were harvested after 0, 24, 48 h drug exposure, and cellular lysates were prepared in RIPA buffer (150 mM NaCl, 50 mM Tris pH 7.4, 0.1% SDS, 1% NP-40, 2 mM EDTA, 1 mM Na3VO4, 1 mM PMSF, 1 mM NaF) supplemented with protease inhibitors (Roche) according to the manufacturer’s instruction. Protein extracts were sonicated for 15 s, and protein concentrations were determined with a BCA assay (Thermo Scientific, Pierce) prior to dilution in 5x Laemmli loading buffer. Equal amounts of protein (30 µg) were resolved on SDS-PAGE gels and transferred onto PVDF or nitrocellulose membranes. Membranes were probed with the relevant antibodies followed by incubation with a horseradish peroxidase-conjugated secondary antibody (1/2000, Cell Signaling). Bands were revealed with an enhanced chemiluminescence detection system (Biorad) and visualized on a Chemidoc system58 Protein expression was quantified by densitometric analysis of the immunoblots using Image Lab software (Bio-Rad).

ShRNA targeting of autophagy gene BECN1

HT-29 cells were transfected with 10 µg of shBeclin1 vector or empty vector control using Lipofectamine 2000 (InVitrogen). After 48 h, cells were selected for resistance to puromycin (6 µg/ml).

Fresh puromycin-containing medium was added every 2–3 days, until individual resistant colonies could be identified and expanded. The silencing of Beclin1 was assessed by qRT-PCR and immunoblotting.

qRT-PCR analysis

Total RNA was extracted from HT-29 parental cells and HT-29-sh-Beclin1 clones with MRC Tri-reagent, according to the manufacturer’s instructions. RNA quantity and purity were determined by using a NanoDrop ND-1000 and 1 µg of total RNA from each sample was reverse transcribed using Revertaid H Minus First Strand cDNA Synthesis Kit (Thermo Scientific) followed by amplification of the DNA product with a SYBR Green kit (Promega). PCR primers were designed with the primer3 program and gene expression was normalized to beta-actin housekeeping gene as previously described (Ferrand et al., 2014). The threshold was set above the non-template control background and within the linear phase of target gene amplification to calculate the number of cycles at which the transcript was detected. Gene expression values were calculated based on the comparative delta CT method and normalized to the housekeeping gene beta-actin.

Colony formation assay

Cells (500–1000) were cultured in the absence or presence of different concentration of Nintedanib or Regorafenib for 14 days at 37 °C. Colonies were then fixed with 10% (v/v) methanol for 15 min and stained with 5% Giemsa (Sigma) for 30 min. The number of colonies per condition was determined, and the cell survival fraction was calculated in comparison with the corresponding untreated control. All values are averages of at least 3 independent experiments, each done in triplicate.

Statistical analysis

The statistical analysis of experimental data was performed using two-way ANOVA, and the results are presented as mean ± standard deviation (SD).

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 (264.4KB, pdf)
Supplementary Material 2 (4.5MB, pdf)

Acknowledgements

We are grateful to Romain Morichon and Annie Munier, CISA (Cytométrie Imagerie Saint-Antoine) at the Saint-Antoine Research Center for their competent assistance with confocal microscopy and flow cytometry analysis. We warmly thank Guido Lenz, Faculty of Universidade Federal do Rio Grande do Sul, Porto Alegre, Basil, who generously provided us with the short hairpin RNA for Beclin1 silencing. We specifically thank Paul Mésange for his early work on autophagy induction by Nintedanib. The authors would like to acknowledge Frank Hilberg from Boehringer Ingelheim for stimulating and fruitful discussions. This work was financed in part by research funding from Boehringer Ingelheim. Lila Louadj was supported by GERCOR and by the RSFN association, France.

Author contributions

L.L.: conceptualized, performed experiments, analyzed the data and wrote the manuscript. A.K.L.: conceptualized, analyzed the data, wrote the manuscript and acquired funding. G.G, and M.S: writing, review and editing.

Funding

This work was financed in part by research funding from Boehringer Ingelheim. Lila Louadj was supported by GERCOR and by the RSFN association, France.

Data availability

All data supporting the findings of this study are presented in the paper and the [Supplementary Information](https:/www.nature.com/articles/s41467-025-59605-5) . All data are available upon request to the corresponding autors L.L: **louadjlila@hotmail.fr.**.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Michèle Sabbah and Annette K. Larsen.

