all-trans retinoic acid inhibits cobalt chloride-induced apoptosis in pc12 cells: role of the...

All-Trans Retinoic Acid Inhibits CobaltChloride-Induced Apoptosis in PC12 Cells:Role of the DimethylarginineDimethylaminohydrolase/AsymmetricDimethylarginine Pathway

Shan Wang,1 Chang-Ping Hu,2 De-Jian Jiang,2 Jun Peng,2 Zhi Zhou,2

Qiong Yuan,2 Sheng-Dan Nie,2 Jun-Lin Jiang,2 Yuan-Jian Li,2

and Ke-Long Huang1*1Department of Pharmaceutical Engineering, College of Chemistry and Chemical Engineering,Central South University, Changsha, China2Department of Pharmacology, School of Pharmaceutical Sciences, Central South University,Changsha, China

Previous studies have shown that the endogenous ni-tric oxide synthase inhibitor asymmetric dimethylargi-nine (ADMA) and its specific hydrolase dimethylargininedimethylaminohydrolase (DDAH) are involved in the reg-ulation of apoptosis in different cell types. In the pres-ent study, we investigated the role of the DDAH/ADMApathway in cobalt chloride (CoCl2)–induced apoptosisand the antiapoptotic effect of all-trans retinoic acid(atRA) in undifferentiated pheochromocytoma (PC12)cells. Treatment of CoCl2 (125 lM) for 48 hr significantlyinduced the apoptosis of PC12 cells, concomitantlywith increased intracellular reactive oxygen species(ROS) production and caspase-3 activity. CoCl2 treat-ment also decreased the activity of DDAH and theexpression of DDAH2 (mRNA and protein), resulting inan increased level of ADMA. All these alterationsinduced by CoCl2 were attenuated by atRA (0.1, 1, or10 lM). Interestingly, the antiapoptotic effects of atRAwere inhibited by DDAH2 small RNA interference. Incontrast, DDAH2 overexpression inhibited the proapop-totic effects of CoCl2. We also found that treatment ofexogenous ADMA (3, 10, or 30 lM) induced the apo-ptosis of PC12 cells in a concentration- and time-de-pendent manner, which was inhibited by the antioxidantor the caspase-3 inhibitor. These findings suggest thatthe modulation of the DDAH/ADMA/ROS pathway playsan important role in CoCl2-induced apoptosis and theantiapoptotic effects of atRA in undifferentiated PC12cells. VVC 2009 Wiley-Liss, Inc.

Key words: caspase-3; nitric oxide; reactive oxygenspecies

Cobalt is an essential element for human life (Kar-ovic et al., 2007). However, excessive cobalt is harmfulto the organism. It has been reported that excessive

cobalt has neurotoxic effects, such as inhibiting synaptictransmission, blocking postsynaptic responses (Gerberand Gahwiler, 1991), and depleting neurotransmitter(Hasan et al., 1980). Furthermore, cobalt chloride(CoCl2) can directly induce neuronal cell apoptosis,accompanying increased production of reactive oxygenspecies (ROS; Wang et al., 2000; Olivieri et al., 2001;Zou et al., 2001).

Although CoCl2 induces overproduction of nitricoxide (NO; Ciafre et al., 2007; Karovic et al., 2007),preincubation of CoCl2-treated astrocytes with the selec-tive inducible NO synthase (iNOS) inhibitors nx-nitro-D-arginine methyl ester hydrochloride or l-N6-(1-imi-noethyl)lysine hydrochloride fails to protect the cellsagainst apoptosis (Karovic et al., 2007). These findingssuggest that overproduction of NO may not play themajor role in CoCl2-induced apoptosis.

Abbreviations used: ADMA, asymmetric dimethylarginine; atRA, all-trans

retinoic acid; DCF, 20,70-dichlorofluorescein; DDAH, dimethylarginine

dimethylaminohydrolase; FITC, fluoresceinisothiocyanate; Hcy, homo-

cysteine; NO, nitric oxide; NOS, nitric oxide synthase; PDTC, pyrroli-

dine dithiocarbamate; RAR, retinoic acid receptor; ROS, reactive oxy-

gen species; RXR, retinoid X receptor; VSMC, vascular smooth muscle

cell.

Contract grant sponsor: National Natural Science Foundation of China;

Contract grant number: 30600817.

*Correspondence to: Ke-Long Huang, PhD, Department of Pharmaceu-

tical Engineering, College of Chemistry and Chemical Engineering, Cen-

tral South University, P. O. Box 58, 110 Xiang-Ya Road, Changsha

410078, China. E-mail: [email protected]

Received 6 September 2008; Revised 3 November 2008; Accepted 16

November 2008

Published online 20 January 2009 in Wiley InterScience (www.

interscience.wiley.com). DOI: 10.1002/jnr.21999

Journal of Neuroscience Research 87:1938–1946 (2009)

' 2009 Wiley-Liss, Inc.

