l-Mimosine blocks cell proliferation via upregulation of B-cell ...brated in the hypoxic incubator (APM-30D, Astec, Fukuoka, Japan). Cells were then incubated in the hypoxic incubator

doi:10.1152/ajpcell.00180.2011 302:C676-C685, 2012. First published 23 November 2011;Am J Physiol Cell Physiol

Chang and Horng-Heng JuangLi-Chuan Chung, Ke-Hung Tsui, Tsui-Hsia Feng, Shiow-Ling Lee, Phei-Langregulated gene 1 in prostate carcinoma cells

downstreammycB-cell translocation gene 2 and N-l-Mimosine blocks cell proliferation via upregulation of

You might find this additional info useful...

35 articles, 16 of which can be accessed free at:This article cites http://ajpcell.physiology.org/content/302/4/C676.full.html#ref-list-1

including high resolution figures, can be found at:Updated information and services http://ajpcell.physiology.org/content/302/4/C676.full.html

can be found at:AJP - Cell Physiologyabout Additional material and information http://www.the-aps.org/publications/ajpcell

This infomation is current as of February 3, 2012.

American Physiological Society. ISSN: 0363-6143, ESSN: 1522-1563. Visit our website at http://www.the-aps.org/.a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2012 by the

is dedicated to innovative approaches to the study of cell and molecular physiology. It is published 12 timesAJP - Cell Physiology

on February 3, 2012

ajpcell.physiology.orgD

ownloaded from

http://ajpcell.physiology.org/content/302/4/C676.full.html#ref-list-1

http://ajpcell.physiology.org/content/302/4/C676.full.html

http://ajpcell.physiology.org/

L-Mimosine blocks cell proliferation via upregulation of B-cell translocationgene 2 and N-myc downstream regulated gene 1 in prostate carcinoma cells

Li-Chuan Chung,1,2* Ke-Hung Tsui,3,4* Tsui-Hsia Feng,5 Shiow-Ling Lee,1 Phei-Lang Chang,3,4

and Horng-Heng Juang2,4

1Department of Bioengineering, Tatung University, Taipei, 2Department of Anatomy, Chang Gung University, 3Departmentof Urology and 4Bioinformation Center, Chang Gung Memorial Hospital, and 5School of Nursing, Chang Gung University,Kwei-Shan, Tao-Yuan, Taiwan, Republic of China

Submitted 3 June 2011; accepted in final form 8 November 2011

Chung LC, Tsui KH, Feng TH, Lee SL, Chang PL, Juang HH.L-Mimosine blocks cell proliferation via upregulation of B-cell translo-cation gene 2 and N-myc downstream regulated gene 1 in prostatecarcinoma cells. Am J Physiol Cell Physiol 302: C676–C685, 2012. Firstpublished November 23, 2011; doi:10.1152/ajpcell.00180.2011.—L-Mimosine, an iron chelator and a prolyl 4-hydroxylase inhibitor,blocks many cancer cells at the late G1 phase. B-cell translocationgene 2 (Btg2) regulates the G1/S transition phases of the cell cycle.N-myc downstream regulated gene 1 (Ndrg1) is a differentiation-inducing gene upregulated by hypoxia. We evaluated the molecularmechanisms of L-mimosine on cell cycle modulation in PC-3 andLNCaP prostate carcinoma cells. The effect of L-mimosine on cellproliferation of prostate carcinoma cells was determined by the[3H]thymidine incorporation and flow cytometry assays. L-Mimosinearrested the cell cycle at the G1 phase in PC-3 cells and at the S phasein LNCaP cells, thus attenuating cell proliferation. Immunoblot assaysindicated that hypoxia and L-mimosine stabilized hypoxia-induciblefactor-1� (HIF-1�) and induced Btg2 and Ndrg1 protein expression,but downregulated protein levels of cyclin A in both PC-3 and LNCaPcells. L-Mimosine treatment decreased cyclin D1 protein in PC-3cells, but not in LNCaP cells. Dimethyloxalylglycine, a pan-prolylhydroxylase inhibitor, also induced Btg2 and Ndrg1 protein expres-sion in LNCaP cells. The transient gene expression assay revealed thatL-mimosine treatment or cotransfection with HIF-1� expression vec-tor enhanced the promoter activities of Btg2 and Ndrg1 genes.Knockdown of HIF-1� attenuated the increasing protein levels of bothBtg2 and Ndrg1 by hypoxia or L-mimosine in LNCaP cells. Ourresults indicated that hypoxia and L-mimosine modulated Btg2 andNdrg1 at the transcriptional level, which is dependent on HIF-1�.L-Mimosine enhanced expression of Btg2 and Ndrg1, which attenu-ated cell proliferation of the PC-3 and LNCaP prostate carcinomacells.

hypoxia; hypoxia-inducible factor-1�; iron; chetomin; cell cycle

IRON CHELATORS WITH HIGH ANTIPROLIFERATION activities upregu-late the expression of growth-inhibitory and metastatic sup-pressor genes mediated by hypoxia-inducible factor-1� (HIF-1�)-dependent and independent mechanisms (16, 22). L-Mimosine {(S)-�-amino-�-[1-(3-hydroxy-4-oxopyridine)] propionicacid; C8H10N2O4}, a plant amino acid, acts as an iron chelatorand reversibly blocks mammalian cell proliferation at late G1

phase (5). An earlier study indicated that numbers of G0/G1

cells were elevated after incubation with L-mimosine in humanprostate carcinoma DU145 cells (32). L-Mimosine also acts as

a prolyl 4-hydroxylase inhibitor and has similar effects on thehypoxic induction of HIF-1� protein in human and rodent cells(3, 31).

Expression of B-cell translocation gene 2 (Btg2) in cyclingcells induced accumulation of growth-inhibitory forms of ret-inoblastoma protein and led to G1 cell cycle arrest (8). Immu-nohistochemistry and proliferative indexes of selected humanprostate peripheral zone lesions indicated that a reduction in, orloss of, Btg2 expression is associated with progression tomalignancy (6). Additionally, forced overexpression of Btg2attenuated cell proliferation in prostate carcinoma PC-3 cells,whereas thyroid hormone downregulated Btg2 gene expressionin LNCaP cells (25). A study using bone marrow-derivedmesenchymal stem cells revealed that hypoxia induced mes-enchymal stem cell differentiation in neuronlike cells by in-creasing the expression of Btg2 and decreasing cyclin D1expression (20).

