Non-canonical Activation of RSK1 Induces EphA2-Mediated Cell Migration under Cellular Stress Conditions

Abstract

p90 ribosomal S6 kinase 1 (RSK1) regulates various cellular events involved in cell growth and migration. We previously demonstrated that RSK1 catalyzes ephrin receptor A2 (EphA2) phosphorylation at Ser-897 to promote cancer cell migration, and that this pathway is regulated independently by extracellular signal-regulated kinase (ERK) and mitogen-activated protein kinase (MAPK)-activated protein kinase 2 (MK2). Although the activation mechanism of RSK1 via ERK has been extensively examined, the mechanism for MK2 remains unclear. In the present study, we showed that MK2-mediated phosphorylation at Ser-380 in the linker region, a key mechanism of RSK1 activation, was dependent on the basal phosphorylation of Ser-221 in the N-terminal kinase domain. This basal phosphorylation was catalyzed by 3-phosphoinositide-dependent kinase 1 (PDK1) and was independent of ERK-catalyzed Ser-380 phosphorylation. The PDK1–MK2–RSK1–EphA2 axis promoted glioblastoma cell migration induced by temozolomide, a chemotherapeutic agent. Collectively, these results reveal a novel activation mechanism of RSK1 in cancer malignancy.

INTRODUCTION

p90 ribosomal S6 kinase 1 (RSK1) is a downstream kinase of extracellular signal-regulated kinase (ERK) that regulates various cellular events involved in the cell growth and migration of various carcinomas.^1–3) RSK1 contains two distinct kinase domains: the carboxyl-terminal kinase (CTK) and amino-terminal kinase (NTK). In the canonical activation of RSK1 by growth factors, ERK phosphorylates Thr-573 and Thr-359/Ser-363 (Supplementary Fig. S1). Phosphorylation at Thr-573 induces the activation of CTK, which in turn phosphorylates Ser-380 in the linker region between CTK and NTK. This phosphorylation creates a docking site for 3-phosphoinositide-dependent kinase 1 (PDK1), which subsequently phosphorylates Ser-221 in NTK to activate downstream signaling.

We recently reported that the non-canonical activation of RSK1, in which mitogen-activated protein kinase (MAPK)-activated protein kinase 2 (MK2), a molecule downstream of p38, regulates RSK1 under cellular stress conditions.⁴⁾ MK2 catalyzes the phosphorylation of Ser-380 to activate NTK in a CTK-independent manner. This non-canonical p38-MK2-RSK1 pathway promotes ephrin receptor A2 (EphA2) phosphorylation at Ser-897 (pS-EphA2) to trigger cancer cell migration.

This study examined the phosphorylation of RSK1 at Ser-221 (pS-221) to determine its role in non-canonical activation. Basal pS-221, rather than inducible phosphorylation, plays a pivotal role in this activation mechanism.

MATERIALS AND METHODS

Antibodies and Reagents

Total antibodies against EphA2 (#6997) and ALK (#3633), and phospho-specific antibodies against EphA2 (Ser-897; #6347), ALK (Tyr-1604; #3341), RSK1 (Ser-380, cross-reacting with RSK2 Ser-386; #11989; Thr-359/Ser-363; #9344), p38 (Thr-180/Tyr-182; #4511), MK2 (Thr-334; #3007), and ERK (Thr-202/Tyr-204; #9101) were purchased from Cell Signaling Technology; antibodies against total ERK (C-9), p38α (A-12), RSK1 (C-21), MK2 (A-11), heat shock protein 27 (HSP27) (F-4), and β-actin (C-4) were obtained from Santa Cruz Biotechnology (Dallas, TX, U.S.A.); antibodies against FLAG (F1804), SB203580, and TMZ were purchased from Merck KGaA (Darmstadt, Germany); phospho-antibodies against RSK1 (Ser-221)/RSK2 (Ser-227) (#AF892) and HSP27 (Ser-78/Ser-82; #MAB23141), as well as recombinant human epidermal growth factor (EGF), were obtained from R&D Systems; a phos-tag ligand and anisomycin were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan); MK2 inhibitor III and GSK2334470 were obtained from MedChemExpress (Monmouth Junction, NJ, U.S.A.); and trametinib was purchased from AdooQ BioScience (Irvine, CA, U.S.A.). All chemical inhibitors were dissolved in dimethyl sulfoxide.

