1

3

4

5

6

7

8

91011

1213

1 5

1617

18192021222324

2 5

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

Chemico-Biological Interactions xxx (2015) xxx–xxx

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

Contents lists available at ScienceDirect

Chemico-Biological Interactions

journal homepage: www.elsevier .com/locate /chembioint

Selected 4-phenyl hydroxycoumarins: In vitro cytotoxicity, teratogeniceffect on zebrafish (Danio rerio) embryos and molecular docking study

http://dx.doi.org/10.1016/j.cbi.2015.02.0110009-2797/� 2015 Published by Elsevier Ireland Ltd.

⇑ Corresponding author at: Faculty of Medicine, Department of Chemistry,University of Niš, Bulevar DrZorana Ðin -dica 81, 18000 Niš, Serbia. Tel.: +381 184570029; fax: +381 18 4238770.

E-mail address: aveselinovic@medfak.ni.ac.rs (A.M. Veselinovic).

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydroxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafishrerio) embryos and molecular docking study, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

Jovana B. Veselinovic a, Gordana M. Kocic b, Aleksandar Pavic c, Jasmina Nikodinovic-Runic c,Lidija Senerovic c, Goran M. Nikolic a, Aleksandar M. Veselinovic a,⇑a Faculty of Medicine, Department of Chemistry, University of Niš, Niš, Serbiab Faculty of Medicine, Department of Biochemistry, University of Niš, Niš, Serbiac Institute of Molecular Genetics and Genetic Engineering, University of Belgrade, Vojvode Stepe 444a, 11000 Belgrade, Serbia

a r t i c l e i n f o a b s t r a c t

262728293031323334

Article history:Available online xxxx

Keywords:4-Phenyl hydroxycoumarinsCytotoxicityMolecular dockingTeratogenic potentialProtein kinase inhibitors

353637383940

A study of structure cytotoxic–activity relationship of three hydroxy 4-phenyl-coumarins and basiccoumarin molecule against two human cell lines (MRC5 fibroblasts and A375 melanoma cells) is present-ed. Of all investigated compounds the highest cytotoxic activity in both cell lines was determined for 7,8-dihydroxy-4-phenyl coumarin. SAR studies revealed the influence of phenyl group and hydroxyl group’snumber and position on cytotoxic activity. In addition, to get an insight about their binding preferences atthe active site of the receptor (catalytic subunit of cAMP-dependent protein kinase) molecular dockingstudies were performed. Docking studies suggest that 4-phenyl hydroxycoumarins are potent cAMP-de-pendent protein kinase inhibitors, better than their analogs without phenyl group. The teratogenic poten-tial was assessed in zebrafish embryo toxicity test and results showed that 4-phenyl dihydroxycoumarinswere more while 7-hydroxy-4-phenyl coumarin was less embryo toxic in comparison to coumarin. Inorder to examine selected 4-phenyl hydroxycoumarins as a new lead compounds the druglikeness ofselected 4-phenyl hydroxycoumarins was estimated by using Lipinski’s ‘‘rule of five’’. All selected 4-phe-nyl hydroxycoumarins proved to have satisfying pharmacokinetic profile.

� 2015 Published by Elsevier Ireland Ltd.

41

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

1. Introduction

Cancers figure among the leading causes of morbidity and mor-tality worldwide with enormous increase in patient number [1,2].According to World Cancer Report from the International Agencyfor Research on Cancer, cases of cancer doubled globally between1975 and 2000, will double again by 2020, and will nearly tripleby 2030 [3]. Cancer can be characterized by the arrest of cell differ-entiation, the inhibition of apoptosis and the accelerated prolif-eration of cloned cells. The understanding of the mechanisms ofcell-death execution and the role that they play in different dis-eases opens new therapeutic strategies [4–7]. Today, most of clin-ical drugs used for cancer treatment show poor curative effect,high toxicity, low selectivity and severe drug resistance [8]. Novelapproaches in cancer treatment include therapies like differen-tiation therapy, angiogenesis inhibition and hormone or tyrosinekinase inhibition. Adenosine 3050-cyclic monophosphate (cyclic

76

77

78

79

80

AMP, cAMP) is a second messenger that plays an important rolein intracellular signal transduction of various stimuli [9,10]. How-ever, the most important function of cAMP in eukaryotes is activa-tion of cAMP-dependent protein kinase (PKA), which is involved inthe control of a variety of cellular processes and is considered asthe best understood member of the serine–threonine proteinkinase superfamily [11]. The cAMP/PKA pathway has been report-ed to stimulate cell growth in many cell types while inhibiting it inothers [12]. Since it has been implicated in the initiation andprogression of many tumors, PKA has been suggested as a novelmolecular target for cancer therapy [13–15].

Coumarins are a large class of naturally occurring phenolic com-pounds [16,17]. Coumarin (2H-1-benzopyran-2-one) is the basiccompound of the coumarin family and chemically can be describedas a fusion of benzene and a 2-pyrone ring. In nature, on basiccoumarin structure hydroxyl groups are present on C-7 and lessfrequently on C-5, C-6, and C-8. Coumarin compounds have theability to exert noncovalent interactions (hydrophobic, p–p andelectrostatic interactions, hydrogen bonds, metal coordination,van der Waals forces, etc.). The unique coumarin structure isresponsible for their various biological activities that include anti-coagulant, estrogenic, vasodilator, hypothermic, anthelmintic,

(Danio

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

2 J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

sedative, analgesic, anti-inflammatory, and antiulcer activity [17–19]. More importantly, a number of coumarin compounds have sig-nificant pharmacological activity, low toxic and side-effects, lowdrug resistance, high bioavailability, broad spectrum of activityand good curative effects which make them very good candidatesfor medicinal use [16,18,19]. Coumarin compounds show anti-cancer activity against various cell lines [20,21]. Despite the differ-ences between the cell lines, the relationship between thestructure and the activity can be established. High cytotoxicity ofcoumarins depends on the presence of at least two polar functionalgroups, particularly phenolic groups at positions 6 and 7 or 7 and 8,which means that coumarin derivatives bearing the ortho-dihy-droxy substitution exert a higher cytotoxic effect [22,23]. However,the underlying mechanisms of the tumor-selectivity and cytotoxicactivity of coumarins are not well understood yet. One of suggestedmechanism is the inhibition of protein kinases [24].