Asymmetric dimethylarginine (ADMA) has recentlyemerged as a major endogenous inhibitor of nitric oxidesynthase (NOS). There is growing evidence thatADMA, in addition to regulating NO production, couldalso play a key role in inducing proinflammatory cyto-kine secretion (Wang et al., 2007), intracellular ROSproduction, and cell apoptosis (Jiang et al., 2006; Yuanet al., 2007). Most of ADMA is directly degradedby dimethylarginine dimethylaminohydrolase (DDAH),which hydrolyzes ADMA to l-citrulline and dimethyl-amine. It is well documented that DDAH activity is sen-sitive to oxidative stress. It has been shown that L-homo-cysteine (Hcy) can inactivate DDAH in neurons byreacting with the cysteine residue in its active site andlead to the accumulation of ADMA, which is attenuatedby the antioxidant pyrrolidine dithiocarbamate (PDTC;Selley, 2004). Decreased DDAH activity and increasedADMA levels have also been reported to occur underanother severe oxidative stress (Stuhlinger et al., 2007).More recently, we have found that dysfunction of theDDAH/ADMA pathway is involved in the apoptoticdeath of vascular smooth muscle cells (VSMCs; Yuanet al., 2007).

Retinoic acid (RA), including all-trans retinoicacid (atRA) and 9-cis retinoic acid (9-cis RA), can bindto the retinoic acid receptor (RAR) and the retinoid Xreceptor (RXR), resulting in transcriptional stimulationor repression of numerous target genes (Chambon, 1969;Maden, 2001). It has been shown that the effect ofatRA on cell apoptosis is cell type dependent. AtRAinduces apoptosis in multiple cell types such as tumorand embryonic cells, whereas it exerts antiapoptoticaction in embryonic neurons and mesangial cells (Ahle-meyer et al., 2001; Kitamura et al., 2002; Xu et al.,2002). It has been reported that atRA induces theexpression of DDAH2 in endothelial cells (Achan et al.,2002) and that the atRA-mediated protective effects invarious cells are related to the inhibition of oxidativestress (Ahlemeyer and Krieglstein, 1998; Jackson et al.,1991; Moreno-Manzano et al., 1999). Because DDAHactivity is sensitive to oxidative stress and atRA is able toup-regulate the expression of DDAH2, in this study, wetherefore investigated the role of the DDAH/ADMApathway in CoCl2-induced apoptosis and the antiapop-totic effect exerted by atRA in undifferentiated pheo-chromocytoma (PC12) cells.

MATERIALS AND METHODS

Materials

PC12 cells (American Type Culture Collection, CRL-1721) were obtained from the Academia Sinica (Shanghai,China). The RPMI 1640, horse serum, and fetal bovine se-rum were from Hyclone (Laboratories, Inc., Logan, UT). Theantibodies against DDAH2 and b-actin were purchased fromSanta Cruz Biotechnology (Santa Cruz, CA). The CoCl2, col-lagen, ADMA, atRA, PDTC, and caspase-3 activity assay kitswere purchased from Sigma (St. Louis, MO). Acetyl-DEVD-CHO was purchased from Calbiochem (San Diego, CA); flu-

oresceinisothocyanate (FITC)–annexin V and PropidiumIodide (PI) were obtained from Jingmei Biotech (Shenzheng,China). Hoechst33342 and the fluorescent ROS detection kitwere purchased from Beytime Biotechnology (Jiangsu, China).All other biochemicals used were of the highest purity avail-able.

Cell Culture

Undifferentiated PC12 cells (2.5 3 105 cells/mL) wereseeded on collagen-precoated tissue culture dishes with RPMI1640 supplemented with 10% horse serum, 5% fetal bovineserum, and 2 mM l-glutamine at 378C in 5% CO2/95% O2

air. CoCl2 was dissolved in distilled H2O. AtRA was dissolvedin dimethyl sulfoxide (DMSO). The final concentration ofDMSO was less than 0.1%. The cells was treated with CoCl2(125 lM) for 48 hr to induce apoptosis. For some experi-ments, AtRA (0.1, 1, 10 lM), PDTC (5 lM), or acetyl-DEVD-CHO (100 lM) was used for 1 hr before the CoCl2treatment.

MTT Assay

The MTT assay was performed as described previously(Vengellur and LaPres, 2004).

Apoptosis Analysis

Hoechst33342 staining assay. The cells were cul-tured in the 24-well plate and treated with Hoechst33342(100 lg/mL) for 10 min at 378C in the dark. Then the sam-ples were analyzed by fluorescence microscopy at 521 nm ofemission wavelength.

