EGCG Disrupts the LIN28B/Let-7 Interaction and Reduces Neuroblastoma Aggressiveness

Abstract

Neuroblastoma (NB) is the most commonly diagnosed extracranial solid tumor in children, accounting for 15% of all childhood cancer deaths. Although the 5-year survival rate of patients with a high-risk disease has increased in recent decades, NB remains a challenge in pediatric oncology, and the identification of novel potential therapeutic targets and agents is an urgent clinical need. The RNA-binding protein LIN28B has been identified as an oncogene in NB and is associated with a poor prognosis. Given that LIN28B acts by negatively regulating the biogenesis of the tumor suppressor let-7 miRNAs, we reasoned that selective interference with the LIN28B/let-7 miRNA interaction would increase let-7 miRNA levels, ultimately leading to reduced NB aggressiveness. Here, we selected (-)-epigallocatechin 3-gallate (EGCG) out of 4959 molecules screened as the molecule with the best inhibitory activity on LIN28B/let-7 miRNA interaction and showed that treatment with PLC/PLGA-PEG nanoparticles containing EGCG (EGCG-NPs) led to an increase in mature let-7 miRNAs and a consequent inhibition of NB cell growth. In addition, EGCG-NP pretreatment reduced the tumorigenic potential of NB cells in vivo. These experiments suggest that the LIN28B/let-7 miRNA axis is a good therapeutic target in NB and that EGCG, which can interfere with this interaction, deserves further preclinical evaluation.

Keywords: (−)-epigallocatechin 3-gallate; AlphaScreen; EGCG; LIN28B/let-7 interaction inhibitors; PLC/PLGA-PEG nanoparticles; differentiation therapy; neuroblastoma; target therapy.

Figure 1

LIN28B downregulation increases let-7 miRNA levels and promotes differentiation in NB cells. (A) Representative immunoblot showing MYCN and LIN28B levels in KELLY, SK-N-BE(2), CHP134, and NB69 NB cell lines. β-TUBULIN was used as a loading control. n = 3 biological replicates. (B) LIN28B mRNA expression levels analyzed by qPCR in shLIN28B CHP134 and shLIN28B NB69 cell lines. Data were normalized to shCTRL cells. n = 3 biological replicates. Mean ± SD. Unpaired two-tailed Welch’s t-test analysis (* p < 0.05). (C) Representative immunoblots showing LIN28B levels in shLIN28B and shCTRL CHP134 and NB69 cell lines. β-TUBULIN was used as a loading control. (D) LIN28B protein levels in shLIN28B CHP134 and shLIN28B NB69 cells. Data were normalized to shCTRL cells. n = 3 biological replicates. Mean ± SD. Unpaired two-tailed Welch’s t-test analysis (** p < 0.01; *** p < 0.001). (E) Let-7 miRNA expression levels analyzed by qPCR in shLIN28B CHP134 and shLIN28B NB69 cell lines. Expression levels are shown as fold change relative to shCTRL cells (dashed line). Data were normalized to the internal reference gene U6. n = 2 biological replicates, n = 3 technical replicates each. Mean ± SD. Unpaired two-tailed t-test analysis (ns = not significant; * p < 0.05; ** p < 0.01). (F) SOX2, SOX9, NESTIN, and GAP43 mRNA expression levels analyzed by qPCR in shLIN28B CHP134 and shLIN28B NB69 cell lines. Fold change relative to shCTRL cells is shown (dashed line). SDHA was used as an internal reference gene. n = 3 biological replicates, n = 3 technical replicates each. Mean ± SD. Unpaired two-tailed t-test analysis (ns = not significant; * p < 0.05; ** p < 0.01). (G) Representative immunoblot showing LIN28B, MYCN, NESTIN, SOX9, and GAP43 levels in shCTRL and shLIN28B CHP134 cells. GAPDH was used as a loading control. n = 3 biological replicates. (H) LIN28B, MYCN, NESTIN, SOX9, and GAP43 protein levels in shLIN28B CHP134 cells. Data were normalized to shCTRL cells (dashed line). n = 3 biological replicates. Mean ± SD. Unpaired two-tailed Welch’s t-test analysis (* p < 0.05; ** p < 0.01; *** p < 0.001).

Figure 2

Identification of candidate molecules that interfere with the LIN28B/let-7 interaction: screening and validation. (A) Schematic representation of the AlphaScreen technique. (B) A dot plot summarizing the screening results expressed as a percentage of the mean of the negative controls. A biotinylated pre-let-7g miRNA was used as a substrate for interaction with the rLIN28B. No drug addition was used as a negative control (highlighted in purple), while a biotinylated pre-let-7g mut miRNA was used instead of the biotinylated pre-let-7g miRNA as a positive control (highlighted in light blue). Compounds that differed by two times the standard deviation from the mean of the negative controls were selected as hits (highlighted in orange). (C) Schematic representation of the REMSA. (D) Representative REMSA results for the validation of the hits selected by AlphaScreen. The rLIN28B protein plus a Cy3-labelled pre-let-7g miRNA probe was used as a negative control, while the free Cy3-labelled pre-let-7g miRNA probe was used as a positive control. (1) Terbutaline hemisulfate, (2) thioguanine, (3) thioridazine hydrochloride, (4) suramin, (5) diflubenzuron, (6) N-hydroxymethylnicotinamide, (7) salicylanilide, (8) dibutyl phthalate, (9) aminosalicylate sodium, (10) amoxicillin, (11) amphotericin B, (12) anthralin, (13) chloramphenicol, (14) chlorcyclizine hydrochloride, (15) dapsone, (16) ethionamide, (17) telenzepine hydrochloride, (18) medroxyprogesterone acetate, (19) piperazine, (20) procaine hydrochloride, (21) acedapsone, (22) doxorubicin, (23) dehydro (11,12)ursolic acid lactone, (24) coralyne chloride, (25) 2′,5′-dihydroxy-4-methoxychalcone. (E) Molecular structures of EGCG, TFMG, GA, and ATA. TFMG was present in the screened library as mixed isomers from black tea. Theaflavin 3-gallate isomer is shown here. The degree of structural similarity between EGCG and TFMG is highlighted in pink.