Cell Cultures

HeLa and HEK293 cells (Manassas, VA, U.S.A.) were cultured in Dulbecco’s modified Eagle’s medium (Shimadzu Diagnostics, Tokyo, Japan) supplemented with 10% fetal calf serum (Merck KGaA), 2 mM l-glutamine (Thermo Fisher Scientific, Waltham, MA, U.S.A.), 100 U/mL penicillin (Meiji Seika Pharma, Tokyo, Japan), and 100 μg/mL streptomycin (Meiji Seika Pharma) at 37°C in 5% CO₂. U87-MG cells (provided by Dr. Tsuneo Imanaka, University of Toyama, Toyama, Japan) were cultured in Eagle’s minimum essential medium (Shimadzu Diagnostics) supplemented with 10% fetal calf serum, 2 mM l-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in 5% CO₂. Before the drug treatment, HeLa cells were starved in the culture medium containing 1% fetal calf serum for 24 h.

RNA Interference

HeLa cells were transfected with small interfering RNAs at a final concentration of 20–100 nM using Lipofectamine RNAiMAX (Thermo Fisher Scientific), according to the manufacturer’s instructions. The target sequences were as follows:

p38α, 5′-GCAUUACAACCAGACAGUUGAUAUU-3′

Negative control, 5′-UAAUGUACUGCGCGUGGAGAGGAA-3′

Transfection of Plasmid DNAs

The expression vectors for FLAG-tagged human RSK1, RSK2, and echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase (EML4-ALK) variant 5a were previously reported.^5–7) The expression vectors for enhanced green fluorescent protein (EGFP)-tagged kinase-dead EphA2 (EphA2-KD-EGFP), RSK1 with CTK-dead (CTKm: K447R), CTKm-S221A, and CTKm-S380A mutants, and the constitutively active (CA) mutation of p38α (D176A and F327L) were previously generated.⁴⁾ The expression vectors for RSK1 containing the S380A, S221A, and CTKm-T359S363AA (TSAA) mutants, as well as RSK2 with the CTKm (K451R) mutant, were generated by RT-PCR using KOD-Plus-Neopolymerase. HEK293 cells were transfected using Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s instructions.

Immunoblotting, Zn²⁺ Phos-Tag Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis, and Migration Assay

Experiments were performed as previously reported.^4,8) Band intensities were quantified using ImageJ software (NIH). Full scans of immunoblotting data are shown in Supplementary Fig. S5.

Statistical Analysis

Statistical analyses were performed using JMP software (SAS Institute, Cary, NC, U.S.A.). Assay values are shown as the mean ± standard deviation (S.D.) or as box-and-whisker plots. The significance of differences was analyzed using the Tukey–Kramer honestly significant difference test or the Wilcoxon rank-sum test with exact p-values. p-Values <0.05 were considered to be significant.

RESULTS

Different Roles of Ser-221 Phosphorylation in RSK1 Activation via ERK and MK2

We previously reported that EGF and the protein synthesis inhibitor anisomycin induce the RSK1-EphA2 signaling pathway mainly via ERK and MK2, respectively.⁴⁾ Serum-starved HeLa cells were stimulated with EGF or anisomycin, and RSK1 phosphorylation at Ser-221 (pS-221) and Ser-380 (pS-380) was detected by Western blot analysis. EGF enhanced both pS-380 and pS-221, resulting in NTK-mediated pS-EphA2 (Fig. 1A). The increase in pS-221 was blocked by the ression of ERK using the mitogen-activated protein kinase kinase (MEK) inhibitor trametinib (Fig. 1B). On the other hand, anisomycin induced pS-380 and pS-EphA2, whereas pS-221 showed only a modest increase (Fig. 1A and Supplementary Fig. 2). In addition, pS-221 was largely unaffected by chemical inhibition of MK2 (MK2-IN-3) and p38 (SB203580), or by the knockdown of p38α (Figs. 1C–1E). Furthermore, the pS-221/pS-380 ratio was high under EGF stimulation but significantly lower with anisomycin treatment (Fig. 1F). These results indicate the different roles of pS-221 in RSK1 activation via ERK and MK2.