Annexin V–PI analyzed by flow cytometry. Thecells were collected and washed with PBS 2 times and thenwere resuspended in 250 lL of binding buffer. Then 5 lL ofFITC–annexin V and 10 lL of PI (20 lg/mL) were addedper 100 lL of cell suspension. The cells were incubated atroom temperature for 15 min. After 400 lL of PBS wasadded into the mixed liquor, the cells were analyzed forannexin V binding within 1 hr with a flow cytometer(Becton-Dickinson).

Measurement of caspase-3 activity. The activity ofcaspase-3-like protease in the lysate was measured using a col-orimetric caspase-3 assay (Jiang et al., 2006). The caspase-3 ac-tivity was expressed by value of OD405.

ROS Assay

Intracellular ROS generation was measured with the useof the fluorescent signal 20,70-dichlorodihydrofluorescein diac-etate (H2DCF-DA, 10 lM, 20 min), a cell-permeable indica-tor of ROS (Yuan et al., 2007). H2DCF-DA is nonfluores-cent until the acetate groups are removed by intracellularROS. The ROS-mediated fluorescence was observed under afluorescent microscope (Backman) with excitation set at 502nm and emission set at 523 nm and results expressed as arbi-trary units.

DDAH/ADMA Pathway and Apoptosis of PC12 Cells 1939

Journal of Neuroscience Research

Determination of ADMA Concentration

The level of ADMA in the conditioned medium wasmeasured by HPLC as described previously (Jiang et al., 2002;Wang et al., 2007).

DDAH Activity Assay

The activity of DDAH in PC12 cells was estimated bydirectly measuring the amount of ADMA metabolized byDDAH as previously described (Lin et al., 2002) and expressedas the percentage of the ADMA metabolized compared withthe control.

DDAH2 RNA Interference

Small interfering RNA (siRNA; Dharmacon, Lafayette,CO) was generated against the following rat DDAH2 mRNAsequences: siRNA-1, 50-GCAGACACGUUCCGGGACUUU;siRNA-2, 50-GGAGUAGAAUCAGGUAAUAUU; siRNA-3, 50-GGUUGAUGGAG- UCCGCAAAUU; and siRNA-4,50-CAGAUGACGCAGCGAGUGAUU. Dharmacon SMART-pool used an algorithm to combine four or more SMARTselected siRNA duplexes in a single pool. PC12 cells wereplated in six-well plates and transfected with 100 nM Dhar-macon SMARTpool siRNA along with 5 lL of Dharma-FECT transfection reagent 3. As a control, scrambled siRNAnot exhibiting homology to any coding region was used.CoCl2 was added 24 hr after transfection.

Plasmid Construction and Cell Transfection

RNA was reverse-transcribed using an oligo(dT) primer,and a DDAH2 open-reading frame (ORF) was PCR-amplifiedusing primers 50-GATCGAATTCAGGATG GGGACG-CCGGGG-30 (underlined encodes an EcoRI site,) and 50-GATCTCTAGAT CAGCTGTGGGGGCGTGTG-30 (under-lined encodes an XbaI site). Product was gel-purified andcloned into the EcoRI/XbaI site of vector pcDNA3.1 (Invi-trogen). We named the expression vector pcDNA3.1/DDAH2.

Undifferentiated PC12 cells were transfected with 2 lgof pcDNA3.1/DDAH2 or pcDNA3.1 by Lipofectamin 2000(Invitrogen Corporation, Carlsbad, CA). Forty-eight hoursafter transfection, cells containing expression plasmids wereselected by incorporating 800 lg/mL G418 sulfate (Sigma)into culture medium. Then individual colonies of resistantcells were transferred to a 24-well plate containing RPMI1640 supplemented with 300 lg/mL G418 sulfate. DDAH2expression was screened by Western blotting.

Real-Time Quantitative RT-PCR Analysis

Total RNA was extracted using TRIzol Reagent (Invi-trogen). Total RNA (2 lg) was used as a template in 20 lLof a cDNA synthesis reaction. The cDNA was synthesizedusing a RevertAidTM First Stand cDNA Synthesis Kit (Invi-trogen). PCR was performed using Power SYBR Green PCRMaster Mix (Applied Biosystems). The final reaction con-tained 10.7 lL of SYBR green/enzyme reaction mix, 0.4 lMof primer and 1 lL of cDNA in a total volume of 25 lL.PCR conditions were 508C for 2 min, 958C for 10 min, fol-lowed by 45 cycles of 958C for 15 sec and 608C for 1 min.GAPDH was used as an internal standard. All results were

repeated in six independent experiments and performed intriplicate each time. The primers were as follows: GAPDH,50-TGGCCTCCAAGGAGTAAGAAAC and 50-GGCCTCT-CTCTTGCTCTCAGTATC; and DDAH2, 50-AAAGCCGT-CAGGGCAATG and 50-CGTCATCTGGGAGGGTCAGA.

