Comparison of differentiation protocols of SH-SY5Y cells for use as an in vitro model of Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is caused by decreased levels of acetylcholine (ACh) and accounts for 60-80% of dementia cases worldwide. Current AD treatment modalities (i.e. cholinesterase inhibitors and N-methyl-D-aspartic acid receptors antagonists) only offer symptomatic relief as there is currently no cure. In vitro studies are more economical as they are capable of large-scale production of cell cultures, and are less labour intensive. In an attempt to make neurotoxicity models more representative, neuronal cell lines (such as SH-SY5Y neuroblastoma cells) are differentiated to a more mature phenotype. However, literature describes several different differentiation procedures, and thus it becomes necessary to assess their applicability in determining specific neurotoxicity parameters. The aim of the study was to compare two differentiation protocols of SH-SY5Y neuroblastoma cells for the purpose of developing an in vitro AD drug development platform. Two published methods were compared for their ability to yield differentiated cells: a six-day and an eighteen-day differentiation procedure. The most representative differentiation model was determined by assessing various endpoints relating to maturity of both differentiation models in comparison to undifferentiated cells: morphological and viability changes of the cells were monitored via light microscopy, sulforhodamine B staining, protein content determination, and live-dead staining. The ACh/acetylcholinesterase (AChE) levels were measured by the Amplex Red assay; and protein expression relating to mature (amyloid precursor protein [APP] and β-secretase APP-cleaving enzyme-1 [BACE-1]) and immature (proliferating cell nuclear antigen [PCNA]) phenotypes by flow cytometry. Live-dead staining confirmed that cells remained viable after treating with the differentiation cocktails. The cellular density of differentiated cells of both differentiation methods by Shipley et al. and Forster et al. significantly (p ≤ 0.001) increased, thus confirming that there are more cells at the end of the differentiation processes which also correlates to the increased protein content of cultures. The six-day differentiation method had a significantly (p ≤ 0.01) higher cellular density of differentiated cells (2.3-fold) than the eighteen-day differentiation method. Despite the difference in cellular densities between the two methods, there were sufficient cells at the end of the differentiation methods to continue with further testing. The protein content of differentiated cells of both differentiation methods significantly (six-day: p ≤ 0.001; eighteen-day: p ≤ 0.01) increased compared to the undifferentiated cells (six-day: 2.1-fold; eighteen-day: 1.3-fold). There was a significant (p ≤ 0.001) decrease in protein content in the supernatant of both differentiation methods (six-day: 1.2-fold; eighteen-day: 1.4-fold). There was a significant (p ≤ 0.001) decrease in ACh levels in the cell lysates (six-day: 0.57-fold; eighteen-day: 0.64-fold) and supernatants (six-day: 0.60-fold; eighteen-day: 0.65-fold) of both differentiation methods. Similar to ACh levels, a significant (p ≤ 0.001) decrease in the AChE level was observed between respective undifferentiated and differentiated cell lysates (six-day: 0.58-fold; eighteen-day: 0.51-fold) and supernatants (six-day: 0.48-fold; eighteen-day: 0.50-fold). The differentiation methods did not produce the expected increase in ACh and AChE levels after cellular differentiation. All three biomarkers (PCNA, APP, BACE-1) were expressed in undifferentiated, six-day differentiation method and the eighteen-day differentiation method. The undifferentiated cells and the cells differentiatited via the six-day differentiation method expressed higher levels of PCNA and APP than BACE-1. However, lower levels of PCNA expression were exhibited after cellular differentiation, with the eighteen-day differentiation method expressing the lowest PCNA levels. The undifferentiated cells expressed higher BACE-1 levels than the differentiated cells. The differentiation methods did not produce the expected increase in mature biomarkers after cellular differentiation. However, morphologically both differentiation methods produced a high number of long neuritic projections that connected to other cells which were evenly distributed, exhibiting a pyramidal shaped cell body, although a reduction in cellular proliferation was observed. The results support morphological differentiation, but it is not recommended to use these preparations to investigate the mature cholinergic system. Further method development will be needed to ascertain the correct cocktail of differentiation factors to yield a mature phenotype expressing markers relevant to AD, including APP and BACE-1, and functional neurotransmitter systems, such as ACh and AChE.