N-myc downstream regulated gene 1 (Ndrg1) plays an im-portant role in both androgen-induced cell differentiation andinhibition of prostate cancer metastasis (19, 28). In vivo studiesshowed that expression of Ndrg1 was inversely associated withGleason grading and overall survival rates of prostate cancerpatients; moreover, Ndrg1 inhibited metastasis of prostatecarcinoma PC-3 cells in vitro (1). Previous study indicated thatiron chelators induced Ndrg1 gene expression via the HIF-1�pathway in several types of cancer cells (34).

The objectives of this study were to determine the regulatoryeffects of L-mimosine on HIF-1� and its association with geneexpression of Btg2 and Ndrg1 in human prostate carcinoma cells.

MATERIALS AND METHODS

Cell culture and chemicals. Human prostate carcinoma LNCaP andPC-3 cells were obtained from the American Type Culture Collection(Manassas, VA) and were maintained as previously described (25).Cells were incubated under normoxia using a standard CO2 incubatorat 37°C in a humidified atmosphere with 5% CO2 and 95% room air(21% O2) until the cells grew to 70%-80% confluence in RPMI-1640medium with 10% fetal cal serum (FCS). On the day of hypoxictreatment (1% O2, 5% CO2, and 94% N2), growth medium wasswitched to fresh medium with 10% FCS that had been preequili-brated in the hypoxic incubator (APM-30D, Astec, Fukuoka, Japan).Cells were then incubated in the hypoxic incubator for another 24 h.L-Mimosine, chetomin, dimethyloxalylglycine (DMOG), and ferricammonium citrate (FAC) were purchased from Sigma Chemical (St.Louis, MO) and dissolved in the suggested solvent, according to themanufacturer’s instructions. The bicinchoninic acid protein assay kitwas purchased from Pierce Protein Research (Rockford, IL). FCS waspurchased from HyClone (Logan, UT), and RPMI-1640 medium waspurchased from GIBCO, Invitrogen (Grand Island, NY).

* L.-C. Chung and K.-H. Tsui contributed equally to this work.Address for reprint requests and other correspondence: H.-H. Juang, Dept.

of Anatomy, Chang Gung Univ., 259 Wen-Hua 1st Rd., Kwei-Shan, Tao-Yuan333, Taiwan, ROC (e-mail: [email protected]).

Am J Physiol Cell Physiol 302: C676–C685, 2012.First published November 23, 2011; doi:10.1152/ajpcell.00180.2011.

0363-6143/12 Copyright © 2012 the American Physiological Society http://www.ajpcell.orgC676

on February 3, 2012


ownloaded from


Cell proliferation assay with [3H]thymidine incorporation. Cellproliferation in response to L-mimosine was measured using a[3H]thymidine incorporation assay, as previously described (12). Inthis assay, 1 � 104 cells were cultured in each well of a 12-well platein RPMI-1640 medium with 10% FCS and different concentrations(0–800 �M) of L-mimosine. After the required incubation periods (24and 48 h), 0.5 �Ci/ml of [3H]thymidine was added to each well of the12-well plate. The cells were then incubated at 37°C in a humidified5% CO2 atmosphere for 4 h. Cells were then washed twice with coldphosphate-buffered saline (PBS) and then with cold 5% trichloro-acetic acid. Cells were lysed by adding 0.5 ml of 0.5 N NaOH. Then400 �l of the solubilized cell solution were mixed with 4 ml ofscintillation cocktail and counted in a liquid scintillation analyzer(Packard BioScience, Meriden, CT). Each sample was tested inquadruplicate.

Flow cytometry. Cells were serum starved for 24 h and thencultured in RPMI-1640 medium with 10% FCS, with or withoutdifferent concentrations (0–800 �M) of L-mimosine for another 24 h.The cells were collected, washed with cold PBS, and centrifuged. Thepellet was resuspended in 300 �l of cold PBS and fixed in ethanol.The cells were centrifuged at 300 g for 10 min, the fixative wasdiscarded, and then the cells were washed with cold PBS. The cellswere resuspended in 1 ml of PBS, and 50 �l of Triton X-100 and 50�l of ribonuclease (10 mg/ml) were added and incubated for 1 h at37°C. After further centrifugation, cells were resuspended in PBS,

stained with propidium iodide, and incubated at 4°C overnight.Cell cycle analysis was performed using the FACS-Calibur cytom-eter and CellQuestPro software (BD Biosciences, San Jose, CA);the data were analyzed using ModFit LT Mac 3.0 software, aspreviously described (26).

Immunoblot assay. Equal quantities of cell extract (40 �g) wereloaded onto 12% sodium dodecyl sulfate-polyacrylamide gels andanalyzed using the Western lightning plus-Enhanced Chemilumines-cence detection system (Perkin Elmer, Waltham, MA) and wereviewed using ChemiGenius image capture system (Syngene, Cam-bridge, UK). Polyclonal rabbit anti-human Btg2 serum was preparedas previously described (25). The blot membranes were probed with1:500 HIF-1� (610958, BD Biosciences), 1:3,000 diluted �-actinantiserum (I-19, Santa Cruz Biotechnology, Santa Cruz, CA), 1:100diluted vascular endothelial growth factor (VEGF; A-20, Santa CruzBiotechnology), 1:2,000 diluted cyclin D1 (DCS6, Cell Signal, MA),1:100 diluted cyclin A (C-19, Santa Cruz Biotechnology), 1:3,000diluted Ndrg1 antiserum (42–6200, Invitrogen, Carlsbad, CA), and1:1,500 diluted Btg2 antiserum. The intensities of the different bandswere analyzed using the GeneTools program of ChemiGenius (Syn-gene).

