Morphological Differentiation Towards Neuronal Phenotype of SH-SY5Y Neuroblastoma Cells by Estradiol, Retinoic Acid and Cholesterol

ORIGINAL PAPER Morphological Differentiation Towards Neuronal Phenotype of SH-SY5Y Neuroblastoma Cells by Estradiol, Retinoic Acid and Cholesterol Heidi Teppola 1 •Jertta-Riina Sarkanen 2 •Tuula O. Jalonen 3 •Marja-Leena Linne 1 Received: 19 July 2015 / Revised: 16 October 2015 / Accepted: 17 October 2015 / Published online: 30 October 2015 ÓThe Author(s) 2015. This article is published with open access at Springerlink.com Abstract Human SH-SY5Y neuroblastoma cells maintain their potential for differentiation and regression in culture conditions. The induction of differentiation could serve as a strategy to inhibit cell proliferation and tumor growth. Previous studies have shown that differentiation of SH-SY5Y cells can be induced by all-trans-retinoic-acid (RA) and cholesterol (CHOL). However, signaling pathways that lead to terminal differentiation of SH-SY5Y cells are still largely unknown. The goal of this study was to examine in the RA and CHOL treated SH-SY5Y cells the additive impacts of estradiol (E 2 ) and brain-derived neurotrophic factor (BDNF) on cell morphology, cell population growth, synaptic vesicle recycling and presence of neurofilaments. The above features indicate a higher level of neuronal differentiation. Our data show that treatment for 10 days in vitro (DIV) with RA alone or when combined with E 2 (RE) or CHOL (RC), but not when combined with BDNF (RB), significantly (p\0.01) inhibited the cell population growth. Synaptic vesicle recycling, induced by high-K ? depolarization, was significantly increased in all treatments where RA was included (RE, RC, RB, RCB), and when all agents were added together (RCBE). Specifically, our results show for the first time that E 2 treatment can alone increase synaptic vesicle recycling in SH-SY5Y cells. This work contributes to the understanding of the ways to improve suppression of neuroblastoma cells’ population growth by inducing maturation and differentiation. Keywords Brain-derived neurotrophic factor  Cholesterol Differentiation Estradiol Retinoic acid  SH-SY5Y Abbreviations AChE Acetylcholinesterase Arc Activity-regulated cytoskeleton associated protein BDNF Brain-derived neurotrophic factor CHOL Cholesterol CO 2 Carbon dioxide CTNF Corrected total neurofilament fluorescence DIV Days in vitro E 2 Estradiol ER Estrogen receptor LTP Long-term potentiation NF-68 Neurofilament 68 kD PBS Phosphate-buffered saline RA All-trans retinoic acid RB All-trans retinoic acid with BDNF RC All-trans retinoic acid with CHOL RCB All-trans retinoic acid with CHOL and BDNF RCBE All-trans retinoic acid with CHOL, BDNF and E 2 RE All-trans retinoic acid with E 2 RT Room temperature TrkB Tropomyosin-related kinase B Tuula O. Jalonen and Marja-Leena Linne are Co-senior authors. &Heidi Teppola [email protected] 1 Department of Signal Processing, Tampere University of Technology, P.O. Box 553, 33101 Tampere, Finland 2 Department of Cell Biology, School of Medicine, University of Tampere, Tampere, Finland 3 Department of Physiology and Neuroscience, St. George’s University School of Medicine, St. George’s, Grenada, West Indies 123 Neurochem Res (2016) 41:731–747 DOI 10.1007/s11064-015-1743-6 Introduction Neuroblastoma is the most common extra-cranial solid malignant tumor of sympathetic nervous system in infants and young children [1]. Regardless of its stage, until today, there is no cure or treatment, which could offer good prognosis for patients [2,3]. Human SH-SY5Y cell line, used in this study, is a subclone of SK-N-SH cell line which was isolated from a bone marrow of a 4 year-old female patient [4]. SH-SY5Y cells maintain their potential for regression, which results in aggressive proliferation of these cells [5]. Novel therapeutic treatments inducing differentiation into neuronal cell types could help to improve the prognosis of children suffering from neuroblastoma [6]. The induction of differentiation could serve as a strategy to inhibit cell population growth and eventually stop the tumor growth, as well as induce healthy mature neurons in patients. Previous studies have presented that differentiation of SH-SY5Y cells can be induced by dibutyryl cyclic AMP [5,7], 12-o-tetradecanoyl-phorbol-13-acetate [8–13], alltrans-retinoic-acid (RA) [9,14–16], brain-derived neurotrophic factor (BDNF) [17–19], vanadate [20], nerve growth factor [21,22], cholesterol (CHOL) [23], vitamin D3, and neuregulin beta1 [24]. The signaling pathways that lead to terminal differentiation of SH-SY5Y cells, however, are still largely unknown. The retinoic acid (RA) is a potent cell differentiating factor, which through its nuclear receptors affects a vast range of promoter sites in the neuronal and glial cells in every step of embryonic and postnatal life [25]. RA-induced differentiation has been shown to inhibit cell proliferation [9], change cellular sodium conductance [15], enhance the outgrowth of neurites [16], increase the acetylcholinesterase (AChE) activity [26,27], and enhance the synaptic vesicle recycling [23]. However, clinical trials have demonstrated that treatment with RA alone, or in combination with interferon alpha, is not enough against recurrent neuroblastoma in children [28,29]. Therefore, new alternative resources for more effective neuronal differentiation are needed. Cholesterol (CHOL) is a necessary component in cell membranes and important for synaptic structure and function [30]. It is synthesized by neurons themselves for their survival and growth. The development of active synapses requires additional amount of cholesterol that is shown to be secreted by glial cells (specifically astrocytes) in the central nervous system [30–32] and by Schwann cells in the peripheral nervous system [33]. The gliaderived cholesterol has also been shown to be crucial for differentiation of dendrites, synaptogenesis, increase in synaptic protein expression (synapsin-1) and neuronal activity, and for transmitter release [30,32,34,35]. In pure human SH-SY5Y cell cultures, the glia-derived cholesterol is non-existing, and addition of cholesterol is needed in order to achieve conditions resembling normal neuronal environment with surrounding glial cells [23]. The brain-derived neurotrophic factor (BDNF) has been shown to support the survival of neurons and stimulate the growth and differentiation of new neurons and synapses [36]. BDNF is a ligand for tropomyosin-related kinase B (TrkB) receptor, expression of which is lacking in naı ¨ve neuroblastoma cells. However, TrkB receptor expression and responsiveness to BDNF is induced by RA [37]. The activation of TrkB by BDNF has been suggested to enhance neuroblastoma cell survival and resistance to chemotherapy [38]. BDNF has also been shown to expose only a modest benefit for RA-induced arrest in a dormant state [6]. However, the sequential treatment of SH-SY5Y cells with RA and BDNF has been reported to induce differentiated, neurotrophic factor-dependent neuron-like cells [18] and sustained treatment has been reported to enhance neuronal differentiation of neural progenitor cells [39]. Moreover, RA-BDNF treatment induces significant increase in the expression of synaptic genes, brain miRNA, miRNA biogenesis machinery, and AChE activity, in comparison to sole RA treatment [19]. These studies stimulate the interest to further examine the potential therapeutic competence of BDNF for RA-induced SHSY5Y cell differentiation and for treatment of neuroblastoma. Disruption of BDNF and its downstream signaling pathways have been observed in many neurodegenerative diseases such as Alzheimer’s, Parkinson’s and Huntington’s diseases [40–42], underlining the importance of BDNF. However, the results of the role of added exogenous BDNF for differentiation of RA treated SH-SY5Y cells are still controversial. Estradiol (17-beta-estradiol, E 2 ), a form of estrogen hormone, has both acute and long-term effects on a variety of neuronal cell types. Increase in the number of dendritic spines and number of excitatory synapses, which are the slow long-term effects of estradiol, were first detected [43– 45]. The acute effects, which alter the intrinsic and synaptic physiology of neurons within minutes (reversible depolarization and increased input resistance with a latency of \1 min in 19.8 % of CA1 neurons tested) were detected later [46,47]. Several studies have shown that estrogen enhances synaptogenesis and synaptic plasticity [48–55], which properties may be crucial for example in enhancing memory consolidation [43,56]. Additionally, estrogen has been demonstrated to induce synaptic connectivity [52], enhance NMDA receptor expression and activity [57–66], and long-term potentiation (LTP) [58,59,64,67]. Several earlier studies have addressed possible signaling 732 Neurochem Res (2016) 41:731–747 123 mechanisms associated with estrogen-induced cellular functions. Estrogen activates these functions through the activation of estrogen receptors (ERs), ERaand ERb, which serve as transcription factors modifying the activity of target genes [68,69]. Estrogen has been shown to increase the key synaptic proteins, e.g. PSD-95, via either activation of ERa[70], ERb[51], or both [71]. Estrogen is thought to use both nuclear ERs and plasma membrane ERs which are usually referred to as classical genomic and non-genomic pathways [69]. In classical genomic action, ERs are thought to translocate into nucleus in ligand-dependent manner and acting as a transcription factor of target genes after prolonged estrogen exposure [72]. In non-genomic action, estrogen has been shown to activate the membrane ERs, which rapidly stimulate the membraneassociated signaling molecules such as PI-3K and MAPK, resulting in quick increase in protein expression [73,74]. SH-SY5Y cells have been shown to express both ERaand ERb[75]. Estrogen has also been shown to stimulate the activity-regulated cytoskeleton associated protein (Arc) expression via the MAPK and PI-3K dependent pathway in SH-SY5Y cells [48]. Arc is known to be induced by neuronal activity and playing a key role in activity-dependent synaptic plasticity [76]. Its knockdown has been shown to lead to impairment of long-term