Fig. 1. Different Roles of Ser-221 Phosphorylation in RSK1 Activation via ERK and MK2

(A–E) Serum-starved HeLa cells were stimulated with 100 ng/mL EGF for 10 min or 50 μM anisomycin for 20 min (A). Serum-starved HeLa cells were treated with 0.03 μM trametinib, 10 μM MK2-IN-3, or 10 μM SB203580 for 30 min, and then stimulated with 100 ng/mL EGF for 10 min or 50 μM anisomycin (Ani) for 30 min (B–D). HeLa cells were transfected with siRNAs against p38α or a negative control. At 72 h post-transfection, serum-starved cells were stimulated with 50 μM anisomycin for 30 min (E). Whole-cell lysates were immunoblotted with the indicated primary antibodies. (F) The intensities of pS-221 and pS-380 bands were quantified, and the ratio of the increase in pS-221 relative to that in pS-380 was calculated. Data are plotted as the mean ± S.D. of 3 (EGF treatment for 10 min) or 5 (anisomycin treatment for 20–30 min) independent experiments. ^*p < 0.05, as determined by the Wilcoxon rank-sum test with exact p-values.

Fig. 2. PDK1 Induces Basal pS-221

(A) Serum-starved HeLa cells were stimulated with 100 ng/mL EGF for 10 min or 50 μM anisomycin (Ani) for 20 min. Whole-cell lysates were separated by Zn²⁺ phos-tag SDS-PAGE, followed by immunoblotting with the indicated primary antibodies. (B) Serum-starved HeLa cells were treated with 10 μM GSK2334470 for 1 h. Whole-cell lysates were immunoblotted with the indicated primary antibodies. (C) HEK293 cells were transfected with expression vectors for FLAG-tagged RSK1 (WT), the Ser-380-to-Ala substitution (S380A), the CTK dead mutant (CTKm), and CTKm with the S380A mutant (CTKm-S380A). At 24 h post-transfection, whole-cell lysates were immunoblotted with the indicated primary antibodies.

PDK1-Induced Basal pS-221 Is Crucial for the Induction of pS-380 via MK2

To further investigate the relationship between pS-221 and pS-380, we performed Zn²⁺ phos-tag immunoblotting, which enabled the separation of proteins based on their phosphorylation status, visualized as distinct band shifts. In the non-stimulated state, RSK1 was detected at positions (b) and (c) (Fig. 2A). Some bands at position (b) contained pS-221, indicating that pS-221 partially occurred at the basal level. EGF induced a significant band shift, particularly at position (a), containing both pS-221 and pS-380. However, upon anisomycin stimulation, the main bands were still at positions (b) and (c), whereas only some of the bands shifted to position (a) with both pS-221 and pS-380. Based on this result, together with those shown in Fig. 1, basal pS-221 appeared to be necessary for pS-380 upon the activation of MK2. To prove this hypothesis, we attempted to identify the kinase responsible for basal pS-221. PDK1 has been shown to catalyze pS-221 of pS-380-RSK1 upon the activation of ERK. The PDK1 inhibitor GSK2334470 blocked basal pS-221 in HeLa cells (Fig. 2B). In addition, constitutive pS-221 was observed in HEK293 cells overexpressing wild-type RSK1 (WT), but not in a mutant with the Ser-380-to-Ala substitution (S380A) (Fig. 2C, left). We also employed the CTK-dead mutant of RSK1 (CTKm), which lacked classical ERK-mediated RSK1 activation.⁴⁾ Similar to the WT, pS-221 was still observed in this mutant, and basal pS-221 was reduced in the CTKm-S380A mutant (Fig. 2C, right). These results indicated that PDK1 constitutively phosphorylates a part of RSK1 at Ser-221 under basal conditions.

We then clarified the effects of pS-221 on pS-380. The activation of ERK, not p38, by the EML4-ALK fusion protein strongly induced canonical RSK1 activation sequentially via pS-380 and pS-221 (Fig. 3A). Therefore, pS-380 was not blocked by the PDK1 inhibitor because preceding pS-380 provided PDK1 docking to catalyze pS-221. In contrast, p38α-CA, which activates the non-canonical p38–MK2 axis to induce pS-380, was unable to increase pS-221 in RSK1-CTKm (Fig. 3B). Most importantly, GSK2334470 blocked pS-380, suggesting that basal pS-221 affected pS-380 upon the activation of MK2. Similar results were obtained for the mutant RSK1 with the Ser-221-to-Ala substitution (S221A) (Figs. 3C, 3D). Another important isoform of the RSK family is RSK2.⁹⁾ We and others previously demonstrated that MK2 phosphorylates RSK2 at Ser-386, which corresponds to Ser-380 of RSK1.^4,10,11) Interestingly, active p38 promoted RSK2 phosphorylation at not only Ser-386 but also Ser-227, indicating that the activation mechanism of RSK1 and RSK2 via MK2 is different even though they have high homology in NTK, CTK, and the linker region (Fig. 3E). Taken together, these results indicate that ERK promoted pS-380 to induce pS-221, whereas basal pS-221 promoted MK2-induced pS-380 (Fig. 3F).

Fig. 3. Basal pS-221 Is Crucial for the Induction of pS-380 via MK2

(A–E) HEK293 cells were transfected with the indicated expression vectors for FLAG-tagged RSK1 (WT or CTKm), EML4-ALK, FLAG-tagged constitutively active p38α (p38α-CA-FLAG), and/or an empty vector, and at 24 h post-transfection, cells were treated with 10 μM GSK2334470 (A, B). HEK293 cells were transfected with the indicated expression vectors for FLAG-tagged RSK1 (WT or CTKm with Ser-221-to-Ala substitution), EML4-ALK, p38α-CA-FLAG, and/or an empty vector (C, D). HEK293 cells were transfected with the indicated expression vectors for FLAG-tagged RSK1 or RSK2 (CTKm), p38α-CA-FLAG, and/or an empty vector (E). At 24 h post-transfection, whole-cell lysates were immunoblotted with the indicated primary antibodies. n.s., Nonspecific band. (F) A schematic diagram of the activation mechanism of RSK via ERK (canonical activation, upper) and MK2 (non-canonical activation, lower).

Basal pS-221-Dependent pS-380 Promotes pS-EphA2 to Induce Cell Motility

EphA2 is one of the substrates of RSK1.^{4,6,8,12–15)} We previously reported that pS-EphA2 induced cell migration through the p38-MK2-RSK1 pathway.⁴⁾ We herein elucidated the contribution of pS-221 to cell migration. Kinase-dead EphA2, RSK1-CTKm, and p38α-CA induced pS-EphA2 and enhanced cell migration, both of which were blocked by GSK2334470 (Figs. 4A, 4B). Similar results were obtained for the expression of the RSK1-S221A mutant (Figs. 4C, 4D). Collectively, these results suggest that basal pS-221 was required for MK2 to induce pS-380, which triggered the activation of NTK, resulting in pS-EphA2-induced cell motility.

Fig. 4. Basal pS-221-Dependent pS-380 Promotes pS-EphA2-Mediated Cell Motility

HEK293 cells were transfected with the expression vectors for EGFP-tagged kinase-dead EphA2 (EphA2-KD-EGFP), p38α-CA-FLAG, RSK1-CTKm-FLAG (WT or S221A), and/or an empty vector. At 24 h post-transfection, cells were treated with 10 μM GSK2334470 for 2 h (A, B). Whole-cell lysates were immunoblotted with the indicated primary antibodies (A, C). Cell migration was observed using a time-lapse imaging system for 2 h (B, D). The accumulated distance of cell migration (μm) was calculated and is shown as box-and-whisker plots. ^*p < 0.05, as determined by the Tukey–Kramer HSD test.

pT-359/pS-363 Is Associated with the Non-canonical Activation of RSK1

Thr-359 are Ser-363 are the major RSK1 phosphorylation sites mediated by ERK.¹⁾ Previous reports showed that the phosphorylation (pT-359/pS-363) was not involved in ERK-induced RSK1 activation.¹⁶⁾ We found that anisomycin also induced pT-359/pS-363 (Fig. 5A). Zn²⁺ phos-tag immunoblotting showed that the bands at position (a) contained pT-359/pS-363 with pS-380 and pS-221, suggesting that pT-359/pS-363 regulates MK2-mediated RSK1 activation (Fig. 5B). Therefore, we generated mutant RSK1 with the Thr-359/Ser-363-to-Ala substitution (TSAA). This mutant was not phosphorylated at Ser-380 and failed to promote pS-EphA2 and cell migration (Figs. 5C, 5D). In contrast, pS-221 was unaffected by the TSAA mutation (Fig. 5C), and pT-359/pS-363 was also not inhibited in the S221A mutant (Supplementary Fig. S3), indicating the absence of crosstalk between pS-221 and pT-359/S363 in the non-canonical activation of RSK1. These results indicate the distinct roles of pT-359/pS-363 in the activation of RSK1 via the ERK and MK2 pathways.

Fig. 5. pT-359/pS-363 Is Associated with the Non-canonical Activation of RSK1

(A, B) Serum-starved HeLa cells were stimulated with 100 ng/mL EGF for 10 min or 50 μM anisomycin (Ani) for 20–30 min. Whole-cell lysates were separated by normal or Zn²⁺ phos-tag SDS-PAGE, followed by immunoblotting with the indicated primary antibodies. (C, D) HEK293 cells were transfected with the expression vectors for EphA2-KD-EGFP, p38α-CA-FLAG, RSK1-CTKm-FLAG (WT or T359/S363-to-A/A substitution), and/or an empty vector. At 24 h post-transfection, whole-cell lysates were immunoblotted with the indicated primary antibodies (C). Cell migration was observed using a time-lapse imaging system for 2 h (D). The accumulated distance of cell migration (μm) was calculated and is shown as box-and-whisker plots. ^*p < 0.05, as determined by the Tukey–Kramer HSD test.

TMZ Induces Glioblastoma Cell Migration via Basal pS-221-Dependent pS-EphA2

The alkylating chemotherapeutic agent temozolomide (TMZ) provokes cellular stress, triggering the p38-MK2 pathway to promote RSK1 (pS-380)-dependent pS-EphA2, which enhances the motility of human glioblastoma U87-MG cells.^4,17) Similar to anisomycin-treated HeLa cells, pS-221 was not induced in TMZ-treated U87-MG cells (Fig. 6A). Unexpectedly, pT-359/pS-363 was also not promoted (data not shown). Treatment with TMZ induced pS-EphA2 and enhanced cell migration, both of which were blocked by GSK2334470 (Figs. 6A, 6B). These results demonstrate that basal pS-221-dependent pS-EphA2 also promotes cell motility under physiological conditions.

Fig. 6. TMZ Induces Glioblastoma Cell Migration via Basal pS-221-Dependent pS-EphA2

U87-MG cells were treated with 100 μM temozolomide (TMZ) for 48 h and then with 10 μM GSK2334470 for 2 h. Whole-cell lysates were immunoblotted with the indicated primary antibodies (A). Cell migration was observed using a time-lapse imaging system for 2 h (B). The accumulated distance of cell migration (μm) was calculated and is shown in box-and-whisker plots. ^*p < 0.05, as determined by the Tukey–Kramer HSD test. TMZ: temozolomide.

DISCUSSION

We herein showed that under cellular stress conditions, basal pS-221 was crucial for the phosphorylation of Ser-380 by MK2 in non-canonical RSK1 activation. Since PDK1 is a constitutively activated enzyme, its binding to substrates is the most important step for their phosphorylation.¹⁸⁾ pS-380 serves as the hydrophobic motif for PDK1 docking to RSK1.¹⁹⁾ As shown in Fig. 2C, the S380A substitution abolished basal pS-221, suggesting that Ser-380 itself has potential as a hydrophobic motif even though its hydrophobic effect is markedly weaker than that of pS-380-RSK1 (Fig. 3B). The main limitation of the present study is that we did not elucidate the mechanisms by which pS-221 initiates pS-380 via MK2. We speculated that pS-221 enables active MK2 to access RSK1. To validate our hypothesis, a docking simulation of RSK1 by AlphaFold2 with active MK2 was performed, and one model showed that the activation loop of MK2 was located near Ser-221 of RSK1, suggesting that the phosphorylation of Ser-221 affected MK2-RSK1 binding (Supplementary Fig. S4). Since the linker region around Ser-380 contains multiple phosphorylation sites,²⁰⁾ this region is structurally flexible; therefore, difficulties are associated with elucidating its crystal structure. Further studies are needed to clarify the specific regulatory mechanism after the real RSK1–MK2 complex has been confirmed.

We previously reported that the induction of pS-EphA2 via anisomycin was weaker than that of EGF, and the present results revealed the underlying mechanisms.⁴⁾ As shown in Fig. 2A, the strong band shift of RSK1 was induced upon EGF stimulation, indicating that the majority of RSK1 was activated. In contrast, only some of the shifted bands were induced upon anisomycin treatment because a small amount of RSK1 possessed basal pS-221. Position (b) contained several bands, implying that other phosphorylation sites induced at the basal level also regulated RSK1 activation via MK2. Even though the induction of pS-EphA2 by MK2 was not as strong as that by ERK, it efficiently initiated cell migration (Figs. 4, 6). Therefore, the combination therapy of anti-cancer drugs with an MK2 inhibitor has the potential to completely block pS-EphA2-initiated cancer malignancy.

In conclusion, we herein demonstrated that basal pS-221-induced RSK1 via PDK1 was phosphorylated by MK2 at Ser-380 under cellular stress conditions, which resulted in the phosphorylation of EphA2 at Ser-897 and ultimately increased cell motility. pT-359/pS-363 may be associated with the noncanonical activation of RSK1 as well. These results provide a more detailed understanding of the regulation of RSK1. In addition, NTK of RSK1 is a member of the AGC kinase family. Akt and S6 kinase, which also belong to the AGC family, are known to be phosphorylated by PDK1.^21,22) Further studies are needed to clarify whether the regulatory mechanisms discovered in the present study are conserved in these molecules, which will provide novel insights into the activation mechanisms of AGC kinases.

Acknowledgments

We are grateful to Drs. Hiroyuki Mano (National Cancer Center Research Institute), Kunio Matsumoto (Kanazawa University), Yoshikazu Sugimoto (Keio University), Kazuhiro Katayama (Nihon University), and Tsuneo Imanaka (University of Toyama) for providing materials.

Funding

This work was supported by JSPS KAKENHI Grants (Nos. 22K06612 to Y.Z., 25K09937 to Y.Z., and 23K24026 to H.S.), a JST Moonshot R&D Grant (No. JPMJMS2021 to H.S.), the Pharmacodynamics Research Foundation (Y.Z.), the Takeda Science Foundation (Y.Z., H.S.), and the Extramural Collaborative Research Grant of the Cancer Research Institute, Kanazawa University (Y. Z., Seiji Yano, and H. S.).

Author Contributions

Yue Zhou: Writing—original draft, writing—review and editing, project administration, conceptualization, formal analysis, funding acquisition, investigation, and methodology. Fang Zhang: Writing—original draft, writing—review and editing, and investigation. Akihiro Tanaka and Takahiro Yokota: Writing—review and editing and investigation. Satoru Yokoyama: Writing—review and editing and supervision. Seiji Yano: Writing—review and editing and funding acquisition. Hiroaki Sakurai: Writing—review and editing, supervision, project administration, conceptualization, and funding acquisition.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Carriere A, Ray H, Blenis J, Roux PP. The RSK factors of activating the Ras/MAPK signaling cascade. Front. Biosci., 13, 4258–4275 (2008).
• 2)  Anjum R, Blenis J. The RSK family of kinases: emerging roles in cellular signalling. Nat. Rev. Mol. Cell Biol., 9, 747–758 (2008).
• 3)  Lara R, Seckl MJ, Pardo OE. The p90 RSK family members: Common functions and isoform specificity. Cancer Res., 73, 5301–5308 (2013).
• 4)  Zhou Y, Oki R, Tanaka A, Song L, Takashima A, Hamada N, Yokoyama S, Yano S, Sakurai H. Cellular stress induces non-canonical activation of the receptor tyrosine kinase EphA2 through the p38-MK2-RSK signaling pathway. J. Biol. Chem., 299, 104699 (2023).
• 5)  Katayama K, Yoshioka S, Tsukahara S, Mitsuhashi J, Sugimoto Y. Inhibition of the mitogen-activated protein kinase pathway results in the down-regulation of P-glycoprotein. Mol. Cancer Ther., 6, 2092–2102 (2007).
• 6)  Zhang F, Zhou Y, Hamada N, Tanaka A, Yokoyama S, Yano S, Matsumoto K, Mano H, Sakurai H. Stress response kinase MK2 induces non-canonical activation of EphA2 in EML4-ALK lung cancer cells. Biol. Pharm. Bull., 48, 172–176 (2025).
• 7)  Takeuchi K, Choi YL, Soda M, Inamura K, Togashi Y, Hatano S, Enomoto M, Takada S, Yamashita Y, Satoh Y, Okumura S, Nakagawa K, Ishikawa Y, Mano H. Multiplex reverse transcription-PCR screening for EML4-ALK fusion transcripts. Clin. Cancer Res., 14, 6618–6624 (2008).
• 8)  Zhou Y, Yamada N, Tanaka T, Hori T, Yokoyama S, Hayakawa Y, Yano S, Fukuoka J, Koizumi K, Saiki I, Sakurai H. Crucial roles of RSK in cell motility by catalysing serine phosphorylation of EphA2. Nat. Commun., 6, 7679 (2015).
• 9)  Lannigan DA. ERK1/2-RSK2 Signaling in Regulation of ERα-Mediated Responses. Endocrinology (United States), 163, (2022).
• 10)  Zaru R, Ronkina N, Gaestel M, Arthur JSC, Watts C. The MAPK-activated kinase Rsk controls an acute Toll-like receptor signaling response in dendritic cells and is activated through two distinct pathways. Nat. Immunol., 8, 1227–1235 (2007).
• 11)  Zaru R, Edgar AJ, Hanauer A, Watts C. Structural and functional basis for p38-MK2-activated Rsk signaling in toll-like receptor-stimulated dendritic cells. Mol. Cell. Biol., 35, 132–140 (2015).
• 12)  Song L, Yang F, Wang Z, Yang L, Zhou Y. Ginsenoside Rg5 inhibits cancer cell migration by inhibiting the nuclear factor-κB and erythropoietin-producing hepatocellular receptor A2 signaling pathways. Oncol. Lett., 21, 452 (2021).
• 13)  Yonehara K, Zhou Y, Takahashi JI, Yokoyama S, Tomihara K, Noguchi M, Sakurai H. RSK-mediated non-canonical activation of EphA2 by tamoxifen. Biol. Pharm. Bull., 45, 162–168 (2022).
• 14)  Yasuta Y, Kaminaka R, Nagai S, Mouri S, Ishida K, Tanaka A, Zhou Y, Sakurai H, Yokoyama S. Cooperative function of oncogenic MAPK signaling and the loss of Pten for melanoma migration through the formation of lamellipodia. Scientific Reports, 14, 1525 (2024).
• 15)  Zhou Y, Sakurai H. Emerging and diverse functions of the EphA2 noncanonical pathway in cancer progression. Biol. Pharm. Bull., 40, 1616–1624 (2017).
• 16)  Utepbergenov D, Hennig PM, Derewenda U, Artamonov MV, Somlyo AV, Derewenda ZS. Bacterial expression, purification and in vitro phosphorylation of full-length ribosomal S6 kinase 2 (RSK2). PLOS ONE, 11, e0164343 (2016).
• 17)  Haryuni RD, Tanaka T, Takahashi JI, Onuma I, Zhou Y, Yokoyama S, Sakurai H. Temozolomide induces endocytosis of EGFRvIII via p38-mediated non-canonical phosphorylation in glioblastoma cells. Biol. Pharm. Bull., 44, 1681–1687 (2021).
• 18)  Casamayor A, Morrice NA, Alessi DR. Phosphorylation of Ser-241 is essential for the activity of 3-phosphoinositide-dependent protein kinase-1: identification of five sites of phosphorylation in vivo. Biochem. J., 342, 287–292 (1999).
• 19)  Frödin M, Jensen CJ, Merienne K, Gammeltoft S. A phosphoserine-regulated docking site in the protein kinase RSK2 that recruits and activates PDK1. EMBO J., 19, 2924–2934 (2000).
• 20)  Hornbeck PV, Zhang B, Murray B, Kornhauser JM, Latham V, Skrzypek E. PhosphoSitePlus, 2014: mutations, PTMs and recalibrations. Nucleic Acids Res., 43 (D1), D512–D520 (2015).
• 21)  Ikuta M, Kornienko M, Byrne N, Reid JC, Mizuarai S, Kotani H, Munshi SK. Crystal structures of the N-terminal kinase domain of human RSK1 bound to three different ligands: implications for the design of RSK1 specific inhibitors. Protein Sci., 16, 2626–2635 (2007).
• 22)  Biondi RM. Phosphoinositide-dependent protein kinase 1, a sensor of protein conformation. Trends Biochem. Sci., 29, 136–142 (2004).