References

  • 1.Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J. Clin.71, 209–249. 10.3322/caac.21660 (2021). [DOI] [PubMed] [Google Scholar]
  • 2.Morgan, E. et al. Global burden of colorectal cancer in 2020 and 2040: incidence and mortality estimates from GLOBOCAN. Gut72, 338–344. 10.1136/gutjnl-2022-327736 (2023). [DOI] [PubMed] [Google Scholar]
  • 3.Burada, F. Autophagy in colorectal cancer: An important switch from physiology to pathology. WJGO7, 271. 10.4251/wjgo.v7.i11.271 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Glick, D., Barth, S. & Macleod, K. F. Autophagy: cellular and molecular mechanisms. J. Pathol.221, 3–12. 10.1002/path.2697 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Cooper, K. F. Till Death Do Us Part: The Marriage of Autophagy and Apoptosis. Oxidative Med. Cell. Longev.2018, 4701275. 10.1155/2018/4701275 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Vargas_The mechanisms. and roles of selective autophagy in mammals_combined.pdf n.d.
  • 7.Zhu, Y., Liu, F., Jian, F. & Rong, Y. Recent progresses in the late stages of autophagy. Cell. Insight. 3, 100152. 10.1016/j.cellin.2024.100152 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Khalafiyan, A. et al. Highlighting roles of autophagy in human diseases: a perspective from single-cell RNA sequencing analyses. Drug Discovery Today. 29, 104224. 10.1016/j.drudis.2024.104224 (2024). [DOI] [PubMed] [Google Scholar]
  • 9.Taucher, E., Mykoliuk, I., Fediuk, M., Smolle-Juettner, F-M. & Autophagy Oxidative Stress and Cancer Development. Cancers14, 1637. 10.3390/cancers14071637 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Saha, S., Panigrahi, D. P., Patil, S. & Bhutia, S. K. Autophagy in health and disease: A comprehensive review. Biomed. Pharmacother.104, 485–495. 10.1016/j.biopha.2018.05.007 (2018). [DOI] [PubMed] [Google Scholar]
  • 11.Udristioiu, A. & Nica-Badea, D. Autophagy dysfunctions associated with cancer cells and their therapeutic implications. Biomed. Pharmacother.115, 108892. 10.1016/j.biopha.2019.108892 (2019). [DOI] [PubMed] [Google Scholar]
  • 12.Jiang, P. & Mizushima, N. Autophagy and human diseases. Cell. Res.24, 69–79. 10.1038/cr.2013.161 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Chavez-Dominguez, R., Perez-Medina, M., Lopez-Gonzalez, J. S., Galicia-Velasco, M. & Aguilar-Cazares, D. The Double-Edge Sword of Autophagy in Cancer: From Tumor Suppression to Pro-tumor Activity. Front. Oncol.10, 578418. 10.3389/fonc.2020.578418 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.White, E. & DiPaola, R. S. The Double-Edged Sword of Autophagy Modulation in Cancer. Clin. Cancer Res.15, 5308–5316. 10.1158/1078-0432.CCR-07-5023 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Liang, X. H. et al. Induction of autophagy and inhibition of tumorigenesis by beclin 1. Nature402, 672–676. 10.1038/45257 (1999). [DOI] [PubMed] [Google Scholar]
  • 16.White, E. Deconvoluting the context-dependent role for autophagy in cancer. Nat. Rev. Cancer. 12, 401–410. 10.1038/nrc3262 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Chen, H-T. et al. Crosstalk between autophagy and epithelial-mesenchymal transition and its application in cancer therapy. Mol. Cancer. 18, 101. 10.1186/s12943-019-1030-2 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Akalay, I. et al. Epithelial-to-Mesenchymal Transition and Autophagy Induction in Breast Carcinoma Promote Escape from T-cell-Mediated Lysis. Cancer Res.73, 2418–2427. 10.1158/0008-5472.CAN-12-2432 (2013). [DOI] [PubMed] [Google Scholar]
  • 19.Duan, Y. et al. Role of autophagy on cancer immune escape. Cell. Commun. Signal.19, 91. 10.1186/s12964-021-00769-0 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Gewirtz, D. A. The Four Faces of Autophagy: Implications for Cancer Therapy. Cancer Res.74, 647–651. 10.1158/0008-5472.CAN-13-2966 (2014). [DOI] [PubMed] [Google Scholar]