memory [77,78]. However, the specific role of E 2 for inducing differentiation in human SH-SY5Y neuroblastoma cells is still unknown. The main goal of the current study was to find a functional combination of substances for effective induction of differentiation of the SH-SY5Y cells. Based on our earlier studies, we used RA and CHOL as primary differentiation treatments [23]. We further investigated the ability of E 2 and BDNF to support, and possible enhance, the RA and CHOL induced neuronal differentiation. We quantified the individual and additive impacts of BDNF and E 2 on the RA and CHOL-induced neurite outgrowth, presence of neurofilament 68, synaptic vesicle recycling and arrest in the population growth rate of SH-SY5Y neuroblastoma cells in vitro. Methods Maintenance and Differentiation of Cell Cultures The human SH-SY5Y neuroblastoma cell line (CRL-2266; American Type Culture Collection, Manassas, VA, USA) was cultured as previously described [23]. Briefly, the cells were plated at passage 29–30 with density of 5000 cells/ cm 2 on 48-well culture dishes. Cells were cultured and maintained in 5 % CO 2 humidified incubator at 37 °Cin 1:1 nutrient mixture F-12K Kaighn’s modification, and minimum essential medium supplemented with 10 % fetal bovine serum, 2 mM/L L-glutamine, 1 % antibiotic–antimycotic mixture and 1 % non-essential amino acids (all reagents from GIBCO, Invitrogen, Carlsbad, CA, USA, unless otherwise stated). Cell differentiation was induced with 10 lM/L all-trans retinoid-acid (RA), 1 nM/L 17-beta-estradiol (E 2 ), 50 ng/mL brain-derived neurotrophic factor (BDNF), 10 lg/mL cholesterol (3b-hydroxy-5-cholestene, CHOL), or with combinations such as (i) 5 lM/L RA with 5 lg/mL CHOL (RC), (ii) 5 lM/L RA with 50 ng/mL BDNF (RB), (iii) 5 lM/L RA with 1 nM/L E 2 (RE), (iv) 5 lM/L RA with 5 lg/mL CHOL, and 50 ng/ mL BDNF (RCB), and (v) 5 lM/L RA with 5 lg/mL CHOL, 50 ng/mL BDNF, and 1 nM/L E 2 (RCBE) for 10 DIV (all differentiation reagents from Sigma-Aldrich, St Louis, MO, USA, unless otherwise stated). Stock solutions of differentiation substances were diluted in 96 % ethanol; the final ethanol concentration never exceeded 0.1 % in cell culture. Control cells were treated with \0.1 % ethanol. All used substance concentrations were carefully evaluated according to already published literature. Suitable, least toxic concentrations, also used by other laboratories, were used to enable comparison of our results with others. All differentiation substances (except BDNF when used in combinations) were applied with medium exchange at 1, 3 and 7 DIV. BDNF was applied at 4 and 7 DIV when used together with RA (RB), RA and CHOL (RCB) or RA, CHOL and E 2 (RCBE). The cell growth, condition, and morphology were observed with culture microscope (Olympus CK40) and images were taken at 10 DIV DP10 microscope digital camera system (Olympus, Tokyo, Japan). Neurofilament Staining For detecting the level of differentiation in the neuroblastoma cell cultures, the cells were stained at 10 DIV with neuronal marker NF-68 for neurofilament light polypeptide (68 kDa, Sigma-Aldrich). Cells were first fixed for 20 min with 4 % paraformaldehyde (Sigma-Aldrich) in phosphate buffered saline solution (PBS), washed three times with PBS and permeabilized in 0.5 % Triton X-100 (J.T. Baker, Phillipsburg, NJ, USA) for 15 min. After washing with PBS, the non-specific antibody binding sites were blocked with 10 % bovine serum albumin (GIBCO) in PBS for 30 min to reduce the background. Cells were then incubated with the primary antibody mouse monoclonal antiNF-68 1:200 for 1 h at room temperature (RT; ?22 °C), rinsed three times with PBS, and then incubated with a secondary antibody FITC-conjugated goat anti-mouse IgG 1:100 (Sigma-Aldrich) for 30 min at RT. Fluorescence was visualized with Nikon Eclipse TS100 microscope equipped with Nikon DS Camera Control Unit DS L-1 and images were organized with Visio 2010 (Microsoft, WA, USA). Neurochem Res (2016) 41:731–747 733 123 The intensity of total neurofilament fluorescence (NF-68) and the intensity of total background fluorescence were measured from each fluorescence image with ImageJ software (National Institute of Mental Health, Bethesda, Maryland, USA) [79]. Corrected total neurofilament fluorescence (CTNF) was calculated from the gathered data in Excel 2010 (Microsoft, WA, USA) with the method used previously [80,81], as follows: The fluorescence of the neurofilaments of interest was selected using the selection tool. Area of interest, integrated density, and mean gray value were calculated from selected areas with ImageJ software. A region next to the selected neurofilament was selected as a background value. The CTNF was calculated by using the following equation CTNF =integrated density -(area of selected neurofilaments 9mean fluorescence of background readings). Quantification of Cell Population Growth The substance-induced changes in the growth rate were quantified by counting the nuclei of 10 DIV cultured SHSY5Y cells in each treatment group. Cell nuclei were stained with 10 lg/mL Hoechst 33258 (Sigma-Aldrich) for 5 min. Cultures were washed