In order to examine 4-phenyl hydroxycoumarins as a new leadcompounds for future development as pharmacophores the drug-likeness of selected 4-phenyl hydroxycoumarins were establishedusing Lipinski’s ‘‘rule of five’’ [25,26]. Following molecular analysis,the aim of this research was to determine in vitro cytotoxic(antiproliferative) activity against two human cell lines (MRC5fibroblasts and A375 melanoma cells) and to establish structure–activity relationships (SAR) of selected 4-phenyl hydroxycoumar-ins in comparison to basic coumarin molecule. To further evaluatechemotherapeutic potential of these molecules and the influenceof phenyl group and number and position of hydroxyl groupsadded to basic coumarin structure, the teratogenic potential wasalso assessed in zebrafish embryo toxicity test. In addition, mole-cular docking studies were performed to get an insight about theirbinding preferences at the active site of the receptor (catalytic sub-unit of cAMP-dependent protein kinase) [27].

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

2. Material and methods

2.1. Chemicals

Coumarin (Cm), 7-hydroxy-4-phenyl coumarin (7C), 5,7-dihy-droxy-4-phenyl coumarin (5,7C), 7,8-dihydroxy-4-phenyl coumar-in (7,8C) and dimethyl sulfoxide (DMSO) of cell tissue grade wereobtained from Sigma (Sigma–Aldrich GmbH, Sternheim, Germany).Chemical structures of investigated molecules are presented inFig. 1. Fish media components and other chemicals were purchasedeither from Sigma–Aldrich or Fisher Scientific.

2.2. Cytotoxicity assay

Melanoma (A375) and fibroblasts (MRC5) were obtained fromthe ATCC culture collection. The cells were cultured in Roswell Park

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186Fig. 1. Molecular structures of coumarin (A), 7-hydroxy-4-phenylcoumarin (B), 5,7-dihydroxy-4-phenylcoumarin (C) and 7,8-dihydroxy-4-phenylcoumarin (D).

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

Memorial Institute medium RPMI-1640 supplemented with100 lg/mL streptomycin, 100 U/mL penicillin and 10% (v/v) fetalbovine serum (FBS) (all from Sigma, Munich, Germany). In orderto test the cytotoxic effect, cells were maintained as a monolayer(1 � 104 cells well�1) in RPMI-1640 and grown in humidifiedatmosphere of 95% air and 5% CO2 at 37 �C.

The viability of cells was evaluated with 3-(4,5-dimethylth-iazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reductionassay [28]. Assay was carried out after 48 h of cell incubation inthe media containing test compounds at different concentrationsand the viability was measured as described in the literature[29]. Stock solutions of tested compounds were prepared in DMSOand diluted in cell culture medium ensuring that maximum con-centration of DMSO in the wells was 0.1% (v/v). The results are pre-sented as percentage of the control (untreated cells) that wasarbitrarily set to 100% and IC50 values were calculated as concen-trations at which 50% cell growth inhibition occurred. Results arepresented as means ± SD.

2.3. Zebrafish embryo toxicity (teratogenicity) test

For zebrafish embryo teratogenicity assessment the generalrules of the OECD 2013 standard procedure were followed [30].

2.3.1. Zebrafish husbandry, eggs collection, and embryo exposureAdult, wild type zebrafish (Danio rerio) were obtained from a

commercial supplier (Pet Centar, Belgrade, Serbia) and weremaintained in the fish medium (2 mM CaCl2, 0.5 mM MgSO4,0.7 mM NaHCO3, 0.07 mM KCl). Fish were maintained in aquaria(70 l volume), with a maximum density of 1 g fish per 1 L of water,at temperature of 26 ± 1 �C, under continuous water aeration andfiltering and under an artificial dark/light cycle of 12:12 h. Femaleand male were continuously kept separated, and regularly fedtwice daily with commercially dry flake food (TetraMin™ flakes;Tetra Melle, Germany) supplemented with Artemia nauplii[30,31]. The day before a spawning, males and females in a rationof 2:1 were placed in breeding aquarium before the onset of dark-ness and left undisturbed overnight. The eggs were collected30 min after the onset of light, twice rinsed from debris using freshfish medium and transferred into Petri dish with the fish medium.Prior to egg transfer, the fish medium was aerated at least for 1 hand pre-warmed to 27 �C.

Within 1.5 h post-fertilization (hpf), fertilized eggs which werefrom 4- to 16-cell stage were selected under binocular stereomi-croscope (PXS-VI, Optica) and transferred into plastic Petri dishescontaining different concentrations of test substance in the fishmedium (treatments) and the fish medium without and with0.1% (v/v) DMSO (negative controls). Then, embryos were indi-vidually transferred into 96-well plates containing 200 ll test solu-tion, one embryo per well, and incubated at 27 �C. Sixteen ortwenty embryos were used per group.

Each substance was tested in three concentrations (1 lg/mL,10 lg/mL, 100 lg/mL) prepared by dissolving the stock solution(in DMSO) in the fish medium. Final concentration of DMSO didnot exceed 0.1% in treatments and negative control with DMSO.

2.3.2. Evaluation of developmental effects and determination ofteratogenicity index (TI)

Fish embryo toxicity assay was performed if fertilization rateswere P80%. An assay was considered valid if overall survival ofembryos in negative controls was P90% until hatching. The apicalendpoints for the assessment of embryotoxicity (mortality) andteratogenicity were evaluated at 24, 48, 72 and 96 hpf using aninverted microscope (CKX41; Olympus, Tokyo, Japan). Theobserved endpoints at different times are given in SupplementaryTable S1.

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx 3

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

All embryos were inspected for morphological characteristics atdifferent developmental stages as described by Kimmel et al. [32].Lethal and teratogenic effects were recorded according to OECD(236) guidelines for testing of the chemicals [30]. Teratogeniceffects were recorded if fingerprint endpoint was observed in atleast 50% of all embryos showing teratogenic effects in test groupand if concentration-response was present [31,33].

