Bioorganic Chemistry 93 (2019) 103321

Contents lists available at ScienceDirect

Bioorganic Chemistry

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

Identification of a potent and selective gametocytocidal antimalarial agent T from the stem barks of Lophira lanceolata Annalisa Lopatrielloa,1, Harouna Soréb,1, Annette Habluetzelc, Silvia Parapinid, ⁎ ⁎ Sarah D'Alessandrod, Donatella Taramellie, , Orazio Taglialatela-Scafatia, a Department of Pharmacy, School of Medicine and Surgery, University of Naples Federico II, Via D. Montesano 49, 80131 Naples, Italy b Centre National de Recherche et de Formation sur le Paludisme, 01 BP 2208 Ouagadougou, Burkina Faso c School of Pharmacy, University of Camerino, Via Madonna delle Carceri 9, 62032 Camerino, Italy d Department of Biomedical, Surgical and Dental Sciences, Università degli Studi di Milano, Via Pascal 36, 20133 Milano, Italy e Department of Pharmacological and Biomolecular Sciences, Università degli Studi di Milano, Via Pascal 36, 20133 Milano, Italy

ARTICLE INFO ABSTRACT

Keywords: Bioassay-guided fractionation of the organic extract obtained from stem barks of the African plant Lophira Malaria lanceolata has led to the isolation of seven biflavonoids, including the new α′-chlorolophirone E(5) and 5′- Transmission-blocking drugs chlorolophirone D (6). Among the isolated compounds, the bichalcone lophirone E was identified as a potent Gametocytocites gametocytocidal agent with an IC50 value in the nanomolar range and negligible cytotoxicity (selectivity Lophira lanceolata index = 570). Lophirone E proved to be about 100 times more active against P. falciparum stage V gametocytes Biflavonoids than on asexual blood stages, thus exhibiting a unique stage-specific activity profile. The isolation of structural Lophirone E analogues allowed to draw preliminary structure-activity relationships, identifying the critical positions on the chemical scaffold of lophirone E.

1. Introduction malaria management programs, moving from control toward elimination of the disease. Accordingly, the transmissible Plasmodium stages i.e. Despite the huge efforts for its control and elimination, malaria the gametocytes, have been identified as a priority target for reducing remains an overwhelming parasitic disease: according to WHO esti- the transmission of the disease. Rationalizing the requirements for mates, about 219 million malaria cases and 429,000 deaths have oc- novel pharmaceutical tools suited to malaria elimination, Medicines for curred worldwide in 2017. Among these, more than 90% were recorded Malaria Venture (MMV) has emphasized the need to develop molecules in sub-Saharan Africa, the remaining occurring in South-East Asia and active not only on asexual Plasmodium blood stages that cause the South America [1]. disease, but also on gametocytes, to block the transmission of the The protozoan parasites of the genus Plasmodium, causative agents parasite [3,4]. However, to follow this strategy, there is the need to of malaria, have a complex life cycle including asexual phases, devel- increase efforts in the identification and development of safe andef- oping in the vertebrate host, and a sexual phase, which initiates in the fective transmission-blocking candidate drugs. While Artemisinin-based vertebrate host (gamogony) and is completed in the mosquito vector Combination Therapy (ACT) seems to reduce gametocyte density [5], it (sporogony). In the vertebrate host, a small proportion of asexual blood cannot provide a sterilizing effect on mature stage-V gametocytes and stage parasites undergo sexual differentiation into dimorphic micro- consequently cannot block malaria transmission. For the management and macrogametocytes. Gametocytes undergo five stages of maturation of malaria patients, the World Health Organization (WHO) currently (from stage I to V) and only mature (stage V) gametocytes are capable recommends the addition of a single dose of primaquine (0.25 mg/kg of of infecting Anopheles mosquitoes. Once taken up by the blood feeding body weight) to the ACT administration to eliminate mature P. falci- vector, the sexual reproduction of gametocytes occurs followed by the parum gametocytes [1]. However, the use of primaquine is restricted by sporogonic development ending up with the production of thousands of its adverse effects in patients with G6PD deficiency [6]. infective sporozoites, which migrate to the insect salivary gland [2]. The need of expanding the arsenal of transmission-blocking drugs is The past few years have witnessed an ambitious shift in global widely recognized by the scientific community. Among the various