five times in PBS and mounted on cover slips. Fluorescence results were visualized with Nikon DS Camera Control Unit DS L-1. Images of each treatment group were analyzed with CellC analysis software [82], which corrects the image background for auto-fluorescence by fitting a two-dimensional quadratic polynomial to the image and subtracts the fitted polynomial surface from the original image. After this the algorithm separates the nuclei pixels from background pixels by global thresholding and produces a binarized image with white nuclei on a black background. It furthermore separates clustered nuclei from each other by marker-controlled watershed segmentation, which is based on nuclei intensity. Eventually the software removes artifacts, such as staining residues by discarding objects smaller than 1/10 of the mean size of all objects. Images were organized with Microsoft Visio 2010. The obtained nuclei counts and statistics (see section ‘‘Statistical Analysis’’) were analyzed and plotted in MATLAB (version 2013b, The Mathworks Inc., MA, USA). Quantification of Neurite Length The SH-SY5Y cells were cultured in CTRL, CHOL, E 2 , BDNF, RA, RE, RB, RC and RCBE conditions at 10 DIV. Neurites were traced from phase contrast images of each treatment group with NeuronGrowth plugin [83] of the ImageJ software (National Institute of Mental Health, Bethesda, Maryland, USA) [79]. The NeuronGrowth automatically counts the length of traced neurites in pixels and exports the data. The gathered data and statistics were analyzed and plotted in MATLAB (version 2013b, The Mathworks Inc., MA, USA). Synaptic Vesicle Recycling The level of synaptic vesicle recycling was verified by measuring the number of fluorescence puncta in 10 DIV cultured SH-SY5Y cells. Cells were treated either with \0.1 % ethanol (CTRL), RA, CHOL, BDNF, E 2 , or with their combinations. Cultures were stained with AM1-43 styryl dye (Biotium, Hayward, CA, USA) for detecting synaptic exo/endocytosis in cells. AM1-43 is a fixable nerve terminal probe. It is not able to pass through the membranes, but instead, when cells are depolarized with high potassium (K ? )-Tyrode solution, AM1-43 styryl dye attaches inside those vesicles, which are ongoing exocytosis. Staining was modified from method previously described [23,84,85]. In the current experiments, the cells were incubated for 1 min with 4 lmol/L AM1-43, according to manufacturer’s protocol, with the depolarizing Tyrode solution including 80 mmol/L K ? (80 mmol/L KCl, 29 mmol/L NaCl, 2 mmol/L MgCl 2 , 30 mmol/L glucose, 25 mmol/L HEPES, 2 mmol/L CaCl 2 ). Cells were further washed several times with SCAS quencher solution (Biotium, Hayward, CA, USA) at RT to reduce background fluorescence. Cells were fixed for 20 min with 4 % paraformaldehyde (GIBCO), permeabilized in 0.01 % Triton X-100 (J.T. Baker) for 12 min and washed three times for 1 min in cold PBS. All reagents were from Sigma Aldrich unless otherwise stated. The fluorescence was visualized with Nikon Eclipse TS100 microscope equipped with Nikon DS Camera Control Unit DS L-1 and images were organized with Microsoft Visio 2010. Fluorescence images of each treatment group were analyzed with ImageJ analysis software [79] using the following procedural steps specifically designed to this study: background of the image was subtracted by setting a rolling ball radius to 50 pixels, after image was sharpened, and then the maxima of fluorescence puncta were found with noise tolerance of 20 and with the point selection style. This procedure was evaluated by visual inspection and it was found to be the best for finding the correct number of AM1-43 puncta from fluorescence images. The obtained counts of fluorescence puncta per image were divided by the obtained median nuclei number (see section ‘‘Quantification of Cell Population Growth’’) in particular culture in 10 DIV treatments. These obtained counts of fluorescence puncta per median nuclei number and statistics (see section ‘‘Statistical Analysis’’) were analyzed and plotted in MATLAB (version 2013b). 734 Neurochem Res (2016) 41:731–747 123 Summary of the Level of Differentiation Results are summarized in Table 1, which shows the level of differentiation induced by different treatments. The level of differentiation was assessed at least from three samples from two separate experiments by analyzing the following features; neurite length, presence of neurofilaments, inhibition in cell population growth rate and synaptic vesicle recycling. Neurite lengths were detected both visually and, by using automated methods to support the visual detection. Other features were defined according to the methods explained above (see sections ‘‘Neurofilament Staining, Quantification of Cell Population Growth and Synaptic Vesicle Recycling’’). Statistical Analysis Statistical analysis was performed using One-way ANOVA in MATLAB (version 2013b). Differences were considered to be significant when p\0.01 or p\0.05, different significances are indicated with ** or * in the figures, respectively. Results Morphology of Differentiated SH-SY5Y Cells Phase contrast images of SH-SY5Y cultures at 10 