Determination of LC50 and EC50 value was performed by theprogram ToxRatPro (ToxRat�, Software for the Statistical Analysisof Biotests, ToxRat Solution GmbH, Alsdorf, Germany, Version2.10.05) using probit analysis with linear maximum likelihoodregression. In order to characterize teratogenic potential of testedsubstances, the teratogenicity index (TI) was determined. TI, calcu-lated as the ratio of LC50 and EC50, was determined for each timepoint. If the TI of given substance is >1, the substance is consideredto be teratogenic; if TI 6 1, given substance has mainly embryolethal effects. The higher TI indicates more teratogenic effects(higher teratogenicity).

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

2.4. Molecular docking

Three dimensional structures of the compounds for dockingsimulation were constructed using MarvinSketch 6.1.0, 2013,ChemAxon (http://www.chemaxon.com). Geometry optimizationwas carried out by employing MMFF94 molecular force field [34].As a target enzyme for docking studies catalytic subunit ofcAMP-dependent protein kinase was selected (protein data bankcode 1BX6) [27]. The flexible compounds were docked into rigidenzyme binding sites using the Molegro Virtual Docker (MVD v.2013.6.0.1.) [35]. Hydrogen bonds and hydrophobic interactionsbetween residues at the active site were calculated. The bindingsite was computed with a grid resolution of 0.3 Å. The MolDockSE as a search algorithm was used with the number of runs setto 100. The parameters of docking procedure were: population size50, maximum number of iterations 1500, energy threshold 100.00and maximum number of steps 300. The number of generated pos-es was 10. The estimation of ligand–receptor interactions weredescribed by the MVD-related scoring functions: MolDock Score,Rerank Score, Hbond Score, Similarity Score, and Docking Score.The ligand was docked into computed cavity instead of ligand from1BX6 using the MolDock Optimizer algorithm and its interactionswere monitored using detailed energy estimates. A maximumpopulation of 100 and maximum iterations of 10,000 were usedfor each run and five best poses were retained. Two dimensionalrepresentations of the best docking pose for selected coumarinsinside target enzyme were generated using LigPlot+ [36].

Molinspiration tool (Molinspiration Cheminformatics 2013)was used for calculating physicochemical properties of investigat-ed coumarins. For calculating of LogP fragment-based contribu-tions and correlation factors were used.

297

298

299

300

301

302

303

304

305

306

307

308

309

310

3. Results and discussion

3.1. Selected 4-hydroxyphenyl coumarins obey ‘‘Rule of five’’

Establishing druglikeness is a very important step in the searchfor new potential drug candidates. Druglikeness allows the assess-ment of the pharmacokinetic profile of the tested molecules basedon the prediction of their absorption and distribution. Absorptionand distribution are two of the four basic pharmacokinetics para-meters and they are part of the LADMER system which includescomplex processes – the release (liberation) of the active substancefrom the pharmaceutical form, its absorption, distribution, themetabolism in the organism and the processes of extracting(elimination) of the drug from the body and the achievement of

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

the pharmacodynamic response. Many problems in the develop-ment of drugs are the result of pharmacokinetic disadvantagessuch as poor absorption, first pass effect, a high degree of bindingto the protein, etc. Based on the analysis of a large number of drugsLipinski and coworkers [25,26] established an effective method-ology for estimation of potential drug solubility and permeabilitybased on the calculation of molecular weight, octanol/water parti-tion coefficient, number of H-bond donors and number of H-bondacceptors. This methodology is known as ‘‘Rule of five’’ since num-ber five is frequently appearing: poor absorption or permeation aremore likely to occur when the molecule has molecular weightmore than 500, log P over 5, and contains more than 5 H-bonddonors or 10 (2 � 5) H-bond acceptors. The critical limit for accept-able drug-likeness is that no more one violation of the rule exists inmolecule [26].

Calculated physicochemical properties of investigated coumar-ins are shown in Table 1. Data indicate that none of the compoundsviolate the ‘‘Rule of five’’.

For all investigated compounds LogP are below 4 and thereforethey have favorable physicochemical profiles for oral bioavail-ability. Topological polar surface area (TPSA) is a good descriptorfor the drug transport properties, drug absorption, includingintestinal absorption, bioavailability, Caco-2 permeability [37–39]. TPSA can be defined as a sum of the surface areas occupiedby the oxygen and nitrogen atoms and the hydrogen atomsattached to them and represent the hydrogen bonding capacityof the molecules. Molecules with TPSA < 140 Å2 have good intesti-nal absorption, while those with TPSA < 60 Å2 show good blood–brain barrier penetration [37,38]. Results presented at Table 1 indi-cate that all investigated coumarin compounds satisfy the criterionfor good intestinal absorption, while TPSA for 7C and Cm indicatedgood blood–brain barrier penetration. The number of H-bonddonors and acceptors can be used for establishing hydrogen bond-ing capacity of the investigated coumarins. Basic coumarin mole-cule has two hydrogen-bond acceptor groups. Substitutedcoumarins have three (7C) and four (5,7C and 7,8C) hydrogen-bondacceptor groups. Also, substituted coumarins have hydrogen-bonddonor groups (hydroxyl group), one in 7C and two in 5,7C and 7,8C.An important factor for oral bioavailability as well as for the effi-cient bonding to receptors and channels is the conformational flex-ibility of the molecules described by the number of rotatable bonds[39]. Sufficient oral bioavailability is expected for molecules with10 rotatable bonds or fewer. Basic coumarin molecule does nothave rotatable bonds, while in all substituted coumarins there isat least one rotatable bond.

Summarizing the physicochemical properties of investigatedcoumarin compounds, conclusion can be made that they obeythe ‘‘Rule of five’’ and meet all criteria for good solubility and per-meability in such a way that further structural modification byintroduction of particularly interesting structural motives can bemade for achieving desired pharmacological properties.

