In vitro optimization of retinoic acid–induced neuritogenesisand TH endogenous expression in human SH-SY5Yneuroblastoma cells by the antioxidant Trolox

Mario Luiz Conte da Frota Junior • Andre Simoes Pires • Fares Zeidan-Chulia •

Ivi Juliana Bristot • Fernanda M. Lopes • Matheus Augusto de Bittencourt Pasquali •

Alfeu Zanotto-Filho • Guilherme Antonio Behr • Fabio Klamt • Daniel Pens Gelain •

Jose Claudio Fonseca Moreira

Received: 26 April 2011 / Accepted: 29 June 2011 / Published online: 7 July 2011

� Springer Science+Business Media, LLC. 2011

Abstract Though, it is quite well-known how retinoic

acid (RA) is able to induce neuritogenesis in different in

vitro models, the putative role exerted by reactive oxygen

species (ROS) during this process still need to be further

studied. For such purpose, we used a neuronal-like cell line

(SH-SY5Y cells) in order to investigate whether the anti-

oxidant Trolox (a hydrophilic analog of alpha-tocopherol)

could have any effect on the number of RA-induced neu-

rites, and how significant changes in cellular redox

homeostasis may affect the cellular endogenous expression

of tyrosine hydroxylase (TH). Our results show a signifi-

cant enhancement of RA (10 lM)-induced neuritogenesis

and TH endogenous expression, when cells were co-treated

with Trolox (100 lM) for 7 days. Moreover, this effect

was associated with an improvement in cellular viability.

The mechanism seems to mainly involve PI3 K/Akt rather

than MEK signaling pathway. Therefore, our data dem-

onstrate that concomitant decreases in basal reactive oxy-

gen species (ROS) production could exert a positive effect

on the neuritogenic process of RA-treated SH-SY5Y cells.

Keywords SH-SY5Y cells � Retinoic acid � Trolox �Neurite outgrowth � Reactive oxygen species �Parkinson’s disease

Abbreviations

RA Retinoic acid

ROS Reactive oxygen species

TH Tyrosine hydroxylase

PD Parkinson’s disease

CAT Catalase

SOD Superoxide dismutase

GPx Glutathione peroxidase

hESCs Human embryonic stem cells

iPS cells Induced pluripotent stem cells

MSCs Mesenchymal stem cells

Introduction

Parkinson’s disease (PD) is a neurodegenerative disorder of

the central nervous system (CNS), caused by chronic loss

of dopamine-producing neurons in the substantia nigra pars

compacta and subsequent depletion of dopamine in the

striatum, the main projection area of the substantia nigra [1,

2]. Over the last decades, cell transplantation has become

an alternative approach for the treatment of neurological

diseases, including PD [3]. In fact, successful in vitro

generation of dopaminergic neurons from human embry-

onic stem cells (hESCs) [4], induced pluripotent stem cells

(iPS cells) [5, 6], and some types of mesenchymal stem

cells (MSCs) [7, 8] has raised even more expectations

when it comes to the use of cell-based therapy for PD.

Either the use of MSCs or reprogramming fibroblasts into

iPS cells potentially solve the ethical issues associated with

hESCs [9], but researchers in the field are still concerned

about the potential risk of tumor generation that accom-

panies the transplantation of these cells in clinical trials

[10]. Nevertheless, the use of different in vitro models

M. L. C. d. Frota Junior (&) � A. S. Pires � F. Zeidan-Chulia �I. J. Bristot � F. M. Lopes � M. A. de Bittencourt Pasquali �A. Zanotto-Filho � G. A. Behr � F. Klamt �D. P. Gelain � J. C. F. Moreira

Center of Oxidative Stress Research (CEEO), Department

of Biochemistry, Institute of Basic Health Sciences (ICBS),

Federal University of Rio Grande do Sul (UFRGS), Rio Grande

do Sul, Rua Ramiro Barcelos 2600—ANEXO, Porto Alegre,

RS 90035-003, Brazil

e-mail: mfrotajr@yahoo.com.br

123

Mol Cell Biochem (2011) 358:325–334

DOI 10.1007/s11010-011-0983-2

(e.g., human SH-SY5Y cells), for better comprehension of

the mechanisms by which dopaminergic differentiation is

promoted, might contribute to the design of novel thera-

peutic strategies, or at least, to improve the already existing

ones.

Human SH-SY5Y cells are a neuroblastic subclone of

the neuroblastoma cell line SK-N-SH [11], and despite

their tumoral origin, they are able to display a functional

and morphological neuronal phenotype upon in vitro

exposure to several agents [12–16], such as retinoic acid

(RA), an active metabolite of retinol [17]. The biological

effects of RA are mediated by specific nuclear receptors,

which function as transcriptional regulators and act by

modifying the transcriptional activity of specific genes [18,

19]. However, the wide spectrum of physiological and

pharmacological effects of retinoids is attributed to both

receptor-dependent and receptor-independent mechanisms

[20]. In this context, previous works have already dem-

onstrated that RA is able to generate reactive oxygen

species (ROS) in several cells types—an effect coupled to

differentiation [21–24].

