432 489.494

The Benzophenanthridine Alkaloid Fagaronine Induces Erythroleukemic Cell Differentiation by Gene of GATA-1 gene by 3.2-fold. A mutation within the GATA-1 bind-ing sites strongly decreased the promoter activation induced by Fagaronine, a benzophenanthridine alkaloid from Fagara zan- FAG. Taken together, our results represent a demonstration that thoxyloides Lam. (Rutaceae), has been tested on the erythroleu- FAG exerts its differentiating activity by a specific activation of kemic cell line K562 in order to explain some previous results the regulating GATA-1 regions of genes involved in the erythroid on cell differentiation. In this study we showed that fagaronine induces a significant hemoglobinization of the human erythro-leukemic cell line K562. This hemoglobin synthesis was accom- panied by a strong increase of erythroid mRNA expression such Fagaronine ´ K562 cell differentiation ´ leukemia ´ erythroid gene as g- and a-globin, and PBGD, an enzyme of heme synthesis. In expression ´ GATA-1 transcription factor addition, the Epo-R transcripts were also stimulated indicatingthat cells are engaged in a maturation process. Both transcription factors GATA-1 and NF-E2, which play an important role in the regulation of genes involved in the erythroid differentiation, were also transcriptionally up-regulated. To elucidate the possi- ble role of GATA-1 in the FAG-induced differentiation of K562 cells, we transfected reporter constructs containing regulatory GAPDH: glyceraldehydes 3-phosphate dehydrogenase regions of erythroid genes encompassing GATA-1 binding sites.
After 48 hours of treatment, FAG stimulated the EPO-R and g-glo-bin promoters by 2- to 3-fold and the promoter/enhancer region the normal regulatory pathways controlling cell proliferationand differentiation. Numerous physiological as well as non- Leukemic cells can be considered as maturation-arrested cells physiological agents, including antitumor drugs, have been de- that continue to proliferate and rapidly accumulate, escaping scribed to induce differentiation of leukemic cells [1]. A complete 1 JE 2428 Onco-Pharmacologie, UFR Pharmacie, IFR 53 BiomolØcules, UniversitØ de Reims 2 Laboratoire de Stress, DØfenses et Reproduction des Plantes, URVVC EA2069, UFR Sciences, UniversitØ de 3 Laboratoire de Pharmacognosie, FRE 2715 CNRS UFR Pharmacie, UniversitØ de Reims Champagne-Ardenne, Chantal Trentesaux ´ JE 2428 Onco-Pharmacologie ´ UFR Pharmacie ´ IFR 53 BiomolØcules ´ UniversitØ de Reims Champagne-Ardenne ´ 51 rue Cognacq Jay ´ 51096 Reims cedex ´ France ´ Phone: +33-326-918-045 ´ Fax: +33-326-918926 ´ E-mail: [email protected] Received August 9, 2004 ´ Accepted January 6, 2005 Planta Med 2005; 71: 489±494 ´  Georg Thieme Verlag KG Stuttgart ´ New York remission by differentiation therapy was obtained in patients perature. The MeOH solution, on evaporation under reduced with acute promyelocytic leukemia treated by all-trans-retinoic pressure gave an extract (2.2 g) which was dissolved in 0.02 N HCl (50 mL). The aqueous solution was precipitated by Mayer'sreagent (20 mL) and the precipitate (390 mg) was dissolved in Our group previously demonstrated that anthracycline antitumor MeOH-Me2CO-H2O (6:2:1). The alkaloids were converted to drugs such as aclacinomycin (ACLA) and doxorubicin (DOX), at the chlorides by passage through an Amberlite IRA 400 (60 subtoxic concentrations, induced in vitro the erythroid differentia- mL) column. After concentration under reduced pressure, a resi- tion of human leukemic K562 cells, leading to the appearance of due (212 mg) was obtained. This gave pure fagaronine (25 mg) as hemoglobinized cells. ACLA stimulated the transcription of genes bright yellow needles after three crystallizations from a mixture involved in hemoglobin synthesis, by the recruitment of ery- of ethyl acetate-methanol: m.p. 202 8C followed by solidification throid-specific transcription factors, notably GATA-1 and NF-E2 and melting again at 255 8C; spectral data (UV, 1H-NMR 500 MHz [3], [4], [5]. In contrast, DOX induced the hemoglobinization of these cells by a post-transcriptional mechanism leading to an in- creased stability of the erythroid transcripts [5].