3.2. 7C, 5,7C and 7,8C exhibited in vitro cytotoxic potential againstmelanoma and fibroblast human cell lines

The anti-proliferative activity of coumarins has been reportedand extensively studied [21,40–43]. Some of the tested cell linesinclude lung carcinoma (A549), melanin pigment producing mousemelanoma (B16), human skin malignant melanocytes (SK-MEL-31), human T-cell leukemia (CCRF-HSB-2), and human gastric can-cer, lymph node metastasized (TGBC11TKB). The established struc-ture–activity relationship revealed that the 6,7-dihydroxy moietyhad an important role for coumarin derivatives antiproliferativeactivity [40]. Coumarin was previously shown to be cytotoxicand to cause cell cycle arrest and apoptosis with IC50 values of54.2 lM against cervical cancer HeLa cells [44]. On the contrary,

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

Table 1Calculated molecular properties of investigated coumarin compounds for the assessment of druglikeness.

Comp. miLogPa TPSAb Natomsc MWd NON

e NOHNHf Nviol.

g Nrotb.h Voli

7C 3.234 50.439 18.0 238.242 3 1 0 1 208.0135,7C 2.943 70.667 19.0 254.241 4 2 0 1 216.0317,8C 2.974 70.667 19.0 254.241 4 2 0 1 216.031Cmj 2.014 30.211 11 146.145 2 0 0 0 128.587

‘‘Rule of five’’ 6 5.a Octanol–water partition coefficient.b Topological polar surface area (Å2).c Number of nonhydrogen atoms.d Molecular weight.e Number of hydrogen-bond acceptors (O and N atoms).f Number of hydrogen-bond donors (OH and NH groups).g Number of ‘‘Rule of five’’ violations.h Number of rotatable bonds.i Molecular volume (Å3).j Cm – unsubstituted coumarin.

4 J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

it was not cytotoxic to malignant melanocytes at concentrations of500 lM [45] confirming that its cytotoxicity was highly dependenton the cell type.

Selected 4-phenyl hydroxycoumarins (7C, 5,7C and 7,8C) weretested for anti-proliferative activity against non-transformed(MRC5) and transformed (A375) cell lines (Fig. 2). Cytotoxicitywas compared to that of coumarin molecule. All tested compoundsexhibited cytotoxic effect on both cell lines when applied in highconcentrations of 500 lg/mL (data not shown). When supplied in

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

Fig. 2. In-vitro antiproliferative effect of 4-phenylhydroxycoumarins on (A) humanfibroblasts and (B) melanoma cell line, following (h – 0.1 lg/mL; – 1 lg/mL; –10 lg/mL; j – 100 lg/mL).

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

concentration of 100 lg/mL, 4-phenyl hydroxycoumarins inhibitedproliferation between 40–70% of fibroblasts and 60–80% of mela-noma cell line (Fig. 2). At this concentration, coumarin was lesscytotoxic inhibiting proliferation of 20% and 40% of MRC5 andA375 cells, respectively. Compound 7,8C (7,8-dihydroxy-4-phenyl-coumarin) showed the highest cytotoxic effect with IC50 values of12 lg/mL (47 lM) for fibroblasts and 10 lg/mL (39 lM) for mela-noma cells. Generally, low selectivity was observable betweenMRC5 and A375 cell lines for all tested compounds (Fig. 2). Thein vitro antiproliferative effect of 7,8C against A375 melanoma cellswas 6-fold higher in comparison to that reported for 7,8-dixy-droxycoumarin (7,8-OHC) against SK-MEL-31 (skin malignant mel-anocytes) [45]. In the same study, Finn and coworkers found thatIC50 values of hydroxycoumarins (including 7,8-OHC) were onaverage 10-fold lower against normal skin cell line (HS613) indi-cating higher cytotoxicity against normal cells [45]. In our study,selected 4-phenyl hydroxycoumarins were not more toxic tonon-transformed human fibroblasts. Improved antiproliferativeeffect in comparison to 7,8-OHC indicated the importance of phe-nyl moiety, while slightly higher antiproliferative activity of 7,8Cin comparison to 7C and 5,7C indicated the importance of the num-ber and position of hydroxyl groups. Previously, structural–activityrelationship analyses of coumarin derivatives including 5,7-dihy-droxy-4-methyl-6-(3-methylbutanoyl)-coumarin revealed that analkyl substitution at position 6 of the coumarin ring was criticalfor inducing apoptosis, and the phenyl group at position 4 mighthave enhanced its bioactivity [46].

Molecular targets for coumarin based anticancer drugs are stillbeing elucidated. 6-Nitro-7-hydroxycoumarin (6-NO2-7-OHC) and3,6,8-trinitro-7-hydroxycoumarin (3,6,8-NO2-7-OHC) were shownto function by decreasing DNA synthesis, through an inhibition ofthe S phase regulatory protein, cyclin A [45]. 7-Hydroxycoumarinand both nitro-derivatives were found to alter the phosphorylationstatus of ERK1/ERK2 and p38, while only nitro derivatives caused adramatic increase in tyrosinase activity in melanoma cells [45].

Some studies indicated that relationship between ROS gen-eration and the pro-apoptotic activity of hydroxy coumarins canbe established [47]. This assumption is based on fact that ROSmay stimulate and inhibit distinct signaling pathways, since ROSgeneration show high dependence on the nature of the oxidativestressor and its cellular location [48]. Further, data from the lit-erature show significant correlation between the cytotoxicity con-centration and the following descriptors: absolute hardness,ionization potential and highest occupied molecular orbital(HOMO) energy [49]. Based on this findings correlation betweencytotoxicity of certain hydroxycoumarins and electronic propertiesof the molecule can be established. 4-phenyl hydroxycoumarins

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

Fig. 3. Images of zebrafish embryos exposed to 10 lg/mL of coumarin and 4-phenyl hydroxycoumarins at different time points. (The most occurring malformations: EM –eyes malformation, OS – malformation of sacculi/otoliths, HM – head malformation, ChM – malformation of notochord, SM – somite malformation, SD – skeletal deformity,TM – tail malformation, PE – pericardial edema, YSE – yolk sac edema, YO – yolk opacity, DE – decomposing embryos.)

Table 2Overview of lethal and teratogenic effects of 4-phenylhydroxycoumarins and coumarin in zebrafish (Danio rerio) embryos at 0–96 hpf (hours post fertilization).

