A Focus on Regulatory Networks Linking MicroRNAs, Transcription Factors and Target Genes in Neuroblastoma

Abstract

Neuroblastoma (NB) is a tumor of the peripheral sympathetic nervous system that substantially contributes to childhood cancer mortality. NB originates from neural crest cells (NCCs) undergoing a defective sympathetic neuronal differentiation and although the starting events leading to the development of NB remain to be fully elucidated, the master role of genetic alterations in key oncogenes has been ascertained: (1) amplification and/or over-expression of MYCN, which is strongly associated with tumor progression and invasion; (2) activating mutations, amplification and/or over-expression of ALK, which is involved in tumor initiation, angiogenesis and invasion; (3) amplification and/or over-expression of LIN28B, promoting proliferation and suppression of neuroblast differentiation; (4) mutations and/or over-expression of PHOX2B, which is involved in the regulation of NB differentiation, stemness maintenance, migration and metastasis. Moreover, altered microRNA (miRNA) expression takes part in generating pathogenetic networks, in which the regulatory loops among transcription factors, miRNAs and target genes lead to complex and aberrant oncogene expression that underlies the development of a tumor. In this review, we have focused on the circuitry linking the oncogenic transcription factors MYCN and PHOX2B with their transcriptional targets ALK and LIN28B and the tumor suppressor microRNAs let-7, miR-34 and miR-204, which should act as down-regulators of their expression. We have also looked at the physiologic role of these genetic and epigenetic determinants in NC development, as well as in terminal differentiation, with their pathogenic dysregulation leading to NB oncogenesis.

Keywords: ALK; LIN28B; MYCN; PHOX2B; let-7; miR-204; miR-34; microRNAs; neuroblastoma; transcription factors.

Figure 1

Schematic representation of the most common miRNA-TF-TG auto-regulatory networks. (A) Seven main types of regulatory relationships are considered among TFs, miRNAs and TGs. (1) TF self-transcriptional activation; (2–4) Direct reciprocal feedback between miRNAs and TFs; (5,6) A TF can activate or repress the TG transcription; (4,7) A miRNA can simultaneously suppress the TF and many of its targets. (B) Common miRNA–TF–TG auto-regulatory network motifs. Left side (a–c): Coherent feedforward loops, in which a TF and miRNA regulate a TG in a complementary direction (either activating or repressing). Right side (d–f): Incoherent feedforward, in which the TFs and miRNAs have opposing (buffering) effects.

Figure 2

Schematic representation of the most common miRNA-TF-TG feedforward loop disrupted in cancer. One of the most common coherent feedforward loops (see Figure 1(Ba)) frequently disrupted in cancer, especially in NB (A) may generate an enhanced oncogene activation when the target gene (TG) is an oncogene (B) or a decreased tumor-suppressor function when the TG is a tumor-suppressor (TS) gene (C).

Figure 3

Physiological interactions between miR-34a, let-7b, miR-204, the transcription factors MYCN and PHOX2B and their target genes ALK and LIN28B through feedback and coherent feedforward loops. During physiologic regulation, miR-34a, let-7b and miR-204 exert effective downregulation of MYCN, PHOX2B and LIN28B (bold inhibitory lines) through feedback and coherent feedforward loops. In turn, MYCN and PHOX2B carry out a controlled activation of the transcription of their target genes (dashed arrows). The effect of MYCN activity over its targets and of LIN28B repression of let-7 biogenesis are also under control (dashed inhibitory lines). All three miRNAs are engaged with the TFs in feedback and coherent feedforward loops and the global effects are the reduced levels of all proteins encoded by target genes (large blue arrows).

Figure 4

Disruption of physiologic regulation and generation of aberrant transcription activations and feedback loops between miR-34a, let-7b, miR-204, the transcription factors MYCN and PHOX2B and their target genes ALK and LIN28B in neuroblastoma. In NB, underexpression of miR-34a, let-7b and miR-204 (dashed boxes) affects their inhibitory functions (red crosses) and leads to increased levels of TFs and target genes. TF overexpression strongly activates target transcription (bold black arrows) with the consequent disruption of the physiologic regulation and feedback loops (red crosses), generation of enhanced transcription and aberrant feedback loops (red arrows and red inhibitory lines) with a global upregulation of target genes and increased levels of their encoded proteins (large red arrows). De-acetylated TP53 is indicated with an asterisk (TP53*).

Figure 5

Transcriptional and post-transcriptional relationships between ALK, PHOX2B, MYCN, LIN28, miR-204, mir-34 and let-7. Genes are schematically represented by colored coding regions inserted between grey upstream (promoter) and downstream (3′UTR) regulatory regions. For each gene, transcription factors (circles) acting on its promoter and/or microRNAs (rectangles) acting on the 3′UTR, are shown in both physiological (left) and neuroblastoma (right) conditions. Thick or dashed arrows indicate an increased or decreased level of expression, respectively, of the gene in NB with respect to the physiological corresponding condition, in which gene expression level is represented by thin arrows. The protein products for each gene are shown as circles painted like the corresponding gene. The number of circles/rectangles represents the amount of product reflecting these levels of gene expression. Activation or inhibition effects of TFs are represented by “+” or “−”, respectively, within the circles on the promoter regions. In NB condition, the colored up and down arrows on the right indicate the resulting protein expression levels for each gene.

Figure 6

Schematic illustration of MYCN enhancer invasion/co-occupancy effects. (A) The two images show regulation of MYCN expression by a general activating TF (“+” white circle) in both physiological and NB conditions, where MYCN is amplified/overexpressed. The thick arrow indicates an increased level of MYCN expression in NB with respect to physiological conditions. The number of circles represents the amount of protein products, reflecting the levels of gene expression. While under physiological conditions there are adequate levels of TS microRNAs (rectangles), low TS miRNA levels are present in NB. (B) From the 5′ to the 3′end, a general MYCN target gene (TG) is represented by its enhancer (faded blue), high and low affinity promoters with E-box (grey), coding region (pink) and 3′UTR (grey). Circles represent a general TF (white), TWIST1 (yellow), HAND2 (cyan) and MYCN (red). The different levels of TG products are shown as consequences of low, medium and high MYCN levels, the latter in absence and the presence of additional TFs co-occupying the TG enhancer. “+” within circles indicate activating effects of TFs. The number of pink circles represents the amount of TG products reflecting levels of gene expression. Pink arrows on the right indicate the resulting level of TG expression for each condition.

Figure 7

Enhancer/super-enhancer-driven transcription of ADRN CRC TFs by overexpressed MYCN. (A) Schematic representation of the ADRN CRC composed of a set of super-enhancer (SE)-associated lineage TFs (i.e., PHOX2B, GATA3, HAND2, ASCL1) co-occupied/invaded by MYCN (red arrows) following MYCN amplification/overexpression. The TFs (ovals) binds each other’s SE and induce a powerful feed-forward loop over all the other genes. (B) Structural drawing of the molecular interactions within the ADNR CRC: the PHOX2B gene is given as an example. Starting from the 5′ end, PHOX2B SE consists of three E-boxes, bound by all TFs taking part in the CRC and by MYCN. In addition, overexpressed MYCN activates the PHOX2B enhancer and PHOX2B positively self-regulates its transcription by interacting with its own promoter in a positive autoregulatory loop.