Background: The oncogene MYCN is critical for tumorigenesis of several types of cancers including neuroblastoma. We previously reported that miR-506-3p repressed MYCN expression in neuroblastoma cells. However, the mechanism underlying such regulation was undetermined since there is no miR-506-3p target site in MYCN 3'UTR. Methods: By a systematic investigation combining microarray, informatics and luciferase reporter assay, we identified that the transcriptional factor pleiomorphic adenoma gene-like 2 (PLAGL2) is a direct target of miR-506-3p that mediates its regulation on MYCN expression. Using CHIP-PCR and luciferase reporter assay, we validated the transcriptional regulation of MYCN by PLAGL2 and we further demonstrated the transcriptional regulation of PLAGL2 by MYCN. We examined the function of PLAGL2 in regulating neuroblastoma cell fate by cell viability assay, colony formation and Western blotting of differentiation markers. We examined the effect of retinoic acid, the differentiation agent used in neuroblastoma therapy, on miR-506-3p, PLAGL2 and MYCN expressions by quantitative PCR and Western blots. We investigated the clinical relevance of PLAGL2 expression by examining the correlation of tumor PLAGL2 mRNA levels with MYCN mRNA expression and patient survival using public neuroblastoma patient datasets. Results: We found that miR-506-3p directly down-regulated PLAGL2 expression, and we validated a PLAGL2 binding site in the MYCN promoter region responsible for promoting MYCN transcription, thereby establishing a mechanism through which miR-506-3p regulates MYCN expression. Conversely, we discovered that MYCN regulated PLAGL2 transcription through five N-Myc-binding E-boxes in the PLAGL2 promoter region. We further confirmed the reciprocal regulation between endogenous PLAGL2 and MYCN in multiple neuroblastoma cell lines. Moreover, we found that PLAGL2 knockdown induced neuroblastoma cell differentiation and reduced cell proliferation, and combined knockdown of PLAGL2 and MYCN showed a synergistic effect. More strikingly, we found that high tumor PLAGL2 mRNA levels were significantly correlated with high MYCN mRNA levels and poor patient survival in neuroblastoma patients. Furthermore, we found that retinoic acid increased expression of miR-506-3p and repressed expression of MYCN and PLAGL2. Conclusions: Our findings altogether suggest that the interplay network formed by PLAGL2, MYCN and miR-506-3p is an important mechanism in regulating neuroblastoma cell fate, determining neuroblastoma prognosis, and mediating the therapeutic function of retinoic acid.

The DNA within cells is constantly subjected to chemical modifications resulting from exposure to endogenous and exogenous DNA damaging agents. Damaged or oxidized bases are primarily repaired by the base excision repair (BER) pathway, crosslinks induced by ultraviolet (UV) light and many bulky lesions are repaired by nucleotide excision repair (NER), mismatches are resolved by the mismatch repair system (MMR), and double‐strand breaks in DNA are repaired by either the homologous recombination (HR) pathway or the nonhomologous end‐joining (NHEJ) pathway. Inactivation of NER genes such as RAD2, RAD7 or RAD14 leads to extreme UV sensitivity in yeast cells. Mutants with defects in BER genes OGG1 or UNG1 are defective in repair of oxidized guanines and the removal of uracil bases from DNA, respectively. Yeast rad52 mutants are deficient in HR and are hypersensitive to ionizing radiation and chemicals that induce double‐strand breaks.Our laboratory has previously observed that rad52 haploid and diploid mutant cell cultures have high levels of distended large‐budded G2 phase cells and long cell cycle transit times (see poster by C. England et al.). This work also demonstrated that the high G2 phase cell phenotype of rad52 cells was abolished when any of seven known DNA damage checkpoint genes were co‐inactivated. The current project has extended these studies and asked whether defects in other DNA repair pathways also lead to accumulation of G2 phase cells and increases in doubling times. Initial studies revealed that log phase cultures of mutants deficient in NER (rad2, rad7 and rad14) or MMR (msh2, msh3 and msh6) do not have high levels of large‐budded cells. By contrast, such cells were strongly elevated in mutant cultures defective in the BER pathway (ogg1 and ung1 cells). Recent experiments are attempting to determine the fraction of these large‐budded cells that are G2 vs. M phase cells via DAPI staining and whether they are specifically associated with activation of the DNA damage checkpoint system. Cell cultures containing mutants defective in other DNA repair processes, such as Rad6‐mediated postreplication repair, translesion synthesis and nuclease processing of DNA ends, are also being analyzed using phase contrast and fluorescence microscopy to identify unusual cell morphology and cell cycle phase characteristics.This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.