Percent (%) hpfa 7C (lg/mL) 5,7C (lg/mL) 7,8C (lg/mL) Coumarin (lg/mL)

1b 10c 100c 1d 10c 100c 1c 10c 100c 1c 10d 100c

Normal embryos 24 81.8 0.0 0.0 87.5 10.0 0.0 70.0 0.0 0.0 75.0 56.2 15.048 81.8 0.0 0.0 87.5 0.0 0.0 70.0 0.0 0.0 75.0 56.2 0.072 81.8 0.0 0.0 56.2 0.0 0.0 70.0 0.0 0.0 60.0 18.7 0.096 81.8 0.0 0.0 56.2 0.0 0.0 70.0 0.0 0.0 45.0 18.7 0.0

Lethal embryos 24 18.2 20.0 20.0 12.5 25.0 80.0 30.0 90.0 100.0 25.0 43.8 55.048 18.2 20.0 25.0 12.5 55.0 100.0 30.0 100.0 100.0 25.0 43.8 60.072 18.2 20.0 55.0 31.3 65.0 100.0 30.0 100.0 100.0 35.0 43.8 60.096 18.2 20.0 55.0 37.5 85.0 100.0 30.0 100.0 100.0 55.0 43.8 85.0

Teratogenic embryos 24 0.0 80.0 80.0 0.0 65.0 20.0 0.0 10.0 0.0 0.0 0.0 30.048 0.0 80.0 75.0 0.0 45.0 0.0 0.0 0.0 0.0 0.0 0.0 40.072 0.0 80.0 45.0 12.5 35.0 0.0 0.0 0.0 0.0 5.0 37.5 40.096 0.0 80.0 45.0 6.3 15.0 0.0 0.0 0.0 0.0 0.0 37.5 15.0

Affected embryos 24 18.2 100.0 100.0 12.5 90 100 30.0 100.0 100.0 25.0 43.8 85.048 18.2 100.0 100.0 12.5 100 100 30.0 100.0 100.0 25.0 43.8 100.072 18.2 100.0 100.0 43.8 100 100 30.0 100.0 100.0 40.0 81.3 100.096 18.2 100.0 100.0 43.8 100 100 30.0 100.0 100.0 55.0 81.3 100.0

a hpf = hours post fertilization.b Tested 22 embryos per concentration.c Tested 20 embryos per concentration.d Tested 16 embryos per concentration.

J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx 5

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

used in presented study were evaluated for antioxidant activityand their electronic properties have been calculated using DensityFunctional Theory [50].

3.3. 7C, 5,7C and 7,8C exhibited embryo toxicity in zebrafish model

Zebrafish embryos have increasingly been accepted as suitablemodel for teratogenicity testing [51,52]. Recently, effects ofcoumarin and warfarin on zebrafish embryos were evaluated[31]. Expanding on the findings of Weight et al. [31], our studyinvestigated embryotoxicity of 4-phenylhydroxylated coumarins

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

(Fig. 3). In 4-day assay, all compounds exhibited teratogenic andlethal effects in concentration dependant manner (Table 2, Supple-mentary material Table S1).

The highest concentration (100 lg/mL) of 4-phenyldihy-droxylated coumarins (5,7C and 7,8C) resulted in early lethal effect,while 7-hydroxy-4-phenyl coumarin (7C) induced malformation of80% zebrafish embryos (Table S2–S4). At the same concentration,coumarin induced disintegration of notochord, edemas, yolk sacischemia, head and skeletal deformations, resulting in lethal out-come for the most of zebrafish embryos up to 96 hpf (Table S5).At 10 times lower concentration (10 lg/mL), exposure to

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

Fig. 4. Best docking pose of studied coumarins inside active site of enzyme. Studiedcoumarins have different color: grey – 7,8C; blue – 5,7C; violet – 7C; green – Dapand purple – Cm. (For interpretation of the references to color in this figure legend,the reader is referred to the web version of this article.)

6 J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

compound 7,8C resulted in 100% mortality of zebrafish embryosduring early embryogenesis, while compounds 5,7C and 7Cinduced teratogenic effects on the majority of embryos already at24 hpf (Fig. 3). By 96 hpf affected embryos exposed to the com-pound 5,7C displayed sub-lethal points such as pericardial edema,slow circulation, yolk sac edema and yolk opacity (Fig. 3; Tables S4and S5), reaching 85% mortality. On the other hand, embryosexposed to 7C did not have lethal outcome and 80% stayed terato-genic with small heads and eyes, edemas and skeletal deformities(Fig. 3), but no circulatory problems were observed (Table 2, TablesS2 and S3). In comparison, when supplied with 10 lg/mL of

Fig. 5. Two dimensional representations of the best docking pose for (A) 7,8-dihydroxy-coumarin, (D) 7,8-dihydroxycoumarinand, (E) coumarin inside binding pocket.

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

coumarin, embryos displayed teratogenic defects at 72 hpf, whichwere solely manifested as slight tail malformation (Fig. 3) withoutfurther lethal outcome (Table S5).

Overall 4-phenyldihydroxylated derivatives were more embryotoxic in comparison to coumarin, at 72 h calculated LC50 valueswere 1.20 lg/mL, and 3.22 lg/mL for 7,8C, and 5,7C respectively.EC50 value was 1.32 lg/mL for 5,7C compound, while underconditions tested it was not possible to determine EC50 value forcompound 7,8C due to its prominent embryotoxicity. Notably,compound 7C was significantly less embryotoxic in comparisonto coumarin and other derivatives, with LC50 values of 100.1 lg/mL, but more teratogenic than coumarin itself. It has to be notedthat there are important species differences in the metabolismand hepatotoxicity of coumarin according to literature [53]. There-fore, the toxicity and teratogenicity effects observed in D. reriocannot be extrapolated in a straight forward manner to humans.

3.4. Possible protein kinase inhibitors

Based on observations that the catalytic domain of serine-thre-onine kinases can be highly conserved, proposed inhibition modelstudy may serve for recognition of possible coumarins interactionwith distinct substrates. The results of molecular docking studiesare presented in Figs. 4 and 5. Two dimensional representationsof the best docking pose for selected coumarins (Fig. 5) give moredetailed insight into the interactions with particular amino acids inenzyme binding pocket.