Real-time reverse transcription-polymerase chain reaction. TotalRNA from cells was isolated using Trizol reagent, and cDNA wassynthesized as described previously (12). Real-time quantitative poly-merase chain reaction (qPCR) was performed using the ABI StepOne

Fig. 1. Effect of L-mimosine on cell cycle distribution and cell proliferation in PC-3 cells. A: PC-3 cells were treated with varying concentrations of L-mimosine,as indicated, for 24 h (�) or 48 h (Œ), and cell proliferation was determined by the [3H]thymidine incorporation assay. Each point on the curve represents themean percentage � SE (n � 4) of [3H]thymidine incorporated into the cells relative to the control solvent-treated group (0 �M of L-mimosine treatment) (*P �0.05, �P � 0.01). B: the cell-cycle distribution of PC-3 cells was analyzed by flow cytometry. The values shown in each bar chart represent the meanpercentage � SE (n � 3) of cells in each phase of the cell cycle and are compared with the control solvent-treated group (�P � 0.01). C: the sub-G1 fractionof PC-3 cells was analyzed by flow cytometry. The values shown in each bar chart represent the mean percentage � SE (n � 3) of cells in the sub-G1 phaseof the cell cycle (�P � 0.01). D: PC-3 cells were treated with varying concentrations of L-mimosine for 24 h. Cells were lysed, and the expressions ofhypoxia-inducible factor-1� (HIF-1�), N-myc downstream regulation gene 1 (Ndrg1), cyclin D1, cyclin A, B-cell translocation gene 2 (Btg2), and �-actin weredetermined by immunoblot assay. E: the quantitative analysis was done by determining the intensity of each band for target genes and �-actin from threeindependent experiments. Data are presented as the fold induction (�SE, n � 3) of the relative density of the target gene/�-actin (�SE) in relation to the controlsolvent-treated group (*P � 0.05, �P � 0.01).

C677HYPOXIA INDUCES Btg2 AND Ndrg1 GENE EXPRESSION

AJP-Cell Physiol • doi:10.1152/ajpcell.00180.2011 • www.ajpcell.org

on February 3, 2012


ownloaded from


Plus Real-Time PCR system (Applied Biosystems, Foster City, CA).FAM dye-labeled TaqMan MGB probes and PCR primers were purchasedfor human Ndrg1 (HS00608387-m1) and Btg2 (HS00198887-m1) fromApplied Biosystems. For the internal positive control, glyceraldehyde3-phosphate dehydrogenase (GAPDH; HS99999905-m1) was usedwith a FAM reporter dye-labeled TaqMan MGB probe. The amplifi-cation conditions were as follows: 40 cycles at 95°C for 15 s and 60°Cfor 1 min. Mean cycle threshold (Ct) values were calculated forGAPDH and the reporter gene using StepOne software version 2.0(Applied Biosystems). Ct values for Ndrg1 and Btg2 were normalizedagainst the GAPDH control probe to calculate Ct values. All reac-tions were preformed in triplicate, and each experiment was con-ducted on at least three independent occasions.

HIF-1� expression vector. The human HIF-1� cDNA (MGC:10483) vector was purchased from Invitrogen. Human HIF-1�cDNA was linearized by cutting with BamHI and XbaI and ligationwith the overexpression vector pcDNA3 (Invitrogen), as previ-ously described (3).

Knock-down HIF-1�. LNCaP cells were plated onto six-well plates1 day before transfection. The culture media were replaced withRPMI-1640 medium plus 10% FCS and 5 �g/ml polybrene (SantaCruz Biotechnology) and then transduced with HIF-1� small hairpinRNA lentiviral particles (sc-35561-V; Santa Cruz Biotechnology) asdescribed by the manufacturer. Two days after transduction, the cells(LN-HIF-1�si) were selected by incubation with 10 �g/ml puromycin

dihydrochloride for another 3 days. The mock-transfected LNCaPcells (LN-COLsi) were transduced with control small hairpin RNAlentiviral particles (sc-10808-V, Santa Cruz Biotechnology) and wereclonally selected in same manner as the LN-HIF-1�si cells. Cells thatexpressed HIF-1� in resistant colonies were evaluated by the afore-mentioned immunoblot assay.

Reporter vectors. The reporter vectors containing different frag-ments of the 5=-flanking region of the human Btg2 gene were clonedas previously described (25). The DNA fragment containing thepromoter of the Ndrg1 gene (4,714 to �46) was cloned as previ-ously described (26). The 5=-deletion of the human Ndrg1 reportervectors containing different DNA fragments were constructed bydigesting with different restriction enzymes. Proper ligation andorientation of the reporter vectors were confirmed by extensive re-striction mapping and sequencing.

Transient transfection and reporter assay. LNCaP cells or PC-3cells were plated onto 24-well plates 1 day before transfection. Cellswere transiently transfected using TransFast transfection reagent, aspreviously described (27). For the transient cotransfection of HIF-1�experiments, cells were transfected with the same amount of plasmidin each well by adding pcDNA3 vector to eliminate variable degreesof efficiency of the reporter activities. Reporter vector-transfectedLNCaP or PC-3 cells were then treated with or without L-mimosine inRPMI-1640 medium with 10% FCS for an additional 24 h. The

Fig. 2. Effect of L-mimosine on cell cycle distribution and cell proliferation in LNCaP cells. A: LNCaP cells were treated with varying concentrations ofL-mimosine for 24 h (�) or 48 h (Œ), and cell proliferation was determined by the [3H]thymidine incorporation assay. Each point on the curve represents the meanpercentage � SE (n � 4) of [3H]thymidine incorporated into the cells relative to the control solvent-treated group (�P � 0.01). The cell-cycle distribution ofLNCaP cells was analyzed after treatments L-mimosine 24 h (B) and 48 h (C) by flow cytometry. The values shown in each bar chart represent the meanpercentage � SE (n � 3) of cells in each phase of the cell cycle and are compared with the control solvent-treated group (*P � 0.05). D: immunoblotting forHIF-1�, Ndrg1, cyclin D1, cyclin A, Btg2, and �-actin using lysates from varying concentrations of L-mimosine-treated LNCaP cells. E: quantitative results wereanalyzed by determining the intensity of each band for the target genes and �-actin from three independent experiments. Data are presented as the fold induction(�SE, n � 3) of the relative density of the target gene/�-actin (�SE) in relation to the control solvent-treated group (*P � 0.05, �P � 0.01).

C678 HYPOXIA INDUCES Btg2 AND Ndrg1 GENE EXPRESSION


on February 3, 2012


ownloaded from


luciferase activity was adjusted for transfection efficiency using thenormalization control plasmid pCMVSPORT�gal.

Statistical analysis. Results are expressed as means � SEs of atleast three independent replications of each experiment. Statisticalsignificance was determined using one-way ANOVA and Student’st-test using the SigmaStat program for Windows, version 2.03 (SPSS,Chicago, IL).

