Role of kynurenines and oxidative stress in the differentiation of SH-SY5Y cells

Q Science (General);QH301 Biology;QH345 Biochemistry

Hernandez-Martinez, Juan-Manuel ; Stone, Trevor W.

University:University of Glasgow

Department:Institute of Neuroscience and Psychology

关键词: SH-SY5Y, cancer, quinolinate, differentiation, NAD+;   

Others  :  http://theses.gla.ac.uk/6133/1/2015hernandez-martinezPhD.pdf

来源: University of Glasgow

【 摘 要 】

Neuroblastoma is the most common solid extracranial tumour in children. The neuroblastoma SH-SY5Y cell line is a third successive subclone established from a metastatic bone tumour biopsy. It can be induced to differentiate (regress) into a neuronal phenotype when treated with any of several molecules including retinoic acid (RA). This characteristic has been exploited in several studies that use the SH-SY5Y cell line as a neuronal model. These studies have had far- reaching implications in shaping our understanding of certain key aspects of neurotoxicity and neurodevelopment yet their genuine relevance becomes evident when approached from an oncological point of view, as they provide information about the process underlying tumour regression which in turn can lead to the development of better therapies for the clinical management of this malignancy.It has been shown both in vitro and in vivo that several tumours constitutively catabolize the essential amino acid tryptophan (Trp) promoting cancer- associated inflammation, immune response suppression, immune escape and tumour outgrowth. The main degradation pathway of Trp is the kynurenine pathway: it involves its transformation into several bioactive compounds such as kynurenic acid (KA) and quinolinic acid (QA). QA has been implicated in several neurodegenerative diseases where it is believed to induce excitotoxic neuronal death through the activation of the N-methyl-D-aspartate (NMDA) receptor, a type of ionotropic glutamate receptor, as well as by causing oxidative stress and energy metabolism disruption. Conversely, KA acts as an NMDA receptor antagonist and exerts neuroprotection.Similarly, glutamate signaling and its dysregulation has been implicated in the development and progression of cancer. Furthermore, several glutamate receptor antagonists, including kynurenic acid, have been shown to inhibit the proliferation and migration of neoplastic cells. Conversely, it has recently been reported that QA increases the proliferation of SK-N-SH cells and protects gliomas against oxidative stress by acting as a precursor of NAD+.In view of all that has been mentioned thus far, the SH-SY5Y cell line was used as a model to investigate the effect of certain Trp metabolites such as QA, KAand 3-hydroxyanthranilic acid (3-HAA) on cellular morphology, viability and neurite extension. An important part of this study was to determine whether the available methods could reliably be employed to investigate these parameters in the SH-SY5Y cell line. It was confirmed that the acquired SH-SY5Y cell line retains its ability to differentiate and to die, and that both processes can be accurately quantified. Additionally, the optimal culturing conditions for the SH- SY5Y cell line were determined.Treatment with RA (10 μM) was used as a positive control of differentiated SH- SY5Y cells. Overall, the morphology adopted by cells after QA (50 μM) treatment was similar to the one that follows RA-induced differentiation. It was demonstrated that QA caused an increase in the neurite/soma ratio in SH-SY5Y cells, which was confirmed by Western blot analysis as evidenced by an increase in the total cellular content of β3-tubulin. These results were also confirmed by a neurite outgrowth assay that selectively quantified the neuritic mass present in cultures. However, unlike RA, QA did not decrease the levels of the neuronal proliferation marker doublecortin; the term neuritogenesis is therefore more appropriately used to refer to the series of morphological and molecular changes induced by QA in SH-SY5Y cells.The morphological changes induced by QA were not reproduced by application of NMDA, nor were they inhibited by blockade of the NMDA receptor with MK-801. Furthermore, SH-SY5Y cells were not susceptible to NMDA excitotoxic death. In view of this, the expression of GluN1 protein was determined by Western blot. GluN1 could not be detected in either undifferentiated or differentiated SH-SY5Y cells, confirming that QA-induced neuritogenesis occurs through a mechanism independent of NMDAR activation. The results herein contained suggest that the SH-SY5Y cell line does not have functional NMDARs, nonetheless it is recognized that a more exhaustive study would be necessary to fully establish which glutamate receptor subtypes are found in the SH-SY5Y cell line.The effect of QA on the production of reactive oxygen species (ROS) was also investigated. QA caused an increase in the intracellular levels of ROS as evidenced by an increase in the fluorescence of oxidised ethidium. Additionally QA-treatment caused an increase in the expression of NRF2, a transcriptionfactor that responds to oxidative stress and which has been implicated in ROS- induced differentiation in SH-SY5Y cells.In contrast, superoxide dismutase (SOD; 300 U/ml) significantly reduced the levels of ROS induced by QA treatment, which in turn caused an increase in cell proliferation and a reduction in the number of neurites. Similarly, diphenylene iodonium (DPI; an inhibitor of NADPH oxidase) also inhibited QA-induced neuritogenesis. These results suggest that the action mechanism of QA is mainly via the production of ROS, most likely superoxide (O2•-) through NADPH-oxidase.Interestingly, nicotinamide (1 nM-1mM; another precursor of NAD+) caused a dose dependent increase in the number of neurites and in the expression of β3- tubulin, which suggests that the action mechanism of QA may be mediated by metabolites of the nicotinate and Nam pathway, including NAD+ either before or after the induction of ROS.Cells were treated with 3-hydroxyanthranilic acid (3-HAA) in order to ascertain whether other pro-oxidant molecules could induce neuritogenesis as well. Single and repeated application of 3-HAA (100 μM) induced cell death in SH-SY5Y cells. Furthermore, when 3-HAA was delivered in combination with SOD, there was a shift in the IC50 values indicating that toxicity was potentiated by SOD. Catalase (CAT; 100 U/ml) afforded complete protection from the exacerbated damage induced by the single co-application of 3-HAA + SOD. However, when repeated treatments were performed, CAT no longer afforded any protection. Interestingly, the serum concentration in the medium did not affect the IC50 of 3-HAA but it did modulate the response to CAT, indicating that the specific ROS produced after 3-HAA treatment depend on the medium in which 3-HAA is delivered. At sublethal doses, 3-HAA interfered with the expression of NeuN (neuronal marker) through a mechanism that involves high production of ROS. The ability of some kynurenines to induce differentiation and cell death in SH- SY5Y cells may open new and exciting avenues of research. If these results can be confirmed in vivo they could impact the way in which certain neuroblastomas are treated.