Figure 3

EGCG disrupts the LIN28B/let-7 interaction by binding to LIN28B. (A) Molecular structure of (−)-epigallocatechin 3,5-digallate (EGCDG). The degree of structural similarity with EGCG and TFMG is highlighted in pink. (B) Dose-dependent titration experiments performed using the Alpha assay showing the effect of increasing concentrations of EGCG, TFMG, GA, and EGCDG on the interaction between rLIN28B and biotinylated pre-let-7g miRNA. (C) Dose-dependent titration experiments performed using the Alpha assay showing the effect of increasing concentrations of EGCG and TFMG on the interaction between HuR and the TNFα AU-rich element. (D–H) Molecular modeling studies: the binding conformation of the pre-let-7 miRNA (AGGAGAU) (D) and the obtained poses by molecular docking for EGCG (E), TFMG (theaflavin 3-gallate) (F), EGCDG (G), and GA (H).

Figure 4

Evaluation of EGCG stability by HPLC, schematic representation of the structure of the EGCG-NPs, and assessment of their penetration and effect in NB cells. (A) Percentage of EGCG amount over time (0, 15, 30, 45, 60, and 75 min) under cell culture conditions. The area of the EGCG peak was normalized to t = 0. n = 3 replicates, mean ± SD. (B) Schematic representation of the EGCG-NPs. Adapted from [32]. (C) Immunofluorescence analysis of NB69 cells treated with different amounts of Cou6-containing NPs. Images acquired using the Operetta-High Content Imaging System were analyzed using the Harmony software 4.1, and the average fluorescence intensity of the Cou6 fluorescent dye (green) was quantified. n = 6 technical replicates. Two-way ANOVA followed by Fisher’s LSD test (**** p < 0.0001). (D) Representative confocal images of CHP134 cells treated with 0.003 µg/µL of Cou6-NPs (green). Nuclei were stained with Hoechst 33342 (blue), and cytoplasm was stained with the CellMask™ Deep Red Stain (red). Scale bar = 10 µm. (E) IC₅₀ values for NB69, KELLY, and CHP134 after treatment with non-encapsulated EGCG or EGCG-NPs. n = at least 3 biological replicates, n = 3 technical replicates each. Mean ± SD. Unpaired two-tailed Welch’s t-test (* p < 0.05, ** p < 0.01).

Figure 5

EGCG-NP treatment increases let-7 miRNA levels, decreases proliferation, promotes differentiation in NB cells, and reduces their engraftment ability in zebrafish. (A) qPCR analysis of let-7d, let-7f, and let-7g miRNAs in NB69, KELLY, and CHP134 cell lines treated for 48 h with empty-NPs or EGCG-NPs. The expression level is shown as fold change relative to non-treated cells, and the data were normalized to the internal reference gene U6. n = 3 biological replicates, n = 3 technical replicates each. Mean ± SD. Unpaired two-tailed t-test analysis (* p < 0.05; ** p < 0.01; *** p < 0.001, ns p ≥ 0.05). (B) Proliferation curves of NB69, KELLY, and CHP134 cells treated with different doses of EGCG-NPs around or below the IC₅₀ values. Cell viability was measured using the CellTiter-Glo^® Luminescent Cell Viability Assay and normalized to the day of the treatment (day 0). NT = non-treated cells. n = 3 biological replicates, n = 3 technical replicates each. Mean ± SEM. Two-way ANOVA followed by Fisher’s LSD test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). (C) MYCN, SOX2, SOX9, TUBB3, GAP43, and TH mRNA expression levels analyzed by qPCR after 96 h of EGCG-NP treatment in CHP134 cells. Expression level is shown as fold change relative to empty-NP treatment (dotted line). SDHA was used as a reference gene. n = 3 biological replicates, n = 3 technical replicates each. Mean ± SD. Unpaired two-tailed t-test analysis (ns = not significant; * p < 0.05; ** p < 0.01, *** p < 0.001). (D) Upper panel shows a transgenic zebrafish embryo with NB cells (red signal) at the injection site in the duct of Cuvier (dashed lines) and in the caudal region (white dashed square). Representative fluorescence microscopy images of the caudal region of Tg(fli1:EGFP) zebrafish embryos analyzed 2 h and 24 h after injection with non-treated or pretreated SK-N-BE(2) cells (empty-NPs and EGCG-NPs) and labeled with the Vybrant^® DiI (red signal). Scale bar = 100 µm. (E) Absolute fluorescence intensity of SK-N-BE(2) cells pretreated with either empty-NPs or EGCG-NPs measured at time 0 or 24 h after injection. Each dot represents the value from a single embryo. A.U. = arbitrary units. Mean ± SEM. n ≥ 32 zebrafish embryos analyzed per condition. Unpaired two-tailed t-test analysis (**** p < 0.0001).