RESULTS

To explore the antiproliferative effects of L-mimosine onprostate cancer cells, we compared and analyzed inhibitoryeffects and cell-cycle distributions in L-mimosine-treated cells.The [3H]thymidine incorporation assay revealed that inhibitionof PC-3 cell growth occurred initially at 50 �M of L-mimosine,increasing as the dose increased. The 400 �M of L-mimosinesignificantly blocked 49 and 97% of [3H]thymidine incorpora-tion in PC-3 cells after treatment with L-mimosine for 24 and48 h, respectively (Fig. 1A). In the flow cytometric analysis,100–800 �M of L-mimosine induced a 24.4% increase in G1

arrest, together with a decrease in G2/M phase and S phasecells after 24 h of incubation (Fig. 1B). However, L-mimosineinduced cell apoptosis in PC-3 cells after 48 h of incubationsince 100–800 �M of L-mimosine increased the sub-G1 frac-

tion of cells by 5.9–20.2% (Fig. 1C). The immunoblottingassays showed that L-mimosine upregulated Btg2 and Ndrg1protein levels, but downregulated protein levels of cyclin D1and cyclin A (Fig. 1D). The quantitative analysis was done bydetermining the intensity of each band of different genes and�-actin from three independent experiments (Fig. 1E).

Similar antiproliferative results were also found in the L-mimosine-treated LNCaP cells. L-Mimosine (400 �M) blocked�30% of [3H]thymidine incorporation into LNCaP cells after24 or 48 h of incubation (Fig. 2A). However, results from flowcytometric analysis of LNCaP cells revealed that 800 �M ofL-mimosine induced a 6.5% increase in S phase cell arrest,together with a decrease in G2/M phase cells after 24 h ofincubation (Fig. 2B). Eight hundred micromoles of L-mimosineinduced a 10.7% increase in S phase arrest after 48 h ofincubation (Fig. 2C). Unlike those studies of PC-3 cells,L-mimosine did not induce cell apoptosis in LNCaP cells, sincethe sub-G1 fraction of cells did not differ significantly amongcells treated with 0–800 �M of L-mimosine after 24 or 48 h ofincubation (Fig. 2, B and C).

Results of the immunoblotting assays showed that L-mi-mosine upregulated HIF-1�, Btg2, and Ndrg1 protein levels in

Fig. 3. Hypoxia modulates gene expressionof LNCaP and PC-3 cells. LNCaP (A) orPC-3 (B) cells were cultured under hypoxic(HP; 1% O2) or normoxic (N; 21% O2)conditions for 24 h. The cells were thenlysed, and the expressions of HIF-1�,Ndrg1, Btg2, vascular endothelial growthfactor (VEGF), cyclin D1, cyclin A, and�-actin were determined by immunoblot as-say. C: LNCaP cells were treated with0–500 �M of dimethyloxalylglycine(DMOG) for 24 h. Cells were lysed, andexpressions of HIF-1�, Ndrg1, Btg2, and�-actin were determined by immunoblot as-say. Quantitative results were analyzed bydetermining the intensity of each band fortarget genes and �-actin from three indepen-dent experiments. Data are presented as thefold induction (�SE, n � 3) of the relativedensity of the target gene/�-actin (�SE) inrelation to the control solvent-treated group(*P � 0.05, �P � 0.01).



on February 3, 2012


ownloaded from


LNCaP cells. However, L-mimosine treatments decreased theprotein levels of cyclin A, but did not significantly affectthe protein expression of cyclin D1 (P � 0.079, Fig. 2D). Theresults of the quantitative analysis are presented in Fig. 2E.

Because we found that L-mimosine stabilized HIF-1� inboth LNCaP and PC-3 cells, we further determined whether thegenes induced by L-mimosine could also be induced by hyp-oxia. The immunoblotting assays revealed that hypoxia, com-pared with normoxia, significantly upregulated protein levelsof HIF-1� in both LNCaP and PC-3 cells (Fig. 3, A, left and B,left). The results of quantitative analysis indicated that hypoxiastabilized HIF-1� and upregulated gene expression of VEGF,Btg2, and Ndrg1 in both LNCaP and PC-3 cells (Fig. 3, A, rightand B, right). Hypoxic treatment downregulated protein ex-pression of the cyclin A in both LNCaP and PC-3 cells.Interestingly, the quantitative results indicated cyclin D1 levelswere downregulated by 51% after 24 h of hypoxic culturecompared with PC-3 cells incubated under normoxia (Fig. 3B).However, we did not find the same phenomenon in the LNCaPcells. Hypoxic treatment increased minimally, but not signifi-cantly (P � 0.647), cyclin D1 levels in LNCaP cells (Fig. 3A).

The results of the immunoblotting assays showed thatDMOG, a prolyl 4-hydroxylase inhibitor, at 100–500 �M,

stabilized HIF-1� and induced Ndrg1 and Btg2 gene expres-sion in LNCaP cells (Fig. 3C, left). The results of the quanti-tative analysis indicated that Btg2 and Ndrg1 protein levelswere upregulated 2.4-and 3.8-fold, respectively, after treatedwith 500 �M of DMOG for 24 h (Fig. 3C, right).

The immunoblotting assay revealed that L-mimosine-inducedincrease in Btg2 gene expression was significantly blocked (P �0.01) after cotreatment with FAC, a cell-permeable iron donor(Fig. 4A). The results of reverse transcription (RT)-qPCR indi-cated that L-mimosine (400 �M) induced a 1.6-fold increase inBtg2 gene expression compared with the solvent-treated group.The transient gene expression assay showed similar results, sug-gesting that iron blocks the elevation of L-mimosine-induced Btg2(Fig. 4B). Furthermore, transient gene expression assays indicatedthat transient overexpression of HIF-1� under normoxia inducedBtg2 promoter activity (Fig. 4C). Induction of transient geneexpression using a 5=-deletion assay indicated that HIF-1� in-creased Btg2 promoter activity is dependent on region 101 to1 upstream of the translational initiation site of Btg2 gene (Fig.4D). Results of immunoblotting assays and RT-qPCR showedhypoxia induced 1.5- and 2.3-fold increases in Btg2 gene expres-sion, respectively, while cotreatment with chetomin did not sig-