Based on the presented results it can be concluded thathydrophobic interactions between investigated coumarins and

4-phenyl coumarin, (B) 5,7-dihydroxy-4-phenyl coumarin, (C) 7-hydroxy-4-phenyl

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506507508509510511512513514

J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx 7

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

binding pocket play an important role. However, number, bondlength, and bond energy of hydrogen bonds formed betweencoumarin molecules and enzyme has a substantial role in effecton enzyme inhibition [54]. It was observed from in silico studiesof compounds binding to 1BX6 that basic coumarin molecule(Cm) does not form hydrogen bonds with active enzyme center.7C forms three hydrogen bonds so that hydroxyl group formstwo, with Ser53 (2.63 Å) and Gln84 (4.87 Å), while oxygen fromcarbonyl group forms one hydrogen bond with Ile73 (3.30 Å).5,7C forms three hydrogen bonds so that hydroxyl group at posi-tion 7 forms two, with Ser53 (2.61 Å) and Gln84 (4.87 Å), whileoxygen from carbonyl group forms one hydrogen bond with Ile73(3.40 Å). 7,8C forms up to five hydrogen bonds so that hydroxylgroup at position 7 forms one with His87 (4.66 Å), while hydroxylgroup at position 8 forms two hydrogen bonds with Gln84 (3.07and 4.99 Å). Lacton oxygen forms one hydrogen bond with Gln84(3.16 Å) and oxygen from carbonyl group forms one hydrogenbond with Ser53 (2.61 Å). Naturally occurring coumarin, Daphnet-in (7,8-dihydroxycoumarin, Dap), is found to be Protein Kinaseinhibitor [24]. In order to compare Dap inhibition activity towardProtein Kinase to the activity of 4-phenyl hydroxycoumarins inves-tigated in this study, Dap was docked into same enzyme active site(Fig. 4). Two dimensional representation of the best docking pose ispresented at Fig. 5. Dap forms four hydrogen bonds so that hydrox-yl group at position 8 forms one with Ser53 (4.29 Å), while hydrox-yl group at position 7 forms two hydrogen bonds, with Ser53(3.10 Å) and Gln84 (4.81 Å). Oxygen from carbonyl group formsone hydrogen bond with Val57 (4.07 Å). For activity assessmentdocking score was employed. Both MolDock score and Dockingscore revealed the same activity order: 7,8C > 5,7C > 7C > Dap >Cm. The most potent inhibitor was 7,8C which is in correlation tothe experimental results for their cytotoxicity. These results indi-cate the importance of phenyl group addition to basic coumarinstructure for inhibition of cAMP-dependent protein kinase.

515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556

4. Conclusion

A study of structure–cytotoxic activity relationship of threehydroxy 4-phenylcoumarins and basic coumarin molecule againsttwo human cell lines (MRC5 fibroblasts and A375 melanoma cells)is presented. Of all investigated compounds the highest cytotoxicactivity in both cell lines was determined for 7,8-dihydroxy-4-phenylcoumarin. SAR studies revealed the influence of phenylgroup and hydroxyl group number and position on cytotoxic activ-ity. Selected 4-phenyl hydroxycoumarins were not more toxic tonon-transformed human fibroblasts. Improved antiproliferativeeffect in comparison to alkyl analogs indicates the importance ofphenyl moiety, while slightly higher antiproliferative activity of7,8C in comparison to 7C and 5,7C indicates the importance ofthe number and position of hydroxyl groups. Molecular dockingstudies were performed at the active site of the receptor (catalyticsubunit of cAMP-dependent protein kinase) for the cytotoxicmechanism elucidation. Docking studies suggest that 4-phenylhydroxycoumarins are potent cAMP-dependent protein kinaseinhibitors, better than their analogs without phenyl group. Themost potent inhibitor was 7,8C which is in correlation toexperimental results for their cytotoxicity. The teratogenic poten-tial was assessed in zebrafish embryotoxicity test and resultsshowed that 4-phenyl-dihydroxycoumarins were more embry-otoxic while 7-hydroxy-4-phenyl coumarin was less toxic in com-parison to coumarin. In order to examine selected 4-phenylhydroxycoumarins as a new lead compounds, the druglikeness ofselected 4-phenyl hydroxycoumarins was established by using Lip-inski’s ‘‘rule of five’’. All selected 4-phenyl hydroxycoumarins havesatisfying pharmacokinetic profile. In summary, selected 4-phenyl

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

hydroxycoumarins can be considered as good molecular templatesfor potential cytotoxic agents. Future perspectives for these com-pounds are clear – their chemical modifications can lead to reduc-ing embryotoxicity and improving cytotoxic effect and selectivity.

Transparency Document

The Transparency document associated with this article can befound in the online version.

Acknowledgments

Authors would like to thank anonymous reviewers whose sug-gestions have improved our manuscript. This work has been finan-cially supported by the Ministry of Education and Science, Republicof Serbia, under Grant Nos. 31060 and 173048. Authors acknowl-edge Dr. BrankaVasiljevic for careful review of the manuscript.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.cbi.2015.02.011.

References

[1] A. Jemal, F. Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancerstatistics, Cancer J. Clin. 61 (2011) 69–90.

[2] C.J.L. Murray, A.D. Lopez, Alternative projections of mortality and disability bycause 1990–2020: global burden of disease study, Lancet 349 (1997) 1498–1504.

[3] C.D. Mathers, D. Loncar, Projections of global mortality and burden of diseasefrom 2002 to 2030, PLoS Med. 3 (2006) 2011–2030.

[4] D.R. Green, G. Kroemer, The pathophysiology of mitochondrial cell death,Science 305 (2004) 626–629.

[5] D. Hanahan, J. Folkman, Patterns and emerging mechanisms of the angiogenicswitch during tumorigenesis, Cell 86 (1996) 353–364.

[6] L.H. Hartwell, M.B. Kastan, Cell cycle control and cancer, Science 266 (1994)1821–1828.