Western Blot Analysis

Cells were lysed in iced lysis buffer. Total protein(50 lg/lane) was separated by SDS-PAGE and transferred to anitrocellulose membrane. After incubation in blocking solu-tion (5% nonfat milk; Sigma), membranes were incubatedwith primary antibodies for DDAH2 (1:400) and b-actin(1:1,000) overnight at 48C. Membranes were washed andthen incubated with a 1:2,000 dilution of horseradish peroxi-dase–conjugated secondary antibody (Santa cruz). Relativedensity of each protein band was normalized to that of b-actin. All results are representative of at least five independentexperiments.

Statistical Analysis

Results are expressed as means 6 SEMs. Data were ana-lyzed by ANOVA followed by the Student-Newman-Keulstest. The significance level was chosen as P < 0.05.

RESULTS

Effect of atRA on CoCl2-Induced Loss of PC12Cell Viability

The MTT assay showed that CoCl2 (125 lM,48 hr) treatment led to a significant decrease in cellviability. Different concentrations of atRA (0.1, 1, or10 lM) improved cell viability. AtRA (10 lM) itself didnot have a significant effect on cell viability (Fig. 1).

Fig. 1. Effect of atRA on CoCl2-induced loss of cell viability. Dataexpressed as mean 6 SEM of five independent experiments (*P <0.01 vs. control, #P < 0.01 vs. CoCl2-treated group).

1940 Wang et al.


Induction of Apoptosis of PC12 Cells by CoCl2 viaROS-dependent Pathway and AntiapoptoticEffects of atRA

Hoechst33342 staining assay showed that treatmentwith CoCl2 (125 lM) for 48 hr increased the ratio ofcells with a profile of cell shrinkage, chromatin conden-sation, and fragmented fluorescent nuclei (Fig. 2A).Annexin V–PI staining assay with flow cytometry alsoshowed that treatment with CoCl2 (125 lM) for 48 hrsignificantly increased the ratio of apoptotic cells (Fig.2B). Caspase-3 activity, a marker of apoptosis, was alsofound to be increased in CoCl2-treated cells in a time-dependent manner (Fig. 2C). CoCl2 increased intracellu-lar ROS generation, as indicated by the increase in20,70-dichlorofluorescein (DCF) fluorescence (Fig. 2D).Pretreatment of PC12 with the intracellular antioxidantPDTC (5 lM) or the caspase-3-specific inhibitorDEVD-CHO (100 lM) significantly attenuated the apo-ptosis elicited by CoCl2 (Fig. 2B). The number of

Hoechst33342-positive cells and the ratio of apoptosiswere significantly reduced by atRA (0.1, 1 or 10 lM),accompanying by reductions in ROS level and caspase-3activity (Fig. 2A–D).

Role of DDAH/ADMA Pathway inCoCl2-Induced Apoptosis and AntiapoptoticEffects of atRA

To explore the role of the DDAH/ADMA path-way in CoCl2-induced apoptosis, the activity/expressionof DDAH and the level of ADMA were determined.The results showed that incubation of PC12 cells withCoCl2 for 0–48 hr led to a significant decrease inDDAH activity and DDAH2 expression (mRNA andprotein) in a time-dependent manner (Fig. 3A–C). Incontrast, the level of ADMA in the medium was signifi-cantly elevated after incubation of PC12 cells withCoCl2 (Fig. 3D). All these alterations induced by CoCl2were inhibited by atRA (Fig. 3A–D).

Fig. 2. Induction of apoptosis of PC12 cells by CoCl2 via ROS-dependent pathway and antiapop-totic effects of atRA. A: Representative images of Hoechst33342 staining. B: Ratio of apoptoticcells. C: Caspase-3 activity. D: Level of intracellular ROS as indicated by dichlorofluorescein(DCF) fluorescence. Data expressed as mean 6 SEM of five independent experiments (*P < 0.01vs. control, #P < 0.01 vs. CoCl2-treated group).



To further confirm the role of the DDAH/ADMApathway, the strategies of DDAH2 siRNA and overex-pression were applied. Transfection of PC12 withDDAH2 siRNA successfully knocked down DDAH2expression at both the mRNA and protein levels inCoCl2/atRA-treated (Fig. 4A,B) and untreated PC12cells (data not shown). In contrast, scrambled siRNA,not targeting to any known cDNA, did not affectDDAH2 expression. As a result, the protective effects ofatRA (10 lM) against CoCl2-induced apoptosis wereblunted by DDAH2 siRNA, as shown by the increase inthe number of apoptotic cells and increased caspase-3 ac-tivity (Fig. 4C,D). Interestingly, the inhibitory effects ofatRA on CoCl2-induced increases in ROS generationand ADMA level were also abolished by DDAH2siRNA (Fig. 4E,F). We successfully constructed DDAH2overexpression in PC12 cells (Fig. 5A). CoCl2-inducedapoptosis assessed by the number of apoptotic cells andthe activity of caspase-3 was significantly inhibited byDDAH2 overexpression (Fig. 5B,C). DDAH2 overex-

pression also abolished CoCl2-induced increases in ROSgeneration and ADMA level (Fig. 5D,E).