Fig. 4. Hypoxia and L-mimosine modulate the gene expression of Btg2 in prostate carcinoma cells. A: PC-3 cells were treated with 400 �M of L-mimosine and/or100 �g/ml of ferric ammonium citrate for 24 h. Cells were then lysed, and expressions of Btg2 and �-actin were determined by the immunoblot assay (top) orRT-quantitative PCR (RT-qPCR; bottom). Data are presented as the fold induction (�SE, n � 3) of the relative mRNA levels in relation to the controlsolvent-treated group (*,�P � 0.01). C, control solvent; L, L-mimosine; F, ferric ammonium citrate. B: the Btg2 reporter vector-transfected PC-3 cells weretreated with L-mimosine (400 �M) and/or ferric ammonium citrate (100 �g/ml) for 24 h. Values are the mean percentage � SE (n � 6) of the luciferase activityin relation to the control solvent-treated group (*,�P � 0.01). C: the Btg2 reporter vector-transfected LNCaP cells were cotransfected with HIF-1� expressionvector. Values are the mean percentage (�SE, n � 6) of the luciferase activity in relation to the mock-transfected group (�P � 0.01). D: nested deletionconstructs of Btg2 reporter vectors were cotransfected with control expression vector (pcDNA3; open bars) or HIF-1� expression vector (solid bars) into LNCaPcells. Values are the mean percentage � SE (n � 6) of the luciferase activity of the reporter vectors induced by the HIF-1� expression vector relative to theluciferase activity associated with group that was cotransfected with control expression vector (�P � 0.01). E: LNCaP cells were cultured under N or HPconditions and treated with or without 50 nM of chetomin for 24 h. Cells were then lysed, and expressions of HIF-1�, Btg2, and �-actin were determined byimmunoblot assay (top) or RT-qPCR (bottom). Data are presented as the fold induction (�SE, n � 3) of the relative mRNA levels in relation to the group culturedunder N conditions (�P � 0.01).



on February 3, 2012


ownloaded from


nificantly attenuate (P � 0.714) Btg2 gene expression by hypoxia(Fig. 4E).

Transient gene expression assays indicated that L-mimosineinduced Ndrg1 promoter activity. L-Mimosine (400 �M) in-duced a 2.09-fold increase in Ndrg1 promoter activity com-pared with the solvent-treated group (Fig. 5A). The results ofthe immunoblotting assay and RT-qPCR indicated that 400�M L-mimosine increased Ndrg1 mRNA levels 1.99-fold com-pared with the solvent-treated group. Furthermore, this in-crease was significantly blocked (P � 0.01) by cotreatmentwith FAC, indicating that iron blocks the elevation of L-mimosine-induced Ndrg1 (Fig. 5B). Similar results were foundin the transient gene expression assays (Fig. 5C). Induction oftransient gene expression using a 5=-deletion assay indicatedthat L-mimosine induced an increase in Ndrg1 promoter activ-ity, which was dependent on the DNA fragment located at4,714 to 1,319 upstream of the transcriptional initiation siteof the Ndrg1 gene (Fig. 5D).

The results of RT-qPCR showed that hypoxia increasedNdrg1 mRNA levels 6.35-fold compared with cells culturedunder normoxia; however, this increased effect was signifi-cantly attenuated (P � 0.01) by cotreatment with chetomin(Fig. 6A). Similar results were found in the immunoblottingassays. Hypoxia increased Ndrg1 and VEGF protein levels,

while chetomin blocked the increases (Fig. 6B). Transient geneexpression assays revealed that transient overexpression ofHIF-1� enhanced Ndrg1 promoter activity (Fig. 6C). Inductionof transient gene expression using a 5=-deletion assay indicatedthat transient overexpression of HIF-1� increased Ndrg1 pro-moter activity, which was also dependent on region 4,714 to1,319 upstream of the transcriptional initiation site of theNdrg1 gene (Fig. 6D).

To determine whether the upregulation of Btg2 and Ndrg1 inprostate carcinoma cells by hypoxia or L-mimosine was HIF-1�dependent, we transiently knocked down HIF-1� in LNCaP(LN-HIF-1�si) cells. The immunoblotting assays revealed thathypoxia, compared with normoxia, significantly upregulated (P �0.001) protein levels of Btg2 and Ndrg1 in mock-transfectedLNCaP (LN-COLsi) cells. However, the increase in protein levelsof both Btg2 and Ndrg1 by hypoxia was significantly attenuated(P � 0.001) when the HIF-1� gene was knocked down fromLNCaP (LN-HIF-1�si) cells (Fig. 7A). Similar results were alsofound for L-mimosine treatment (Fig. 7B).

DISCUSSION

L-Mimosine, a plant amino acid, reversibly blocks mamma-lian cell cycle at the late G1 phase (5). Earlier studies using

Fig. 5. L-Mimosine modulates the gene expression of Ndrg1 of prostate carcinoma cells. A: the Ndrg1 reporter vector-transfected LNCaP cells were treated withvarying concentration of L-mimosine, as indicated, for 24 h. Values are the mean percentage � SE of the luciferase activity induced by the L-mimosine treatmentsrelative to the control solvent-treated group (*P � 0.05, �P � 0.01). B: gene expression of Ndrg1 was determined by immunoblot assay (top) or RT-qPCR(bottom). Data are presented as the fold induction (�SE, n � 3) of the relative mRNA levels in relation to the control solvent-treated group (*,�P � 0.01). C: theNdrg1 reporter vector-transfected PC-3 cells were treated with L-mimosine (400 �M) and/or ferric ammonium citrate (100 �g/ml) for 24 h. Values are the meanpercentage � SE (n � 6) of the luciferase activity in relation to the control solvent-treated group (*,�P � 0.01). PC-3 cells were treated with L-mimosine (400�M) and/or ferric ammonium citrate (100 �g/ml) for 24 h. D: luciferase activity of nested deletion constructs of Ndrg1 reporter vectors after treatment withcontrol solvent (open bars) or 400 �M of L-mimosine (solid bars). Valeus are the mean percentage � SE (n � 6) of the Ndrg1 reporter activity induced by theL-mimosine treatments in relation to the control solvent-treated groups (�P � 0.01).



on February 3, 2012


ownloaded from


prostate DU145 carcinoma cells revealed that numbers ofG0/G1 cells were elevated after incubation with L-mimosine,resulting in upregulation of p27 (32). Our results indicated thatL-mimosine inhibited PC-3 cell proliferation with increasingnumbers of cells accumulating in the G1 phase of the cell cycle.Moreover, L-mimosine induced cell apoptosis when PC-3 cellswere treated with L-mimosine for long period of incubation (48h). Consistent with previous studies, these results showed thatL-mimosine exerted apoptotic activity in human pancreaticcancer and 937 leukemia cells (9, 35). With LNCaP cells, thenumbers of S phase cells were elevated after incubation withL-mimosine. These results are in agreement with early studiesindicating that L-mimosine reversibly arrested the cell cyclelate in the G1 phase or at the beginning of the S phase (10). Ourflow cytometric analysis indicated that L-mimosine (800 �M)suppressed cell growth but did not exert apoptotic activity inhuman prostate carcinoma LNCaP cells in vitro. Results of our[3H]thymidine incorporation and flow cytometric assays indi-cated that LNCaP cells were more resistant to antiproliferationand apoptosis inducing by L-mimosine compared with PC-3cells.