DIV were first visually analyzed for morphological assessment. In the visual analysis, control cells showed no particular neurite outgrowth (Fig. 1a), whereas CHOL-treated cells had a number of short neurites (green arrows in Fig. 1b; see also Table 1). Cells treated with E 2 had very short neurites (Fig. 1c), which were both fewer and shorter than the CHOL-induced neurites. The morphology of BDNF treated SH-SY5Y was relatively polar and cells grew more spread in the culture dish (Fig. 1d). This differed from control cells, which grew in clusters (Fig. 1a). No significant outgrowth of neurites was observed after BDNF treatment. Treatment with RA alone (Fig. 1e) as well as with combinations such as RE, RB, RC, RCB or RCBE, induced branching of longer neurites and detectable network formation (Fig. 1f–j, respectively). The data show other morphological differences between the treatments as well. The RE treated SH-SY5Y cells had Table 1 Summary of differentiation Treatment  Neurite length Total neurofilament fluorescence Inhibition of cell population growth Synaptic vesicle recycling CTRL -- - - CHOL ??** -- - E 2 ?*-- ?* BDNF -- - - RA ???** ???** ???** ??** RE ???** ??*???** ??** RC ???** ???** ???** ???* RB ???** ???** ?? ??** RCB ???** àà ???* RCBE ???** ???** à ???* The criteria for categorizing the neurite length were as follows: -neurites similar to control, ?short neurites without branching and significantly longer than in control [p\0.05 (*)], ?? intermediate neurites without branching and significantly longer than in control [p\0.01 (**)], ??? long neurites with branching and significantly longer than in control [p\0.01 (**)] and in CHOL or E 2 treatment conditions [p\0.01 (**)]. The criteria for categorizing the total neurofilament fluorescence were as follows: -no neurofilament fluorescence, ?? significantly [p\0.05 (*)] increased neurofilament fluorescence compared to control, ??? significantly [p\0.01 (**)] increased neurofilament fluorescence compared to control. The criteria for categorizing the inhibition of cell population growth were as follows: -the number of cells has not changed after the treatment, ?? the number of cells decreased (not significantly), ??? the number of cells significantly [p\0.01 (**)] decreased. The criteria for categorizing the amount of synaptic vesicle recycling were as follows: -no significant change in the counts of puncta in comparison to control, ?significant [p\0.05 (*)] increase in the counts of fluorescence puncta in comparison to control, ?? significant [p\0.01 (**)] increase in the counts of fluorescence puncta in comparison to control, ??? significant [p\0.05 (*)] increase in the counts of fluorescence puncta in comparison to RE and RB  Control (CTRL, \0.1 % ethanol), cholesterol (CHOL; 10 lg/ml), 17-beta-estradiol (E 2 ; 1 nM/L), brain derived neurotrophic factor (BDNF; 50 ng/mL), all-trans retinoic acid (RA; 10 lg/mL), RA with E 2 (RE; RA 5 lg/mL, E 2 1 nM/L), RA with CHOL (RC; RA 5 lg/mL, CHOL 5 lg/ mL), RA with BDNF (RB; RA 5 lg/mL, BDNF 50 ng/mL), RA with CHOL and BDNF (RCB; RA 5 lg/mL, CHOL 5 lg/mL, BDNF 50 ng/ mL), RA with CHOL, BDNF and E 2 (RCBE; RA 5 lg/mL, CHOL 5 lg/mL, BDNF 50 ng/mL, E 2 1 nM/L) à Conclusive data not available Neurochem Res (2016) 41:731–747 735 123 Fig. 1 Morphology and network formation of SH-SY5Y neuroblastoma cells at 10 DIV. aSH-SY5Y cells were grown for 10 DIV in control conditions (CTRL) and with bcholesterol (CHOL), cestradiol (E 2 ), dbrain-derived neurotrophic factor (BDNF), eall-trans retinoic-acid (RA), or with their combinations f–jRE, RB, RC, RCB, RCBE, respectively. Data show that the CHOL treatment induced short neurites (green arrows) with many varicosities (red arrows)(b). E 2 induced few very short neurites in comparison to CHOL induced neurites (c). BDNF treatment did not induce significant growth of neurites or network formation (d). RA treatment generated thin branching neurites and promoted network formation (blue arrows,e). Networks of cells with cell-to cell contacts (blue arrows) were always detected when treated with RA together with fE 2 (RE), gBDNF (RB), hCHOL (RC), iCHOL and BDNF (RCB), and jCHOL, BDNF, and E 2 (RCBE). Flat substrateadherent (S-type) cells were detected especially when cells were treated with E 2 or BDNF (orange arrows)(c,d) but also when cells were treated with RE (data not shown) or RB (orange arrows)(g). The RE treated cells (f) had thinner neurites than those treated with RCBE (j) (Color figure online) 736 Neurochem Res (2016) 41:731–747 123 networks of roundish cells with thin neurites without heavy branching (Fig. 1f). A small number of substrate-adherent (S-type) flat cells [86] were observed in cultures treated with RE (data not shown), RB (orange arrows in Fig. 1g) and RA (data not shown). The RB treatment induced networks that consisted of extended contacting neurites, as well as of cells in direct contact with each other without neurites. The neurites of the RB treated cells were thicker in comparison to RE induced neurites. The RC treatment induced cells with long, branching and connecting neurites (blue arrows in Fig. 1h). Other