  • 21.Kim, Y. C. & Guan, K-L. mTOR: a pharmacologic target for autophagy regulation. J. Clin. Invest.125, 25–32. 10.1172/JCI73939 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Aveic, S. et al. Autophagy inhibition improves the cytotoxic effects of receptor tyrosine kinase inhibitors. Cancer Cell. Int.18, 63. 10.1186/s12935-018-0557-4 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Gewirtz, D. A. When cytoprotective autophagy isn’t… and even when it is. Autophagy10, 391–392. 10.4161/auto.27719 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Gewirtz, D. A. The Switch between Protective and Nonprotective Autophagy; Implications for Autophagy Inhibition as a Therapeutic Strategy in Cancer. Biology9, 12. 10.3390/biology9010012 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Tanaka, H. et al. Comparison of autophagy inducibility in various tyrosine kinase inhibitors and their enhanced cytotoxicity via inhibition of autophagy in cancer cells in combined treatment with azithromycin. Biochem. Biophys. Rep.22, 100750. 10.1016/j.bbrep.2020.100750 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Shi, Y-H. et al. Targeting autophagy enhances sorafenib lethality for hepatocellular carcinoma via ER stress-related apoptosis. Autophagy7, 1159–1172. 10.4161/auto.7.10.16818 (2011). [DOI] [PubMed] [Google Scholar]
  • 27.Selvakumaran, M., Amaravadi, R. K., Vasilevskaya, I. A. & O’Dwyer, P. J. Autophagy Inhibition Sensitizes Colon Cancer Cells to Antiangiogenic and Cytotoxic Therapy. Clin. Cancer Res.19, 2995–3007. 10.1158/1078-0432.CCR-12-1542 (2013). [DOI] [PubMed] [Google Scholar]
  • 28.Ikeda, T. et al. Inhibition of Autophagy Enhances Sunitinib-Induced Cytotoxicity in Rat Pheochromocytoma PC12 cells. J. Pharmacol. Sci.121, 67–73. 10.1254/jphs.12158FP (2013). [DOI] [PubMed] [Google Scholar]
  • 29.Grothey, A. et al. Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): an international, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet381, 303–312. 10.1016/S0140-6736(12)61900-X (2013). [DOI] [PubMed] [Google Scholar]
  • 30.Boland, P. M. et al. A phase I/II study of nintedanib and capecitabine for refractory metastatic colorectal cancer. JNCI Cancer Spectr.8, pkae017. 10.1093/jncics/pkae01 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Poindessous, V. et al. EGFR- and VEGF(R)-Targeted Small Molecules Show Synergistic Activity in Colorectal Cancer Models Refractory to Combinations of Monoclonal Antibodies. Clin. Cancer Res.17, 6522–6530. 10.1158/1078-0432.CCR-11-1607 (2011). [DOI] [PubMed] [Google Scholar]
  • 32.Lauzier, A. et al. Colorectal cancer cells respond differentially to autophagy inhibition in vivo. Sci. Rep.9, 11316. 10.1038/s41598-019-47659-7 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Jacinto, E. et al. SIN1/MIP1 Maintains rictor-mTOR Complex Integrity and Regulates Akt Phosphorylation and Substrate Specificity. Cell127, 125–137. 10.1016/j.cell.2006.08.033 (2006). [DOI] [PubMed] [Google Scholar]
  • 34.Guertin, D. A. et al. Ablation in Mice of the mTORC Components raptor, rictor, or mLST8 Reveals that mTORC2 Is Required for Signaling to Akt-FOXO and PKCα, but Not S6K1. Dev. Cell. 11, 859–871. 10.1016/j.devcel.2006.10.007 (2006). [DOI] [PubMed] [Google Scholar]
  • 35.Figueiredo, V. C., Markworth, J. F. & Cameron-Smith, D. Considerations on mTOR regulation at serine 2448: implications for muscle metabolism studies. Cell. Mol. Life Sci.74, 2537–2545. 10.1007/s00018-017-2481-5 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hilberg, F. et al. BIBF 1120: Triple Angiokinase Inhibitor with Sustained Receptor Blockade and Good Antitumor Efficacy. Cancer Res.68, 4774–4782. 10.1158/0008-5472.CAN-07-6307 (2008). [DOI] [PubMed] [Google Scholar]
  • 37.Mésange, P. et al. Intrinsic bevacizumab resistance is associated with prolonged activation of autocrine VEGF signaling and hypoxia tolerance in colorectal cancer cells and can be overcome by Nintedanib, a small molecule angiokinase inhibitor. Oncotarget510.18632/oncotarget.1671 (2014). [DOI] [PMC free article] [PubMed]
  • 38.Zhang, Y. & Zheng, X. F. S. mTOR-independent 4E-BP1 phosphorylation is associated with cancer resistance to mTOR kinase inhibitors. Cell. Cycle. 11, 594–603. 10.4161/cc.11.3.19096 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Gao, M. et al. Aberrant modulation of ribosomal protein S6 phosphorylation confers acquired resistance to MAPK pathway inhibitors in BRAF-mutant melanoma. Acta Pharmacol. Sin. 40, 268–278. 10.1038/s41401-018-0020-z (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Bruneau, S., Nakayama, H., Woda, C. B., Flynn, E. A. & Briscoe, D. M. DEPTOR regulates vascular endothelial cell activation and proinflammatory and angiogenic responses. Blood122, 1833–1842. 10.1182/blood-2013-03-488486 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wang, S. et al. Regulation of Endothelial Cell Proliferation and Vascular Assembly through Distinct mTORC2 Signaling Pathways. Mol. Cell. Biol.35, 1299–1313. 