Cyclin D1 serves as a key sensor and an integrator ofextracellular signals and promotes progression through G1-Sphases of the cell cycle (7). Our in vitro study using PC-3 cellsshowed that hypoxia or L-mimosine treatment inhibited theexpression of cyclin D1. This result is similar to the results ofearly studies that indicated L-mimosine blocked cell cycle

progression and suppressed proliferation of H226 lung cancercells and human breast cancer MDA-MB-453 cells by inhibi-tion of cyclin D1 expression (2, 14). Our results also are inagreement with results of other studies that indicated theinverse association between cyclin D1 and HIF-1� in A549pulmonary cancer cells and bone marrow-derived mesenchy-mal stem cells (20, 33). However, unlike in PC-3 cells, hypoxiaor L-mimosine treatment increased cyclin D1 expressionslightly but not significantly in LNCaP cells. In 3T3 cells,L-mimosine arrested cells in the G1 phase, but did not affect theprotein levels of cyclin D1 (29). Additionally, hypoxia en-hanced cyclin D1 gene expression in human breast carcinomaMCF-7 cells (11). It appears that hypoxic effects on cyclin D1expression are cell type dependent. It is worthy to note thatLNCaP cells are p53-wild-type cells, and PC-3 cells are p53-null cells (27). p53 is a well-known regulator of the G1/S andG2/M check points (18). The difference in p53 expressionbetween LNCaP and PC-3 cells could account for the differ-ences in cyclin D1 expression and cell cycle arrest underhypoxic conditions. Our results also showed that L-mimosinetreatment inhibited the expression of cyclin A in both PC-3 andLNCaP cells, which is consistent with results of other studiesthat indicated hypoxia or L-mimosine treatment downregulatedthe expression of cyclin A (21, 24). These results may alsoaccount for cell cycle arrest in S phase by L-mimosine inLNCaP cells.

Fig. 6. Hypoxia modulates the gene expression of Ndrg1 of prostate carcinoma cells. LNCaP cells were cultured under N or HP conditions and treated with orwithout 50 nM of chetomin for 24 h. Gene expression of Ndrg1 was determined by RT-qPCR (A), and the expressions of HIF-1�, Ndrg1, VEGF, and �-actinwere determined by immunoblot assay (B). Data are presented as the fold induction (�SE, n � 3) of the relative mRNA levels in relation to the group culturedunder N conditions (*,�P � 0.01). C: the LNCaP cells were transient transfected with HIF-1� expression vector (LN-HIF-1�) or mocked expression vector(LN-DNA) for 36 h. Cells were then lysed, and expressions of HIF-1� and �-actin were determined by the immunoblot assay (top). The Ndrg1 reportervector-transfected LNCaP cells were cotransfected with varying concentrations of HIF-1� expression vector. Values are the mean percentage � SE (n � 6) ofthe luciferase activity in relation to the mock-transfected group (bottom; *P � 0.05, �P � 0.01). D: luciferase activity of nested deletion constructs of Ndrg1reporter vectors cotransfected with control expression vector (pcDNA3; open bars) or HIF-1� expression vector (solid bars) into LNCaP cells. Values are themean percentage � SE (n � 6) of the luciferase activity of the reporter vectors induced by the HIF-1� expression vector relative to the luciferase activityassociated with group that was cotransfected with the control expression vector (�P � 0.01).



on February 3, 2012


ownloaded from


Induction of Btg2 led to G1 cell cycle arrest in several celllines (8, 23, 25). Loss of Btg2 expression is the earliest knownindicator that a cell cycle regulator is lost in prostate carcino-genesis (6); nonetheless, the function and regulatory mecha-nisms of Btg2 in the human prostate are still unclear. In aprevious study using stable Btg2 overexpressing PC-3 cells, wefound that overexpression of Btg2 attenuated cell proliferation;moreover, Btg2 gene expression was downregulated by thyroidhormone in prostate carcinoma LNCaP cells (25). In thepresent study, upregulation of Btg2 gene expression by L-mimosine in PC-3 cells led to attenuated cell proliferation andcell cycle arrest in phase G1. The effect of L-mimosine on Btg2in our study was mediated by iron depletion, since its modu-lation was reversed by adding iron. We also found that hypoxiaenhanced Btg2 gene expression, and HIF-1� enhanced thepromoter activity of the Btg2 reporter vector. Our results are inagreement with another study in which hypoxia induced in-creased Btg2 and decreased cyclin D1 expression in bonemarrow-derived mesenchymal stem cells (20). Induction oftransient gene expression by 5=-deletion assay indicated thatHIF-1�-increased Btg2 promoter activity was dependent on the101 to 1 region upstream of the translational initiation siteof the Btg2 gene. Nonetheless, our RT-qPCR and immunoblotdata showed that chetomin did not significantly affect hypoxia-induced Btg2 expression. Chetomin is a small molecule thatdisrupts the CH1 domain of p300, precluding its interactionwith HIF-1�, and thus its subsequent transcriptional activity

(15). Additionally, we did not find a conserved DNA sequence(5=-RCGTG-3=) of HIF-1� response element (30) in the Btg2promoter region (297 to 1) when using simple sequenceanalysis with the GCG program (http://bioinfo.nhri.org.tw).Taken together, we suggest that enhancement of Btg2 geneexpression by L-mimosine occurs via an HIF-1� indirect path-way.