Cell line and induction of erythroid differentiationThe human erythroleukemic K562 cells were cultured in RPMI Among the other compounds able to stimulate erythroid differen- 1640 Glutamax medium, 10 % FCS (Invitrogen) as previously de- tiation, the benzo[c]phenanthridine alkaloid fagaronine (FAG) scribed [3]. FAG chloride was reconstituted in 70% ethanol as a (Fig.1) was also reported to induce the hemoglobinization of 0.1 M stock solution and diluted in the culture medium immedi- K562 cells [6] but through an unknown mechanism. Other works ately before use. Various FAG concentrations were added to the have established that FAG displays an antileukemic activity K562 cell suspensions at the beginning of the exponential against murine leukemia P388 in vivo [7], inhibits DNA polymer- growth phase. Cell cultures were incubated in a 5% CO2 humidi- ase activity in murine embryos [8], nucleic acid and protein syn- fied atmosphere at 37 8C during 72 hours. Growth inhibition and thesis in KB cells, respectively [9]. These biological activities are cell viability were evaluated as previously described [3]. After 3 related to its properties to intercalate DNA and to interact with days of treatment, the number of erythroid differentiated cells the ribosomal system [10]. FAG also inhibits the activities of the was determined by scoring benzidine-positive cells. K562 cells DNA topoisomerases I and II [11], [12], human DNA ligase I [13] were stained using a benzidine-H2O2 method and gave an in- and reverse transcriptases from RNA virus [14], especially the hu- tense blue cytoplasmic staining known to correlate with hemo- man HIV-1 reverse transcriptase in vitro [15].
globin synthesis. As previously described [6], an average of 300cells was scored for benzidine-positivity and the results are Since ACLA was also reported to intercalate DNA and to inhibit topoisomerase I activity [16], we wondered whether the differ-entiating activity of FAG is triggered by similar molecular events related to transcriptional mechanisms.
Total RNA was extracted from 5”106 cells using the TriZOL re- agent (Invitrogen) as recommended by the manufacturer. One We have studied the erythroid gene expression induced by FAG microgram of total RNA was reverse transcribed in a 20 mL reac- in the human erythroid K562 cell line using RT-PCR and reporter tion volume using oligo-dT primers, Superscript II reverse tran- gene analysis. Our results clearly indicate that FAG stimulates scriptase (Invitrogen) according to the manufacturer's instruc- erythroid gene transcription through a mechanism involving tions. At the end of the reaction, the volume of the RT products was adjusted to 200 mL with DNase RNase-free water. A 10 mL ali-quot of cDNA was used for PCR amplification using a -[32P]-dCTP(NEN) and gene specific primers of: g-globin (forward: 5¢- GGCAACCTGTCCTCTGCCTC-3¢; reverse: 5¢-GCCAGGAAGCTTGCA-CCTCA-3¢) [17]; a-globin (forward: 5¢-TGGGGTAAGGTCGGCGCG- CA-3¢; reverse: 5¢-TGCACCGCAGGGGTGAACTC-3¢) [18]; PBGD The roots of Fagara zanthoxyloides Lam. were collected in 1999, (forward: 5¢-GGTCCTACTATCGCCTCCCTC-3¢; reverse: 5¢-AGAAT- in the Ivory Coast and identified by Dr. C. Moretti. A voucher spe- CTTGTCCCCTGTGGTG-3¢); Epo-R (forward: 5¢-AGCCTGTGTCGC- cimen (No. 15042) is kept at the Herbarium of the National Cen- TGCTGACGC-3¢; reverse: 5¢-GGTCCTCCGTGAAGGGGGTGC-3¢); ter of Floristics, University of Cocody, Abidjan, Ivory Coast.