⁎ Corresponding authors. E-mail addresses: [email protected] (D. Taramelli), [email protected] (O. Taglialatela-Scafati). 1 These authors contributed equally. https://doi.org/10.1016/j.bioorg.2019.103321 Received 1 July 2019; Received in revised form 18 September 2019; Accepted 26 September 2019 Available online 26 September 2019 0045-2068/ © 2019 Elsevier Inc. All rights reserved. A. Lopatriello, et al. Bioorganic Chemistry 93 (2019) 103321 possible approaches, investigation of medicinal plants has proven to be 2.3. Extraction and isolation a valid strategy for the discovery of new transmission-blocking compounds active against gametocytes and possibly also against the spor- The powder of L. lanceolata stem barks (1.1 Kg) was exhaustively ogonic stages, for the inclusion in multi-stage combination medicines extracted by maceration in methanol (5 × 2 L) to afford 157 g ofa [7–9]. As taught by the history of modern medicine, the link between crude extract after removal of the solvent under vacuum. The methanol the plant kingdom and malaria dates back to the discovery of the first extract thus obtained was dissolved in water and then partitioned antimalarial drug quinine, isolated from Cinchona barks in 1820 [10] subsequently against n-hexane, EtOAc, and n-butanol to yield 3.0, 9.1, and continues, in more recent times, with the identification of the and 18.5 g of organic phases, respectively. The EtOAc extract (9.1 g) highly effective antimalarial compound artemisinin derived from Ar- was subjected to MPLC over a silica gel column (70-230 mesh) and temisia annua leaves [11]. In effect, the two most significant break- eluted with a solvent gradient of increasing polarity from n-hexane to throughs in the fight against malaria are the result of a multi-dis- EtOAc, EtOAc-MeOH (1:1), and finally MeOH. Altogether, ten fractions ciplinary study approach on traditional medicine practices and plants were obtained, which were then subjected to bioactivity evaluation. used for the treatment of malaria fevers. Fraction 2 (41 mg), eluted with n-hexane-EtOAc, 6:4, was separated by Preliminary screening of plants used for treatment of malaria in normal-phase HPLC (n-hexane-EtOAc, 65:35, flow rate 0.8 mL/min), Burkina Faso has selected the stem barks of Lophira lanceolata as a yielding lophirone D (4, 1.3 mg), lophirone E (3, 1.8 mg), 5′-chlor- promising source of bioactive compounds, worthy of further in- olophirone D (6, 1.0 mg) and α′-chlorolophirone E (5, 0.9 mg). Fraction vestigation. Lophira lanceolata Van Tiegh Ex Keay, belonging to the 3 (175 mg), eluted with n-hexane-EtOAc 1:1, was partially separated by family Ochnaceae, is a tall tree reaching up to 60 feet, distributed in HPLC (n-hexane-EtOAc, 1:1, flow rate 0.8 mL/ min) to afford pure lo- West and Central Africa [12]. A concoction prepared from the bark of phirone C (7, 7.6 mg) and lophirone F (2, 9.8 mg). Fraction 4 (2 g), the roots or stems is traditionally used to treat malaria, menstrual pains, eluted with n-hexane-EtOAc, 4:6, was first purified by GCC over a silica intestinal troubles, dysentery, diarrhea in children, general tiredness, gel column (70–230 mesh) using n-hexane-EtOAc (from 8:2 to EtOAc) pulmonary diseases, rheumatism, toothache, and yellow fever. In pre- and then EtOAc-MeOH (1:1) and finally MeOH, to afford lophirone A vious studies conducted with crude extracts, L. lanceolata revealed ac- (1, 121.0 mg). tivity on asexual blood stages of the parasite, confirming the malaria α′-chloro-Lophirone E (5): Colorless amorphous solid. ESIMS m/z 405 − − curative potential of the plant [12]. [M−H] ; HR-ESIMS: m/z [M−H] 405.0533 (calcd for C23H13ClO5, 1 13 Several polyphenols and flavonoids have been isolated from the 405.0535). H and C NMR [(CD3)2CO]: see Table 2. stem bark of L. lanceolata [13–16] but little information is available 5′-chloro-Lophirone D (6): Colorless amorphous solid. ESIMS m/z about the biological or pharmacological profile on the isolated sec- 433 [M−H]−; HR-ESIMS: m/z [M−H]− 433.0480 (calcd for 1 13 ondary metabolites of this plant. The aim of the present study was to C24H13ClO6, 433.0484). H and C NMR [(CD3)2CO]: see Table 2. characterize in vitro the anti-plasmodial activity of L. lanceolata stem barks against P. falciparum and isolate and identify the molecules ef- 2.4. Plasmodium falciparum parasite cultures fective against mature gametocytes and/or asexual blood stages. Asexual blood-stage cultures. P. falciparum CQ-sensitive 3D7 and CQ- 2. Material and methods resistant W2 strains were cultured according to the method of Trager & Jensen, with minor modifications [17]. Briefly, parasites were cultured 2.1. General experimental procedures using human erythrocytes (type O-positive red blood cells for 3D7 and A-positive red blood cells for W2) at 5% hematocrit in parasite culture 1H (700 MHz) and 13C (175 MHz) NMR spectra were measured on a medium (RPMI 1640 supplemented with 20 mM HEPES, 2 mM Gluta- Varian INOVA NMR spectrometer. Chemical shifts are referenced to the mine, 0.01% hypoxanthine and 1% of lipid-rich bovine serum albumin- residual solvent signal (CDCl3: δH 7.26, δC 77.0; (CD3)2CO: δH 2.09, δC Albumax II). Parasite cultures were kept at 37 °C in the presence of a 1 205.9, 30.6). Homonuclear H connectivities were determined by COSY fixed gas composition (1%2 O , 5% CO2, 94% N2). experiments; one-bond heteronuclear 1He13C connectivities by the Gametocyte cultures. The transgenic 3D7 strain 3D7elo1-pfs16- HSQC experiment; and two - and three-bond 1He13C connectivities by CBG99 expressing the luciferase CBG99 under the gametocyte specific gradient-HMBC experiments optimized for a 2,3J value of 8 Hz. Low- and promoter pfs16 was used in all the gametocytocidal experiments and high-resolution ESIMS were performed on LTQ OrbitrapXL (Thermo the luciferase activity was taken as measure of gametocyte viability Scientific) mass spectrometer. Medium-pressure liquid chromatography [18]. To obtain constant and high gametocyte production, culture was performed on a Büchi (Switzerland) apparatus using a silica gel (7- medium was supplemented with 10% naturally clotted heat-inactivated 230 mesh) column; separations were monitored by TLC on Macherey- 0-positive human serum (Interstate Blood Bank, Inc.) instead of Al-