Previous study identified that induction of Ndrg1, which isessential for the G1 phase arrest, is an early event afterL-mimosine treatment (4). Although L-mimosine is similar toother iron chelators inducing Ndrg1 gene expression in severalcell lines (34), our study is the first to characterize L-mimosine-induced Ndrg1 gene expression in prostate carcinoma cells.Moreover, we found that gene expression of Ndrg1 increasedwhen LNCaP cells were treated with other prolyl hydroxylaseinhibitors, such as DMOG. The effect of L-mimosine on Ndrg1in our study was mediated by iron depletion, because it wasreversed by adding iron, which is in agreement with otherstudies (16, 17). RT-qPCR and immunoblotting assays showedthat chetomin suppressed hypoxia-induced Ndrg1 expression.In combination with the transient gene expression assay, ourdata suggested that direct HIF-1� binding to the putativeresponse element on the Ndrg1 genes is involved in themediation of HIF-1� on Ndrg1 gene expression. The actualmechanism responsible for the effect of HIF-1� on the Ndrg1gene is still not well known. A previous study indicated thathypoxia-inducible transcription of the Ndrg1 gene is dependent

Fig. 7. HIF-1� knockdown attenuates the activation of hypoxia and L-mimosine on gene expression of Ndrg1 and Btg2 of LNCaP cells. A: the mock-knockdownLNCaP cells (LN-COLsi) and HIF-1�-knockdown LNCaP cells (LN-HIF-1�si) were cultured under N or HP conditions for 24 h. B: LN-COLsi cells andLN-HIF-1�si were treated without () or with (L) 400 �M of L-mimosine for 24 h. Cells were then lysed, and expressions of HIF-1�, Btg2, Ndrg1, and �-actinwere determined by the immunoblot assay (left). Quantitative results were analyzed by determining the intensity of each band for the target genes (open bars,Btg2; solid bars, Ndrg1; and hatched bars, HIF-1�) and �-actin from three independent experiments (right). *P � 0.01, �P � 0.01.



on February 3, 2012


ownloaded from


on the activation of c-Jun/AP-1 transcription factor or Egr1/Sp1 binding site located 80 to 50 upstream of the tran-scriptional initiation site of the Ndrg1 gene (36). Our resultssuggest that induction of Ndrg1 gene expression by hypoxia isdependent on the binding of the HIF-1� on the putative HIF-1�response element located 4,714 to 1,319 bp upstream of thetranscriptional initiation site of the Ndrg1 gene. Our results arein agreement with previous studies that determined the putativeHIF-1� response element of the Ndrg1 gene based on bioin-formatic analysis (4, 13, 16). Moreover, our results (Fig. 7)indicated that the increasing of protein levels of both Btg2 andNdrg1 by hypoxia is attenuated when the HIF-1� gene wasknocked down in LNCaP cells, suggesting that the upregula-tion of Btg2 and Ndrg1 in prostate carcinoma cells by hypoxiais dependent on HIF-1�.

In conclusion, the results suggest that L-mimosine treatmentand hypoxia modulate the expression of Btg2 and Ndrg1 genesthrough different mechanisms. L-Mimosine stabilizes HIF-1�protein and enhances expression of Btg2 and Ndrg1 at thetranscriptional level, which attenuates cell proliferation ofprostate carcinoma cells in vitro.

ACKNOWLEDGMENTS

The authors thank Yu-Fen Lin for technical assistance.

GRANTS

This research was supported by grants from Chang Gung Memorial Hos-pital (CMRP-D190542, CMRP-D190612, and G392142), and the NationalScience Council, Taiwan, ROC (NSC 100-2320-B-182-006 and NSC 98-2314-B-182-042-MY3).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: L.-C.C., T.-H.F., and H.-H.J. performed experiments;L.-C.C., K.-H.T., T.-H.F., S.- L. L., P.-L.C., and H.-H.J. analyzed data;L.-C.C., K.-H.T., T.-H.F., S.-L.L., and H.-H.J. interpreted results of experi-ments; L.-C.C., K.-H.T., and T.-H.F. prepared figures; L.-C.C., K.-H.T.,T.-H.F., S.-L.L., and H.-H.J. drafted manuscript; L.-C.C., K.-H.T., T.-H.F.,and H.-H.J. edited and revised manuscript; L.-C.C., K.-H.T., T.-H.F., P.-L.C.,and H.-H.J. approved final version of manuscript; K.-H.T. and H.-H.J. con-ception and design of research.

REFERENCES

1. Bandyopadhyay S, Pai SK, Gross SC, Hirota S, Hosobe S, Miura K,Saito K, Commes T, Hayashi S, Watabe M, Watabe K. The Drg-1 genesuppresses tumor metastasis in prostate cancer. Cancer Res 63: 1731–1736, 2003.

2. Chang HC, Lee TH, Chuang LY, Yen MH, Hung WC. Inhibitory effectof mimosine on proliferation of human lung cancer cells is mediated bymultiple mechanisms. Cancer Lett 145: 1–8, 1999.

3. Chung LC, Tsui KH, Feng TH, Lee SL, Chang PL, Juang HH.Curcumin provides potential protection against the activation of hypoxiaand prolyl 4-hydroxylase inhibitors on prostate-specific antigen expressionin human prostate carcinoma cells. Mol Nutr Food Res 55: 1666–1676,2011.

4. Dong Z, Arnold RJ, Yang Y, Park MH, Hrncirova P, Mechref Y,Novotny MV, Zhang JT. Modulation of differentiation-related gene 1expression by cell cycle blocker mimosine, revealed by proteomic analy-sis. Mol Cell Proteomics 4: 993–1001, 2005.

5. Dong Z, Zhang JT. EIF3 p170, a mediator of mimosine effect on proteinsynthesis and cell cycle progression. Mol Biol Cell 14: 3942–3951, 2003.

6. Ficazzola MA, Fraiman M, Gitlin J, Woo K, Melamed J, Rubin MA,Walden PD. Antiproliferative B cell translocation gene 2 protein isdown-regulated post-transcriptionally as an early event in prostate carci-nogenesis. Carcinogenesis 22: 1271–1279, 2001.

7. Fu M, Wang C, Li Z, Sakamaki T, Pestell RG. Minireview: Cyclin D1:normal and abnormal functions. Endocrinology 145: 5439–5447, 2004.

8. Guardavaccaro D, Corrente G, Covone F, Micheli L, D’Agnano I,Starace G, Caruso M, Tirone F. Arrest of G(1)-S progression by thep53-inducible gene PC3 is Rb dependent and relies on the inhibition ofcyclin D1 transcription. Mol Cell Biol 20: 1797–1815, 2000.

9. Hallak M, Vazana L, Shpilberg O, Levy I, Mazar J, Nathan I. Amolecular mechanism for mimosine-induced apoptosis involving oxida-tive stress and mitochondrial activation. Apoptosis 13: 147–155, 2008.

10. Hughes TA, Cook PR. Mimosine arrests the cell cycle after cells enterS-phase. Exp Cell Res 222: 275–280, 1996.

11. Joung YH, Lim EJ, Lee MY, Park JH, Ye SK, Park EU, Kim SY,Zhang Z, Lee KJ, Park DK, Park T, Moon WK, Yang YM. Hypoxiaactivates the cyclin D1 promoter via the Jak2/STAT5b pathway in breastcancer cells. Exp Mol Med 37: 353–364, 2005.

