We also compared protein abundances in

We also compared protein abundances in primary Merlin-deficient human meningioma T0901317 against human meningeal cells, and primary human schwannoma cells against primary human Schwann cells. We identified numerous novel upregulated and downregulated proteins and phospho-proteins, performed Gene Ontology (GO) mapping and functional enrichment analyses for GO and pathway terms. We identified proteins common to both Merlin-deficient tumour types. Several of the upregulated proteins contained either a PDZ/LIM domain, or both. These proteins have been shown to have a wide range of biological functions including roles in cell signalling (Te Velthuis et al., 2007). We found PDZ and LIM domain protein 2 (PDLIM2/ mystique/SLIM) commonly upregulated in both tumour types compared to the normal controls. Previous experiments on PDLIM2 suggested a role in cytoskeletal organization as it was co-immunoprecipitated together with alpha-actinin-1, alpha-actinin-4, filamin A, and myosin heavy polypeptide 9 in rat corneal epithelial cells (Loughran et al., 2005a; Torrado et al., 2004). PDLIM2 was also identified at the nuclear level exerting tumour suppressive functions by terminating NF-κB activation during inflammation (Tanaka et al., 2007) and in breast cancer (Qu et al., 2010). PDLIM2 overexpression was found in metastatic cancer cells (Loughran et al., 2005b) and androgen-independent prostate cancer cell lines (Kang et al., 2016). Using our primary human cultures we performed PDLIM2 silencing in primary human schwannomas and meningiomas and observed a statistically significant reduction in cell proliferation in both tumour types.
Materials and Methods
Results
Discussion The aim of this study was to decipher the proteome and phospho-proteome of Merlin-deficient schwannomas and meningiomas relative to normal controls. Prior to this study, there was only one comparative analysis between meningioma and schwannoma at the genomic level reported in the literature (Torres-Martin et al., 2013b, 2014). Our proteomic analysis was highly informative and revealed many proteins of potential interest in each dataset. However, despite the vast amount of information provided by this study there are also some limitations that have to be considered. Firstly, this research approach provides a general overview about dysregulated proteins and pathways; however, it is impossible to detect the whole proteome as non-abundant proteins cannot reach the level of detection. Secondly, phosphoproteomic studies require a large amount of starting material prior phospho-enrichment; schwannoma and especially human primary meningioma cells grow at slow rate for a limited number of passages (<7) making extremely difficult to obtain the required amount of proteins. Like every enrichment technique, there are possible false-positives in the dataset and additional validation experiments are needed. The slow proliferation rate of our primary cells likely explain the reduced dataset obtained from the analysis of primary meningiomas which were cultured without the addition of external growth factors to avoid artificial manipulation of the protein signalling. In schwannomas, by functional enrichment analysis, we identified several factors related to the cytoskeleton and its regulation, in line with the pivotal role of Merlin as cytoskeletal regulator (Gladden et al., 2010; Johnson et al., 2002; Lallemand et al., 2003; McClatchey and Giovannini, 2005). MAPK signalling was also found enriched, in agreement with previous studies (Ammoun et al., 2008; Fraenzer et al., 2003). Endocytosis, possibly clathrin-mediated, was listed among the upregulated pathways and cellular components in schwannoma, as well as the AP-2 adaptor complex required to internalize cargo in clathrin-mediated endocytosis (McMahon and Boucrot, 2011). This is in keeping with previous data in flies that showed Merlin is important for controlling membrane protein turnover in part by regulating endocytosis (Maitra et al., 2006). When the proteome and phospho-proteome were compared in order to identify highly activated proteins, we identified several cytoskeletal proteins like PDLIM2, Filamin B, Vinculin and the kinase ROCK1, a key regulator of the actin cytoskeleton and cell polarity, and previously associated with the ERM family (Hebert et al., 2008). Again, we recognized proteins related to endocytosis and vesicle transport like PACSIN3, NAPA, CHMP2B and VPS29.