cholesterol treated cultures, such as RCB and RCBE, contained neurons with long branching neurites and network formation without S-type cells. More varicosities (red arrows in Fig. 1b) and small cell clusters (data not shown) were observed in CHOL treated cultures (CHOL, RC, RCB, RCBE) in comparison to the control, RE and RB treated groups, in which cells were more uniformly distributed (data not shown). Inhibition of Cell Population Growth The ability of a substance to inhibit the population growth of human SH-SY5Y cells is one of the indicators of increased level of differentiation. Therefore, we counted the numbers of the Hoechst 33258 stained nuclei at 10 DIV in RA, CHOL, BDNF, E 2 , RE, RB, and RC treated cell cultures and compared the results to the number of nuclei in control conditions. The data demonstrated that CHOL, E 2 or BDNF treatments on their own did not inhibit the cell population growth (Fig. 2), which, however, was seen when treated with RA, as well as with RA together with CHOL (RC) [p\0.01(**)] when compared to controls. Moreover, significant (p\0.01 (**) inhibition was also detected with RA together with E 2 (RE) treatment, when compared to controls (Fig. 2). Interestingly, when cells were treated with RA and BDNF, no inhibition of growth was detected (Fig. 2). Neurite Lengths The neurites of SH-SY5Y cells were traced from phase contrast images taken from each experiment at 10 DIV with NeuronGrowth (see section ‘‘Methods’’), which provides supportive information of the lengths in addition to the visual inspection of the cell morphology. In addition to RA [p\0.01 (**)], also with CHOL alone [p\0.01 (**)] and E 2 alone [p\0.05 (*)] treatments, induced a significant increase in the length of neurites in comparison to control cells at 10 DIV (Fig. 3). Furthermore, the significant increase in the neurite length was seen in all 0 50 100 150 200 250 300 350 400 CTRL CHOL ES BDNF RA RE RB RC TREATMENTS CELL NUCLEI COUNTS ** ** ** Fig. 2 Inhibition of cell population growth. Cells were cultured in control conditions [CTRL (n =7)], and with CHOL (n =6), E 2 (n =6), BDNF (n =6), RA (n =6), RE (n =6), RB (n =6), and RC (n =8) (n is the number of analyzed microscopy images) and the cell nuclei were counted at 10 DIV. In the boxplot representation the obtained median nuclei counts, 25th and 75th percentiles, extreme data points, and outliers of the data are shown with red line,blue edges,black whiskers, and red asterisks, respectively. Each differentiation agent is shown on the x-axis and the cell nuclei counts on the y-axis. The statistically significant differences (p\0.01) are shown with asterisks (**). Significantly lower cell numbers were detected when cells were treated with RA, RE, or RC in comparison to CTRL data. Slight increase in cell numbers (nuclei counts) were observed when cells were treated with RB, in comparison to the cells treated solely with RA. The nuclei counts of RB treated cultures were not significantly lower in comparison to CTRL (Color figure online) Neurochem Res (2016) 41:731–747 737 123 combination treatments such as RE, RB, RC, RCBE [p\0.01 (**)] relative to control conditions (Fig. 3). With the BDNF treatment alone no increase in the neurite length was detected. Presence of Neurofilaments in SH-SY5Y Neuroblastoma Cells The level of differentiation of the SH-SY5Y neuroblastoma cells at 10 DIV was further verified by imaging the NF-68 neurofilaments. The neurofilament fluorescence was defined by visual inspection and by measuring the intensity of CTNF, when cells were treated solely with RA, BDNF, CHOL, or E 2 (Fig. 4b–e, respectively) or in combination with RA and E 2 (RE), RA and BDNF (RB), RA and CHOL (RC), and RA, CHOL, BDNF, and E 2 (RCBE) (Fig. 4g–j, respectively). Both visual and automated image analyses showed that the presence of the NF-68 neurofilaments was clearly induced by RA (Fig. 4b). No major increase in NF68 fluorescence levels was observed visually or automatically, when cells were treated with BDNF, CHOL or E 2 (Fig. 4c–e). However, the intensity of NF-68 fluorescence was significantly increased in all of the combination treatments relative to control cells, as shown in Fig. 4g–j, and in Fig. 5for RA [p\0.01 (**)], RE [p\0.05(*)], RC [p\0.01(**)], RB [p\0.01(**)] and RCBE [p\0.01(**)]. Differentiation-Induced Synaptic Vesicle Recycling Our group has earlier shown that RA and RC treated human SH-SY5Y cells show intense Synaptophysin I (SypI) fluorescence in cell somata, along the neurites and at the sites of the cell-to-cell contacts. Furthermore, we have shown co-localization of SypI and AM1-43 at the end of the neurites at the cell-to-cell contacts of the RA and RC differentiated and high K ? depolarized human SH-SY5Y cells [23]. It has been also shown elsewhere that the SHSY5Y cells are capable of depolarization with high K ? stimulation [87]. Therefore, high K ? stimulation was used for studying the stimulation-related synaptic vesicle recycling also in this study. The SH-SY5Y neuroblastoma cells were incubated with E 2 , CHOL, BDNF, or RA or with their combinations (RE, RB, RC, RCB and RCBE) and stained at 10 DIV