Although, potential activation of signaling pathways

upon RA stimulation have extensively been studied in the

process of neuritogenesis, to the best of our knowledge, a

possible link between cellular redox status and the acqui-

sition of both morphological and functional neuronal

characteristics still remains unclear. Thus, we herein used

human neuroblastoma SH-SY5Y cells as a neuronal-like

model to study this possible relation. Specifically, we

treated the cells with 10 lM of RA for 7 days, either in the

presence or absence of the antioxidant Trolox (100 lM),

MEK inhibitor UO126 (10 lM), or PI3 K inhibitor wort-

mannin (500 nM). Our results suggest that concomitant

decreases in basal ROS production can potentiate RA-

induced effects on neurite development in SH-SY5Y cells,

associated with a higher viability and reduced cytotoxicity

of the culture. Furthermore, our in vitro results suggest that

a coordinated regulation of ROS production and activation

of PI3 K signaling pathway could have a significant posi-

tive effect on RA-derived neuritogenesis in this experi-

mental model.

Materials and methods

Cell culture and treatments

Human neuroblastoma SH-SY5Y cells were obtained from

the American Type Culture Collection (ATCC, Rockville,

MD) and were previously maintained in 75 cm2 flasks in

DMEN:F12 (1:1) supplemented with 10% heat-inactivated

fetal bovine serum, L-glutamine (2 mM), and 0.28 mg/ml

of gentamycin sulfate in a 5% CO2 humidified incubator at

37�C. For Western blot analysis, evaluation of neurito-

genesis and flow cytometer analysis, cells were grown in

6-well cluster dishes, whereas cell viability and ROS pro-

duction assays were performed by using 96-well plates.

Cells for drug treatment were initially seeded at 1 9 104

cells/well (96-well plates) or 2 9 105 cells/well (6-well

plates) and cultured for 24 h. Later, concentration of fetal

bovine serum (FBS) was reduced to 1% in the culture

medium, RA alone (10 lM) or RA plus Trolox 100 lM

were added, and the cells were left for 1 week in the

presence or absence of MEK inhibitor UO126 (10 lM) or

PI 3-K inhibitor wortmannin (500 nM). The culture med-

ium was replaced every 3 days. To reveal the effects of the

kinase inhibitors on RA-induced neurite formation, cul-

tures were pretreated for 45 min before addition of RA and

Trolox.

Cell morphology analysis and evaluation

of neuritogenesis

To evaluate the differences in cell morphology, cultures

were observed in an inverted microscope (Nikon Eclipse

TE300) and images were captured by using a digital

camera connected to the microscope. Cell bodies and

neurites present in 10 randomly selected fields were

counted. The ratio between cell bodies and neurites was

calculated yielding the average of neurites per neuron and

was expressed as the mean ± SEM value. Differences

were considered to be significant when P \ 0.05;

Immunoblotting

Proteins (20 lg) were separated by SDS–PAGE—10% (w/

v) acrylamide, 0.275% (w/v) bisacrylamide gels—and

electrotransferred onto nitrocellulose membranes. Mem-

branes were then incubated in Tris-buffered saline Tween-

20 [TBS-T; 20 mM Tris–HCl, pH 7.5, 137 mM NaCl,

0.05% (v/v) Tween 20] containing 1% (w/v) non-fat milk

powder for 1 h at room temperature. Subsequently, the

membranes were incubated for 12 h with polyclonal rabbit

anti-TH (1:10.000 dilution; Sigma). After washing in TBS-

T, blots were incubated with horseradish peroxidase-linked

anti-immunoglobulin G (IgG) antibodies (1:10.000 dilu-

tion) for 1.5 h at room temperature. Chemiluminescent

bands were detected, and densitometric analysis was per-

formed by Image-J� software;

ROS production

Intracellular ROS production was determined by the

DCFH-DA-based real-time assay using intact living cells

[25]. Briefly, SH-SY5Y cells were initially seeded onto

326 Mol Cell Biochem (2011) 358:325–334

123

96-well plates and incubated for 24 h. Later, concentration

of fetal bovine serum (FBS) was reduced to 1% in the

culture medium, and cells were incubated for 1 h with

DCFH-DA 100 lM (stock solution in DMSO, 10 mM) at

5% CO2 and 37�C. Then, cells were washed, and treat-

ments were carried out. During the treatment, changes in

fluorescence by the oxidation of DCFH into the fluorogen

DCF were monitored in a microplate fluorescence reader

(F2000, Hitachi Ltd., Tokyo, Japan) at 37�C. H2O2 1 mM

was used as positive control for ROS production (data not

shown). Excitation filter was set at 485 ± 10 nm, and the

emission filter was set at 530 ± 12.5 nm. Data were

recorded every 30 s and plotted in Excel software.

Total reactive antioxidant potential (TRAP assay)

The antioxidant potential of cells was estimated by the total

reactive antioxidant potential parameter (TRAP), which

determines the non-enzymatic antioxidant potential of the

cell. Briefly, the reaction was initiated by adding luminol

(4 mM)—as an external probe to monitoring radical pro-

duction—and AAPH (10 mM)—a free radical source that

produces peroxyl radical at a constant rate—in glycine

buffer (0.1 M) pH 8.6 at room temperature that resulted in

steady luminescence emission. Afterward, SH-SY5Y cell

samples were suspended in glycine buffer and mixed into

the reaction vial. The decrease in luminescence, which is

proportional to the non-enzymatic antioxidant potential,

was monitored in a liquid scintillation counter (Wallace

1409) as counts per minute (CPM). The luminescence

emission was followed for 30 min after the addition of

sample homogenates (150 lg of protein). Chemilumines-

cence values were standardized against protein content.

