Sialidases and cancer: human sialidase neu3 enhances egfr activation in colorectal cancer

Abnormal glycosylation is known to be associated with cancer malignancy. In the cell, this process is finely regulated by many enzymes, including sialidases or neuraminidases; these are glycohydrolases widely distributed in nature that remove sialic acid residues from glycoproteins and glycolipids. In mammals, four sialidases with different subcellular localizations and biochemical features have been described: a lysosomal sialidase (NEU1), a cytosolic sialidase (NEU2), a plasma membrane-associated sialidase (NEU3) and a mitochondrial sialidase (NEU4). Studies performed over the last decade have focused on the involvement of sialylation in the progression of cancer and moreover on the association of sialidase deregulation to the tumorigenic transformation. In particular, recent studies on the Japanese population have shown that sialidase NEU3 is often deregulated in colorectal cancer and also that it co-immunoprecipitates with the epidermal growth factor receptor (EGFR), the molecular target of the most recent therapies based on monoclonal antibodies. In collaboration with the Istituto Nazionale dei Tumori of Milan (Italy) (IRCCS) and with the Istituto Cantonale di Patologia of Locarno (Switzerland) we recruited a cohort of 85 Caucasian patients resected for a colorectal cancer. By real-time PCR experiments we observed a deregulation of the mitochondrial sialidase transcripts and, on the contrary, an up regulation of NEU3 mRNA in tumor tissues compared to paired normal mucosa. Moreover, by comparing NEU3 and EGFR mRNA levels, we observed a statistically significant correlation, suggesting that the increase in EGFR expression could be associated with NEU3 increment. Vice versa, no correlation was observed between the overexpression of NEU3 sialidase and mutations in KRAS, BRAF, PIK3CA and PTEN diagnostic markers. Experiments performed on colorectal cancer cell lines have demonstrated that overexpression of wild type NEU3 enhanced EGFR activation, compared to colon normal mucosa CCD841 cell line, irrespectively of mRNA and protein levels, mutational and gene status of the receptor in all the cell lines tested. The only exception was represented by SW620 cells, that are commonly used as EGFR negative control. Moreover, Western Blots of wild type NEU3 overexpressing cells revealed increased EGFR and ERK1/2 phosphorylation in SW480 colorectal cell line and also in DIFI cells, which represents the best cellular model to study the EGFR pathway, while mRNA and total EGFR protein contents remained constant. On the contrary, we could not detect any EGFR activation in cells overexpressing a totally inactive mutant of NEU3. In addition we performed MTT based test in transfected cells. Western blot analyses showed a significant increase of cell viability, only upon overexpression of wild type NEU3, the inactive mutant being completely ineffective in this respect. Having demonstrated that the human NEU3 sialidase is more strongly anchored to the membrane its murine counterpart, we proved, not only by the lectin binding assay but also by mass spectrometry, that this sialidase directly modified the sialylation level of EGFR extracellular domain. Moreover we also showed that NEU3 overexpression modulated the response to the pharmacological treatment with Cetuximab. The overexpression of the active form of NEU3 sialidase lead to a significant increase in cell viability in all tested cell lines, also under pharmacological treatment with Cetuximab, with the exception of SW48 cells that presented a hyperactivating mutation in the tyrosine kinase domain of EGFR, causing the receptor to act independently from the dimerization. Our data suggest that, in the cell lines in which EGFR acts correctly as a dimer, sialidase overexpression caused an increment of viability even in those presenting hyperactivating mutations in the downstream pathways, that influence the efficacy of the therapy. We decided to extend the analysis of the deregulation of human sialidases to other types of cancer, in order to identify and deepen the knowledge of common variations of these enzymes also in the Western population. On the whole, by demonstrating the role of sialidase NEU3 in CRC, our work strongly suggests that this enzyme might be taken into consideration as a new effective molecular marker for CRC diagnosis and treatment. Furthermore, these data confirm the need of further studies concerning the role played by sialidases as a defining factor in cancer progression, opening up potential applications in diagnosis and therapy.