The potential of anti-malarial compounds derived from African medicinal plants, part II: a pharmacological evaluation of non-alkaloids and non-terpenoids

Malaria is currently a public health concern in many countries in the world due to various factors which are not yet under check. Drug discovery projects targeting malaria often resort to natural sources in the search for lead compounds. A survey of the literature has led to a summary of the major findings regarding plant-derived compounds from African flora, which have shown anti-malarial/antiplasmodial activities, tested by in vitro and in vivo assays. Considerations have been given to compounds with activities ranging from “very active” to “weakly active”, leading to >500 chemical structures, mainly alkaloids, terpenoids, flavonoids, coumarins, phenolics, polyacetylenes, xanthones, quinones, steroids and lignans. However, only the compounds that showed anti-malarial activity, from “very active” to “moderately active”, are discussed in this review.


Background
Malaria is caused by protozoans of the genus Plasmodium (Plasmodium falciparum, Plasmodium malariae, Plasmodium ovale, and Plasmodium vivax) [1,2]. According to the World Health Organization (WHO), about half of the world's population is at risk of malaria and one to two million annual deaths (mostly among African children) can be attributed to malaria alone [3,4]. The causative agent is transmitted by the female Anopheles mosquito species, which has also developed resistance against insecticides, such as dichlorodiphenyltrichloroethane (DDT), and chemoprophylaxis has not often yielded the expected results [2]. Additionally, the diseasecausing protozoans have developed resistance against most of the drugs currently used to treat malaria. There is an urgent need to discover new active compounds. Nature, and particularly plants are a potential source of new anti-malarial drugs, since they contain a quantity of metabolites with a great variety of structures and pharmacological activities. Traditional preparations (with the use of macerations, extracts, steam baths, concoctions, and decoctions from plant materials) have been the main source of treatment of malaria in Africa [5] and other continents where the disease is endemic [6,7]. Thus, with failing treatment regimens, many research groups in Africa (African indigenous research groups and their foreign collaborators) have resorted to plant sources in the quest to expand the anti-malarial chemotherapeutic arsenal [1,8,9]. This effort has been motivated by the use of these plant materials in the treatment of malaria and fevers in African traditional medicine (ATM). The results from Africa and other continents have been quite promising and hence there has been a general call for the use of natural products as drugs or as sources of inspiration for the development of novel anti-malarials, in order to possibly avoid problems related to drug resistance [10][11][12].
It is believed that the next generation anti-malarials or the scaffolds necessary for their synthesis may be found in the plants currently used in ATM [13,14]. However, the last review on anti-malarial compounds from African flora dates back about ten years [13], with other reviews focusing on plant-screening campaigns in particular regions and/or countries in Africa  or on active compounds obtained by bioassay-guided fractionation efforts from given countries and/or regions, not covering an entire continent [36][37][38][39][40][41]. Even though natural products that are active against P. falciparum have been discussed in a number of review papers [1,[42][43][44][45][46][47][48], the goal has been to provide an coverage of the most promising anti-malarials from the entire African continent, by giving an overview of the most pertinent in vitro and in vivo screening results reported in the literature. The most successful anti-malarials in use to date have been derived from natural product sources (quinolones/artemisinins). It is indeed a glaring omission that the African continent, despite its rich ethno-pharmocological heritage, is yet to yield a significant contribution in this respect. Clearly, as a first step, a systematic review of the many traditional therapeutic options is needed and this review addresses an important issue in this aspect. In part I, the most promising alkaloids and terpenoids were presented [49], while in this part the most interesting findings for flavonoids, coumarins, phenolics, polyacetylates, xanthones, quinones, steroids, and lignans are shown. The last part of the work is essentially focused on the cheminformatic analysis of >500 natural products (NPs), derived from African medicinal plants, which have demonstrated from weak to very good in vitro anti-malarial activities, with a focus on molecular descriptors related to "drug-likeness", drug metabolism and pharmacokinetics (DMPK). The predicted properties of plant-derived anti-malarials are those related to drug absorption, distribution, metabolism, elimination, and toxicity (ADMET) based on in silico computed molecular descriptors.