10.1128/MCB.00306-14 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Kim, J-H. et al. Raf/MEK/ERK can regulate cellular levels of LC3B and SQSTM1/p62 at expression levels. Exp. Cell Res.327, 340–352. 10.1016/j.yexcr.2014.08.001 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Martinez-Lopez, N., Athonvarangkul, D., Mishall, P., Sahu, S. & Singh, R. Autophagy proteins regulate ERK phosphorylation. Nat. Commun.4, 2799. 10.1038/ncomms3799 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Roux, P. P. et al. RAS/ERK Signaling Promotes Site-specific Ribosomal Protein S6 Phosphorylation via RSK and Stimulates Cap-dependent Translation. J. Biol. Chem.282, 14056–14064. 10.1074/jbc.M700906200 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Corcoran, R. B. et al. TORC1 Suppression Predicts Responsiveness to RAF and MEK Inhibition in BRAF- Mutant Melanoma. Sci. Transl Med. 5. 10.1126/scitranslmed.3005753 (2013). [DOI] [PMC free article] [PubMed]
  • 46.Winter, J. N., Jefferson, L. S. & Kimball, S. R. ERK and Akt signaling pathways function through parallel mechanisms to promote mTORC1 signaling. Am. J. Physiology-Cell Physiol.300, C1172–C1180. 10.1152/ajpcell.00504.2010 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Alimonti, L. et al. Inhibition of mTORC1 leads to MAPK pathway activation through a PI3K-dependent feedback loop in human cancer. J. Clin. Invest.118 (9), 3065–3074. 10.1172/JCI34739 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Turke, A. B. et al. MEK inhibition leads to PI3K/AKT activation by relieving a negative feedback on ERBB receptors. Cancer Res.72 (13), 3228–3237. 10.1158/0008-5472.CAN-11-3747 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Stulpinas, A. et al. Crosstalk between protein kinases AKT and ERK1/2 in human lung tumor-derived cell models. Front. Oncol.12, 1045521. 10.3389/fonc.2022.1045521 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Tufail, M. et al. Hallmarks of cancer resistance. iScience27, 6109979. 10.1016/j.isci.2024.109979 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Cancer Genome Atlas Network 2012, (2015).
  • 52.Agostini, G. et al. Autophagy inhibition improves sensitivity to the multi-kinase inhibitor regorafenib in preclinical mouse colon tumoroids. Front. Cell. Dev. Biol.13, 1631116. 10.3389/fcell.2025.1631116 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Pienkowski, K. W. P. T., Tankiewicz-Kwedlo, A., Ciborowski, M., Kurek, K. & Pawlak, D. Differences in treatment outcome between translational platforms in developing therapies for gastrointestinal cancers,European. J. Pharmacol. Volume. 99110.1016/j.ejphar.2025.177309 (2025). [DOI] [PubMed]
  • 54.Isinelli, G., Failla, S., Plebani, R. & Prete A.Exploring oncology treatment strategies with tyrosine kinase inhibitors through advanced 3D models. Med. Int.5 (2), 13. 10.3892/mi.2024.212 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Bingel, C. et al. Three-dimensional tumor cell growth stimulates autophagic flux and recapitulates chemotherapy resistance. Cell. Death Dis.8, e3013. 10.1038/cddis.2017.398 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Soares, D. G. et al. Replication and homologous recombination repair regulate DNA double-strand break formation by the antitumor alkylator ecteinascidin 743. Proceedings of the National Academy of Sciences. ;104:13062–7. (2007). 10.1073/pnas.0609877104 [DOI] [PMC free article] [PubMed]
  • 57.Escargueil, A. E. et al. Influence of irofulven, a transcription-coupled repair-specific antitumor agent, on RNA polymerase activity, stability and dynamics in living mammalian cells. J. Cell Sci.121, 1275–1283. 10.1242/jcs.023259 (2008). [DOI] [PubMed] [Google Scholar]
  • 58.Ferrand, N. et al. WISP1/CCN4 inhibits adipocyte differentiation through repression of PPARγ activity. Sci. Rep.7, 1749. 10.1038/s41598-017-01866-2 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (264.4KB, pdf)
Supplementary Material 2 (4.5MB, pdf)

Data Availability Statement

All data supporting the findings of this study are presented in the paper and the [Supplementary Information](https:/www.nature.com/articles/s41467-025-59605-5) . All data are available upon request to the corresponding autors L.L: **louadjlila@hotmail.fr.**.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES