Identification of Inhibitors for eIF5A2-Dependent Translation Elongation by Monitoring the Translational Efficiency of Polyproline Motif in Mitochondrial Fission Regulator 1

Abstract

We previously reported that eukaryotic translation initiation factor 5A2 (eIF5A2), rather than eIF5A1, is important for the proliferation of HeLa S3 and MDA-MB-231 cells, despite the 84% amino acid sequence identity between eIF5A1 and eIF5A2. In addition, individual upregulated genes, including mitochondrial fission regulator 1 (MTFR1), which has a proline-rich motif, by eIF5A2 were different from those in eIF5A1. Thus, eIF5A2-dependent translational elongation is a promising target for cancer treatment with minimal side effects. In this study, we constructed a high-throughput screening system to identify eIF5A2-dependent translation elongation inhibitors by monitoring the translation efficiency of the MTFR1-luciferase fusion protein in HeLa S3 cells. Orlistat and andrographolide (AGP) were suggested as inhibitors of eIF5A2-dependent translation elongation among 1744 compounds from libraries. In addition to the findings related to AGP, the present study revealed that orlistat and the silencing of eIF5A2 suppressed the invasive activity of MDA-MB-231 cells. Downregulation of heparanase 1 expression, but not of matrix metalloproteinase-2 (MMP2) and MMP9, by eIF5A2 silencing was similar to that by treatment with orlistat and AGP, suggesting that orlistat and AGP were, at least in part, capable of attenuating eIF5A2-dependent translation elongation through the repression of eIF5A2 expression. Based on these observations, monitoring the translational efficiency of MTFR1 synthesis may be useful for identifying new eIF5A2 inhibitors.

INTRODUCTION

Eukaryotic translation initiation factor 5A (eIF5A) is the only hypusinated protein at a specific lysine residue (Lys⁵⁰ in humans) using the polyamine spermidine and plays an important role in the proliferation of eukaryotic cells. eIF5A is a homolog of elongation factor P and is adjacent to the P site of tRNA, overlapping with the E site in 80S ribosomes. Moreover, its hypusine (N^e-(4-amino-2-hydroxybutyl)lysine) residue interacts with the CCA end of the P-site tRNA,^1–3) stimulating the translational elongation of polyproline or other difficult-to-translate peptide motifs and termination.^4,5) Two EIF5A genes, eIF5A1 and eIF5A2, are conserved in most eukaryotes, with amino acid (AA) sequence identities of 84% in humans and 82% in mice.^6,7) eIF5A1 and eIF5A2 show similar hypusination efficiencies mediated by deoxyhypusine synthase and protein stability.⁸⁾ Notably, eIF5A1 is ubiquitously expressed, and its homozygous knockout shows embryonic lethality, whereas eIF5A2 is expressed only in the brain and testes of mice and is not essential for normal development and viability.^9,10) Although human eIF5A2 has gained attention as an oncogene in cancer development and progression,⁷⁾ it remains unclear why eIF5A1 and eIF5A2 are highly conserved with different functions in eukaryotes.

We previously found that protein synthesis of eIF5A2 was stimulated by polyamines via the inhibition of miR6514-5p function and that silencing eIF5A2, but not eIF5A1, suppressed the proliferation of HeLa S3 and breast cancer cell lines.¹¹⁾ The genes, including MTFR1, which has a proline-rich motif, that were upregulated by eIF5A2 were different from those upregulated by eIF5A1.¹¹⁾ Furthermore, the expression levels of ribosomal proteins, including RPL22L1,^12–14) RPS27A,¹⁵⁾ and RPL36A,^16,17) which are associated with cancer malignancy, were upregulated by polyamines, and the affinity of RPL10 for eIF5A2 was distinct from that of eIF5A1.¹¹⁾ Considering the alterations in ribosomal composition and upregulation of eIF5A2 expression in various cancers,^18,19) it is possible that the set of genes required for cancer cell growth is regulated by onco-ribosomes and eIF5A2 during translation elongation.¹¹⁾ Thus, eIF5A2-dependent translational elongation is a potential target for cancer treatment with minimal side effects. In this study, a high-throughput luciferase (Luc) screening assay system using MTFR1 was used to identify a selective inhibitor of eIF5A2-dependent translation elongation.

MATERIALS AND METHODS

Materials

Screen-Well U. S. Food and Drug Administration (FDA) approved drug library V2 (BML-2843J Version 1.3: 765 drugs), Screen-Well New ICCB (Institute of Chemistry and Cell Biology) known Bioactives Library (L2000-Z752018: 478 compounds), and Screen-Well Natural Product Library (BML-2865 v.7.4 Rev.07-Mar-12: 501 compounds) were obtained from Enzo Life Sciences, Inc. (Farmingdale, NY, U.S.A.). Orlistat and andrographolide (AGP) were obtained from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). α-Solanine was purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.).

Plasmid Construction

pMTFR1-Luc was purchased from Vector Builders, Inc. (Chicago, IL, U.S.A.). Overlap extension PCR was performed using the primer set 5′- TGGTACCATAGCACTGGGGCTCCACCAAAG-3′ (forward) and 5′-GCCCCAGTGCTATGGTACCAAATGTGGTAG-3′ (reverse), and pMTFR1-Luc was used as a template to generate pMTFR1 Δpolyproline-rich sequence (PRS)-Luc. Plasmid sequences were confirmed using DNA sequencing (Eurofins Genomics K.K., Tokyo, Japan).

High-Throughput Screening of the eIF5A2 Inhibitors from the Drug Library and Western Blotting

HeLa S3 and MDA-MB-231 cells were cultured as described previously.¹¹⁾ Transfection of pMTFR1-Luc and pMTFR1-Luc ΔPRS to HeLa S3 cells was performed according to the method by Fukumoto et al.,²⁰⁾ with minor modifications. Transfected cells (5.0 × 10³ cells of HeLa S3 in 50 μL of Dulbecco’s Modified Eagle Medium [DMEM]/well) were inoculated into a cell culture microplate (96-well; Greiner Bio-One GmbH, Frickenhausen, Germany) and further cultured for 24 h in the presence of 5 μM of compounds. Luc activity was measured as described previously.²¹⁾ Anti-matrix metalloproteinase-9 (MMP9) (10375-2-AP) and anti-MMP2 (10373-2-AP) antibodies were purchased from Proteintech (Rosemont, IL, U.S.A.). The anti-heparanase 1 (HPSE1) antibody (ab254254) was purchased from Abcam (Cambridge, U.K.). Western blotting for eIF5A1, eIF5A2, MTFR1, and β-actin was performed as described previously.¹¹⁾

Invasion Assay

To examine the effect of drug candidates on invasive activity, MDA-MB-231 cells (1.0 × 10⁵ cells/mL) were cultured for 2 d in the presence of drugs at the specified concentrations. Silencing of eIF5A1 and eIF5A2, and mRNA measurements to check the knockdown efficiency, were performed as described previously.¹¹⁾ Forty microliters of Matrigel Basement Membrane Matrix Growth Factor Reduced (Catalog No. 356230; Corning Inc., Corning, NY, U.S.A.) diluted to 0.25 mg/mL in DMEM was coated onto the top membrane surface of Chemotaxicell (8-μm pore size; Kurabo Industries Ltd., Osaka, Japan) and incubated for 2 h at 37°C and 5% CO₂. After coating, the Chemotaxicell was transferred to a 24-well plate containing 600 μL of DMEM supplemented with 10% fetal bovine serum and orlistat at the specified concentrations. MDA-MB-231 cells cultured with or without orlistat at the specified concentrations for 48 h were seeded at 2.0 × 10⁵ cells in 200 μL of DMEM, with or without orlistat, onto the Chemotaxicell upper layer. After 24 h of culture at 37°C and 5% CO₂, cells infiltrating beneath the Chemotaxicell membrane were fixed for 15 min in 4% paraformaldehyde-phosphate-buffered saline solution. Residual cells and medium on the membrane surface were removed using a cotton swab. Cells that infiltrated beneath the membrane were stained for 30 min with a 0.2% Crystal Violet Water Solution (catalog no. 15192; Muto Pure Chemicals Co., Ltd., Tokyo, Japan). After washing with water and removing the residual staining solution from the membrane surface with a cotton swab, the samples were observed using a BZ-X800 microscope (Keyence Corp., Osaka, Japan).

Statistical Analysis

The Molecular Taxonomy of Breast Cancer International Consortium dataset (n = 2509)^22,23) was downloaded from cBioportal (https://www.cbioportal.org/) on September 19, 2025. Prognostic analyses were performed using the Kaplan–Meier method as previously described.^24,25) The patients were classified into high- and low-EIF5A2 expression groups. Receiver-operating characteristic curves were constructed using disease-specific survival data, and the Youden index was used to determine the optimal cutoff value. Kaplan–Meier survival curves were plotted using the R software version 4.4.1 (R Core Team, R Foundation for Statistical Computing, Vienna, Austria).

RESULTS

Identification of the Potential Inhibitors for eIF5A2-Dependent Translation Elongation

The human MTFR1 gene (NM_001145838), which encodes hMTFR1, consists of 348 AAs, and a polyproline motif is present within the sequence of 151–164 AAs. The protein synthesis of MTFR1 was upregulated by eIF5A2 rather than eIF5A1 in HeLa S3 cells¹¹⁾; therefore, the MTFR1 coding region (1–1044 bp) was fused to Luc reporter genes to construct the plasmid pMTFR1-Luc (Fig. 1A). HeLa S3 cells transfected with pMTFR1-Luc were inoculated into a 96-well microplate and further cultured with 5 μM of 765 compounds from an FDA-approved drug library, 478 compounds from a new ICCB Known Bioactives Library, and 501 compounds from a Natural Products Compound Library. After 24 h, the cells were mixed with a luciferin solution containing Triton X-100, and Luc activity was measured at 562 nm. Luc activity was decreased by >75% by 17 FDA-approved drugs, 24 ICCB known bioactives, and 35 natural products (Figs. 1B–1D and Supplementary Table 1).

Fig. 1. High-Throughput Screening of Inhibitors for the Protein Synthesis of the MTFR1-Luc Fusion Protein in HeLa S3 Cells

(A) Structure of pMTFR1-Luc. (B–D) Effects of FDA-approved drugs (B), New ICCB known bioactives (C), and natural products (D) on the translational efficiency of the MTFR1-Luc fusion protein. High-throughput screening by luciferase assay was performed as described under “Materials and Methods.” The compounds showed 75% inhibition compared to the control.

To increase the specificity for the selection process for potential eIF5A2 inhibitors, the coding region of PRS in the MTFR1 gene was deleted to establish pMTFR1ΔPRS-Luc (Fig. 2A). Although deletion of the polyproline motif did not affect the translational efficiency of the MTFR1-Luc protein (Fig. 2B), we continued to search for drug candidates. When HeLa S3 cells transfected with pMTFR1ΔPRS-Luc were cultured with orlistat, AGP, and α-solanine, inhibition of Luc activity was partially recovered compared with that of other candidates (Figs. 2C–2E).

Fig. 2. Effects of Potential Inhibitors on the Protein Synthesis of the MTFR1ΔPRS-Luc Fusion Protein in HeLa Cells

(A) Structure of pMTFR1Δ-Luc. (B) Luciferase activities of HeLa cells transfected with pMTFR1-Luc or pMTFR1ΔRPS-Luc. (C–E) Effects of 17 FDA-approved drugs (C), 25 New ICCB known bioactives (D), and 37 natural products (E) on the translational efficiency of MTFR1ΔRPS-Luc fusion protein. High-throughput screening by luciferase assay was performed as described under “Materials and Methods.” PRS: polyproline-rich sequence; ns: not significant.

Cell growth of HeLa S3 and MDA-MB-231 cells was inhibited by orlistat, AGP, and α-solanine in a dose-dependent manner (Fig. 3A). Expression levels of eIF5A1, eIF5A2, and MTFR1 in HeLa S3 cells were investigated using Western blotting and quantitative PCR. As a result, the levels of eIF5A1, eIF5A2, and MTFR1 proteins were decreased by orlistat and AGP, but not by α-solanine (Fig. 3B). The levels of eIF5A1 mRNA in HeLa S3 cells treated with orlistat and AGP were almost the same as those in untreated HeLa S3 cells (Fig. 3C). Increased eIF5A2 mRNA levels were observed despite the reduction of eIF5A2 protein levels following orlistat treatment (Figs. 3B and 3C). Considering that elevated levels of eIF5A2 mRNA caused by eIF5A1 silencing did not influence the protein level,¹¹⁾ eIF5A2 expression appears to be strictly regulated at the translational level or via protein degradation. Taken together, these results suggest that orlistat and AGP are capable of inhibiting eIF5A-dependent translational elongation by downregulating eIF5A expression.

Fig. 3. Orlistat and AGP, but Not α-Solanine, Suppressed eIF5A1 and eIF5A2 Expressions

(A) Effect of orlistat, AGP, and α-solanine on cell growth of HeLa S3 and MDA-MB-231 cells. HeLa S3 cells (1.0 × 10⁵ cells/mL) and MDA-MB-231 cells (1.0 × 10⁵ cells/mL) were cultured in DMEM medium with 10% FBS. After 48 h, cell numbers of untreated HeLa S3 and MDA-MB-231 cells increased to 1.05 × 10⁶ and 4.09 × 10⁵ cells/mL, respectively. (B) Effects of 100 μM orlistat, 25 μM AGP, and 15 μM α-solanine on the protein levels of eIF5A1, eIF5A2, and MTFR1 in HeLa S3 cells. Cell-lysate protein (20 μg) was loaded for Western blot analysis. (C) Effects of 100 μM orlistat, 25 μM AGP, and 15 μM α-solanine on the mRNA levels of eIF5A1 and eIF5A2 in HeLa S3 cells. *p < 0.05; ns: not significant; AGP: andrographolide.

Orlistat Suppressed the Invasive Activity of MDA-MB-231 Cells by Downregulating HPSE1 Expression

Orlistat, an FDA-approved pancreatic lipase inhibitor, exhibits anticancer activity by inactivating the fatty acid synthase, which is upregulated in malignant cells with altered lipid metabolism.²⁶⁾ AGP, a natural terpenoid from Andrographis paniculata, has numerous therapeutic activities, including anticancer, anti-inflammatory, antioxidant, and protective effects on the bones and organs.²⁷⁾ Although AGP inhibits proliferation and invasion of several cancer cell lines, including MDA-MB-231 cells, by targeting phosphatidylinositol 3-kinase/AKT, extracellular signal-regulated kinase, nuclear factor-κB, and hypoxia-inducible factor-1 mediated MMP expression,^28–33) the effect of orlistat on the cell invasion of MDA-MB-231 has not been examined. We examined whether orlistat inhibits the invasive activity of MDA-MB-231 cells using Matrigel containing collagen and heparan sulfate, which are major components of the basement membrane. Orlistat suppressed the invasiveness of MDA-MB-231 cells in a dose-dependent manner (Fig. 4A). MMP2, MMP9, and HPSE1, a mammalian endo-β-d-glucuronidase that cleaves the glycosidic bond of heparan sulfate,^34–36) are required for invasive activity; therefore, the effects of orlistat and AGP on their expression, together with eIF5As, were examined. The expression levels of the eIF5A1 and eIF5A2 proteins were suppressed by orlistat and AGP in MDA-MB-231 cells (Fig. 4B). Interestingly, orlistat and AGP suppressed the protein levels of HPSE1 and proHPSE1, and orlistat decreased the level of MMP9 (Fig. 4C).

Fig. 4. Decreased Levels of MMP2 and HPSE1 by Orlistat and AGP Resulted in the Suppression of Invasive Activity of MDA-MB-231

(A) Effect of orlistat on the invasive activity of MDA-MB-231 cells. (B, C) Effects of 100 μM orlistat, 10 μM AGP, and 5 μM α-solanine on the expression levels of eIF5A1 and eIF5A2 (B), MMP9, MMP2, and HPSE1 (C) in MDA-MB-231 cells. Cell-lysate protein (20 μg) was loaded for Western blot analysis. Scale bar, 500 μm. AGP: andrographolide.

Silencing of eIF5A2 but Not eIF5A1 Suppressed the Invasive Activity of MDA-MB-231 Cells by Downregulated HPSE1 Expression

Effects of eIF5A silencing on the cell invasion of MDA-MB-231 cells were examined. Silencing eIF5A2, but not eIF5A1, reduced the invasive activity (Figs. 5A and 5B). Notably, silencing of eIF5A2 failed to downregulate the expression of MMP2 and MMP9 proteins, although MMP2 and MMP9 are required for the invasion of MDA-MB-231 cells³⁶⁾ (Fig. 5C) and regulation of their proteins by eIF5A2 in other cancer cell lines was reported.^37,38) Additionally, the expression levels of HPSE1 and proHPSE1 proteins were significantly decreased by eIF5A2 silencing (Fig. 5C), while the expression level of HPSE1 mRNA was almost similar to that in MDA-MB-231 cells with or without eIF5As silencing (Fig. 5D), suggesting that the expression of HPSE1 was, at least in part, regulated by eIF5A2 during translation elongation. Further studies are needed to clarify the detailed regulatory mechanism of HPSE1 expression by eIF5A2; however, the regulation of invasive activity and gene expression by eIF5A2 silencing is similar to that of orlistat and AGP. Based on these observations, we concluded that orlistat and AGP suppress the invasive activity via suppression of eIF5A2 expression; furthermore, monitoring MTFR1 synthesis is useful for identifying new eIF5A2 inhibitors.

Fig. 5. Silencing of eIF5A2 Suppressed the Invasive Activity of MDA-MB-231 Cells via Decreased Levels of MMP2 and HPSE1 Proteins

(A) Effect of eIF5As silencing on the invasive activity of MDA-MB-231 cells. (B) Levels of eIF5A1 and eIF5A2 proteins in MDA-MB-231 cells treated with siRNAs. (C) Effect of silencing of eIF5A1 or eIF5A2 on the expression level of MMP9, MMP2, and HPSE1 in MDA-MB-231 cells. (D) Effect of silencing of eIF5As on the expression level of HPSE1 mRNA. HPSE1 mRNA was measured as described previously.⁴⁸⁾ Scale bar, 500 μm. ns: not significant.

DISCUSSION

The human eIF5A2 isoform is frequently upregulated in 27 types of cancers and is a useful prognostic marker.^19,39) We also observed that eIF5A2 expression significantly affected the survival of patients with luminal B breast cancer treated with endocrine therapy (Supplementary Fig. S1). Thus, eIF5A2 and cancer-associated translational elongation are promising targets for developing cancer therapies with minimal side effects. However, the high AA identity at the N-terminal coding region, which contains the hypusinated Lys⁵⁰ residue, between eIF5A1 and eIF5A2 makes it difficult to develop a selective eIF5A2 inhibitor. Recently, girolline has been reported to be a sequence-selective modulator of eIF5A⁴⁰⁾; however, its selectivity for eIF5A inhibition remains unclear.

In this study, orlistat and AGP inhibited eIF5A2-dependent translation elongation in the PRS by reducing the expression levels of eIF5A1 and eIF5A2 proteins, likely due to translational regulation¹⁹⁾ or protein degradation^41,42) (Figs. 3B and 3C). Notably, the decrease in eIF5A2 protein levels by orlistat and AGP was stronger than that of eIF5A1, despite the expression level of eIF5A1 being higher than that of eIF5A2. Similarly, a reduction in eIF5A1 and eIF5A2 expression was observed when HeLa S3 and breast cancer cell lines were treated with α-difluoromethylornithine.¹¹⁾ Several reports have shown that eIF5A2 expression is upregulated by hypoxia,⁴³⁾ reactive oxygen species-related pathways,⁴⁴⁾ transforming growth factor-β signaling,⁴⁵⁾ AKT signaling,⁴⁶⁾ and androgen receptor,⁴⁷⁾ which exacerbate cancer prognosis. However, eIF5A1 protein expression was observed in all human cancer cell lines, whereas eIF5A2 expression was cell type-specific, despite eIF5A2 mRNA being constitutively expressed.⁸⁾ Interestingly, eIF5A2 mRNA has 4 different polyadenylation signals that form multiple mRNAs (0.7–5.6 kb) that encode identical proteins, and cis-elements in the 5′- or 3′-untranslated region (UTR) may trigger inefficient translation.⁸⁾ In fact, eIF5A2 expression is regulated by multiple long noncoding RNAs and microRNAs (miRNAs), and the dysregulation of their function is involved in cancer progression.¹⁹⁾ We previously found that the translation initiation of eIF5A2 mRNA is suppressed by miR6514-5p via interaction with the 5′-UTR, and that polyamines inhibit miRNA functions, thereby stimulating eIF5A2 synthesis.¹¹⁾ However, miR6514-5p showed no influence on eIF5A1 expression in HeLa S3 cells.¹¹⁾ miRDB (https://mirdb.org/index.html) suggests that there are common binding sites for miR1207-3p and miR3663-3p in the 5′-UTR and coding region, and for 8 miRNAs in the 3′-UTR of both eIF5A1 and eIF5A2 mRNA. Experiments are in progress to clarify the suppression of eIF5A1 and eIF5A2 synthesis by orlistat and AGP.

We also found that eIF5A2 silencing suppressed HPSE1 synthesis during translational elongation (Figs. 5C and 5D). Although the tripeptide motif (PPP), which induces ribosome stalling,⁵⁾ is found in the N-terminal domain of the HPSE1 protein, the detailed regulatory mechanism remains unclear. Further experiments are underway to clarify the regulatory mechanisms of HPSE1 expression by eIF5A2.

In conclusion, we identified orlistat and AGP as inhibitors of eIF5A2-dependent translational elongation of proline-rich motifs via suppression of eIF5A2 expression. In addition, the suppression of eIF5A2 expression by orlistat impeded HPSE1 expression, resulting in a reduction in the invasive activity of MDA-MB-231 cells. These findings reveal a new therapeutic potential for orlistat and AGP. Monitoring MTFR1 synthesis will be useful for identifying new eIF5A2 inhibitors.

DECLARATIONS

Funding

This study was supported in part by a Grant-in-Aid for Scientific Research (C) (No. 23K06139) from the Japan Society for the Promotion of Science (to K.H.) and an Extramural Collaborative Research Grant from the Cancer Research Institute, Kanazawa University (to K.H.).

Author Contributions

Conceptualization: K.H. Funding acquisition: K.H. Investigation: M.S., M.K., R.Y., A.M., M.E., and H.K. Resources: K.M., K.A., C.T., and K.I. Writing—original draft: M.S., M.K., K.A., and K.H. Writing—review and editing: M.S. and K.H. Visualization: M.S. and K.H. Supervision: K.M., K.A., C.T., and K.I.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Schmidt C, Becker T, Heuer A, Braunger K, Shanmuganathan V, Pech M, Berninghausen O, Wilson DN, Beckmann R. Structure of the hypusinylated eukaryotic translation factor eIF-5A bound to the ribosome. Nucleic Acids Res., 44, 1944–1951 (2016).
• 2)  Holm M, Natchiar SK, Rundlet EJ, Myasnikov AG, Watson ZL, Altman RB, Wang HY, Taunton J, Blanchard SC. mRNA decoding in human is kinetically and structurally distinct from bacteria. Nature, 617, 200–207 (2023).
• 3)  Beppu T, Shirahata A, Samejima K. Determination of covalently bound hypusine and deoxyhypusine to protein using submilligram of protein samples by HPLC. Biol. Pharm. Bull., 19, 1–5 (1996).
• 4)  Schuller AP, Wu CC, Dever TE, Buskirk AR, Green R. eIF5A functions globally in translation elongation and termination. Mol. Cell, 66, 194–205.e5 (2017).
• 5)  Pelechano V, Alepuz P. eIF5A facilitates translation termination globally and promotes the elongation of many non polyproline-specific tripeptide sequences. Nucleic Acids Res., 45, 7326–7338 (2017).
• 6)  Barba-Aliaga M, Alepuz P. Role of eIF5A in mitochondrial function. Int. J. Mol. Sci., 23, 1284 (2022).
• 7)  Wu GQ, Xu YM, Lau ATY. Recent insights into eukaryotic translation initiation factors 5A1 and 5A2 and their roles in human health and disease. Cancer Cell Int., 20, 142 (2020).
• 8)  Clement PM, Johansson HE, Wolff EC, Park MH. Differential expression of eIF5A-1 and eIF5A-2 in human cancer cells. FEBS J., 273, 1102–1114 (2006).
• 9)  Nishimura K, Lee SB, Park JH, Park MH. Essential role of eIF5A-1 and deoxyhypusine synthase in mouse embryonic development. Amino Acids, 42, 703–710 (2012).
• 10)  Pällmann N, Braig M, Sievert H, Preukschas M, Hermans-Borgmeyer I, Schweizer M, Nagel CH, Neumann M, Wild P, Haralambieva E, Hagel C, Bokemeyer C, Hauber J, Balabanov S. Biological relevance and therapeutic potential of the hypusine modification system. J. Biol. Chem., 290, 18343–18360 (2015).
• 11)  Suzuki M, Suzuki T, Nakano Y, Matsumoto K, Manaka H, Komeno M, Tamori S, Sato A, Dohmae N, Akimoto K, Miyazaki S, Suda T, Toida T, Kashiwagi K, Igarashi K, Higashi K. Polyamines stimulate the protein synthesis of the translation initiation factor eIF5A2, participating in mRNA decoding, distinct from eIF5A1. J. Biol. Chem., 301, 110453 (2025).
• 12)  Rao S, Peri S, Hoffmann J, Cai KQ, Harris B, Rhodes M, Connolly DC, Testa JR, Wiest DL. RPL22L1 induction in colorectal cancer is associated with poor prognosis and 5-FU resistance. PLOS ONE, 14, e0222392 (2019).
• 13)  Zhang D, Zhou Y, Ma Y, Jiang P, Lv H, Liu S, Mu Y, Zhou C, Xiao S, Ji G, Liu P, Zhang N, Sun D, Sun H, Wu N, Jin Y. Ribosomal protein L22-like1 (RPL22L1) mediates sorafenib sensitivity via ERK in hepatocellular carcinoma. Cell Death Discov., 8, 365 (2022).
• 14)  Liang Z, Mou Q, Pan Z, Zhang Q, Gao G, Cao Y, Gao Z, Pan Z, Feng W. Identification of candidate diagnostic and prognostic biomarkers for human prostate cancer: RPL22L1 and RPS21. Med. Oncol., 36, 56 (2019).
• 15)  Wang H, Yu J, Zhang L, Xiong Y, Chen S, Xing H, Tian Z, Tang K, Wei H, Rao Q, Wang M, Wang J. RPS27a promotes proliferation, regulates cell cycle progression and inhibits apoptosis of leukemia cells. Biochem. Biophys. Res. Commun., 446, 1204–1210 (2014).
• 16)  Kim JH, You KR, Kim IH, Cho BH, Kim CY, Kim DG. Over-expression of the ribosomal protein L36a gene is associated with cellular proliferation in hepatocellular carcinoma. Hepatology, 39, 129–138 (2004).
• 17)  Chen TW, Chang KP, Cheng CC, Chen CY, Hong SW, Sie ZL, Cheng HW, Yen WC, Huang Y, Liu SC, Wang CI. Characterization of recurrent relevant genes reveals a novel role of RPL36A in radioresistant oral squamous cell carcinoma. Cancers, 13, 5623 (2021).
• 18)  Xu Y, Ruggero D. The Role of Translation Control in Tumorigenesis and Its Therapeutic Implications. Annu. Rev. Cancer Biol., 4, 437–457 (2020).
• 19)  Xiong X, Du Y, Liu P, Li X, Lai X, Miao H, Ning B. Unveiling EIF5A2: A multifaceted player in cellular regulation, tumorigenesis and drug resistance. Eur. J. Pharmacol., 997, 177596 (2025).
• 20)  Fukumoto Y, Obata Y, Ishibashi K, Tamura N, Kikuchi I, Aoyama K, Hattori Y, Tsuda K, Nakayama Y, Yamaguchi N. Cost-effective gene transfection by DNA compaction at pH 4.0 using acidified, long shelf-life polyethylenimine. Cytotechnology, 62, 73–82 (2010).
• 21)  Kotani H, Tanabe H, Mizukami H, Makishima M, Inoue M. Identification of a naturally occurring rexinoid, honokiol, that activates the retinoid X receptor. J. Nat. Prod., 73, 1332–1336 (2010).
• 22)  Curtis C, Shah SP, Chin SF, et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature, 486, 346–352 (2012).
• 23)  Pereira B, Chin SF, Rueda OM, et al. The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes. Nat. Commun., 7, 11479 (2016).
• 24)  Onaga C, Tamori S, Matsuoka I, Ozaki A, Motomura H, Nagashima Y, Sato T, Sato K, Xiong Y, Sasaki K, Ohno S, Akimoto K. High expression of SLC20A1 is less effective for endocrine therapy and predicts late recurrence in ER-positive breast cancer. PLOS ONE, 17, e0268799 (2022).
• 25)  Nagashima Y, Sasaki K, Chiwaki R, Ishii H, Nohata K, Maemura Y, Kasai T, Ozaki A, Tamori S, Ohno S, Akimoto K. High expression of PKCζ and CTNNBIP1 is associated with poor prognosis in luminal B breast cancer. Cancer Genomics Proteomics, 22, 538–556 (2025).
• 26)  Kridel SJ, Axelrod F, Rozenkrantz N, Smith JW. Orlistat is a novel inhibitor of fatty acid synthase with antitumor activity. Cancer Res., 64, 2070–2075 (2004).
• 27)  Zeng B, Wei A, Zhou Q, Yuan M, Lei K, Liu Y, Song J, Guo L, Ye Q. Andrographolide: a review of its pharmacology, pharmacokinetics, toxicity and clinical trials and pharmaceutical researches. Phytother. Res., 36, 336–364 (2022).
• 28)  Han J, Tan C, Pan Y, Qu C, Wang Z, Wang S, Wang C, Xu K. Andrographolide inhibits the proliferation and migration of vascular smooth muscle cells via PI3K/AKT signaling pathway and amino acid metabolism to prevent intimal hyperplasia. Eur. J. Pharmacol., 959, 176082 (2023).
• 29)  Tohkayomatee R, Reabroi S, Tungmunnithum D, Parichatikanond W, Pinthong D. Andrographolide exhibits anticancer activity against breast cancer cells (MCF-7 and MDA-MB-231 cells) through suppressing cell proliferation and inducing cell apoptosis via inactivation of ER-α receptor and PI3K/AKT/mTOR signaling. Molecules, 27, 3544 (2022).
• 30)  Yang SL, Kuo FH, Chen PN, Hsieh YH, Yu NY, Yang WE, Hsieh MJ, Yang SF. Andrographolide suppresses the migratory ability of human glioblastoma multiforme cells by targeting ERK1/2-mediated matrix metalloproteinase-2 expression. Oncotarget, 8, 105860–105872 (2017).
• 31)  Zhai Z, Qu X, Li H, Ouyang Z, Yan W, Liu G, Liu X, Fan Q, Tang T, Dai K, Qin A. Inhibition of MDA-MB-231 breast cancer cell migration and invasion activity by andrographolide via suppression of nuclear factor-κB-dependent matrix metalloproteinase-9 expression. Mol. Med. Rep., 11, 1139–1145 (2015).
• 32)  Li GF, Qin YH, Du PQ. Andrographolide inhibits the migration, invasion and matrix metalloproteinase expression of rheumatoid arthritis fibroblast-like synoviocytes via inhibition of HIF-1α signaling. Life Sci., 136, 67–72 (2015).