Flow cytometry analysis

SH-SY5Y cells were initially seeded onto 6-well plates and

incubated for 24 h. Later, concentration of fetal bovine

serum (FBS) was reduced to 1% in the culture medium, and

cells were treated for 7 days. At the end of treatment, cells

were washed twice in PBS (137 mM NaCl, 2.7 mM KCl,

4.3 mM Na2HPO4.2H2O, 1.4 mM KH2PO4, pH 7.4) con-

taining 1 mM EDTA (PBS-EDTA) and subsequently

trypsinized with 0.13 g/L trypsin in PBS-EDTA. Trypsin

was inhibited with bovine fetal serum 10%, and medium

washes, and cells were combined and centrifuged (5 min,

200g, 4�C). Cells were allowed to recover from trypsin-

ization in complete medium (30 min, 37�C). Propidium

iodide (2 lM) was added 10 min prior to analysis. PI is

excluded from healthy cells, but following loss of mem-

brane integrity, enters the cells, binds to DNA, and

becomes highly fluorescent. Red fluorescence of DNA-

bound PI in individual cells was measured using a

FACScalibur (BD PharMingen). Ten thousand cells were

analyzed per sample, and data were reported as the per-

centage of necrotic cells. H2O2 1 mM was used as positive

control of cell death (data not shown).

Thiobarbituric acid reactive species (TBARS)

As an index of lipid peroxidation, we used the formation of

TBARS during an acid-heating reaction as previously

described [26]. Briefly, cells were harvested at the end of

differentiation, the samples were mixed with 1 ml of tri-

chloroacetic acid 10% (TCA) and 1 ml of thiobarbituric

acid 0.67% (TBA), then heated in a boiling water bath for

15 min. TBARS were determined by the absorbance at

535 nm and were expressed as TBARS/mg protein.

Antioxidant enzyme activities

To determine CAT activity, cells were sonicated in 50 mM

phosphate buffer (pH 7.0) and the resulting suspension was

centrifuged at 3,000 g for 10 min. The supernatant was

used for enzyme assays. CAT activity was assayed by

measuring the rate of decrease in H2O2 absorbance at

240 nm [27]. SOD activity was assayed by measuring the

inhibition of adrenaline auto-oxidation [28]. For GPx

activity, NADPH oxidation was followed at 340 nm in the

presence of reduced glutathione, tert-butyl hydroperoxide,

and glutathione reductase [29].

Protein determination

For immunoblotting analysis, the protein contents were

measured by the Bradford assay [30]. For TBARS and

antioxidant enzyme assays, protein content of each sample

was measured by Lowry method [31].

Statistical analysis

Results were expressed as the mean ± SEM of at least

three independent experiments. Data were analyzed by a

one-way analysis of variance (ANOVA), using a Newman–

Keuls test to compare mean values across groups. Differ-

ences were considered to be significant when P \ 0.05.

Results

SH-SY5Y cells exposed to 10 lM RA assume a neuronal-

like phenotype by activating neurite production and the

expression of neuronal-specific markers [32]. First, we

evaluated the possible morphological changes in SH-SY5Y

cells when treated with 10 lM RA, in the presence or

absence of the antioxidant Trolox (100 lM). As expected,

Mol Cell Biochem (2011) 358:325–334 327

123

cells clearly underwent neurite-like development and

acquire neuronal-like morphology after 7 days of treatment

with RA (Fig. 1). Interestingly, in the presence of Trolox,

the number of viable cells in culture seems to be higher and

thus, positively affecting neuritogenesis in RA-treated cells

(Fig. 1).

In order to determine whether the presence of Trolox

could have some kind of effect on RA-induced neurite-like

development, we quantified the number of neurites per cell

body in each case. The results show a significant increase

in the number of neurites per cell, when cultures were

co-treated with both RA and Trolox in comparison with

those ones exposed to RA alone (Fig. 2a). Since tyrosine

hydroxylase (TH, EC1.14.16.2) is the rate-limiting enzyme

in the catecholamine synthesis pathway, and its regulation

controls the levels of dopamine, epinephrine, and norepi-

nephrine in tissues [33], we performed immunoblotting

experiments to quantify possible variations at the protein

level of the enzyme. We found that TH immunocontent,

a classical biochemical marker of catecholaminergic

differentiation, was significantly increased in cells treated

with RA plus Trolox when compared to RA-treated cells

and control cells (Fig. 2b).

These first results seemed to point out that ROS pro-

duction may have a possible role in the effectiveness of

RA-induced neurite production. So then, we decided to

investigate whether RA is able to induce ROS production

in our cell model. To achieve this goal, an in vitro real-time

DCFH-DA assay was performed showing that ROS pro-

duction in RA-treated cells was not increased in compari-

son with control cells. However, co-treatment with 100 lM

Trolox significantly decreased the basal levels of free

radical generation, observed under our assays conditions

(Fig. 3a). Next, we conducted parallel experiments for

evaluating possible changes in the cellular redox status

after each different treatment. Figure 3b shows that expo-

sure to RA for 1 week was able to increase the values

obtained by TRAP assay and measured by chemilumines-

cence-enhanced method, all together indicating that RA

treatment for 1 week was pro-oxidant, indeed. Again,

(200 X)(100 X)

Co

ntr

ol

RA

RA

+ T

rolo

x

Fig. 1 Phase-contrast

micrographs of SH-SY5Y cells.

Control, control cells; RA, cells

treated with RA 10 lM for

7 days; RA ? Trolox, cells

treated with RA plus Trolox

100 lM for 7 days. The

medium was replaced every

3 days. Scale bar, 100 lm

328 Mol Cell Biochem (2011) 358:325–334

123

co-treatment with Trolox attenuated the RA-induced

effects, restoring TRAP values to control levels.