GATA-1 (forward: 5¢-GATCCTGCTCTGGTGTCCTCC-3¢; reverse: 5¢-ACAGTTGAGCAATGGGTACAC-3¢) [17]; NF-E2 (forward: 5¢-AT- TTGAGCCCCAAGCCCCAGC-3¢; reverse: 5¢-CCAGCCTCTGTCCCCT- Dried roots of Fagara zanthoxyloides (50 g), defatted with light CCAGC-3¢). Amplification of GAPDH was performed as control petroleum (1 L), were extracted with MeOH (1 L) at room tem- using the same PCR conditions with primer (forward: 5¢-CTC-TGCCC CCTCTGCTGATGC-3¢; reverse: 5¢-CCATCACGCCACAGTTT-CCCG-3¢). PCR was performed using Taq DNA polymerase (Invi- trogen) with the following cycling conditions: 94 8C for 2 min, followed by 15 ± 25 cycles of 94 8C for 30 sec, 60 8C for 30 sec,and 72 8C for 60 sec and at the last cycle the reaction was main-tained at 72 8C for 10 min to finish cDNA chain elongation. Am-plified products were analyzed on 6 % non-denaturating poly-acrylamide gels in 1 ” TBE. After electrophoresis, the gels were Dupont C et al. The Benzophenanthridine Alkaloid ¼ Planta Med 2005; 71: 489±494 exposed and quantified on a GS-363 Molecular Imager (Bio- Western blot analysis of GATA-1 protein expression was per- formed as previously described [19]. Proteins were separated on a 10% gel by SDS-PAGE and blotted onto PVDF membrane (Amer- sham). The membrane was blocked by a 2-h incubation in Tris- buffered saline containing 5% non-fat milk, 0.05% Tween 20. Im- munodetections were performed by incubating the membrane with the specific GATA-1 monoclonal antibody (Santa Cruz) and then with the secondary antisera, conjugated with horse radish peroxidase (HRP). Filters were developed using the ECL Western blotting detection reagent (Amersham).
Transient transfection of K562 cells and analysis of reporter Reporter constructs were prepared by insertion of promoter and enhancer regions in restriction sites of pGL2-basic plasmids,respectively, upstream and downstream of the firefly luciferasegene as described previously [4]. K562 cells were transfected bythese reporter plasmids using the transferrinfection technique(Transferrinfection kit, Serva). Briefly, 6 mg plasmid DNA were The expression of erythroid mRNAs was studied after 3 days of mixed with 10 mg Fe-loaded transferrin-polylysine complex in a treatment by 6 mM FAG (Fig. 3). The PCR products analysis 0.5 mL 200 mM HEPES buffer (pH 7.2). This mix was added to showed that FAG induced the over-expression of g- and a- globin pretreated (24 h incubation in culture medium containing transcripts (2-fold), of porphobilinogene deaminase (PBGD) 50 mM desferroxamine) K562 cells in a proportion of 1 mL of cul- mRNAs (4.5-fold), a key enzyme of the heme synthesis, and of ture medium containing 50 mM desferroxamine and 100 mM Epo-R mRNAs (2.2-fold), a receptor for the erythropoietin hor- chloroquine for 3”105 cells. After 6 h at 37 8C, for DNA capture, mone which regulates the erythroid differentiation process cells were washed once with RPMI, divided into equal parts, and (Fig. 3). Such an increased mRNA expression was found to be then cultivated in the same medium with or without FAG for 24 specific to erythroid genes since the expression of GAPDH ubi- or 48 h. Cells were resuspended in 0.25 mM Tris-HCl (pH 7.5) and quitous transcripts remained constant after FAG treatment.
cellular extracts were obtained by three cycles of freezing-thaw- ing. The amount of protein in the extracts was determined by Taking into account the role of GATA-1 and NF-E2 transcription using the Bio-Rad protein assay. Luciferase activity in the ex- factors in the regulation of erythroid gene expression, we also ex- tracts was tested with the Luciferase Assay Kit (Promega) in amined their expression following FAG treatment (Fig. 3). Simi- accordance with the manufacturer's instructions and quantified larly, FAG induced the over-expression of GATA-1 (2.7-fold) and using a Lumac-3M luminometer (Bertold). An absolute signal NF-E2 transcripts (2.5-fold) after 3 days of treatment, as compar- was determined as the maximal rate of the sample luminescence ed to control cells. These results suggested that FAG may exert its during the first minute of the assay and the activities were finally differentiating effects by a transcriptional regulation of erythroid expressed as light units/mg of protein.
To support these data, we examined whether the over-expres- sion of GATA-1 transcripts was also associated with an increasedGATA-1 protein level. After 3 days of treatment with 6 mM FAG, Hemoglobinized cell content and growth inhibition of human cell lysates were analyzed by Western blot using a monoclonal leukemic K562 cells were evaluated after 3 days of treatment by anti GATA-1 antibody. The results shown in Fig. 4 indicate that various FAG concentrations (Fig. 2). FAG induced a dose-depen- FAG induced a 2- to 3-fold over-expression of GATA-1 protein as dent hemoglobinization of K562 cells. The maximal differentiat- compared to untreated cells, in agreement with the over-expres- ing effect of around 60% was obtained at 6 mM FAG and main- tained at 10 mM, as compared to untreated cells (1% benzidine-positive cells). This induction of hemoglobin synthesis was ac- In order to determine whether the accumulation of erythroid companied by a marked growth inhibition varying from 63% for mRNAs resulted from a transcriptional activation mediated by the lowest concentration to 95% for the highest. In contrast, GATA-1, we have transfected K562 cells with different plasmid these concentrations of FAG had a limited effect on cell viability, constructs containing either the promoter of EPO-R or g-globin as determined by trypan blue dye exclusion, and cell death did genes upstream to the firefly luciferase gene. These 2 promoter not exceed 5% at the optimal differentiating concentration regions contained GATA-1 consensus sequences. Then, we ana- lyzed the effect of FAG treatment (6 mM) for 24 or 48 hours ontothe reporter activity.