Nagel 60 F254 (0.20 mm) plates and were visualized by UV inspection bumax. Briefly, cultures of P. falciparum 3D7 asexual parasites were and/or staining with Pancaldi mixture and heating. HPLC was conducted diluted to 0.5% parasitemia and maintained up to day 15 with daily on a Knauer apparatus equipped with a refractive index detector. media changes without fresh red blood cells (RBC) addition. When early Luna (normal phase, SI60, or reverse-phase RP18, 250 × 4 mm) gametocytes (stages I and II) and dead asexual forms were found, cul- (Phenomenex) columns were used, with 0.8 mL/min as flow rate. tures were treated with 50 mM N-acetyl-D-glucosamine (NAG) for 72 h, to block reinvasion of remaining asexual parasites and obtain pure and 2.2. Plant material nearly synchronous gametocytes. All the cultures were maintained at

37 °C in a standard gas mixture consisting of 1% O2, 5% CO2, and 94% Stem barks of L. lanceolata were collected in Banfora, western N2. Gametocyte stages were routinely checked by Giemsa-stained Burkina Faso on January 2016 and dried at the temperature of 25 °C, smears. The day of the experiment, stage V gametocytes were counted humidity level ranging from 30 to 85%, for two weeks. After drying, the and when their percentage was higher than 90% the experiment was stem bark material was powdered using a food blender and then stored performed. in a clean container, in dry and dark conditions. The plant was identified and authenticated at Comoé regional department of forestry and 2.5. P. falciparum drug susceptibility assay confirmed on www.theplantlist.org. A voucher specimen of thebark (Cnrfp16Ll) was deposited in the herbarium of “Centre National de Extracts, fractions and pure compounds were dissolved in dimethyl Recherche et de Formation sur le Paludisme CNRFP” (Ouagadougou, sulfoxide (DMSO) and diluted with medium to achieve the required Burkina Faso). concentrations (final DMSO or ethanol concentration ≤1%, which is