12. Juang HH, Lin YF, Chang PL, Tsui KH. Cardiac glycosides decreaseprostate specific antigen expression by down-regulation of prostate de-rived Ets factor. J Urol 184: 2158–2164, 2010.

13. Kovacevic Z, Fu D, Richardson DR. The iron-regulated metastasissuppressor, Ndrg-1: identification of novel molecular targets. BiochimBiophys Acta 1783: 1981–1992, 2008.

14. Kulp KS, Vulliet PR. Mimosine blocks cell cycle progression by chelat-ing iron in asynchronous human breast cancer cells. Toxicol Appl Phar-macol 139: 356–364, 1996.

15. Kung AL, Zabludoff SD, France DS, Freedman SJ, Tanner EA, VieiraA, Cornell-Kennon S, Lee J, Wang B, Wang J, Memmert K, NaegeliHU, Petersen F, Eck MJ, Bair KW, Wood AW, Livingston DM. Smallmolecule blockade of transcriptional coactivation of the hypoxia-induciblefactor pathway. Cancer Cell 6: 33–43, 2004.

16. Le NT, Richardson DR. Iron chelators with high antiproliferative activityup-regulate the expression of a growth inhibitory and metastasis suppres-sor gene: a link between iron metabolism and proliferation. Blood 104:2967–2975, 2004.

17. Le NT, Richardson DR. Competing pathways of iron chelation: angio-genesis or anti-tumor activity: Targeting different molecules to inducespecific effects. Int J Cancer 110: 468–469, 2004.

18. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell 88:323–331, 1997.

19. Mostaghel EA, Page ST, Lin DW, Fazli L, Coleman IM, True LD,Knudsen B, Hess DL, Nelson CC, Matsumoto AM, Bremner WJ,Gleave ME, Nelson PS. Intraprostatic androgens and androgen-regulatedgene expression persist after testosterone suppression: therapeutic impli-cations for castration-resistant prostate cancer. Cancer Res 67: 5033–5041,2007.

20. Pacary E, Legros H, Valable S, Duchatelle P, Lecocq M, Petit E,Nicole O, Bernaudin M. Synergistic effects of CoCl2 and ROCK inhi-bition on mesenchymal stem cell differentiation into neuron-like cells. JCell Sci 119: 2667–2678, 2006.

21. Park SH, Cho HN, Lee TH, Lee Y, Park YM, Lee YJ, Cho CK, YooSY, Lee YS. Hsp25-induced radioresistance is associated with reductionof death by apoptosis: involvement of Bcl2 and the cell cycle. Radiat Res154: 421–428, 2000.

22. Richardson DR. Molecular mechanisms of iron uptake by cells and theuse of iron chelators for the treatment of cancer. Curr Med Chem 12:2711–2729, 2005.

23. Schwarze SR, DePrimo SE, Grabert LM, Fu VX, Brooks JD, JarrardDF. Novel pathways associated with bypassing cellular senescence inhuman prostate epithelial cells. J Biol Chem 277: 14877–14883, 2002.

24. Seim J, Graff P, Åmellem Ø, Landsverk KS, Stokke T, Pettersen EO.Hypoxia-induced irreversible S-phase arrest involves down-regulation ofcyclin A. Cell Prolif 36: 321–332, 2003.

25. Tsui KH, Hsieh WC, Lin MH, Chang PL, Juang HH. Triiodothyroninemodulates cell proliferation of human prostatic carcinoma cells by down-regulation of the B-cell translocation gene 2. Prostate 68: 610–619, 2008.

26. Tsui KH, Chung LC, Feng TH, Chang PL, Juang HH. Upregulation ofprostate-derived Ets factor by luteolin causes inhibition of cell prolifera-tion and cell invasion in prostate carcinoma cells. Int J Cancer. In press.

27. Tsui KH, Feng TH, Lin YF, Chang PL, Juang HH. p53 downregulatesthe gene expression of mitochondrial aconitase in human prostate carci-noma cells. Prostate 71: 62–70, 2011.

28. Tu LC, Yan X, Hood L, Lin B. Proteomics analysis of the interactomeof N-myc downstream regulated gene 1 and its interactions with theandrogen response program in prostate cancer cells. Mol Cell Proteomics6: 575–588, 2007.



on February 3, 2012


ownloaded from


29. Wang G, Miskimins R, Miskimins WK. Mimosine arrests cells in G1by enhancing the levels of p27kip1. Exp Cell Res 254: 64 –71,2000.

30. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2

tension. Proc Natl Acad Sci U S A 92: 5510–5514, 1995.31. Warnecke C, Griethe W, Weidemann A, Jurgensen JS, Willam C,

Bachmann S, Ivashchenko Y, Wagner I, Frei U, Wiesener M, EckardtKU. Activation of the hypoxia-inducible factor-pathway and stimulationof angiogenesis by application of prolyl hydroxylase inhibitors. FASEB J17: 1186–1188, 2003.

32. Wartenberg M, Fischer K, Hescheler J, Sauer H. Modulation ofintrinsic p-glycoprotein expression in multicellular prostate tumorspheroids by cell cycle inhibitors. Biochim Biophys Acta 1589: 49 –62,2002.

33. Wen W, Ding J, Sun W, Wu K, Ning B, Gong W, He G, Huang S, DingX, Yin P, Chen L, Liu Q, Xie W, Wang H. Suppression of cyclin D1 byhypoxia-inducible factor-1 via direct mechanism inhibits the proliferationand 5-fluorouracil-induced apoptosis of A549 cells. Cancer Res 70:2010–2019, 2010.

34. Xu X, Sutak R, Richardson DR. Iron chelation by clinically relevantanthracyclines: alteration in expression of iron-regulated genes and atyp-ical changes in intracellular iron distribution and trafficking. Mol Phar-macol 73: 833–844, 2008.

35. Zalatnai A, Bocsi J. Mimosine, a plant-derived amino acid inducesapoptosis in human pancreatic cancer xenografts. Anticancer Res 23:4007–4009, 2003.

36. Zhang P, Tchou-Wong KM, Costa M. Egr-1 mediates hypoxia-inducibletranscription of the NDRG1 gene through an overlapping Egr-1/Sp1binding site in the promoter. Cancer Res 67: 9125–9133, 2007.



on February 3, 2012


ownloaded from


Documents

l-Mimosine blocks cell proliferation via upregulation of B-cell ...brated in the hypoxic incubator (APM-30D, Astec, Fukuoka, Japan). Cells were then incubated in the hypoxic incubator