Dupont C et al. The Benzophenanthridine Alkaloid ¼ Planta Med 2005; 71: 489±494 was performed as described in Materials and Methods using a GATA-1 monoclonal antibody. Lane 1, K562 cells; lane 2, K562 cells + FAG. The amount of protein loaded was normalized by performing an immuno- blot analysis using an anti-actin Mab. Results from one experiment are To determine whether the GATA binding sites contained in thesepromoter sequences played a role in the FAG-mediated gene ac-tivation, we also used a construct with the GATA-1 gene promo- ter/enhancer region, which included two inverted canonical GATA binding sites [4]. This construct was significantly activatedafter 48 hours of FAG treatment (3.2-fold activation, Fig. 5B). Inparallel, a construct containing mutated GATA-1 binding sites lo-cated in the enhancer region was also used. In that case, FAG in-duced transcription activation by 1.9-fold only. These results in-dicated that the GATA-1 binding sites were involved in the pro-moter regulation induced by FAG treatment and thus representa molecular basis to explain the FAG-induced transcriptionalstimulation of erythroid genes.
We have examined here the effects of FAG on erythroid differen-tiation and growth of K562 cells. FAG induced an efficient hemo-globinization of the K562 cell line without subsequent toxicity Fig. 3 FAG increased erythroid gene expression. K562 cells were in- together with a strong inhibition of cell growth (about 80%).
duced by 6 mM FAG during 3 days. RT-PCR analysis of g-globin, a-glo- These observations are in agreement with previous results [6] bin, PBGD, EPO-R, GATA-1, NF-E2 and GAPDH mRNAs was performed and emphasize the interest in FAG which blocks cell division as described in Materials and Methods in the presence of a[32P]-d-CTP.
through a strong growth inhibition and induction of cell differ- After electrophoresis on an 8% polyacrylamide gel, PCR products were detected and quantified by exposure on a Bio-Rad GS-363 Molecular entiation. The relationship between these two effects of the Imager. (A) Lane c, PCR negative control; lane 1, K562 cells; lane 2, drug could be explained by the fact that cell differentiation is ac- K562 cells + FAG. Results from one experiment representative of three.
companied by a loss of proliferation capacity. No lethality was (B) Quantification results obtained in FAG-treated cells are expressed observed at the concentration used, confirming the absence of as percentages of results in control cells. Data are the means  S.D. of an acute toxicity of fagaronine on the K562 cell line.
three independent experiments. ** and ***, values were significantly different from control according to Student's t test with p < 0.01 and p In vitro, numerous compounds are able to induce cancer cell dif-ferentiation [1] and appear to represent an attractive alternativeor adjuvant therapy to the conventional cytotoxic chemotherapy.
As shown in Fig. 5A, luciferase activity for the g-globin construct Up to now, clinical applications of the differentiation therapy was found to be increased in FAG-treated cells and reached a have been successfully achieved with the all-trans-retinoic acid maximum at 48 hours with a 3.3-fold activation, as compared treatment of patients with acute promyelocytic leukemia [2].
to untreated cells. Under the same conditions, the EPO-R promo- More recently, differentiation of the malignant clone and com- ter activation was also found activated by 2.0-fold after 24 hours plete clinical remission has been obtained in an ATRA-refractory of FAG treatment and was maintained at this level after 48 hours, APL patient treated with ATRA in combination with phenyl buty- (2.1-fold, Fig. 5A). As a control, the signal obtained from the rate, an HDAC inhibitor [20]. Similar results were observed in pGL2-basic plasmid never exceeded 0.1% of the activities meas- vitro on an ATRA-resistant cell line NB4 treated by a retinoid/buty- ured for erythroid constructs (Fig. 5A). Therefore, at the optimal ric prodrug, which led to growth inhibition, partial differentia- FAG differentiating concentration, a transcription of the reporter tion and apoptosis of the resistant cells. This ªtranscription ther- gene under the control of erythroid gene promoter regions was apyº combines elements to facilitate transcriptional initiation of blocked differentiation pathways by inhibiting histone deacetyl-ase [21].