with AM1-43, a fluorescent styryl dye (a nerve terminal probe) with the presence of depolarizing high K ? - Tyrode solution. The number of fluorescent puncta, reflecting the recycling synaptic vesicles, was counted after depolarization (see Fig. 6and section ‘‘Methods’’). Treatment with CHOL or BDNF alone does not increase the number of fluorescence puncta in comparison to CTRL. Our data show for the first time, that the treatment with E 2 alone [p\0.05(*)], or RA together with E 2 (RE), BDNF (RB), BDNF and CHOL (RBC) or BDNF, CHOL and E 2 (RBCE), [p\0.01(**)] significantly increases the number CTRL CHOL E2 BDNF RA RE RB RC RCBE 200 400 600 800 1000 1200 TREATMENTS NEURITE LENGTH (PIXELS) ** ** ** ** ** ** * Fig. 3 Neurite lengths in SH-SY5Y cells. Cells were cultured in CTRL conditions, and with CHOL, E 2 , BDNF, RA, or with their combinations; RE, RB, RC, RCBE. In the boxplot representation the obtained median of neurite lengths, the 25th and 75th percentiles, the extreme data points, and the outliers of the data are shown with red line,blue edges,black whiskers and red asterisks, respectively. Each treatment is shown on x-axis and the neurite lengths in pixels on y-axis. The statistically significant differences (p\0.01) and (p\0.05) are shown with asterisks (**) and (*), respectively. The neurite lengths were significantly longer when cells are treated with RA, CHOL, RE, RB, RC, and RCBE (p\0.01). Interestingly, at 10 DIV, E 2 induces only short neurites, but the increase in the neurite length is still significant in comparison to controls (p\0.05). Moreover, RA induces significantly longer neurites when compared to CHOL or E 2 induced neurite lengths (p\0.01) (Color figure online) 738 Neurochem Res (2016) 41:731–747 123 Fig. 4 The presence of NF-68 neurofilaments in SH-SY5Y neuroblastoma cells. Neurofilaments were detected at 10 DIV by immunostaining of NF-68 (green) and cell nuclei with Hoechst 33258 (blue). a– eCombined double staining of neurofilaments and nuclei in the SH-SY5Y cells in control conditions (CTRL), and when treated with RA, BDNF, CHOL and E 2, show that RA induces NF-68, seen also in the long branching neurites. a,fNo NF68 is seen in CTRL. g–jRE, RB, RC and RCBE treatments show presence of neurofilaments similar to those with RA alone (Color figure online) Neurochem Res (2016) 41:731–747 739 123 73. Ivanova T, Mendez P, Garcia-Segura LM et al (2002) Rapid stimulation of the PI3-kinase/akt signalling pathway in developing midbrain neurones by oestrogen. J Neuroendocrinol 14:73–79 74. Mannella P, Brinton RD (2006) Estrogen receptor protein interaction with phosphatidylinositol 3-kinase leads to activation of phosphorylated akt and extracellular signal-regulated kinase 1/2 in the same population of cortical neurons: a unified mechanism of estrogen action. J Neurosci 26:9439–9447 75. Bang OY, Hong HS, Kim DH et al (2004) Neuroprotective effect of genistein against beta amyloid-induced neurotoxicity. Neurobiol Dis 16:21–28 76. Steward O, Worley PF (2001) A cellular mechanism for targeting newly synthesized mRNAs to synaptic sites on dendrites. Proc Natl Acad Sci U S A 98:7062–7068 77. Guzowski JF, Lyford GL, Stevenson GD et al (2000) Inhibition of activity-dependent arc protein expression in the rat hippocampus impairs the maintenance of long-term potentiation and the consolidation of long-term memory. J Neurosci 20:3993–4001 78. Guzowski JF, Setlow B, Wagner EK et al (2001) Experiencedependent gene expression in the rat hippocampus after spatial learning: a comparison of the immediate-early genes arc, c-fos, and zif268. J Neurosci 21:5089–5098 79. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to ImageJ: 25 years of image analysis. Nat Methods 9:671–675 80. Gavet O, Pines J (2010) Progressive activation of CyclinB1Cdk1 coordinates entry to mitosis. Dev Cell 18:533–543 81. Burgess A, Vigneron S, Brioudes E et al (2010) Loss of human greatwall results in G2 arrest and multiple mitotic defects due to deregulation of the cyclin B-Cdc2/PP2A balance. Proc Natl Acad Sci U S A 107:12564–12569 82. Selinummi J, Seppala J, Yli-Harja O et al (2005) Software for quantification of labeled bacteria from digital microscope images by automated image analysis. Biotechniques 39:859–863 83. Fanti Z, De-Miguel FF, Martinez-Perez ME (2008) A method for semiautomatic tracing and morphological measuring of neurite outgrowth from DIC sequences. Conf Proc IEEE Eng Med Biol Soc 2008:1196–1199 84. Croft BG, Fortin GD, Corera AT et al (2005) Normal biogenesis and cycling of empty synaptic vesicles in dopamine neurons of vesicular monoamine transporter 2 knockout mice. Mol Biol Cell 16:306–315 85. Pyle JL, Kavalali ET, Choi S et al (1999) Visualization of synaptic activity in hippocampal slices with FM1-43 enabled by fluorescence quenching. Neuron 24:803–808 86. Biagiotti T, D’Amico M, Marzi I et al (2006) Cell renewing in neuroblastoma: electrophysiological and immunocytochemical characterization of stem cells and derivatives. Stem Cells 24:443–453 87. Daniele F, Di Cairano ES, Moretti S et al (2015) TIRFM and pH-sensitive GFP-probes to evaluate neurotransmitter vesicle