Neuroblastoma tumorigenesis is regulated through the Nm23-H1/h-Prune C-terminal interaction

Abstract

Nm23-H1 is one of the most interesting candidate genes for a relevant role in Neuroblastoma pathogenesis. H-Prune is the most characterized Nm23-H1 binding partner, and its overexpression has been shown in different human cancers. Our study focuses on the role of the Nm23-H1/h-Prune protein complex in Neuroblastoma. Using NMR spectroscopy, we performed a conformational analysis of the h-Prune C-terminal to identify the amino acids involved in the interaction with Nm23-H1. We developed a competitive permeable peptide (CPP) to impair the formation of the Nm23-H1/h-Prune complex and demonstrated that CPP causes impairment of cell motility, substantial impairment of tumor growth and metastases formation. Meta-analysis performed on three Neuroblastoma cohorts showed Nm23-H1 as the gene highly associated to Neuroblastoma aggressiveness. We also identified two other proteins (PTPRA and TRIM22) with expression levels significantly affected by CPP. These data suggest a new avenue for potential clinical application of CPP in Neuroblastoma treatment.

Figure 1. Nm23-H1 and h-Prune gene expression and NBL overall survival.

(a) NME1 expression in human NBL tissues (88 samples) compared to normal adrenal gland (13 samples) searched through a public database (Versteeg database). (b) H-prune expression in human NBL tissues compared to normal adrenal gland (Versteeg database). (c) In the Essen Affymetrix database, Nm23-H1 was detected with a probe which does not discriminate between Nm23-H1 and Nm23-H2. High expression of Nm23-H1/H2 in tumors promotes worse overall survival. (d) High expression of h-Prune in NBL tissues shows a trend towards worse overall survival that does not reach significance.

Figure 2. Three-dimensional model of h-Prune protein based on protein similarities at the N-H2 terminal domain region combined with NMR h-Prune C-terminal structural studies.

(a) The h-Prune C-terminal sequence. The amino acids given in bold represent those that are more exposed according to limited proteolysis experiments. (b) Affinity chromatography. His-tagged amino acids 354-453 of h-Prune was immobilized on the resin. SH-SY5Y and SH-SY5Y-Nm23-H1 total extracts were loaded as control. Elution E3 (from 200 mM imidazole) and E4 (from elution of the complex) were loaded. An anti-His antibody was used as control of immobilized protein. The anti-Nm23-H1 antibody shows that h-Prune C-terminal interacts with Nm23-H1. (c) The secondary chemical shift index, corrected for sequence-dependent contributions, based on¹H_α,¹³Cβ,¹³C_α and¹³CO chemical shifts of the h-Prune C-terminal. Protein regions with a propensity to a helical structure (α1, α2 and α3) are highlighted. The¹H-¹⁵N steady-state heteronuclear NOE values are also reported, according to the h-Prune C-terminal amino acids.

Figure 3. Three-dimensional model of full-length h-Prune.

(a) The amino acids showing large variations upon complex formation with Nm23-H1 and the CPP peptide are mapped in cyan. The regions colored in magenta correspond to the amino acids recognized exclusively by Nm23-H1, as shown in detail in panels b and c.

Figure 4. Migration assays to assert migration properties of differential protein domains of the h-Prune protein.