Further, we observed an increase in lipid peroxidation

when cells were exposed to RA for 1 week. In the same

way, RA-induced oxidative damage was reversed by anti-

oxidant co-treatment (Fig. 4a). Catalase (CAT), superoxide

dismutase (SOD), and glutathione peroxidase (GPx) activ-

ities were also measured. Comparing with control cells,

CAT activity was significantly increased in RA-exposed

cells (Fig. 4b), but this effect seems to be attenuated by

co-treating the cells with RA plus Trolox. On the contrary,

neither SOD (Fig. 4c) nor GPx (Fig. 4d) showed any sig-

nificant change when cells were cultured with RA. Whereas

co-treatment of RA plus Trolox resulted in decreased values

of both SOD and GPx activities.

Several signaling pathways are known to be redox

sensitive [34–37]. This characteristic could be of extreme

importance during cell cycle regulation and cell prolifera-

tion. In fact, whenever the intracellular condition is mod-

ified into a reducing state by adding antioxidant agents like

Trolox (an antioxidant synthetic analog of vitamin E),

cathechin, acetylcystein (NAC), etc., both apoptosis and

survival signaling pathways induced by ROS are sup-

pressed [38–41]. Furthermore, Vitamin E has been shown

to protect cultured cortical neurons from oxidative stress–

induced cell death through the activation of mitogen-acti-

vated protein kinase (MAPK) and PI3 K [42]. Thus, we

wanted to test how Trolox could interfere with the effects

of both MEK and PI3 K/Akt signal transduction–specific

inhibitors on RA-induced neuritogenesis.

In an elegant study, Lopez-Carballo and colleagues [43]

proposed that RA, by activating the PI3 K/Akt signaling

pathway, would have a relevant role in the regulation of

neuronal cell survival. Then, this would positively affect

the rate of cells that would undergo neuritogenesis. In our

experiments, exposure to UO126 (MEK-specific inhibitor)

did not apparently affect RA-induced neurite production at

the end of differentiation, but on the other hand, those ones

A

020406080

100120140160180

RA+TroloxRA

Treatments

Neu

rite

s/C

ell B

od

y (%

)

B

*

0

5000

10000

15000

20000

25000

30000

Control RA RA + Trolox

Treatments

Arb

itra

ry u

nit

s

TH

Control RA

62kD

*

*

RA + Trolox

Fig. 2 Effect of Trolox on RA-induced neuritogenesis and TH

endogenous expression of SH-SY5Y cells. a The antioxidant Trolox

increases RA-induced neurite formation in SH-SY5Y cells. SH-SY5Y

cells were treated for 7 days and neurite formation was scored as

described in ‘‘Materials and methods’’, considering the corresponding

value of RA treatment as 100%. Data are the means (%) ± SEM of

three independent experiments, and differences were considered to be

significant when P \ 0.05 (one-way ANOVA). b Representative

immunoblot and densitometric analysis (means ± SEM) of three

independent experiments showing the levels of tyrosine hydroxylase

immunocontent in SH-SY5Y cells treated with 10 lM RA alone or

RA plus 100 lM Trolox for 7 days

Fig. 3 Generation of ROS in RA-treated cells before and after Trolox

co-treatment. a Kinetics of intracellular ROS production in SH-SY5Y

cells evaluated by the DCFH-DA real-time fluorescence assay as

described in ‘‘Materials and methods’’. The cells were treated with

10 lM of RA or RA plus 100 lM Trolox. Control cells, cells without

RA treatment. Data are the means ± SEM of three independent

experiments, and differences were considered to be significant when

P \ 0.05 (one-way ANOVA). b Cellular redox status in SH-SY5Y

cells at the end of the drug treatment. The cells were treated for

7 days, and the assay was performed. Chemiluminescence was

measured at room temperature in a liquid scintillation counter as

described under ‘‘Materials and methods’’. Data are expressed as

means ± SEM for three individual experiments, and differences were

considered to be significant when P \ 0.05 (one-way ANOVA)

Mol Cell Biochem (2011) 358:325–334 329

123

treated with wortmannin (PI3 K-/Akt-specific inhibitor)

showed an apparent reduction in cell viability and there-

fore, impaired the process of RA-induced neuritogenesis,

as assessed by morphological criteria (Fig. 5a). On the

contrary, when cells were co-treated with RA, wortmannin

and the antioxidant Trolox, we noticed a clear rescue of the

previously observed wortmannin-derived inhibitory effects

on cell viability and thus, the number of cell undergoing

neurite-like development (morphological criteria)

(Fig. 5a). In order to support these observations, we per-

formed further analysis of cell-culture viability by flow

cytometry after 7 days, showing that cells treated with

RA and wortmannin have a significant reduction in cell

viability compared with cells treated with RA alone, as

expected (Fig. 5b). On the other hand, when cells were

co-treated with RA, wortmanin, and Trolox, cell viability

was rescued when comparing to RA/wortmannin-treated

cultures. Moreover, PI incorporation was higher in cells

treated with 10 lM RA than that in control cells, sug-

gesting that RA caused a loss of membrane integrity. Such

effect was abolished by the addition of Trolox.

Discussion

Parkinson’s disease (PD) is a neurodegenerative disease

with no definitive cure so far [44], which is pathologically

characterized by chronic loss of dopamine-producing

neurons in the substantia nigra pars compacta [1, 2]. Cell

transplantation for replacing the damaged neurons repre-

sents a promising approach to restore tissue functionality in

different neurodegenerative diseases, including PD. How-

ever, many are the obstacles to overcome, such as ethical

concerns, selection of the best viable model for cell

transplantation or how to induce and control the differen-

tiation of these cells into specific cellular phenotypes; in

the case of PD, dopamine-producing cells.

