Metallothionein-3 promotes cisplatin chemoresistance remodelling in neuroblastoma

Abstract

Metallothionein-3 has poorly characterized functions in neuroblastoma. Cisplatin-based chemotherapy is a major regimen to treat neuroblastoma, but its clinical efficacy is limited by chemoresistance. We investigated the impact of human metallothionein-3 (hMT3) up-regulation in neuroblastoma cells and the mechanisms underlying the cisplatin-resistance. We confirmed the cisplatin-metallothionein complex formation using mass spectrometry. Overexpression of hMT3 decreased the sensitivity of neuroblastoma UKF-NB-4 cells to cisplatin. We report, for the first time, cisplatin-sensitive human UKF-NB-4 cells remodelled into cisplatin-resistant cells via high and constitutive hMT3 expression in an in vivo model using chick chorioallantoic membrane assay. Comparative proteomic analysis demonstrated that several biological pathways related to apoptosis, transport, proteasome, and cellular stress were involved in cisplatin-resistance in hMT3 overexpressing UKF-NB-4 cells. Overall, our data confirmed that up-regulation of hMT3 positively correlated with increased cisplatin-chemoresistance in neuroblastoma, and a high level of hMT3 could be one of the causes of frequent tumour relapses.

Figure 1

Comparison of WT, mock and hMT3 in UKF-NB-4 neuroblastoma cells. The cells were transfected with either pcDNA3.1-GFP-TOPO (mock transfection) or pcDNA3.1-GFP-hMT3-TOPO (hMT3). (A) Transfection was validated by confocal microscopy through fluorescence of expressed GFP. The length of scale bar is 10 µm. (B) qRT-PCR showing changes in mRNA encoding hMT3 upon transfection. Data were analysed by comparative CT method and presented as relative fold gene expression (2^–ΔΔCT). (C) Whole-cell lysates immunoblots showing higher expression of hMT3 in UKF-NB-4^hMT3 compared to UKF-NB-4^CDDP cells. GAPDH, loading control. The number below the bands show values obtained by densitometric analysis of bands. (D) Percentages of metabolically active cells after treatment with annotated concentrations of CDDP. **p < 0.001, *p < 0.05. WT, wild-type.

Figure 2

Evaluation of efficiency of CDDP to inhibit tumour growth and metastatic spread of UKF-NB-4 and UKF-NB-4^CDDP cells in CAM assay. (A) Representative CAM with the tumor formed 7^th day after induction. (B) Weights of tumors excised from the CAM upon experiment termination (17^th day). Numbers of Nbl cells identified in (C) distal CAM (extravasation), (D) liver, (E) lung and (F) brain (intravasation). Quantitation of Nbl cells was based on qPCR using Alu primers. In addition, qPCR of chicken GAPDH was performed as an internal control to confirm the presence of equivalent quantities of host genomic DNA. Data shows average ± SEM from three (n = 3) independent experiments. *p < 0.05; **p < 0.005, ***p < 0.001.

Figure 3

Evaluation of efficiency of CDDP to inhibit tumour growth and metastatic spread in UKF-NB-4^mock cells and their hMT3 counterparts. (A) Weights of tumours excised from the CAM upon experiment termination (17^th day). Numbers of Nbl cells identified in (B) distal CAM (extravasation), (C) liver, (D) lung and (E) brain (intravasation). Quantitation of Nbl cells was based on qPCR using Alu primers. In addition, qPCR of chicken GAPDH was performed as an internal control to confirm the presence of equivalent quantities of host genomic DNA. Data shows average ± SEM from three (n = 3) independent experiments. *p < 0.05; **p < 0.005, ***p < 0.001, ****p < 0.0001.

Figure 4

MALDI-TOF mass spectra of rMT2-CDDP complex and purified rMT2 standard. Upper insert shows CLUSTAL multiple sequence alignment of rabbit rMT2 and hMT3. Amino acid from the sequences with asterisk (*) indicate 100% of similarity between rMT2 and hMT3. (A) MALDI-TOF spectra of rMT2 and (B) complex between rMT2 and CDDP after 24 h incubation in PBS (pH 7.5). Each spectrum was averaged from 2000 subspectra. In each mass spectrum, the hypothetical models of hMT3 complexed with Zn and Pt are shown. hMT3 protein sequence (P25713) was used to build a homology model based on template structure of rat MT2 (PDB ID 4MT2) via SwissModel server. Assuming that superior electronegativity of Pt ions replaced Zn in a similar fashion to Cd ions, the 3D structures of hMT3 Pt/Zn complexes are shown here. (A.1) Zn7-MT3 and (B.1) Pt2Zn5-hMT3 and (B.2) Pt3Zn4-hMT3. hMT3 is shown in blue ribbon and molecular surface HETATOM colours. Zn and Pt ions are shown in orange and green, respectably. (C) Venn diagram showing exclusiveness of proteins identified in proteomic signatures of parental UKF-NB-4 and UKF-NB-4 cells with up-regulated hMT3. (D) Table showing the numbers of 1481common proteins and the distribution of their regulation (UKF-NB-4-MT3 vs. UKF-NB-4, up-regulation-fold ratio > 2.5; down-regulation-fold ratio < 0.5; 2.5 > no significant differences > 0.5).

Figure 5

(A) STRING interactome network showing the proteins, which were found exclusively expressed and up-regulated in UKF-NB-4^hMT3 cells compared to mock cell line. These are involved in transport (red nodes), cellular response to stress (blue nodes) and negative regulation of biological process (green nodes), respectively. The colour of the line provides evidences of the different interactions among proteins. A red line indicates the presence of fusion evidence; a green line, neighbourhood evidence; a blue line, concurrence evidence; a purple line, experimental evidence; a light blue line, database evidence; a black line, co-expression evidence. Schematic description of (B) vesicles-mediated transport and membrane trafficking, and (C) apoptotic pathways. The numbers in brackets indicate amount of proteins, which were identified exclusively deregulated due to hMT3 overexpression in UKF-NB-4 cells compared to mock cells. RAB: member RAS oncogene GTPases, BLOC-1: Biogenesis Of Lysosomal Organelles Complex 1, BID: BH3 Interacting Domain Death Agonist, BAX: BCL2 Associated X, BAK: BCL2 Antagonist/Killer, and HMGB1/2: High Mobility Group Box). (D) Proteasome activity analysed in parental UKF-NB-4 cells compared to UKF-NB-4^mock, UKF-NB-4^hMT3 and UKF-NB-4^CDDP cells. The data are results from three (n = 3) independent experiments. *p < 0.01, n.s., not significant.

Figure 6

(A) Expression-based heatmap of total proteins identified in UKF-NB-4, hMT3 overexpressing UKF-NB-4 and UKF-NB-4^CDDP cells. (B) Venn diagram showing exclusiveness and commonness of proteins identified in proteomic signatures of parental UKF-NB-4, UKF-NB-4^hMT3 and UKF-NB-4^CDDP cells. (C) Classification summary of pathways of 83 commons proteins identified in UKF-NB-4^CDDP and UKF-NB-4^hMT3 cells, as predicted by DAVID and Reactome software.