(a) Alignment of the h-Prune C-terminal. Amino acids involved in Nm23-H1 binding are highlighted in yellow. (b) Two-dimensional invasion assay. HEK293 cells transfected with h-Prune mutant proteins (D388A and D422A) showed decreased migration ability. Data are represented as relative (fold) increases in the number of cells migrating compared to full-length h-Prune wild-type transfected cells. (c) Affinity chromatography. The His-tagged 354-453 h-Prune D422A mutant (upper) and D388A mutant (lower) were immobilized on the resin. HEK293 total extract was loaded as control. An anti-His antibody was used as control for the protein immobilized. An anti-Nm23-H1 antibody shows that the mutated 354–453 h-Prune interacts weakly with Nm23-H1. (d) Affinity chromatography. SH-SY5Y pre-infected cells (Ad-CPP) were loaded onto the chromatography column and the eluates were loaded onto acrylamide gels. Nm23-H1 was detected using an anti-Nm23-H1 antibody. Protein 373–353 h-Prune was revealed using an antibody against the His tag. (e) Following Ad-CPP and Ad-Mock infections in SH-SY5Y and SK-N-BE cells, the protein levels of phospho-Nm23-H1, Nm23-H1 and anti-h-Prune were analyzed by Western blotting. β-Actin was used as the loading control. (f) Two-dimensional migration assay of NBL cells showing that CPP overexpression reduces the migratory properties of both SH-SY5Y and SK-N-BE cells.

Figure 5. In vivo functional effect of CPP expression.

(a) Representative explanted tumors. Tumor weight data are presented as means ± standard deviation. (b) Proteins extracted from tumor tissues were loaded onto acrylamide gels to evaluate the protein expression levels of phospho-Nm23-H1, Nm23-H1 and h-Prune. β-Actin was used as the loading control. (c, d) Two NOD/SCID mice groups were given intra-adrenal gland injections of SH-SY5Y-luc cells previously infected with Ad-Mock (7 mice) or Ad-CPP (7 mice). Tumorigenesis was followed by in-vivo bioluminescence photon emissions signals (IVIS Imaging System). Cells pre-infected with Ad-CPP showed impaired tumor growth, with respect to controls. (e) Photograph of resected tumors and representative images of the immunostaining for Ha-Tag, Nm23-H1, h-Prune, Tuj1 and Caspase 3.

Figure 6. A common molecular network for NBL aggressiveness, derived by meta-analysis, combining gene expression NBL databases from two different studies.

(a) Two microarray datasets were analysed: Cologne 2-Color (251 samples) and Essen Affymetrix (76 samples). In the network, the NME1 gene is the most connected. Nodes represent genes; node size represents the degree of the node (i.e., the number of connections it has). Nodes are connected with lines, which represent interactions between genes. Line thickness represents the strength of interactions, as measured by mutual information. Statistical confidence of the interactions is represented by the opacity of the color of the lines (strong gray, most reliable interactions; light gray, least reliable interactions). (b) Proteins extracted from tumor tissues and SH-SY5Y cells were loaded on acrylamide gels to determine the protein expression levels of TRIM22 and PTPRA. β-Actin was used as the loading control. Action view of gene networks of genes that are highly correlated with Trim22 (c, green) and PTPRA (d, green). Modes of action are shown in different colors. Genes highlighted in yellow were further investigated. (e) Western blotting showing CPP effects on the Akt pathway in SH-SY5Y cells. β-Actin was used as the loading control. (f) Western blotting for activated β-Catenin, c-Myc, P-Ikbα, and Ikbα protein levels in SH-SY5Y cells after CPP overexpression. β-Actin was used as the loading control. (g) Following Ad-CPP infections in SH-SY5Y cells, protein levels of EGFR, Paxillin, Fak, Fak (Y397), Grb2 and Src (Y527) were analyzed by Western blotting. β-Actin was used as the loading control.

Figure 7. Status of N-MYC/Nm23-H1 high and h-Prune NBL tumors and a model of action of CPP.

(a) Level of phosphorylation of Nm23-H1 and complex formation through the h-Prune C-terminal has prognostic relevance for determining an aggressive NBL phenotype, and thus worse outcome. (b) Use of the mimetic peptide CPP which can impair the binding of Nm23-H1 and h-Prune in vitro and in vivo, resulting in substantial impairment of cell motility and metastatic niche formation in vivo, with therapeutic benefit. This is accompanied by inhibition of the WNT, NF-κb, AKT and FAK intracellular signaling pathways.