2 A. Lopatriello, et al. Bioorganic Chemistry 93 (2019) 103321 not toxic to the parasites). Chloroquine (CQ) or methylene blue (MB) endothelial cells (HMEC-1). The cells were maintained under standard were used as positive control for the assessment on asexual or game- conditions at 37 °C in a 5% CO2 incubator in MCDB 131 medium sup- tocyte stages, respectively. plemented with 10% fetal calf serum, 10 ng/mL epidermal growth Test compounds were placed in 96-well plates (EuroClone) and se- factor, 1 µg/mL hydrocortisone, 2 mM glutamine, 100 U/mL penicillin, rial dilutions were performed in a final volume of 100 µL/well. One 100 mg/mL streptomycin, and 20 mM HEPES buffer (pH 7.3). For the hundred micro-liters of asexual parasite or gametocyte cultures with toxicity experiments, HMEC-1 cells were plated in 96-well plates at 104 parasitemia of 1–1.5% were distributed into each well to achieve a final cells/100 µL/well and incubated at 37 °C and 5% CO2 overnight. Test volume of 200 µL and final hematocrit of 1%. The plates were put into compounds were dissolved in DMSO and diluted with medium to an incubating chamber and inflated with the proper gas mixture for achieve the required concentrations. One hundred µL/well of herbal 2 min before putting in the incubator at 37 °C for 72 h. products were added to the plate to reach a final volume of 200 µL/ The growth of the asexual stages was determined spectro- well. Cells were incubated for 72 h before measuring cell viability photometrically by measuring the activity of the parasite lactate de- by the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium- hydrogenase (pLDH), according to a modified version of Makler’s bromide) test, as described previously [20]. Cytotoxicity was expressed method [19]. Briefly, the drug-treated culture was re-suspended and as the 50% inhibitory concentration (IC50). Three independent experi- 20 µL/well were transferred into a new plate containing 100 µL of ments were performed in duplicate wells. Malstat reagent (0.11% [vol/vol] Triton-100, 115.7 mM lithium L-lactate, 30.27 mM Tris, 0.62 mM 3-acetylpyridine adenine dinucleotide 3. Results and discussion [APAD] [Sigma-Aldrich], adjusted to pH 9 with 1 M HCl) and 25 µL of PES/NBT (1.96 mM nitroblue tetrazolium chloride and 0.24 mM phe- 3.1. Bioassay-guided purification of L. lanceolata extract for activity nazine ethosulfate). The plate was incubated in the dark at room tem- against P. falciparum asexual blood stages and mature gametocytes perature (20 °C) for 15 min and then read at a wavelength of 650 nm using a microplate reader Synergy4 (BioTek). Results were analyzed A panel of twelve medicinal plants, selected among those commonly and expressed as the 50% inhibitory concentration in respect to control used as traditional antimalarial remedies in Burkina Faso, has been wells (IC ). 50 preliminarily screened for activity against Plasmodium gametocytes. For the determination of gametocytes viability, the luciferase assay Different plant parts were extracted with ethanol and water andthe was used as described previously [18]. Briefly, at the end of the in- percentage of gametocyte viability was evaluated at the fixed dose of cubation, 100 µL of supernatant were discarded from each well of the 100 µg/mL (see Table S1 in Supporting Information) using the lumi- experimental plate. The remaining culture at 2% hematocrit was then nescent assay. This preliminary screening allowed to select L. lanceolata re-suspended and 70 µL were transferred into a black plate. The luci- stem barks as the most promising plant species and part, worthy of ferase substrate D-luciferin (1 mM in 0.1 M citrate buffer, pH 5.5) was being further investigated. In fact, L. lanceolata stem bark ethanolic added in each well at the 1:1 vol ratio and the plate was incubated in extract reduced gametocyte viability by more than 90% at a dose of the dark for 10 min. The luminescence signal was read using a lumi- 100 µg/mL. nescence reader Synergy4 (BioTek) with integration time 500 ms. A larger amount of extract was therefore prepared with methanol The results of the chemo-sensitivity assays were expressed as the (characterized by lower cost but practically identical extraction effi- percent viability compared to the untreated controls, calculated with ciency compared to ethanol) and then subjected to a partitioning the following formula: 100 × ([Readout of treated sample- blank]/ scheme that eventually afforded hexane, ethyl acetate (EtOAc) and [Readout of untreated sample- blank]); Readout is OD (Optical Density) butanol phases. The in vitro activity of total extract and the three phases for asexual parasites and ALU (Arbitrary Luminescent Units) for ga- on the viability of P. falciparum asexual blood stages and stage V gametocytes. As blank, uninfected RBC and gametocytes treated with a metocytes was assessed using the pLDH and the luminescent assay, high dose of MB (300 ng/mL) were used for the asexual and gametocyte respectively (Table 1), as illustrated in the Experimental Section. The assay, respectively. The percentage of viability was plotted as a function results confirmed the activity of the total methanolic extract against of drug concentrations and the curve fitting was obtained by nonlinear stage V gametocytes, with an IC value of 7.74 µg/mL (Table 1), while regression analysis using a four-parameter logistic method (software 50 neither the methanolic extract nor any of the three phases showed Gen5 1.10 provided with the Synergy4 plate reader [Biotek]). The IC , 50 significant activity (IC ˃30 μg/mL) against 3D7 and W2 asexual which is the dose capable of inducing 50% inhibition of parasite via- 50 stages. Testing L. lanceolata butanol, hexane and EtOAc phases for ac- bility, was obtained by extrapolation. At least three experiments in tivity against stage V gametocytes revealed that the plant gametocy- duplicate were performed with the test extract/compounds against each tocidal activity was conserved in the EtOAc phase, displaying an IC of investigated parasite stage and strain. 50 13.38 μg/mL (Table 1). Based on these results, the EtOAc phase was selected for further fractionation studies. 2.6. In vitro mammalian cell toxicity assay Chromatographic purification of the EtOAc phase afforded tensub- fractions, that were then tested on asexual stage parasites and stage V Cytotoxicity was evaluated in human dermal microvascular gametocytes (see Table S2 in Supporting Information). As expected,