SH-SY5Y cells respond to RA and give rise to neuronal-

like cells, acquiring morphological and neurochemical

properties characteristic of neurons. In that sense, they may

represent an available and optimal model to determine

what kind of factors affect biochemical routes leading to

neuronal differentiation in vitro, and whether they might

potentially be beneficial in future clinical trials. In general,

promotion of cellular differentiation implies a complex

re-arrangement of intrinsic cellular programs, cell-to-cell

interaction and the effect of diverse growth factors, hor-

mones as well as other signaling molecules. As a matter of

fact, it is well-known the crucial importance RA during

early embryonic development and organogenesis [45–47];

but the possible influence of ROS production during these

functions still needs further studies.

In the present work, we examined the effect of

the antioxidant Trolox on RA-induced neuritogenesis of

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

RA+TroloxRAControl

Treatments

MD

A e

qu

ival

ents

0

0.5

1

1.5

2

2.5

RA+TroloxRAControl

Treatments

CA

T (

U/m

g p

rote

in)

02468

101214

Control RA RA + Trolox

TreatmentsSO

D (

U/m

g p

rote

in)

02468

101214161820

RA+TroloxRAControl

Treatments

GP

x (U

/mg

pro

tein

)

BA

C D

*

*

**

(nm

ol/m

g p

rote

in)

Fig. 4 Lipid peroxidation and oxidative stress enzymatic profile of

RA-treated cells before and after Trolox co-treatment. a Determina-

tion of TBARS in SH-SY5Y cells at the end of the drug treatment.

Data are the means ± SEM of three independent experiments, and

differences were considered to be significant when P \ 0.05 (one-

way ANOVA). b Catalase activity in SH-SY5Y cells at the end of the

drug treatment. Data are the means ± SEM of three independent

experiments, and differences were considered to be significant when

P \ 0.05 (one-way ANOVA). c Superoxide dismutase activity in SH-

SY5Y cells at the end of the drug treatment. Data are the

means ± SEM of three independent experiments, and differences

were considered to be significant when P \ 0.05 (one-way ANOVA).

d Glutathione peroxidase activity in SH-SY5Y cells at the end of the

drug treatment. Data are the means ± SEM of three independent

experiments, and differences were considered to be significant when

P \ 0.05 (one-way ANOVA)

330 Mol Cell Biochem (2011) 358:325–334

123

SH-SY5Y cells. Our results show a significant enhance-

ment of RA (10 lM)-induced neuritogenesis and TH

endogenous expression, when cells were co-treated with

Trolox (100 lM) for 7 days.

To search for a link between cellular redox status and

the induction of neurite-like processes in SH-SY5Y cells,

we performed a real-time DCFH-DA assay. As shown in

Fig. 3a, RA did not induce ROS production when com-

pared to control cells. However, co-treatment with 100 lM

Trolox was able to significantly decrease the basal levels of

free radical generation observed under this condition.

Moreover, evaluation of total reactive antioxidant potential

in RA-treated SH-SY5Y cells by TRAP assay suggested

that RA was leading to partial consumption of cellular non-

enzymatic antioxidants (Fig. 3b). In other words, whenever

RA is inducing neuritogenesis in our cell model, Trolox

optimize this process by increasing, at least in part, the

viability of the culture.

ROS are known to activate an array of intracellular

signaling cascades that are closely associated with both cell

death and cell survival pathways [48]. On one hand, pre-

vious studies have reported that H2O2 induces apoptosis in

SH-SY5Y cells and rat pheochromocytoma PC12 cells (a

model system for catecholamine-containing neurons), but

the mechanisms involved are not fully understood [49, 50].

On the other hand, an interesting work suggested that a

transient enhancement of the intracellular ROS level plays

an important role during docosahexaenoic acid–enhanced

neurite outgrowth [51], and more recently, Franco et al.

[52] demonstrated that SH-SY5Y cells transfected with

human TH isoform 1 were substantially more resistant to

H2O2 and 6-hydroxydopamine-induced cell death, when

A

B

0

5

10

15

20

25

30

35

40

Control RA RA + Trolox RA + UO126 RA + W RA + W +Trolox

Treatments

% P

I po

siti

ve c

ells

*

*

Control RA RA + Trolox

RA + W + TroloxRA + WRA + UO126

100

X

Fig. 5 Effects of MEK and PI3 K/Akt inhibitors on RA-induced

neuritogenesis and cell viability before and after Trolox treatment.

a Phase-contrast micrographs of SH-SY5Y cells. Control, control

cells; RA, RA-treated cells (10 lM) for 7 days; RA ? Trolox,

RA-treated cells (10 lM) plus Trolox (100 lM) for 7 days;

RA ? UO126, RA-treated cells (10 lM) plus MEK inhibitor

UO126 (10 lM) for 7 days; RA ? W, RA-treated cells (10 lM)

plus PI-3 K inhibitor wortmannin (500 nM) for 7 days; RA ?

W ? Trolox, RA-treated cells (10 lM) plus wortmannin (500 nM)

plus Trolox (100 lM) for 7 days. Culture medium was replaced every

3 days. b Effects of RA and specific inhibitors of MEK and PI-3 K/

Akt signal pathways on cell viability. Red fluorescence of DNA-

bound PI in individual SH-SY5Y cells was measured using a flow

cytometer as described under ‘‘Materials and methods’’. The cells

were treated for 7 days, and the assay was performed. Data are the

means (%) ± SEM of three independent experiments, and differences

were considered to be significant when P \ 0.05 (one-way ANOVA)

Mol Cell Biochem (2011) 358:325–334 331

123

compared to proliferative neuroblastoma cells, suggesting

that a preconditioning-like mechanism linked to higher

dopamine levels increased the resistance of cells against

oxidative insults.