Dupont C et al. The Benzophenanthridine Alkaloid ¼ Planta Med 2005; 71: 489±494 and NF-E2 mRNAs as well as by GATA-1 protein accumulation.
These results are in agreement with an erythroid maturationthat seems to occur at the transcriptional level. Indeed, resultsobtained with reporter constructs containing erythroid gene reg-ulatory regions showed that FAG caused an increased transcrip-tional activity of luciferase gene downstream of the g-globin,EPO-R and GATA-1 promoters. Constructs with GATA-1 gene en-hancer region mutated or not at the level of two GATA-1 bindingsites clearly showed that the binding of GATA-1 to its target se-quence was required to stimulate reporter gene. Although wecannot exclude that other factors are involved, it is interestingto note that GATA-1's implication in FAG transcriptional activa-tion was also described for the antitumor drug aclacinomycin [5].
All these data show for the first time that fagaronine exerts itsdifferentiating activity by a specific activation of regulatory re- gions which control the erythroid differentiation program of hu- man erythroleukemic cells. This process involves the participa-tion of erythroid transcription factors such as GATA-1. Fagaro-nine may represent a new family of natural products able to actby modulating the activity of genes important for proliferation,differentiation and apoptosis control.
This work was supported by grants from the region of Cham-pagne-Ardenne, the Ligue Nationale contre le Cancer, ComitØsde la Haute-Marne et de l'Aisne. C.D. was the recipient of a fel-lowship from the Reims city.
Fig. 5 Effects of FAG on the transcriptional activity of reporter con- structs. (A) Activity of g-globin and EPO-R promoter (P). Luciferase ac- tivity in transfected K562 cells was determined as described in Materi- als and Methods after 24 or 48 hours of incubation in the presence or in the absence of FAG and expressed in lights units/mg of proteins. (B) 1 Reiss M, Gamba-Vitalo C, Sartorelli AC. Induction of tumor cell dif- Comparison of the luciferase activity from constructs containing the ferentiation as a therapeutic approach: preclinical models for he- promoter and enhancer (P./E.) of the GATA-1 gene or the promoter matopoietic and solid neoplasms. Cancer Treat Rep 1986; 70: 201 ± and the mutated enhancer (P./mutated E.) of the same gene after 48 hours treatment with 6 mM FAG. Results are the means  S.D. of three 2 Degos L, Dombret H, Chomienne C, Daniel MT, MiclØa JM, Chastang C, independent experiments. ** and ***, values significantly different Castaigne S, Fenaux P. All-trans-retinoic acid as a differentiating agent from untreated control (p < 0.01 and p < 0.001, respectively) according in the treatment of acute promyelocytic leukemia. Blood 1995; 85: to Student's t test. §§, activation values from mutated constructs were significantly different from those obtained in non-mutated constructs Trentesaux C, Ngo Nyoung M, Aries A, Morceau F, Ronchi A, Ottolenghi S, Jardillier JC, Jeannesson P. Increased expression of GATA-1 and NFE- 2 erythroid-specific transcription factors during aclacinomycin-medi- ated differentiation of human erythroleukemic cells. Leukemia 1993; In a previous work, our group was also interested in the molecular 4 Aries A, Trentesaux C, Ottolenghi S, Jardillier JC, Jeannesson P, Dou- mechanisms by which antitumor compounds can induce ery- beikovski A. Activation of erythroid-specific promoters during an- throid differentiation of the human K562 leukemic cells and dem- thracycline-induced differentiation of K562 cells. Blood 1996; 87: onstrated that the anthracycline derivative aclacinomycin acted at 5 Morceau F, Chenais B, Gillet R, Jardillier JC, Jeannesson P, Trentesaux C.
the transcriptional level by stimulating GATA-1 and NF-E2, both Transcriptional and posttranscriptional regulation of erythroid gene specific transcription factors regulating the expression of ery- expression in anthracycline-induced differentiation of human ery- throid genes [3], [4]. Another mechanism involving the erythroid throleukemic cells. Cell Growth Diff 1996; 7: 1023±9 RNA stabilization rather than its transcriptional activation was Como L, Jeannesson P, Trentesaux C, Desoize B, Jardillier JC. The anti- leukemic alkaloid fagaronine and the human K 562 leukemic cells: ef- also observed for doxorubicin, another anthracycline currently fects on growth and induction of erythroid differentiation. Leuk Res used in clinical practice, suggesting that differentiating effects of antitumor agents are mediated by various molecular mechanisms.