Table 1 In vitro antiplasmodial activity of the crude extract and phases of L. lanceolata stem barks against P. falciparum asexual blood stages of the CQ-sensitive 3D7 and CQ- resistant W2 strain and on stage V gametocytes of the 3D7elo1CBG99 strain.

Crude extracts/Phases 3D7 asexual stage IC50 (µg/mL) W2 asexual stage IC50 (µg/mL) 3D7elo1CBG99 stage V gametocyte IC50 (µg/mL)

Methanol extract 38.18 ± 5.6 30.33 ± 1.7 7.74 ± 3.45 Butanol phase 35.59 ± 3.64 33.92 ± 3.00 30.80 ± 7.55 Ethyl Acetate phase 39.77 ± 7.97 33.49 ± 8.39 13.38 ± 2.48 Hexane phase 44.82 ± 5.42 49.44 ± 14.84 54.76 ± 13.49 Chloroquine 0.005 ± 0.002 0.178 ± 0.056 NT Methylene blue NT NT 0.020 ± 0.003

IC50 values are given as the mean ± SD from at least three independent dose-response experiments conducted in duplicate or triplicate wells. Chloroquine was used as positive control against asexual parasites and methylene blue against gametocytes. NT = not tested.

3 A. Lopatriello, et al. Bioorganic Chemistry 93 (2019) 103321

Table 2 sub-fractions were therefore selected for isolation and identification of 1H (700 MHz) and 13C (175 MHz) NMR data of α′-chlorolophirone E (5) and 5′- the bioactive compounds and characterization of their activity on P. chlorolophirone D (6) in (CD3)2CO. falciparum stage V gametocytes. Pos. 5 6

δC, mult. δH, mult., J in Hz δC, mult. δH, mult., J in Hz 3.2. Chemical characterization

1 130.7, C 128.4, C Extensive HPLC purification on normal and reverse-phase columns 2 120.9, CH 8.12, s 121.9, CH 8.62, s 3 118.8, C 131.5, C of the bioactive fractions LL2, LL3 and LL4 afforded seven compounds 4 156.9, C 154.1, C (1–7) in the pure form, whose chemical structures were fully char- 5 112.6, CH 7.68, d, 8.5 111.6, CH 7.73, d, 8.7 acterized by means of NMR and MS analysis. Lophirones A (1) [21],F 6 125.9, CH 7.94, d, 8.5 126.2, CH 8.03, d, 8.7 (2) [22],E(3) [15],D(4) [15], and C (7) [23] were identified on the α 119.8, CH 8.07, overl. 120.8, CH 8.12, d, 16.5 β 145.7, CH 8.05, overl. 143.9, CH 8.08, d, 16.5 basis of the comparison of their spectral data with those reported in the CO 193.7, C 191.2, C literature. HPLC purifications also afforded small amounts of twonew 1′ 114.3, C 116.0, C chlorinated lophirone derivatives, α′-chlorolophirone E (5, 0.9 mg) and 2′ 167.9, C 165.1, C 5′-chlorolophirone D (6, 1.0 mg) (Fig. 1). 3′ 104.6, CH 6.40, bs 108.3, CH 6.59, bs 1D NMR data of compound 5,C H ClO by HR-ESIMS, closely 4′ 167.2, C 160.2, C 23 15 5 5′ 111.2, CH 6.51, dd, 9.0, 2.1 117.3, C paralleled those of lophirone E (C23H16O5)(3), and this greatly helped 6′ 133.3, CH 8.22, d, 9.0 132.3, CH 8.42, s its structural assignment. An evident difference between the two series α' 115.5, C 114.0, s of spectra was lack of the furan proton, likely replaced by a chlorine β' 153.1, C 165.2, C atom. This was also suggested by the concomitant shift of neighboring 1″ 122.8, C 119.2, C 2″-6″ 127.7, CH 8.04, d, 8.6 129.7, CH 7.94, d, 8.3 protons (Fig. 2). Placement of the chlorine atom at position α′ was 3″-5″ 116.9, CH 7.06, d, 8.6 116.1, CH 7.13, d, 8.3 further confirmed by the 2D NMR HMBC correlation of H-2 withthe 4″ 159.8, C 160.8, C relatively deshielded unprotonated carbon α′. Complete assignment of eCHO 187.5, CH 10.39, s NMR data of 5 is reported in Table 2.

