Comparison between proliferative and neuron-like SH-SY5Y cells as an in vitro model for Parkinson disease studies

Abstract

The molecular mechanisms underlying the cellular lost found in the nigrostriatal pathway during the progression of Parkinson's disease (PD) are not completely understood. Human neuroblastoma cell line SH-SY5Y challenged with 6-hydroxydopamine (6-OHDA) has been widely used as an in vitro model for PD. Although this cell line differentiates to dopaminergic neuron-like cells in response to low serum and retinoic acid (RA) treatment, there are few studies investigating the differences between proliferative and RA-differentiated SH-SY5Y cells. Here we evaluate morphological and biochemical changes which occurs during the differentiation of SH-SY5Y cells, and their responsiveness to 6-OHDA toxicity. Exponentially growing SH-SY5Y cells were maintained with DMEM/F12 medium plus 10% of fetal bovine serum (FBS). Differentiation was triggered by the combination of 10 µM RA plus 1% of FBS during 4, 7 and 10 days in culture. We found that SH-SY5Y cells differentiated for 7 days show an increase immunocontent of several relevant neuronal markers with the concomitant decrease in non-differentiated cell marker. Moreover, cells became two-fold more sensitive to 6-OHDA toxicity during the differentiation process. Time course experiments showed loss of mitochondrial membrane potential triggered by 6-OHDA (mitochondrial dysfunction parameter), which firstly occurs in proliferative than neuron-like differentiated cells. This finding could be related to the increase in the immunocontent of the neuroprotective protein DJ-1 during differentiation. Our data suggest that SH-SY5Y cells differentiated by 7 days with the protocol described here represent a more suitable experimental model for studying the molecular and cellular mechanisms underlying the pathophysiology of PD.

Introduction

Parkinson disease (PD) is one of the most common neurodegenerative disorders, affecting about 2% of the population over the age of 60 (Lo Bianco et al., 2004). This chronic disturb causes severe motor dysfunction, such as bradykinesia, resting tremor, rigidity, postural instability, and also affects autonomic function and cognition (Poewe, 2008, Lesage and Brice, 2009). Pathologically, it is associated with the profound loss of dopamine-producing neurons in the substantia nigra pars compacta and the presence of Lewy bodies in affected regions of the central nervous system (Schapira, 2008). Although several factors, including mitochondrial dysfunction, oxidative stress and apoptosis have been suggested to contribute to cell death in PD, its etiology remains unknown (Prabhakara et al., 2008).

6-hydroxydopamine (6-OHDA) is the most used toxin in experimental models of PD (Gomez-Lazaro et al., 2008, Ikeda et al., 2008, Mu et al., 2009). Because this neurotoxin has similar structure to dopamine, it shows high affinity for the dopamine transporter and for this reason selectively destroys dopaminergic/catecholaminergic neurons (Lehmensiek et al., 2006). Once inside the neuron, 6-OHDA accumulates and undergoes non-enzymatic auto-oxidation, promoting free radical formation (Bladini et al., 2008). The inhibitory effect over complex I activity in mitochondria also accounts for the described mechanism of reactive oxygen species (ROS) generation by this neurotoxin (Lehmensiek et al., 2006, Inden et al., 2006, Chin et al., 2008). Moreover, 6-OHDA induces cell death of human neuroblastoma SH-SY5Y (Jordan et al., 2004) and mouse pheochromocytoma PC12 cell lines (Nie et al., 2002) and selectively kills tyrosine hydroxylase (TH)-immunoreactive neurons in substantia nigra and striatum in animal models of intranigral-administration (Inden et al., 2006).

Exponentially growing SH-SY5Y cells treated with 6-OHDA are often used as an in vitro model for PD (Hwang and Jeong, 2008, Lev et al., 2008). This cell line is a human catecholaminergic neuroblastoma derived from SK-N-SH, which resembles immature sympathetic neuroblasts in culture (Biedler et al., 1978). These cells are typically locked in an early neuronal differentiation stage, characterized biochemically by the low presence of neuronal markers (Biedler et al., 1978, Gilany et al., 2008). In this regard, the proliferative SH-SY5Y cells do not represent a suitable experimental model for studying the molecular and cellular mechanisms underlying the pathophysiology of PD, a disease that affects primarily differentiated dopaminergic neurons (Schapira, 2008).

In spite of that, many lines of evidence have indicated that human neuroblastoma SH-SY5Y cells are able to acquire neuron-like phenotypes with neurite outgrowth and branches by all-trans-retinoic acid (RA) treatment (Pahlman et al., 1984, Miloso et al., 2004). RA is essential in embryonic development and maintenance of growth and differentiation of epithelial, fibroblastic and myelomonocytic cells (Bastien and Rochette-Egly, 2004). Hence, RA controls cellular differentiation processes by modulating the expression of several RA-responsive genes by the activation of retinoic acid/retinoid nuclear receptors (Mark et al., 2006). In vitro, RA also plays a role in regulating transition from the proliferating precursor cell to post-mitotic differentiated cell (López-Carballo et al., 2002). Although some previous reports have addressed the RA-differentiation process in SH-SY5Y cells (López-Carballo et al., 2002, Savickiene et al., 2009), there are few studies that compare the changes in neuronal markers in SH-SY5Y cells undergoing the differentiation process, and their cellular response to 6-OHDA (Cheung et al., 2009).

Here we evaluate morphological, biochemical and cytotoxic parameters related to human neuroblastoma SH-SY5Y cells differentiated by RA treatment and challenged with 6-OHDA. Our results suggest that SH-SY5Y cells differentiated by 7 days with 1% FBS and RA represent a more suitable experimental model for studying the molecular and cellular mechanisms underlying the pathophysiology of PD.