[7] W. Risau, Mechanisms of angiogenesis, Nature 386 (1997) 671–674.[8] H.X. Chen, J.N. Cleck, Adverse effects of anticancer agents that target the VEGF

pathway, Nat. Rev. Clin. Oncol. 6 (2009) 465–477.[9] J.A. Beavo, L.L. Brunton, Cyclic nucleotide research – still expanding after half a

century, Nat. Rev. Mol. Cell Biol. 3 (2002) 710–718.[10] K.V. Chin, W.L. Yang, R. Ravatn, T. Kita, E. Reitman, D. Vettori, M.E. Cvijic, M.

Shin, L. Iacono, Reinventing the wheel of cyclic AMP: novel mechanisms ofcAMP signaling, Ann. N.Y. Acad. Sci. 968 (2002) 49–64.

[11] K. Taskén, B.S. Skalhegg, K.A. Taskén, R. Solberg, H.K. Knutsen, F.O. Levy, M.Sandberg, S. Orstavik, T. Larsen, A.K. Johansen, T. Vang, H.P. Schrader, N.T.Reinton, K.M. Torgersen, V. Hansson, T. Jahnsen, Structure, function, andregulation of human cAMP-dependent protein kinases, Adv. Second MessengerPhosphoprotein Res. 31 (1997) 191–204.

[12] P.J. Stork, J.M. Schmitt, Crosstalk between cAMP and MAP kinase signaling inthe regulation of cell proliferation, Trends Cell Biol. 12 (2002) 258–266.

[13] R. Kleppe, C. Krakstad, F. Selheim, R. Kopperud, S.O. Døskeland, The cAMP-dependent protein kinase pathway as therapeutic target: possibilities andpitfalls, Curr. Top. Med. Chem. 11 (2011) 1393–1405.

[14] A. Caretta, C. Mucignat-Caretta, Protein Kinase A in Cancer, Cancers 3 (2011)913–926.

[15] L. Sapio, F. Di Maiolo, M. Illiano, A. Esposito, E. Chiosi, A. Spina, S. Naviglio,Targeting protein kinase a in cancer therapy: an update, EXCLI J. 13 (2014)843–855.

[16] P. Borges, F. Roleira, N. Milhazes, L. Santana, E. Uriarte, Simple coumarins andanalogous in medicinal chemistry: occurrence, synthesis and biologicalactivity, Curr. Med. Chem. 12 (2005) 887–916.

[17] R.D.H. Murray, J. Mendez, S.A. Brown, The Natural Coumarins: Occurrence,Chemistry, and Biochemistry, Wiley, Chichester, 1982.

[18] X.-M. Peng, G.L.V. Damu, C.-H. Zhou, Current developments of coumarincompounds in medicinal chemistry, Curr. Pharm. Des. 19 (2013) 3884–3930.

[19] L. Wu, X. Wang, W. Xu, F. Farzaneh, R. Xu, The structure and pharmacologicalfunctions of coumarins and their derivatives, Curr. Med. Chem. 16 (2009)4236–4260.

[20] I. Kostova, Synthetic and natural coumarins as cytotoxic agents, Curr. Med.Chem. 5 (2005) 29–46.

[21] A. Lacy, R. O’Kennedy, Studies on coumarins and coumarin-related compoundsto determine their therapeutic role in the treatment of cancer, Curr. Pharm.Des. 10 (2004) 3797–3811.

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011

557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607

608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657

658

8 J.B. Veselinovic et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

CBI 7287 No. of Pages 8, Model 5G

24 February 2015

[22] M. Kawase, H. Sakagami, K. Hashimoto, S. Tani, H. Hauer, S.S. Chatterjee,Structure–cytotoxic activity relationships of simple hydroxylated coumarins,Anticancer Res. 23 (2003) 3243–3246.

[23] H. Kolodziej, O. Kayser, H.J. Woerdenbag, W. van Uden, N. Pras, Structure–cytotoxicity relationships of a series of natural and semi-synthetic simplecoumarins as assessed in two human tumour cell lines, Z. Naturforsch C 52(1997) 240–244.

[24] E.B. Yang, Y.N. Zhao, K. Zhang, P. Mack, Daphnetin, one of coumarinderivatives, is a protein kinase inhibitor, Biochem. Biophys. Res. Commun.260 (1999) 682–685.

[25] C.A. Lipinski, Lead- and drug-like compounds: the rule-of-five revolution, DrugDiscovery Today Technol. 1 (2004) 337–341.

[26] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental andcomputational approaches to estimate solubility and permeability in drugdiscovery and development settings, Adv. Drug Deliv. Rev. 46 (2001) 3–26.

[27] N. Narayana, T.C. Diller, K. Koide, M.E. Bunnage, K.C. Nicolaou, L.L. Brunton,N.H. Xuong, L.F. Ten Eyck, S.S. Taylor, Crystal structure of the potent naturalproduct inhibitor balanol in complex with the catalytic subunit of cAMP-dependent protein kinase, Biochemistry 38 (1999) 2367–2376.

[28] M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development ofa precise and rapid dye method for measuring cell growth/cell kill, J. Immunol.Methods 119 (1989) 203–210.

[29] L.E. Mihajlovic, A. Savic, J. Poljarevic, I. Vuckovic, M. Mojic, M. Bulatovic, D.Maksimovic-Ivanic, S. Mijatovic, G.N. Kaluderovic, S. Stosic-Grujicic, D.Miljkovic, S. Grguric-Sipka, T.J. Sabo, Novel methylene modified cyclohexylethylenediamine-N,N0-diacetate ligands and their platinum(IV) complexes.Influence on biological activity, J. Inorg. Biochem. 109 (2012) 40–48.

[30] OECD, OECD guidelines for the testing of chemicals. Fish embryo acute toxicity(FET) test. No. 236, OECD, Paris (2013).

[31] S. Weigt, N. Heubler, R. Strecker, T. Braunbeck, T.H. Broshard, Developmentaleffects of coumarin and the anticoagulant coumarin derivative warfarin onzebrafish (Danio rerio) embryos, Reprod. Toxicol. 33 (2012) 133–141.

[32] C.B. Kimmel, W.W. Ballard, S.R. Kimmel, B. Ullmann, T.F. Schilling, Stages ofembryonic development of the zebrafish, Dev. Dyn. 203 (1995) 253–310.