Similarly, compound 6,C24H15ClO6 by HR-ESIMS, was readily identified as a chlorinated analogue of lophirone D(C24H16O6)(4) by none of them showed significant activity against asexual parasites, inspection of 1D NMR data. The most significant changes of this spec- while sub-fractions LL2-LL4 significantly inhibited the viability of stage trum, compared to lophirone D, were the lack of the doublet signal at δH V gametocytes with IC ranging from 0.79 to 7.94 μg/mL. These three 50 6.50 (H-5′) (Fig. 2) and the replacement of H-6′ doublet with a singlet at

Fig. 1. Chemical structures of compounds 1–7.

4 A. Lopatriello, et al. Bioorganic Chemistry 93 (2019) 103321

Fig. 2. left: comparison between 1H NMR spectra of α′-chlorolophirone E (A) and lophirone E (B), right: comparison between 1H NMR spectra of 5′-chlorolophirone D (C) and lophirone D (D).

Table 3 and C. Relatively more simple analogues, as lophirones D and E, likely In vitro antiplasmodial activity of pure compounds 1–7 against P. falciparum derive from oxidative cleavage on the structure of lophirone C. This asexual blood stage parasites of the CQ-sensitive 3D7 and CQ-resistant W2 family of metabolites has attracted the interest of the scientific com- strain and on stage V gametocytes of the 3D7elo1CBG99 strain. munity for its interesting bioactivities, such as antioxidant and anti- Compounds 3D7 asexual stage W2 asexual 3D7elo1CBG99 mutagenic activities, with protection of liver cells by upregulation of IC50 (µM) stage IC50 (µM) stage V gametocyte redox transcription factors [24]. Not surprisingly, total syntheses have IC50 (µM) been reported for several lophirones, and, among them, the very recent syntheses of lophirones F [25] and H [26] represent noticeable ex- Lophirone A (1) > 50 > 50 49.4 ± 3.20 Lophirone F (2) 29.90 ± 3.94 26.74 ± 2.42 52.9 ± 1.69 amples. Lophirone E (3) 38.47 ± 3.21 12.23 ± 2.08 0.14 ± 0.04 Compounds 5 and 6 are the first chlorinated lophirones to be re- Lophirone D (4) 47.75 ± 4.75 45.47 ± 7.84 > 100 ported and among the very few examples of halogenated flavonoids. Compound 5 > 50 17.88 ± 5.03 4.11 ± 1.14 Data in the literature indicate that the halogenation of flavonoids can Compound 6 > 50 > 50 81.38 ± 9.56 Lophirone C (7) 14.23 ± 3.27 7.34 ± 0.78 8.07 ± 0.97 occur more as a result of microbial biotransformation than as a product Chloroquine 0.0259 ± 0.0059 0.608 ± 0.22 NT of plant enzymes [27]. In our case, we did not use halogenated solvents Methylene blue NT NT 0.08 ± 0.039 for extraction, partitioning or chromatographic purification; however, a halogenated solvent (CDCl3) has been used to register NMR spectra of Data are the mean ± SD of IC50 values derived from at least three independent extracts, phases and sub-fractions. Thus, in principle, we cannot com- dose-response experiments in duplicate or triplicate wells. Chloroquine was pletely exclude that 5 and 6 are artifacts formed in the NMR tube. To used as positive control against asexual parasites and methylene blue against test this hypothesis, we have left lophirones D and E for 24 h in CHCl at gametocytes. NT = not tested. 3 room temperature but no formation of artifacts was observed. Re- gardless the not unambiguous natural origin, compounds 5 and 6 have 140 Lophirone E the advantage of increasing the chemodiversity within the small library Methylene Blue of metabolites obtained from the bioactive sub-fractions and were 120 therefore evaluated for their gametocytocidal activity and effects on 100 asexual blood stages.

80 3.3. Activity of pure compounds against mature gametocytes and asexual 60 blood stages % of control

40 The seven pure compounds (1–7) obtained from purification of bioactive subfractions were evaluated for their activity against asexual 20 stages using a chloroquine-resistant (W2) and a chloroquine-sensitive (3D7) strain (pLDH assay) and against stage V P. falciparum gameto-