Induction of Apoptosis of PC12 Cells byExogenous ADMA via ROS-Dependent Pathway

To seek direct evidence of the role of ADMA inCoCl2-induced apoptosis, the effects of exogenousADMA were tested. As shown in Figure 6A–C, ADMAinduced apoptosis of PC12 cells in a concentration- andtime-dependent manner, assessed by the number of apo-ptotic cells and the activity of caspase-3. Importantly,ADMA increased ROS generation in a time-dependentmanner (Fig. 6D). Furthermore, CoCl2-induced apopto-sis was significantly attenuated in the presence ofDEVD-CHO or PDTC (Fig. 6A).

DISCUSSION

The primary finding of our present study was thatthe DDAH/ADMA pathway plays a key role in CoCl2-

Fig. 3. Effect of CoCl2 and atRA on activity/expression of DDAH and ADMA level. A: DDAHactivity. B: Summary of DDAH2 mRNA expression. C: Representative DDAH2 protein expres-sion. D: ADMA level in conditioned medium. Data expressed as mean 6 SEM of five independ-ent experiments (*P < 0.01 vs. control; #P < 0.05, ##P < 0.01 vs. CoCl2-treated group).

1942 Wang et al.


induced apoptosis of undifferentiated PC12 cells. Inkeeping with previous studies (Zou et al., 2001; Junget al., 2007), the present results showed that CoCl2markedly increased intracellular ROS level, resulting incaspase-3-dependent apoptosis of undifferentiated PC12cells. Interestingly, CoCl2 also significantly increased theaccumulation of ADMA in a cultured medium of PC12cells. As an endogenous NOS inhibitor, ADMA has alsobeen found to induce intracellular ROS production andpromote inflammatory responses (Jiang et al., 2007;Wang et al., 2007; Yuan et al., 2007), suggesting it maybe a novel pro-oxidative and proinflammatory molecule.Our previous studies showed that exogenous ADMAcould directly induce cellular apoptosis via ROS-dependent signaling pathway in several cell types includ-ing endothelial cells and VSMCs (Jiang et al., 2006; Yuanet al., 2007). Thus, it is possible that elevation of ADMAlevel is involved in CoCl2-induced PC12 apoptosis. Thisspeculation was also indirectly evidenced by the presentresults that exogenous ADMA could increase ROS leveland caspase-3 activity, resulting in apoptosis of PC12cells, which was attenuated in the presence of intracellu-lar antioxidant and caspase-3-specific inhibitor. It is ofnote that the effective concentration of exogenousADMA in our present study is similar to that reported inprevious studies (Boger et al., 2000; Smirnova et al.,

2004; Jiang et al., 2006), which is higher than the con-centrations of ADMA measured on CoCl2 treatment.Maybe the concentration of ADMA measured in themedium cannot exactly reflect the intracellular concen-tration of endogenous ADMA and it takes a long timefor exogenous ADMA to enter cells. It is possible thatlower doses of ADMA can induce apoptosis of cells afterlonger times that are not practical to analyze in an invitro system.

DDAH is the specific hydrolase of ADMA and playsan important role in the modulation of ADMA level.There is evidence that DDAH activity is modulated atthe posttranslational or/and transcriptional levels. It hasbeen shown that there exists a sulfhydryl group in thecatalytic region of DDAH through which the DDAHactivity could be directly inactivated by various oxidativeor inflammatory stimuli including Hcy, oxidative low-density lipoprotein (ox-LDL), or TNF-a (Ito et al.,1999; Stuhlinger et al., 2001). On the other hand, theDDAH2 expression could also be down-regulated bylipopolysaccharide (LPS) or high glucose (Sorrenti et al.,2006; Xin et al., 2007). In this study, CoCl2 treatmentled to a significant decrease in DDAH activity beforealteration of its mRNA and protein expression. Then asignificant decrease in the expression of DDAH2mRNA and protein was also observed after CoCl2 treat-

Fig. 4. Effect of DDAH2 siRNA on antiapoptotic effect of atRA. A: Summary of DDAH2mRNA expression. B: Representative of DDAH2 protein expression. C: Ratio of apoptotic cells.D: Caspase-3 activity. E: Level of intracellular ROS as indicated by DCF fluorescence. F: ADMAlevel in conditioned medium. Data expressed as mean 6 SEM of five independent experiments(*P < 0.05, **P < 0.01 vs. CoCl2 1 scrambled siRNA group).



ment. Thus, it is possible that CoCl2 affects DDAHactivity via both posttranslational and transcriptionalregulatory mechanisms, resulting in the change inADMA level.