7 Messmer WM, Tin-WA M, Fong HHS, Bevelle C, Farnsworth NR, Abraham DJ, Trojanek J. Fagaronine, a new tumor inhibitor isolated from Fagara zanthoxyloides Lam. (Rutaceae). JPharm Sci 1972; 61: Here, we showed that hemoglobin synthesis mediated by FAG treatment was preceded by an increased transcription of several 8 Sethi VS. Inhibition of mammalian and oncornavirus nucleic acid genes known as markers of erythroid differentiation (g- and a- polymerase activities by alkoxybenzophenantridine alkaloids. Cancer globins, PBGD, and EPO-R) and by an over-expression of GATA-1 Dupont C et al. The Benzophenanthridine Alkaloid ¼ Planta Med 2005; 71: 489±494 9 Pezzuto JM, Antosiak SK, Messmer WM, Slaytor MB, Honig GR. Inter- 16 Sorensen BS, Jensen PB, Sehested M, Jensen PS, Kjeldsen E, Nielsen OF, action of the antileukemic alkaloid, 2-hydroxy-3,8,9-trimethoxy-5- Alsner J. Antagonistic effect of aclarubicin on camptothecin induced methylbenzo[c]phenanthridine (fagaronine), with nucleic acids.
cytotoxicity: role of topoisomerase I. Biochem Pharmacol 1994; 47: 10 Casiano Torres CA, Baez A. Effects of the antitumor drugs 3-nitroben- 17 Leonard M, Brice M, Engel JD, Papayannopoulou T. Dynamics of GATA zothialozo[3,2-a]quinolium and fagaronine on nucleic acid and pro- transcription factor expression during erythroid differentiation. Blood tein synthesis. Biochem Pharmacol 1986; 35: 679±85 11 Larsen AK, Grondard L, Couprie J, Desoize B, Como L, Jardillier JC, Riou 18 Privitera E, Schiro R, Longoni D, Ronchi A, Rambaldi A, Bernasconi S, J.F. The antileukemic alkaloid fagaronine is an inhibitor of DNA topo- Ottolenghi S, Masera G, Biondi A. Constitutive expression of GATA-1, isomerases I and II. Biochem Pharmacol 1993; 46: 1403 ±12 EPOR, a-globin, and g-globin genes in myeloid clonogenic cells from 12 Fleury F, Sukhanova A, Ianoul A, Devy J, Kudelina I, Duval O, Alix AJ, juvenile chronic myelocytic leukemia. Blood 1995; 86: 323±8 Jardillier JC, Nabiev I. Molecular determinants of site-specific inhibi- 19 Gillet R, Devemy E, Dupont C, Billat C, Jeannesson P, Trentesaux C. Evi- tion of human DNA topoisomerase I by fagaronine and ethoxidine. Re- dence of the role of protein kinase C during aclacinomycin induction lation to DNA binding. JBiol Chem 2000; 275: 3501±9 of erythroid differentiation in K 562 cells. FEBS Lett 1999; 454: 331±4 13 Tan GT, Lee S, Lee IS, Chen J, Leitner P, Bersterman JM, Kinghorn AD, 20 Warrell RPJ, He L, Richon V, Calleja E, Pandolfi PP. Therapeutic target- Pezzuto JM. Natural-product inhibitors of human DNA ligase I. Bio- ing of transcription in acute promyelocytic leukemia by use of an inhibitor of histone deacetylase. JNatl Cancer Inst 1998; 90: 1621±5 14 Sethi ML. Inhibition of reverse-transcriptase activity by benzophenan- 21 Mann KK, Rephaeli A, Colosimo AL, Diaz Z, Nudelman A, Levovich I, tridine alkaloids. JNat Prod 1979; 42: 187±96 Jing Y, Waxman S, Miller Jr WH. A retinoid/butyric acid prodrug over- Tan GT, Pezzuto JM, Kinghorn AD, Hughes SH. Evaluation of natural comes retinoic acid resistance in leukemias by induction of apoptosis.
products as inhibitors of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase. JNat Prod 1991; 54: 143±54 Dupont C et al. The Benzophenanthridine Alkaloid ¼ Planta Med 2005; 71: 489±494

Source: http://jeanfrancois.riou.free.fr/DupontPlantaMedica2005.pdf