[33] R. Nagel, DarT: The embryo test with the zebrafish Danio rerio – a generalmodel in ecotoxicology and toxicology, ALTEX 19 (2002) 38–48.

[34] T.A. Halgren, Merck molecular force field. I. Basis, form, scope,parameterization, and performance of MMFF94, J. Comput. Chem. 17 (1996)490–519.

[35] R. Thomsen, M.H. Christensen, MolDock: a new technique for high-accuracymolecular docking, J. Med. Chem. 49 (2006) 3315–3321.

[36] R.A. Laskowski, M.B. Swindells, LigPlot+: multiple ligand–protein interactiondiagrams for drug discovery, J. Chem. Inf. Model. 51 (2011) 2778–2786.

[37] P. Ertl, B. Rohde, P. Selzer, Fast calculation of molecular polar surface area as asum of fragment-based contributions and its application to the prediction ofdrug transport properties, J. Med. Chem. 43 (2000) 3714–3717.

[38] S. Prasanna, R.J. Doerksen, Topological polar surface area: a useful descriptor in2D-QSAR, Curr. Med. Chem. 16 (2009) 21–41.

[39] D.F. Veber, S.R. Johnson, H.-Y. Cheng, B.R. Smith, K.W. Ward, K.D. Koppl,Molecular properties that influence the oral bioavailability of drug candidates,J. Med. Chem. 45 (2002) 2615–2623.

Please cite this article in press as: J.B. Veselinovic et al., Selected 4-phenyl hydrerio) embryos and molecular docking study, Chemico-Biological Interactions (

[40] S. Kawaii, Y. Tomono, K. Ogawa, M. Sugiura, M. Yano, Y. Yoshizawa, Theantiproliferative effect of coumarins on several cancer cell lines, AnticancerRes. 21 (2A) (2001) 917–923.

[41] C.S. Francisco, L.R. Rodrigues, N.M. Cerqueira, A.M. Oliveira-Campos, L.M.Rodrigues, Synthesis of novel benzofurocoumarin analogues and their anti-proliferative effect on human cancer cell lines, Eur. J. Med. Chem. 47 (2012)370–376.

[42] M. Sumiyoshi, M. Sakanaka, M. Taniguchi, K. Baba, Y. Kimura, Anti-tumoreffects of various furocoumarins isolated from the roots, seeds and fruits ofAngelica and Cnidium species under ultraviolet A irradiation, J. Nat. Med. 68(2014) 83–94.

[43] Y. Ye, X. Han, B. Guo, Z. Sun, S. Liu, Combination treatment with platycodin Dand osthole inhibits cell proliferation and invasion in mammary carcinomacell lines, Environ. Toxicol. Pharmacol. 36 (2013) 115–124.

[44] J.Y. Chuang, Y.F. Huang, H.F. Lu, H.C. Ho, J.S. Yang, T.M. Li, N.W. Chang, J.G.Chung, Coumarin induces cell cycle arrest and apoptosis in human cervicalcancer HeLa cells through a mitochondria- and caspase-3 dependentmechanism and NF-kappa B down-regulation, In Vivo 21 (2007) 1003–1009.

[45] G.J. Finn, B.S. Creaven, D.A. Egan, Activation of mitogen activated proteinkinase pathways and melanogenesis by novel nitro-derivatives of 7-hydroxycomarin in human malignant melanoma cells, Eur. J. Pharm. Sci. 26(2005) 16–25.

[46] M.-H. Lin, C.-H. Cheng, K.-C. Chen, W.-T. Lee, Y.-F. Wang, C.-Q. Xiao, C.-W. Lin,Induction of ROS-independent JNK-activation-mediated apoptosis by a novelcoumarin-derivative, DMAC, in human colon cancer cells, Chem. Biol. Interact.218 (2014) 42–49.

[47] M.E. Riveiro, A. Moglioni, R. Vazquez, N. Gomez, G. Facorro, L. Piehl, E.R. deCelis, C. Shayo, C. Davio, Structural insights into hydroxycoumarin-inducedapoptosis in U-937 cells, Bioorg. Med. Chem. 16 (2008) 2665–2675.

[48] G.I. Giles, The redox regulation of thiol dependent signaling pathways incancer, Curr. Pharm. Des. 12 (2006) 4427–4443.

[49] M. Ishihara, Y. Yokote, H. Sakagami, Quantitative structure cytotoxicityrelationship analysis of coumarin and its derivatives by semiempiricalmolecular orbital method, Anticancer Res. 26 (4B) (2006) 2883–2886.

[50] J.B. Veselinovic, A.M. Veselinovic, Z.J. Vitnik, V.D. Vitnik, G.M. Nikolic,Antioxidant properties of selected 4-phenyl hydroxycoumarins: integratedin vitro and computational studies, Chem. Biol. Interact. 214 (2014) 49–56.

[51] I.W. Selderslaghs, R. Blust, H.E. Witters, Feasibility study of the zebrafish assayas an alternative method to screen for developmental toxicity andembryotoxicity using a training set of 27 compounds, Reprod. Toxicol. 33(2012) 142–154.

[52] I.W. Selderslaghs, A.R. Van Rompay, W. De Coen, H.E. Witters, Development ofa screening assay to identify teratogenic and embryotoxic chemicals using thezebrafish embryo, Reprod. Toxicol. 28 (2009) 308–320.

[53] J.H. Fentem, J.R. Fry, Species differences in the metabolism and hepatotoxicityof coumarin, Comp. Biochem. Physiol. C 104 (1993) 1–8.

[54] J. Veselinovic, A. Veselinovic, A. Toropov, A. Toropova, I. Damnjanovic, G.Nikolic, Monte Carlo method based QSAR modeling of coumarin derivates aspotent HIV-1 integrase inhibitors and molecular docking studies of selected 4-phenyl hydroxycoumarins, Acta Fac. Med. Naiss. 31 (2014) 95–103.

roxycoumarins: In vitro cytotoxicity, teratogenic effect on zebrafish (Danio2015), http://dx.doi.org/10.1016/j.cbi.2015.02.011