The second finding of our present study was thatthe DDAH/ADMA pathway was involved in the antia-poptotic effect of atRA. It has been reported that atRAexerts an antiapoptotic effect in hippocampus cells (Ahle-meyer et al., 2001). In the present study, atRA treatmentsignificantly attenuated CoCl2-induced apoptosis of un-differentiated PC12 cells, which has been widelyemployed as a neuronal cell model by different groups(Caruso et al., 2006; Chen et al., 2008; Misiti et al.,2008). We also found that treatment of undifferentiatedPC12 cells with atRA dramatically decreased ROS leveland caspase-3 activity, which may account, at least inpart, for its antiapoptotic property. This is supported byother evidence that atRA reduces susceptibility to oxida-tive stress in multiple cell types (Jackson et al., 1991;Ahlemeyer and Krieglstein, 1998; Choudhary et al.,2008) and inhibits caspase-3 activity (Aidlen et al.,2007). As discussed above, the DDAH/ADMA pathwayis involved in CoCl2-induced apoptosis, we thereforeexamined the effects of atRA on the DDAH/ADMApathway. The results showed that atRA significantlyincreased DDAH activity and DDAH2 expression,resulting in the decreased level of ADMA in CoCl2-

treated cells. Interestingly, we found that DDAH2 genesilence abolished the antiapoptotic effect of atRA andthat DDAH2 overexpression inhibited the proapoptoticeffect of CoCl2.

It is unclear, however, how atRA regulates DDAHactivity. As mentioned above, DDAH activity is sensitiveto oxidative stress and atRA has antioxidative properties;therefore, we speculated that the inhibitory effect ofatRA on ROS production may be contributing to theup-regulation of DDAH activity exerted by atRA inCoCl2-treated PC12 cells. Also, it has been shown thatatRA can bind and activate two classes of nuclear recep-tors, the RAR family (RARa, b, and g) and the RXRfamily (RXRa, RXRb, and RXRg; Chambon, 1969;Maden, 2001). In the presence of atRA, these receptorsact as transcription factors in the form of RAR/RXRheterodimers or RXR homodimers, which bind to reti-noic acid response elements in the promoter of targetgenes. A recent study showed that the promoter regionof the DDAH2 gene includes a PPAR/RXR bindingsite and that atRA induces the expression of DDAH2 inendothelial cells (Achan et al., 2002). In the presentstudy, pretreatment of PC12 cells with atRA reversedthe reduction of DDAH2 expression induced by CoCl2at both the mRNA and protein levels, suggesting thatatRA may regulate DDAH activity through direct up-regulation of DDAH2 expression.

Fig. 5. Effect of DDAH2 overexpression on CoCl2-induced apoptosis of PC12 cells. A: Repre-sentative of DDAH2 protein expression. B: Ratio of apoptotic cells. C: Caspase-3 activity. D:Level of intracellular ROS as indicated by DCF fluorescence. E: ADMA level in conditioned me-dium. Data expressed as mean 6 SEM of five independent experiments (*P < 0.01 vs. control).

1944 Wang et al.


The relationship between intracellular ROS andDDAH/ADMA deserves to be noted. It has been docu-mented that the DDAH/ADMA pathway is oxidativestress sensitive (Selley, 2004). In the present study,CoCl2 decreased the activity of DDAH, resulting in theincrease in ADMA level and subsequently increased in-tracellular ROS generation. Importantly, DDAH siRNAabolished the antioxidant effect of atRA. In contrast,DDAH overexpression inhibited the increase in ROSgeneration induced by CoCl2. Furthermore, exogenousADMA induced ROS production. These findings sug-gest that there is a positive feedback between ROS andthe DDAH/ADMA pathway, resulting in amplificationof the proapoptotic effects of CoCl2.

In summary, the DDAH/ADMA pathway isinvolved in CoCl2-induced apoptosis and the antiapop-totic effects of atRA in undifferentiated PC12 cells. TheDDAH/ADMA pathway may be a potential therapeutictarget for neuronal injury elicited by excessive cobalt.

ACKNOWLEDGMENTS

This work was supported by a grant from NationalNatural Science Foundation of China (30600817).

REFERENCES

Achan V, Tran CT, Arrigoni F, Whitley GS, Leiper JM, Vallance P.

2002. All-trans-retinoic acid increases nitric oxide synthesis by endothe-

lial cells: a role for the induction of dimethylarginine dimethylaminohy-

drolase. Circ Res 90:764–769.

Ahlemeyer B, Bauerbach E, Plath M, Steuber M, Heers C, Tegtmeier F,

Krieglstein J. 2001. Retinoic acid reduces apoptosis and oxidative stress by

preservation of SOD protein level. Free Radic Biol Med 30:1067–1077.

Ahlemeyer B, Krieglstein J. 1998. Retinoic acid reduces staurosporine-

induced apoptotic damage in chick embryonic neurons by suppressing

reactive oxygen species production. Neurosci Lett 246:93–96.

