Functional Genomics of PRUNE1 in Neurodevelopmental Disorders (NDDs) Tied to Medulloblastoma (MB) and Other Tumors

Abstract

We analyze the fundamental functions of Prune_1 in brain pathophysiology. We discuss the importance and maintenance of the function of Prune_1 and how its perturbation influences both brain pathological conditions, neurodevelopmental disorder with microcephaly, hypotonia, and variable brain anomalies (NMIHBA; OMIM: 617481), and tumorigenesis of medulloblastoma (MB) with functional correlations to other tumors. A therapeutic view underlying recent discoveries identified small molecules and cell penetrating peptides to impair the interaction of Prune_1 with protein partners (e.g., Nm23-H1), thus further impairing intracellular and extracellular signaling (i.e., canonical Wnt and TGF-β pathways). Identifying the mechanism of action of Prune_1 as responsible for neurodevelopmental disorders (NDDs), we have recognized other genes which are found overexpressed in brain tumors (e.g., MB) with functional implications in neurodevelopmental processes, as mainly linked to changes in mitotic cell cycle processes. Thus, with Prune_1 being a significant target in NDDs, we discuss how its network of action can be dysregulated during brain development, thus generating cancer and metastatic dissemination.

Keywords: PRUNE_1; TGF—transforming growth factor; Wnt; medulloblastoma; metastasis; microtubules polymerization; neurodevelopmental disorder with microcephaly, hypotonia, variable brain anomalies (NMIHBA); proliferation rate.

Figure 1

A schematic representation of the Prune_1 protein and PRUNE_1 gene with the mutations and binding regions, as reported in the literature. (A) Prune_1 protein is composed of 453 amino acids harboring the DHH (Asp-His-His) (from amino acid residues 10 to 180) and DHHA2 (from residues 215 to 350) domains. In the DHH domain, the truncating variant p.L18Sfs*8, the missense mutations p.M1?, p.D30N, p.P54T, p.D106N, p.R128Q, p.L172P, and the composite heterozygous mutations p.R128Q;G174X, and p.D106N;C180* have been identified. In the DHHA2 domain, the missense variant p.R297W and the homozygous frameshift variant p.H292Qfs*3 have been identified. (B) Exons are denoted by red boxes, UTRs are denoted by white boxes, and introns are denoted by yellow boxes. In exon 1 are shown the homozygous truncating c.50dup variant and the homozygous mutation c.G88A. The homozygous deletion g.1509-84457–151016-662del involves exons 2–8. Other homozygous mutations are found in exon 1 (c.3G>A) exon 2 (c.160C>A), exon 3 (c.G316A), and exon 4 (c.383G>A, c.515T>C, and G520T). The mutation c.132+2T>C was found in the splice donor site within intron 2, while the c.521-2A>G variant was found in the canonical splice acceptor site in intron 4. The homozygous frameshift variant c.874_875insA and the missense mutation c.C889T were found in exon 7. (C) Three-dimensional representation of Prune_1 protein from the N-terminal (left) to the C-terminal (right). The DHH (from amino acid residues 10 to 180) and DHH2 (from residues 215 to 350) domains are shown in red. The GSK-3β binding domain (from residues 330 to 394) is represented in green. The Nm23-H1 binding domains (from amino acid residues 388 to 402 and from 422 to 446) are depicted in blue. We additionally included in both domains of the protein information about the mutations published in the literature. In the DHH domain the truncating variant p.L18Sfs*8, the missense mutations p.D30N, p.P54T, p.D106N, p.R128Q, and p.L172P, and the composite heterozygous mutations p.R128Q;G174X and p.D106N;C180* have been represented. In the DHHA2 domain, the missense variant p.R297W and the homozygous frameshift variant p.H292Qfs*3 have been represented.