dynamics in SH-SY5Y neuroblastoma cells: Cell imaging and data analysis. J Vis Exp (95). doi:10.3791/52267 88. Simon T, Berthold F, Borkhardt A et al (2011) Treatment and outcomes of patients with relapsed, high-risk neuroblastoma: results of german trials. Pediatr Blood Cancer 56:578–583 89. Berthold F, Boos J, Burdach S et al (2005) Myeloablative megatherapy with autologous stem-cell rescue versus oral maintenance chemotherapy as consolidation treatment in patients with high-risk neuroblastoma: a randomised controlled trial. Lancet Oncol 6:649–658 90. Matthay KK, Villablanca JG, Seeger RC et al (1999) Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. Children’s cancer group. N Engl J Med 341:1165–1173 91. Pearson AD, Pinkerton CR, Lewis IJ et al (2008) High-dose rapid and standard induction chemotherapy for patients aged over 1 year with stage 4 neuroblastoma: a randomised trial. Lancet Oncol 9:247–256 92. Nakamura K, Martin KC, Jackson JK et al (2006) Brain-derived neurotrophic factor activation of TrkB induces vascular endothelial growth factor expression via hypoxia-inducible factor-1alpha in neuroblastoma cells. Cancer Res 66:4249–4255 93. Acheson A, Conover JC, Fandl JP et al (1995) A BDNF autocrine loop in adult sensory neurons prevents cell death. Nature 374:450–453 94. Jaboin J, Kim CJ, Kaplan DR et al (2002) Brain-derived neurotrophic factor activation of TrkB protects neuroblastoma cells from chemotherapy-induced apoptosis via phosphatidylinositol 3’-kinase pathway. Cancer Res 62:6756–6763 95. Matsumoto K, Wada RK, Yamashiro JM et al (1995) Expression of brain-derived neurotrophic factor and p145TrkB affects survival, differentiation, and invasiveness of human neuroblastoma cells. Cancer Res 55:1798–1806 96. Nakagawara A, Azar CG, Scavarda NJ et al (1994) Expression and function of TRK-B and BDNF in human neuroblastomas. Mol Cell Biol 14:759–767 97. Nishida Y, Adati N, Ozawa R et al (2008) Identification and classification of genes regulated by phosphatidylinositol 3-kinaseand TRKB-mediated signalling pathways during neuronal differentiation in two subtypes of the human neuroblastoma cell line SH-SY5Y. BMC Res. Notes 1:95-0500-1-95 98. Takahashi K, Piao S, Yamatani H et al (2011) Estrogen induces neurite outgrowth via rho family GTPases in neuroblastoma cells. Mol Cell Neurosci 48:217–224 99. Muley PD, McNeill EM, Marzinke MA et al (2008) The atRAresponsive gene neuron navigator 2 functions in neurite outgrowth and axonal elongation. Dev Neurobiol 68:1441–1453 100. Fryer RH, Kaplan DR, Kromer LF (1997) Truncated trkB receptors on nonneuronal cells inhibit BDNF-induced neurite outgrowth in vitro. Exp Neurol 148:616–627 101. Messi E, Florian MC, Caccia C et al (2008) Retinoic acid reduces human neuroblastoma cell migration and invasiveness: effects on DCX, LIS1, neurofilaments-68 and vimentin expression. BMC Cancer 8:30-2407-8-30 102. Rebhan M, Vacun G, Bayreuther K et al (1994) Altered ganglioside expression by SH-SY5Y cells upon retinoic acid-induced neuronal differentiation. NeuroReport 5:941–944 103. Hu R, Cai WQ, Wu XG et al (2007) Astrocyte-derived estrogen enhances synapse formation and synaptic transmission between cultured neonatal rat cortical neurons. Neuroscience 144:1229–1240 104. Sodhi RK, Singh N (2014) Retinoids as potential targets for Alzheimer’s disease. Pharmacol Biochem Behav 120:117–123 105. Pfrieger FW (2003) Role of cholesterol in synapse formation and function. Biochim Biophys Acta 1610:271–280 106. Deutsch JW, Kelly RB (1981) Lipids of synaptic vesicles: relevance to the mechanism of membrane fusion. Biochemistry 20:378–385 107. Linetti A, Fratangeli A, Taverna E et al (2010) Cholesterol reduction impairs exocytosis of synaptic vesicles. J Cell Sci 123:595–605 108. Smith AJ, Sugita S, Charlton MP (2010) Cholesterol-dependent kinase activity regulates transmitter release from cerebellar synapses. J Neurosci 30:6116–6121 109. Renner M, Choquet D, Triller A (2009) Control of the postsynaptic membrane viscosity. J Neurosci 29:2926–2937 110. Aoto J, Nam CI, Poon MM et al (2008) Synaptic signaling by all-trans retinoic acid in homeostatic synaptic plasticity. Neuron 60:308–320 746 Neurochem Res (2016) 41:731–747 123 111. Davis GW (2006) Homeostatic control of neural activity: from phenomenology to molecular design. Annu Rev Neurosci 29:307–323 112. Rich MM, Wenner P (2007) Sensing and expressing homeostatic synaptic plasticity. Trends Neurosci 30:119–125 113. Kramar EA, Babayan AH, Gall CM et al (2013) Estrogen promotes learning-related plasticity by modifying the synaptic cytoskeleton. Neuroscience 239:3–16 114. Fester L, Zhou L, Butow A et al (2009) Cholesterol-promoted synaptogenesis requires the conversion of cholesterol to estradiol in the hippocampus. Hippocampus 19:692–705 115. Nilsen J, Mor G, Naftolin F (1998) Raloxifene induces neurite outgrowth in estrogen receptor positive PC12 cells. Menopause 5:211–216 116. Chen G, Li HM, Chen YR et al (2007) Decreased estradiol release from astrocytes contributes to the neurodegeneration in a mouse model of niemann-pick disease type C. Glia 55:1509– 1518 Neurochem Res (2016) 41:731–747 747 123