That is why we decided to study the effects exerted by

our in vitro treatments on the oxidative damage and anti-

oxidant enzyme activities. When neuroblastoma cells were

exposed to RA for 1 week, an increase in lipid peroxidation

was observed (Fig. 4a). The regulation of the antioxidant

enzymes seems to be important for holding the balance

between generation and consumption of ROS. For this

reason, we additionally determined possible changes in

CAT, SOD, and GPx activities. In comparison with pro-

liferative cells, CAT activity was significantly increased in

RA-treated cells (Fig. 4b). In contrast, neither SOD nor

GPx activities significantly vary. However, co-treatment

with Trolox resulted in a significant decrease in SOD

(Fig. 4c) and GPx (Fig. 4d) activities, indicating a reduc-

tion in ROS production.

We further investigated the influence of two signaling

pathways, MEK and PI3 K/Akt pathways in RA-induced

neuritogenesis (Fig. 5a) and cell viability (Fig. 5b) and in

which way co-treatment with Trolox could affect it. Both

signaling pathways are known to be important in mediating

neuronal survival and differentiation [53]. Cells treated

with RA plus wortmannin showed a decreased viability

of the culture and therefore, less efficient neuritogenic

induction when compared to RA-treated cells. However,

when cells were co-treated with RA, wortmanin, and

Trolox, we observed that those negative effects were res-

cued. UO126 did not apparently affect RA-induced neurite

formation (assessed by morphological criteria).

Some authors have reported that RA actives ERK1/2,

PKC, and PI3 K in cells where retinoid RXR/RAR recep-

tors were silenced, suggesting a RA receptor-independent

mechanism [54, 55]. Others demonstrated that RA actives

PI-3 K/Akt pathway by RA-mediated interaction of RAR

with the catalytic subunit of PI-3 K-inducing Akt phos-

phorylation [56]. However, the exact mechanism underly-

ing non-genomic actions of RA and other retinoids still

requires further studies.

In contrast to the physiological effects of RA (10-9–

10-8 M), which are primarily mediated by specific nuclear

receptors, the mechanism underlying RA toxicity upon

pharmacological exposure (10-7–10-5 M) remains unde-

fined [46], and nowadays, it is unclear whether oxidative

stress is the main mechanism of RA toxicity. Our results

suggest that a concomitant decrease in basal ROS pro-

duction can exert a positive effect on RA-induced neurite

development. Moreover, a coordinated regulation of ROS

production and PI3-K activation may play an important

role in RA-induced neuritogenesis of human SH-SY5Y

neuroblastoma cells; it is possible that Trolox-induced

antioxidant effects may counter balance the basal

RA-induced cytotoxicity, and thus, survival effects of

RA-induced phosphorylation of Akt would take place.

Thus, we believe that an antioxidant-based strategy could

avoid a premature death of cells entering in differentiation

programs, allowing the survival of newly formed TH

positive cells. Nevertheless, additional studies are required

for better comprehension of the exact mechanism by which

Trolox affects the RA-induced neurite-like process for-

mation, as well as its possible significance in the modula-

tion of PI 3-K pathway. Since optimal neuritogenesis and

survival of the newly formed neurons is a critical step of

neuronal development, we expect that our results may be

relevant to the development of new strategies involving

cellular transplantation and replacement of dopamine-pro-

ducing cells.

Acknowledgments This work was supported by CNPq, CAPES,

FAPERGS, PROPESQ-UFRGS and ‘‘Rede Instituto Brasileiro deNeurociencia’’ (IBM-Net) # 01.06.0842.00.

Conflict of interest None.

References

1. Bernheimer H, Birkmayer W, Hornykiewiez O, Jellinger K,

Seitelberger F (1973) Brain dopamine and the syndromes of

Parkinson and Huntington. Clinical, morphological and neuro-

chemical correlations. J Neurol Sci 20(4):415–455

2. Hirsch E, Graybiel AM, Agid YA (1988) Melanized dopami-

nergic neurons are differentially susceptible to degeneration in

Parkinson’s disease. Nature 334:345–348

3. Bjorklund A, Lindvall O (2000) Cell replacement therapies for

central nervous system disorders. Nat Neurosci 3:537–544

4. Perrier AL, Taba V, Barberi T, Rubio ME, Bruses J, Topf N,

Harrison NL, Studer L (2004) Derivation of midbrain dopamine

neurons from human embryonic stem cells. Proc Natl Acad Sci

USA 34:12543–12548

5. Wernig M, Zhao JP, Pruszak J, Hedlund E, Fu D, Soldner F,

Broccoli V, Constantine-Paton M, Isacson O, Jaenisch R (2008)

Neurons derived from reprogrammed fibroblasts functionally

integrate into the fetal brain and improve symptoms of rats with

Parkinson’s disease. Proc Natl Acad Sci USA 15:5856–5861

6. Swistowski A, Peng J, Liu Q, Mali P, Rao MS, Cheng L, Zeng X

(2010) Efficient generation of functional dopaminergic neurons

from human induced pluripotent stem cells under defined con-

ditions. Stem Cells 10:1893–1904

7. Trzaska KA, Kuzhikandathil EV, Rameshwar P (2007) Specifi-

cation of a dopaminergic phenotype from adult human mesen-

chymal stem cells. Stem Cells 11:2797–2808

8. Chen S, Xianwen C, Dehua X, Zhenguo L, Lingfei X, Smith SW,

Zhongcheng Z (2003) Behavioral correction of Parkinsonian rats

following the transplantation of immortalized fibroblasts geneti-

cally modified with TH and GCH genes. Parkinsonism Relat

Disord 9(2):91–97

9. Zeidan-Chulia F, Noda M (2009) ‘‘Opening’’ the mesenchymal

stem cell tool box. Eur J Dent 3:240–249

10. Arenas E (2010) Towards stem cell replacement therapies for

Parkinson’s disease. Biochem Biophys Res Commun 396(1):