• 33)  Lu MK, Shih YW, Chang Chien TT, Fang LH, Huang HC, Chen PS. α-Solanine inhibits human melanoma cell migration and invasion by reducing matrix metalloproteinase-2/9 activities. Biol. Pharm. Bull., 33, 1685–1691 (2010).
• 34)  Takaoka M, Naomoto Y, Ohkawa T, Uetsuka H, Shirakawa Y, Uno F, Fujiwara T, Gunduz M, Nagatsuka H, Nakajima M, Tanaka N, Haisa M. Heparanase expression correlates with invasion and poor prognosis in gastric cancers. Lab. Invest., 83, 613–622 (2003).
• 35)  Tang W, Nakamura Y, Tsujimoto M, Sato M, Wang X, Kurozumi K, Nakahara M, Nakao K, Nakamura M, Mori I, Kakudo K. Heparanase: a key enzyme in invasion and metastasis of gastric carcinoma. Mod. Pathol., 15, 593–598 (2002).
• 36)  Mehner C, Hockla A, Miller E, Ran S, Radisky DC, Radisky ES. Tumor cell-produced matrix metalloproteinase 9 (MMP-9) drives malignant progression and metastasis of basal-like triple negative breast cancer. Oncotarget, 5, 2736–2749 (2014).
• 37)  Wang FW, Cai MY, Mai SJ, Chen JW, Bai HY, Li Y, Liao YJ, Li CP, Tian XP, Kung HF, Guan XY, Xie D. Ablation of EIF5A2 induces tumor vasculature remodeling and improves tumor response to chemotherapy via regulation of matrix metalloproteinase 2 expression. Oncotarget, 5, 6716–6733 (2014).
• 38)  Yang SH, Hu S, Kang Q, Liu LX, Wei Q, Song ZM, Chen YH, Liao ZJ, Huan YF, Wang XJ, Zou H, Zhang XW. EIF5A2 promotes proliferation and invasion of intrahepatic cholangiocarcinoma cells. Clin. Res. Hepatol. Gastroenterol., 46, 101991 (2022).
• 39)  Fang J, Yu T, Jiang X, Lu Y, Shang X, Shen H, Lu Y, Zheng J, Fu P. Prognostic value of EIF5A2 in solid tumors: A meta-analysis and bioinformatics analysis. Open Med., 19, 20240962 (2024).
• 40)  Schneider-Poetsch T, Dang Y, Iwasaki W, Arata M, Shichino Y, Al Mourabit A, Moriou C, Romo D, Liu JO, Ito T, Iwasaki S, Yoshida M. Girolline is a sequence context-selective modulator of eIF5A activity. Nat. Commun., 16, 223 (2025).
• 41)  Shang Y, Zhao X, Tian B, Wang Y, Ren F, Jia B, Zhai Y, Chen W, He D, Chang Z. CHIP/Stub1 interacts with eIF5A and mediates its degradation. Cell. Signal., 26, 1098–1104 (2014).
• 42)  Zhang Z, He G, Lv Y, Liu Y, Niu Z, Feng Q, Hu R, Xu J. HERC3 regulates epithelial-mesenchymal transition by directly ubiquitination degradation EIF5A2 and inhibits metastasis of colorectal cancer. Cell Death Dis., 13, 74 (2022).
• 43)  Li Y, Fu L, Li JB, Qin Y, Zeng TT, Zhou J, Zeng ZL, Chen J, Cao TT, Ban X, Qian C, Cai Z, Xie D, Huang P, Guan XY. Increased expression of EIF5A2, via hypoxia or gene amplification, contributes to metastasis and angiogenesis of esophageal squamous cell carcinoma. Gastroenterology, 146, 1701–1713.e9 (2014).
• 44)  Liu RR, Lv YS, Tang YX, Wang YF, Chen XL, Zheng XX, Xie SZ, Cai Y, Yu J, Zhang XN. Eukaryotic translation initiation factor 5A2 regulates the migration and invasion of hepatocellular carcinoma cells via pathways involving reactive oxygen species. Oncotarget, 7, 24348–24360 (2016).
• 45)  Martínez-Férriz A, Gandía C, Pardo-Sánchez JM, Fathinajafabadi A, Ferrando A, Farràs R. Eukaryotic initiation factor 5A2 localizes to actively translating ribosomes to promote cancer cell protrusions and invasive capacity. Cell Commun. Signal., 21, 54 (2023).
• 46)  Khosravi S, Wong RP, Ardekani GS, Zhang G, Martinka M, Ong CJ, Li G. Role of EIF5A2, a downstream target of Akt, in promoting melanoma cell invasion. Br. J. Cancer, 110, 399–408 (2014).
• 47)  Zheng Y, Li P, Huang H, Ye X, Chen W, Xu G, Zhang F. Androgen receptor regulates eIF5A2 expression and promotes prostate cancer metastasis via EMT. Cell Death Discov., 7, 373 (2021).
• 48)  Ko K, Suzuki T, Ishikawa R, Hattori N, Ito R, Umehara K, Furihata T, Dohmae N, Linhardt RJ, Igarashi K, Toida T, Higashi K. Ischemic stroke disrupts the endothelial glycocalyx through activation of proHPSE via acrolein exposure. J. Biol. Chem., 295, 18614–18624 (2020).