0 cytes using the luminescent assay. The results, expressed as IC50, are 0.001 0.01 0.1 1 10 reported in Table 3. None of the tested compounds showed significant conc (µM) activity against asexual stages. Lophirone C (7) was the only molecule that displayed an IC value < 10 µM, with a higher potency against the Fig. 3. Dose-response curve of lophirone E (black circle) compared to methy- 50 lene blue (empty circle) on gametocytes of the 3D7elo1CBG99 strain. CQ-R strain. Remarkably, lophirone E (3) proved to be a very potent Representative experiment out of three. agent against stage V gametocytes with a nanomolar IC50 value (IC50 = 0.14 µM), comparable to that of methylene blue (MB). Fig. 3 shows the dose-response curve of lophirone E compared to methylene δ 8.42. The structure of compound 6 was further secured upon in- H blue. In addition, differently from MB [28], lophirone E exhibited a spection of the network of 2D HMBC cross-peaks, and, in particular, unique stage-specific activity, proving to be about 100 times moreac- correlations from both H-3′ and H-2′ to the chlorinated C-5′. Complete tive on stage V gametocytes than on asexual stages (IC50 = 12.2 and assignment of NMR data of 6 is reported in Table 2. 38.5 µM against D10 and W-2 strains, respectively). Lophirones are biflavonoids typically found in plants ofthe To further investigate the characteristics of the molecule, the cyto- Ochnaceae family, biogenetically arising from dimerization of chal- toxicity of lophirone E was assessed in vitro on the human microvascular cones with different regiochemistry, as exemplified by the different endothelial cell line HMEC-1. Dose-response experiments revealed the scaffolds of lophirone A (also deriving from an aryl shift), lophirones F molecule to be neglectably cytotoxic, as evidenced by an IC50 value of

5 A. Lopatriello, et al. Bioorganic Chemistry 93 (2019) 103321

79.7 µM, that is more than 500-fold higher than that determined against Venture MMV grant RD/18/0047. NMR spectra were recorded at “Centro stage V gametocytes (selectivity index = 570). di Servizio Interdipartimentale di Analisi Strumentale”, Università di The library of tested compounds is somewhat limited, but some Napoli Federico II. preliminary structure-activity relationships can nevertheless be drawn, taking advantage of the close similarity among some of them. Appendix A. Supplementary material Compounds 4 and 5 differ from lophirone E(3) only for the replacement of the furan hydrogen (position α') with a formyl group (in 4) and Additional data on antiplasmodial activity and 1H NMR spectra of a chlorine atom (in 5), respectively. This change causes a marked de- isolated compounds are available as Supplementary Material. crease of activity for 5 (40 times less active than 3) and a complete loss Supplementary data to this article can be found online at https://doi. of activity for 4, thus highlighting the crucial importance of a non- org/10.1016/j.bioorg.2019.103321. substituted furan ring. Lophirone C (7), characterized by a phenylk- etone substituent at α' and saturation of the furan double bond, ex- References hibited a moderate activity against stage V gametocytes