152–156

332 Mol Cell Biochem (2011) 358:325–334

123

11. Ross RA, Spengler BA, Biedler JL (1983) Coordinate morpho-

logical and biochemical interconversion of human neuroblastoma

cells. J Natl Cancer Inst 71:741–747

12. Nicolini G, Miloso M, Zoia C, Di Silvestro A, Cavaletti G,

Tredici G (1998) Retinoic acid differentiated SH-SY5Y human

neuroblastoma cells: an in vitro model to assess drug neurotox-

icity. Anticancer Res 8:2477–2481

13. Pahlman S, Hoehner JC, Nanberg E, Hedborg F, Fagerstrom S,

Gestblom C, Johansson I, Larsson U, Lavenius E, Ortoft E, Soder-

holm H (1995) Differentiation and survival influences of growth

factors in human neuroblastoma. Eur J Cancer 31A:453–458

14. Kim B, Leventhal PS, Saltiel AR, Feldman EL (1997) Insulin-

like growth factor immediate neurite outgrowth in vitro requires

mitogen-activated protein kinase activation. J Biol Chem

272:21268–21273

15. Encinas M, Iglesias M, Llecha N, Comella JX (1999) Extracel-

lular-regulated kinases and phosphatidylinositol 3-kinase are

involved in brain-derived neurotrophic factor mediated survival

and neuritogenesis of the neuroblastoma cell line SH-SY5Y.

J Neurochem 73:1409–1421

16. Olsson AK, Vadhammar K, Nanberg E (2000) Activation and

protein kinase C dependent nuclear accumulation of ERK in

differentiating human neuroblastoma cells. Exp Cell Res 256:

454–467

17. Miloso M, Villa D, Crimi M, Galbiati S, Donzelli E, Nicolini G,

Tredici G (2004) Retinoic acid-induced neuritogenesis of human

neuroblastoma SH-SY5Y cells is ERK independent and PKC

dependent. J Neurosci Res 75:241–252

18. Glass CK, Rosenfeld MG (2000) The coregulator exchange in

transcriptional functions of nuclear receptors. Genes Dev 14:

121–141

19. Aranda A, Pascual A (2001) Nuclear hormone receptors and gene

expression. Physiol Rev 81:1269–1304

20. Radominska-Pandya A, Chen G, Czernik PJ, Little JM, Samo-

kyszyn VM, Carter CA, Nowak G (2000) Direct interaction of

All-trans-retinoic acid with protein kinase C (PKC): implications

for PKC signaling and cancer therapy. J Biol Chem 275:

22324–22330

21. Frota MLC, Silva EG, Behr GA, Oliveira MR, Dal-Pizzol F,

Klamt F, Moreira JCF (2006) All-trans retinoic acid induces free

radical generation and modulate antioxidant enzyme activities in

rat sertoli cells. Mol Cell Biochem 285:173–179

22. Castro-Obregon S, Covarrubias L (1996) Role of retinoic acid

and oxidative stress in embryonic stem cell death and neuronal

differentiation. FEBS Lett 381:93–97

23. Delia D, Aiello A, Meroni L, Nicolini M, Reed JC, Pierotti MA

(1997) Role of antioxidants and intracellular free radicals in re-

tinamide-induced cell death. Carcinogenesis 18:943–948

24. Furuke K, Sasada T, Ueda-Taniguchi Y, Yamauchi A, Inamoto T,

Yamaoka Y, Masutani H, Yodoi J (1997) Role of intracellular

redox status in apoptosis induction of human T-cell leukemia

virus type I- infected lymphocytes by 13 cis-retinoic acid. Cancer

Res 57:4916–4923

25. Wang H, Joseph JA (1999) Quantifying cellular oxidative stress

by dichlorofluorescein assay using microplate reader. Free Radic

Biol Med 27:612–616

26. Draper HH, Hadley M (1990) Malondialdehyde determination as

index of lipid peroxidation. Methods Enzymol 186:421–431

27. Aebi H (1984) Catalase in vitro. Methods Enzymol 105:121–126

28. Bannister JV, Calabrese L (1987) Assays for SOD. Methods

Biochem Anal 32:279–312

29. Flohe L, Gunzler WA (1984) Assays of glutathione peroxidase.

Methods Enzymol 105:114–121

30. Bradford MM (1976) A rapid and sensitive method for the

quantitation of microgram quantities of protein utilizing the

principle of protein-dye binding. Anal Biochem 72:248–254

31. Lowry OH, Rosebrough AL, Farr AL, Randal RJ (1951) Protein

measurement with the Folin phenol reagent. J Biol Chem 193:

265–275

32. Encinas M, Iglesias M, Liu Y, Wang H, Muhaisen A, Cena V,

Gallego C, Comella JX (2000) Sequential treatment of SH-SY5Y

cells with retinoic acid and brain-derived neurotrophic factor

gives rise to fully differentiated, neurotrophic factor-dependent,

human neuron-like cells. J Neurochem 75:991–1003

33. Dunkley PR, Bobrovskaya L, Graham ME, von Nagy-Felsobuki

EI, Dickson PW (2004) Tyrosine hydroxylase phosphorylation:

regulation and consequences. J Neurochem 91:1025–1043

34. Fernandez-Checa JC (2003) Redox regulation and signaling lip-

ids in mitochondrial apoptosis. Biochem Biophys Res Commun

3:471–479

35. Pantano C, Reynaert NL, van der Vliet A, Janssen-Heininger YM

(2006) Redox-sensitive kinases of the nuclear factor-kappaB

signaling pathway. Antioxid Redox Signal 9–10:1791–1806

36. Hansen JM (2006) Oxidative stress as a mechanism of terato-

genesis. Birth Defects Res C Embryo Today 4:293–307

37. Anilkumar N, Sirker A, Shah AM (2009) Redox sensitive sig-

naling pathways in cardiac remodeling, hypertrophy and failure.