Figure 2

Representation of putative transcriptional factors predicted to bind the regulatory regions of the PRUNE_1 gene. (A) Schematic representation of PRUNE_1 gene and its upstream putative regulatory region as reported in the UCSC Genome Browser on Human Dec. 2013 (GRCh38/hg38) Assembly (https://genome.ucsc.edu/). The first exon (in red) is composed of a 5′UTR and a coding region. Introns are depicted in yellow. A CpG island (in green) is found in the promoter region of the gene. (B) The upstream region of PRUNE_1 shows H3K4me3 peaks, as marker of the promoter region. (C) The upstream region of PRUNE_1 shows H3K27Ac peaks, as marker of the enhancer regions. (B, C) Data obtained from chromatin immunoprecipitation sequencing (ChIP-seq) experiments performed on seven different cell lines, as reported in ENCODE (https://www.encodeproject.org/). (D) Peaks of the conserved genomic elements upon species are present in the upstream region of the gene.

Figure 3

Representative expression gene panel list of genes that have been found causative of microcephaly (MCPH). RNA log2 expression of the genes reported as causative of primary MCPH (MCPH1, WDR62, CDK5RAP2, CEP152, ASPM, CENPJ, STIL, CEP135, ZNF335, CASC5, PHC1, CDK6, CENPE, SASS6, MFSD2A, ANKLE2, WDFY3, COPB2, KIF14, NCAPD2, NCAPD3, NCAPH, NUP37, C7orf43, LMNB1, and LMNB2) and PRUNE_1 derived from transcriptome analysis of the primary cohorts of medulloblastoma (MB) in public available datasets (Pfister, Delattre, Gilbertson, and Kool) and normal cerebellum (Root) using the R2 Genomics Analysis and Visualization Platform (http://r2.amc.nl).

Figure 4

Protein network generated via STRING (Search Tool for Retrieval of Interacting Genes/Proteins). Analyses for the STRING database used the genes reported as causative of primary microcephaly (MCPH) and found overexpressed in medulloblastoma (MB) publicly available datasets (i.e., MCPH1, CDK5RAP2, CEP152, ASPM, CENPJ, STIL, CEP135, CASC5, PHC1, CDK6, CENPE, ANKLE2, COPB2, KIF14, NCAPD2, NCAPD3, NCAPH, NUP37, LMNB1, and LMNB2). The network is characterized by proteins involved in cell cycle (red, GO:0007049), mitotic cell cycle process (blue, GO:1903047), microtubule (MT)-based process (green, GO:0007017), and central nervous system (CNS) development (yellow, GO:0007417).

Figure 5

Cartoon representation illustrating the hypothesis of the mechanisms of action of Prune_1 when overexpressed in medulloblastoma (MB) tumorigenic cells (upper panel) or mutated in neural progenitor cells of patients with neurodevelopmental disorder with microcephaly, hypotonia, and variable brain anomalies (NMIHBA) (bottom panel). Upper panel: Prune_1 acting as a microtubule-associated protein (MAP) in the mitotic spindle thus increases the microtubule (MT) polymerization rate. Similarly, Nm23-H1 was also shown to take part in polymerization processes. When Prune_1 is mutated (e.g., p.D30N and p.R297W), MT polymerization is delayed, thus unbalancing the MT dynamics in mitotic cells, causing mitotic defects and decreasing the cell proliferation rate. Bottom panel: Once overexpressed in tumorigenic cells (e.g., MB), Prune_1 was reported to enhance canonical TGF-β pathways via interaction with Nm23-H1, thus increasing phosphorylated SMAD2 and leading to OTX2 upregulation, PTEN inhibition, and epithelial–mesenchymal transition (EMT) activation (through the upregulation of N-cadherin). Prune_1 was also shown to enhance canonical Wnt signaling through its interaction with GSK-β, which causes the increase of its inhibitory phosphorylation on Ser9 and Ser21 residues, thus leading to the activation of β-catenin. Through the activation of these two signaling cascades, Prune_1 can also take part in extracellular pathways, thus increasing the secretion of Wnt3a and IL17F and modulating the protein contents of extracellular vesicles (EVs). These actions are responsible for the increased proliferation rate of tumorigenic cells and their metastatic dissemination.