Aidlen JT, Nazarey PP, Kinane TB, Donahoe PK, Schnitzer JJ, Kling

DE. 2007. Retinoic acid–mediated differentiation protects against nitro-

fen-induced apoptosis. Birth Defects Res B Dev Reprod Toxicol 80:

406–416.

Fig. 6. ADMA-induced apoptosis of PC12 cells via ROS-dependent pathway. A, B: Ratio of ap-optotic cells. C: Caspase-3 activity. D: Level of intracellular ROS as indicated by DCF fluores-cence. Data expressed as mean 6 SEM of five independent experiments [*P < 0.05, **P < 0.01vs. control; ##P < 0.01 vs. ADMA (10 lM)].



Boger RH, Bode-Boger SM, Tsao PS, Lin PS, Chan JR, Cooke JP.

2000. An endogenous inhibitor of nitric oxide synthase regulates endo-

thelial adhesiveness for monocytes. J Am Coll Cardiol 36:2287–2295.

Caruso A, Motolese M, Iacovelli L, Caraci F, Copani A, Nicoletti F,

Terstappen GC, Gaviraghi G, Caricasole A. 2006. Inhibition of the

canonical Wnt signaling pathway by apolipoprotein E4 in PC12 cells.

J Neurochem 98:364–371.

Chambon P. 1996. A decade of molecular biology of retinoic acid recep-

tors. FASEB J 10:940–954.

Chen L, Liu L, Luo Y, Huang S. 2008. MAPK and mTOR pathways

are involved in cadmium-induced neuronal apoptosis. J Neurochem

105:251–261.

Choudhary R, Baker KM, Pan J. 2008. All-trans retinoic acid prevents

angiotensin II- and mechanical stretch-induced reactive oxygen species

generation and cardiomyocyte apoptosis. J Cell Physiol 215:172–181.

Ciafre SA, Niola F, Giorda E, Farace MG, Caporossi D. 2007. CoCl(2)-

simulated hypoxia in skeletal muscle cell lines: Role of free radicals in

gene up-regulation and induction of apoptosis. Free Radic Res 41:391–

401.

Gerber U, Gahwiler BH. 1991. Cobalt blocks postsynaptic responses

induced by neurotransmitters in the hippocampus in vitro. Neurosci

Lett 134:53–56.

Hasan M, Ali S, Anwar J. 1980. Cobalt-induced depletion of dopamine,

norepinephrine & 5-hydroxytryptamine concentration in different

regions of the rat brain. Indian J Exp Biol 18:1051–1053.

Ito A, Tsao PS, Adimoolam S, Kimoto M, Ogawa T, Cooke JP. 1999.

Novel mechanism for endothelial dysfunction: dysregulation of dime-

thylarginine dimethylaminohydrolase. Circulation 99:3092–3095.

Jackson GR, Morgan BC, Werrbach-Perez K, Perez-Polo JR. 1991.

Antioxidant effect of retinoic acid on PC12 rat pheochromocytoma. Int

J Dev Neurosci 9:161–170.

Jiang DJ, Jia SJ, Dai Z, Li YJ. 2006. Asymmetric dimethylarginine indu-

ces apoptosis via p38 MAPK/caspase-3-dependent signaling pathway in

endothelial cells. J Mol Cell Cardiol 40:529–539.

Jiang DJ, Tan GS, Zhou ZH, Xu KP, Ye F, Li YJ. 2002. Protective

effects of demethylbellidifolin on myocardial ischemia-reperfusion

injury in rats. Planta Med 68:710–713.

Jiang JL, Wang S, Li NS, Zhang XH, Deng HW, Li YJ. 2007. The in-

hibitory effect of simvastatin on the ADMA-induced inflammatory

reaction is mediated by MAPK pathways in endothelial cells. Biochem

Cell Biol 85:66–77.

Jung JY, Mo HC, Yang KH, Jeong YJ, Yoo HG, Choi NK, Oh WM,

Oh HK, Kim SH, Lee JH, Kim HJ, Kim WJ. 2007. Inhibition by epi-

gallocatechin gallate of CoCl2-induced apoptosis in rat PC12 cells. Life

Sci 80:1355–1363.

Karovic O, Tonazzini I, Rebola N, Edstrom E, Lovdahl C, Fredholm

BB, Dare E. 2007. Toxic effects of cobalt in primary cultures of mouse

astrocytes. Similarities with hypoxia and role of HIF-1alpha. Biochem

Pharmacol 73:694–708.

Kitamura M, Ishikawa Y, Moreno-Manzano V, Xu Q, Konta T, Lucio-

Cazana J, Furusu A, Nakayama K. 2002. Intervention by retinoic acid

in oxidative stress-induced apoptosis. Nephrol Dial Transplant 17(Suppl

9):84–87.