Front Biosci 14:3168–3187

38. Inanami O, Watanabe Y, Syuto B, Nakano M, Tsuji M, Kuwa-

bara M (1998) Oral administration of (-)catechin protects against

ischemia-reperfusion-induced neuronal death in the gerbil. Free

Radic Res 4:359–365

39. Inanami O, Takahashi K, Kuwabara M (1999) Attenuation of

caspase-3-dependent apoptosis by Trolox post-treatment of

X-irradiated MOLT-4 cells. Int J Radiat Biol 2:155–163

40. Niwa K, Inanami O, Yamamori T, Ohta T, Hamasu T, Kubawara

M (2003) Redox regulation of PI3K/Akt and p53 in bovine aortic

endothelial cells exposed to hydrogen peroxide. Antioxid Redox

Signal 6:713–722

41. Kuwabara M, Asanuma T, Niwa K, Inanami O (2008) Regulation

of cell survival and death signals induced by oxidative stress.

J Clin Biochem Nutr 2:51–57

42. Numakawa Y, Numakawa T, Matsumoto T, Yagasaki Y, Kuma-

maru E, Kunugi H, Taguchi T, Niki E (2006) Vitamin E protected

cultured cortical neurons from oxidative stress-induced cell death

through the activation of mitogen-activated protein kinase and

phosphatidylinositol 3-kinase. J Neurochem 4:1191–1202

43. Lopez-Carballo G, Moreno L, Masia S, Perez P, Barettino D(2002) Activation of the phosphatidylinositol 3-kinase/Akt sig-

naling pathway by retinoic acid is required for neural differen-

tiation of SH-SY5Y human neuroblastoma cells. J Biol Chem

28:25297–25304

44. Kedar NP (2003) Can we prevent Parkinson’s and Alzheimer’s

disease? J Postgrad Med 49:236–245

45. Bain G, Kitchens D, Yao M, Huettner JE, Gottlieb DI (1995)

Embryonic stem cells express neuronal properties in vitro. Dev

Biol 2:342–357

46. Ross SA, McCaffery PJ, Drager UC, De Luca LM (2000) Reti-

noids in embryonal development. Physiol Rev 80:1021–1054

47. Duester G (2008) Retinoic acid synthesis and signaling during

early organogenesis Gregg Duester. Cell 6:921–931

48. Kamata H, Hirata H (1999) Redox regulation of cellular signal-

ling. Cell Signal 11:1–14

49. Zhang L, Zhao B, Yew DT, Kusiak JW, Roth GS (1997) Pro-

cessing of Alzheimer’s amyloid precursor protein during H2O2-

induced apoptosis in human neuronal cells. Biochem Biophys

Res Commun 235:845–848

50. Yamakawa H, Ito Y, Naganawa T, Banno Y, Nakashima S,

Yoshimu S, Sawada M, Nishimura Y, Nozawa Y, Sakai N (2000)

Activation of caspase-9 and 3 during H2O2-induced apoptosis of

PC12 cells independent of ceramide formation. Neurol Res 22:

556–564

Mol Cell Biochem (2011) 358:325–334 333

123

51. Wu H, Ichikawa S, Tani C, Zhu B, Tada M, Shimoishi Y, Murata

Y, Nakamura Y (2009) Docosahexaenoic acid induces ERK1/2

activation and neuritogenesis via intracellular reactive oxygen

species production in human neuroblastoma SH-SY5Y cells.

Biochim Biophys Acta 1791:8–16

52. Franco JL, Posser T, Gordon SL, Bobrovskaya L, Schneider JJ,

Farina M, Dafre AL, Dickson PW, Dunkley PR (2010) Expres-

sion of tyrosine hydroxylase increases the resistance of human

neuroblastoma cells to oxidative insults. Toxicol Sci 113(1):

150–157

53. Kaplan DR, Stephens RM (1994) Neurotrophin signal transduc-

tion by the Trk receptor. J Neurobiol 25:1404–1417

54. Hoyos B, Imam A, Chua R, Swenson C, Tong GX, Levi E, Noy

N, Hammerling U (2000) The cysteine-rich regions of the regu-

latory domains of Raf and protein kinase C as retinoid receptors.

J Exp Med 192:835–845

55. Aggarwal S, Kim SW, Cheon K, Tabassam F, Joon JH, Koo JS

(2006) Nonclassical action of retinoic acid on the activation of

the camp response element-binding protein in normal human

bronchial epithelial cells. Mol Biol Cell 17:566–575

56. Masia M, Alvarez S, de Lera AR, Barettino D (2007) Rapid,

nongenomic actions of retinoic acid on phosphatidylinositol-3-

kinase signaling pathway mediated by the retinoic acid receptor.

Mol Endocrinol 21(10):2391–2402

334 Mol Cell Biochem (2011) 358:325–334

123