(IC50 = 8.1 µM); however, it exhibited an activity of the same magni- [1] World malaria report, (2018). www.who.int, (accessed on June 12, 2019). tude also against asexual stages. [2] C.L. Peatey, D. Leroy, D.L. Gardiner, K.R. Trenholme, Malar. J. 11 (2012) 34. [3] J.N. Burrows, R. Hooft Van Huijsduijnen, J.J. Möhrle, C. Oeuvray, T.N. Wells, Malar. J. 12 (2013) 1–20. 4. Conclusions [4] J.N. Burrows, E. Burlot, B. Campo, S. Cherbuin, S. Jeanneret, D. Leroy, T. Spangenberg, D. Waterson, T.N. Wells, P. Willis, Parasitol. 141 (2014) 128–139. [5] WWARN Gametocyte Study Group, BMC Med 14 (2016) 79. Bioassay-guided fractionation of organic extract obtained from stem [6] N.J. White, L.G. Qiao, G. Qi, L. Luzzatto, Malar. J. 11 (2012) 418. barks of the African antimalarial plant L. lanceolata has led to the [7] S. Tapanelli, G. Chianese, L. Lucantoni, R.S. Yerbanga, A. Habluetzel, identification of the bichalcone lophirone E as a potent, stage specific O. Taglialatela-Scafati, Fitoterapia 114 (2016) 122–126. gametocytocidal agent with an IC value in the nanomolar range. [8] S.M. Abay, L. Lucantoni, N. Dahiya, G. Dori, E.G. Dembo, F. Esposito, G. Lupidi, 50 S. Ogboi, R.K. Ouédraogo, A. Sinisi, O. Taglialatela-Scafati, R.S. Yerbanga, Lophirone E is about 100 times more active against P. falciparum stage M. Bramucci, L. Quassinti, J.B. Ouédraogo, G. Christophides, A. Habluetzel, Malar. V gametocytes (IC50 = 0.14 µM) than on asexual blood stages J. 14 (2015) 288. (IC = 12 µM on W2 strain). Moreover, the molecule displays negli- [9] C. Sirignano, A. Snene, D. Rigano, S. Tapanelli, C. Formisano, P. Luciano, R. El 50 Mokni, S. Hammami, A.R. Tenoh, A. Habluetzel, O. Taglialatela-Scafati, J. Nat. gible cytotoxicity (SI = 570). The isolation of structural analogues has Prod. 80 (2017) 2787–2794. also allowed to identify the critical positions on the chemical scaffold of [10] J. Achan, A.O. Talisuna, A. Erhart, A. Yeka, J.K. Tibenderana, F.N. Baliraine, lophirone E. P.J. Rosenthal, U. D'Alessandro, Malar. J. 10 (2011) 144. [11] P.J. Weathers, P.R. Arsenault, P.S. Covello, A. Mcmickle, K.H. Teoh, D.W. Reed, The existence of strict structure-activity relationships points toward Phytochem. Rev. 10 (2011) 173–183. the existence of a defined molecular target, whose identification would [12] L.P. Ouattara, S. Sanon, V. Maihou-Leddet, A. Gansané, Parasitol. Res. 113 (2014) be of crucial importance. The chemical structure of lophirone E pos- 405–416. [13] D.E. Pegnyemb, B.B. Messanga, R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, sesses several features (MW < 500, relative simplicity, absence of B. Bodo, J. Nat. Prod. 61 (1998) 801–803. stereogenic carbons) that could facilitate a straightforward total [14] R. Ghogomu Tih, A.E. Tih, B.L. Sondengam, M.T. Martin, B. Bodo, J. Nat. Prod. 57 synthesis, thus overcoming to resort to plant material as the unique (1994) 142–145. [15] R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, B. Bodo, J. Nat. Prod. 52 (1989) source. A synthetic approach to lophirone E could also allow the pre- 284–288. paration of analogues aiming at further improving the potency and [16] D.E. Pegnyemb, R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, B. Bodo, J. Nat. optimizing the druggability parameters of the compound. Prod. 57 (1994) 1275–1278. Hopefully, in the near future we will see that another plant-derived [17] W. Trager, J.B. Jensen, Science 193 (1976) 673–675. [18] S. D'Alessandro, G. Camarda, Y. Corbett, G. Siciliano, S. Parapini, L. Cevenini, compound will be available as a drug and included as transmission E. Michelini, A. Roda, D. Leroy, D. Taramelli, P. Alano, J. Antimicrob. Chemother. blocking agent in drug combinations that will contribute to malaria 71 (2016) 1148–1158. control and elimination efforts. [19] M.T. Makler, J.M. Ries, J.A. Williams, J.E. Bancroft, R.C. Piper, B.L. Gibbins, D.J. Hinrichs, Am. J. Trop. Med. Hyg. 48 (1993) 739–741. [20] S. D'Alessandro, M. Gelati, N. Basilico, E.A. Parati, R.K. Haynes, D. Taramelli, Declaration of Competing Interest Toxicology 241 (2007) 66–74. [21] R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, B. Bodo, Tetrahedron Lett. 28 (1987) 2967–2968. The authors declare that they have no conflict of interests. [22] R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, B. Bodo, Phytochemistry 29 (1990) 2289–2293. Acknowledgements [23] R. Ghogomu Tih, B.L. Sondengam, M.T. Martin, B. Bodo, Phytochemistry 28 (1989) 1557–1559. [24] N.O. Aliyu, R.A. Ajala-Lawal, T.O. Ajiboye, J. Biochem. Mol. Toxicol. 32 (2018) This research work was supported by the Italian MIUR, PRIN2015, e22055. project 20154JRJPP “Towards multi-stage drugs to fight poverty related [25] H.M. Le, H.D. Mac, C. Ho Oh, D.T. Do, Eur. J. Org. Chem. 13 (2019) 2362–2367. [26] H. Abas, S.M. Linsdall, M. Mamboury, H.S. Rzepa, A.C. Spivey, Org. Lett. 19 (2017) and neglected parasitic diseases: synthetic and natural compounds di- 2486–2489. rected against Leishmania, Plasmodium and Schistosoma life stages and [27] H. Cao, X. Chen, A.R. Jassbi, J. Xiao. Biotechnol. Adv. 33 (2015) 214–223. assessment of their mechanisms of action“. H.S. was supported in part by [28] M.J. Delves, A. Ruecker, U. Straschil, J. Lelièvre, S. Marques, M.J. López-Barragán, the Italian Ministry of Foreign Affairs/International Cooperation (MAECI) E. Herreros, R.E. Sinden, Antimicrob. Agents Chemother. 57 (2013) 3268–3274. through the note MAE/DGCS/IV N°141706 and the Medicines for Malaria