Lin KY, Ito A, Asagami T, Tsao PS, Adimoolam S, Kimoto M, Tsuji H,

Reaven GM, Cooke JP. 2002. Impaired nitric oxide synthase pathway

in diabetes mellitus: role of asymmetric dimethylarginine and dimethy-

larginine dimethylaminohydrolase. Circulation 106:987–992.

Maden M. 2001. Role and distribution of retinoic acid during CNS de-

velopment. Int Rev Cytol 209:1–77.

Misiti F, Orsini F, Clementi ME, Lattanzi W, Giardina B, Michetti F.

2008. Mitochondrial oxygen consumption inhibition importance for

TMT-dependent cell death in undifferentiated PC12 cells. Neurochem

Int 52:1092–1099.

Moreno-Manzano V, Ishikawa Y, Lucio-Cazana J, Kitamura M. 1999.

Suppression of apoptosis by all-trans-retinoic acid. Dual intervention in

the c-Jun n-terminal kinase-AP-1 pathway. J Biol Chem 274:20251–

20258.

Olivieri G, Hess C, Savaskan E, Ly C, Meier F, Baysang G, Brockhaus

M, Muller-Spahn F. 2001. Melatonin protects SHSY5Y neuroblastoma

cells from cobalt-induced oxidative stress, neurotoxicity and increased

beta-amyloid secretion. J Pineal Res 31:320–325.

Selley ML. 2004. Homocysteine increases the production of asymmetric

dimethylarginine in cultured neurons. J Neurosci Res 77:90–93.

Smirnova IV, Sawamura T, Goligorsky MS. 2004. Upregulation of lec-

tin-like oxidized low-density lipoprotein receptor-1 (LOX-1) in endo-

thelial cells by nitric oxide deficiency. Am J Physiol Renal Physiol

287:F25–32.

Sorrenti V, Mazza F, Campisi A, Vanella L, Li Volti G, Di Giacomo C.

2006. High glucose-mediated imbalance of nitric oxide synthase and

dimethylarginine dimethylaminohydrolase expression in endothelial

cells. Curr Neurovasc Res 3:49–54.

Stuhlinger MC, Conci E, Haubner BJ, Stocker EM, Schwaighofer J,

Cooke JP, Tsao PS, Pachinger O, Metzler B. 2007. Asymmetric di-

methyl l-arginine (ADMA) is a critical regulator of myocardial reperfu-

sion injury. Cardiovasc Res 75:417–425.

Stuhlinger MC, Tsao PS, Her JH, Kimoto M, Balint RF, Cooke JP.

2001. Homocysteine impairs the nitric oxide synthase pathway: role of

asymmetric dimethylarginine. Circulation 104:2569–2575.

Vengellur A, LaPres JJ. 2004. The role of hypoxia inducible factor 1alpha

in cobalt chloride induced cell death in mouse embryonic fibroblasts.

Toxicol Sci 82:638–646.

Wang G, Hazra TK, Mitra S, Lee HM, Englander EW. 2000. Mitochon-

drial DNA damage and a hypoxic response are induced by CoCl(2) in

rat neuronal PC12 cells. Nucleic Acids Res 28:2135–2140.

Wang S, Jiang JL, Hu CP, Zhang XJ, Yang DL, Li YJ. 2007. Relation-

ship between protective effects of rosiglitazone on endothelium and en-

dogenous nitric oxide synthase inhibitor in streptozotocin-induced dia-

betic rats and cultured endothelial cells. Diabetes Metab Res Rev 23:

157–164.

Xin HY, Jiang DJ, Jia SJ, Song K, Wang GP, Li YJ, Chen FP. 2007.

Regulation by DDAH/ADMA pathway of lipopolysaccharide-induced

tissue factor expression in endothelial cells. Thromb Haemost 97:830–

838.

Xu Q, Konta T, Furusu A, Nakayama K, Lucio-Cazana J, Fine LG,

Kitamura M. 2002. Transcriptional induction of mitogen-activated pro-

tein kinase phosphatase 1 by retinoids. Selective roles of nuclear recep-

tors and contribution to the antiapoptotic effect. J Biol Chem 277:

41693–41700.

Yuan Q, Jiang DJ, Chen QQ, Wang S, Xin HY, Deng HW, Li YJ.

2007. Role of asymmetric dimethylarginine in homocysteine-induced

apoptosis of vascular smooth muscle cells. Biochem Biophys Res Com-

mun 356:880–885.

Zou W, Yan M, Xu W, Huo H, Sun L, Zheng Z, Liu X. 2001. Cobalt

chloride induces PC12 cells apoptosis through reactive oxygen species

and accompanied by AP-1 activation. J Neurosci Res 64:646–653.

1946 Wang et al.


all-trans retinoic acid inhibits cobalt chloride-induced apoptosis in pc12 cells: role of the...

Documents