Flavonoids
Several bioactive flavonoids have been derived from medicinal plants growing in Africa. Even though the molecular mechanism of action of anti-malarial activities of flavonoids is not fully elucidated, it is believed that flavonoids act by inhibiting the fatty acid biosynthesis (FAS II) of the parasite [53,54]. Some flavonoids have also been shown to inhibit the influx of L-glutamine and myoinositol into infected erythrocytes [55]. The active anti-malarial flavonoids are summarized in Table 1, while the chemical structures are shown in Figures 1, 2 and 3.
The chalcones bartericin A (16), and 4-hydroxylonchocarpin (17), stipulin (18) and kanzonol B (19) were isolated from the twigs of Dorstenia barteri (Moraceae) from Cameroon by Ngameni et al. [63]. These compounds were evaluated in culture against the W2 strain   and reported the isolation, characterization and antiplasmodial activities of the first 9-hydroxyhomoisoflavo noid (28), 2,3-dihydro-5-hydroxy-7-methoxy-2-phenylchromen-4-one (29), along with the antiplasmodial activities of some of chalconoids and a flavanone isolated along with it from the surface exudate of Polygonum senegalense [61]. The antiplasmodial and cytotoxic activities of flavonoids and arylbenzofuran derivatives from Morus mesozygia were investigated by Zelefack et al. [65]. This plant is used in treating many diseases, including malaria and fever. Fractionation of the methanolic extract of its stem bark led to the isolation and identification of two flavonoids: artocarpesin (31) and kushenol E (32), among other compounds (mulberrofuran F, bartericin A and 4-hydroxylonchocarpin). The methanolic extract and the isolated compounds were tested for antiplasmodial activity against the chloroquine-resistant FcB1 P. falciparum strain and cytotoxicity on MCF-7 human breast cancer cells. It was found that all compounds were active against the FcB1 strain of Plasmodium, with compounds 31, 32 and mulberrofuran F exhibiting particular potency (with the median inhibitory concentrations IC 50 = 2.5-2.6 μg mL −1 ).
The acetone extracts of the root bark and stem bark of Erythrina sacleuxii showed antiplasmodial activities against the D6 and W2 strains of P. falciparum. Further chromatographic separation of the acetone extract of the root bark by Andayi et al. afforded a new isoflavone, 7-hydroxy-4′methoxy-3′-prenylisoflavone, named 5-deoxy-3′-prenylbiochanin A (34) along with known isoflavonoids as the antiplasmodial principles [67]. Flavonoids and isoflavonoids isolated from the stem bark of E. sacleuxii were also tested and showed antiplasmodial activities.

Rotenoids
Milletia usaramensis ssp. usaramensis is a plant growing in East Africa, which is reported to contain anti-malarial flavonoids, particularly rotenoids [58]. Seven rotenoids have been reported from this species, including usararotenoid C (35), usararotenoid A (36), (+)-usararotenoid B (37), and (+)-12a-epi-millettosin (38). These compounds exhibited moderate to weak antiplasmodial activity against the D6 and W2 strains of P. falciparum. Yenesew et al. further established some structure-activity relationships. It was observed that rotenoids containing a prenyl unit or a 2,2-dimethylpyrano substituent were more potent than the non-prenylated rotenoid, e g, usararotenoid A. It was also reported that there is no significant activity for usararotenoid B, suggesting the importance of the carbonyl function at C-12 in usararotenoid A for the weak antiplasmodial activity observed.

Phenolics
Zofou et al. have isolated p-hydroxy-cinnamic acid (39), along with other compounds, atranorin, specicoside, 2β,3β,19α-trihydroxy-urs-12-20-en-28-oic acid, from the stem bark of Kigelia africana (Bignoniaceae), harvested from Cameroon and performed the drug interactions of the isolated compounds among themselves, as well as their combination effects with quinine and artemether [68]. The antiplasmodial activity and drug interactions were evaluated against the multidrug-resistant W2mef Usararotenoid C (35): (37): Figure 3 Promising anti-malarial retenoids from African medicinal plants. strain of P. falciparum. The results equally showed a slight synergistic effect between atranorin and 2β,3β,19αtrihydroxy-urs-12-20-en-28-oic acid (combination index, CI of 0.82) whereas the interaction between specicoside and p-hydroxycinnamic acid was instead antagonistic (CI of 2.67). All three compounds were shown to significantly act in synergy with some first line malaria drugs like artemether (CI of 0.42 to 0.71). More excitingly, none of these four compounds showed any significant sign of toxicity against the monkey kidney cell strains LLC-MK2 (selectivity index below 10). Compound 39 exhibited antiplasmodial activitity against the W2mef strain with an IC 50 value of 2.11 μg mL −1 [69]. The origins of the isolated anti-malarial/antiplasmodial phenolics are shown in Table 2, while the chemical structures are shown in Figure 4.
In an effort to identify a lead compound for antimalarial drug discovery, Kamkumo   fruits of Sorindeia juglandifolia (Anacardiaceae) from Mt Kalla in Cameroon and tested the isolated compounds in vitro against the P. falciparum W2, against field isolates of P. falciparum, and against the P. falciparum recombinant cysteine protease falcipain-2 [70,71]. The main end products of the activity-guided fractionation were 2,3,6-trihydroxy benzoic acid (40) and 2,3,6trihydroxy methyl benzoate (41). Overall, nine fractions tested against P. falciparum W2 and falcipain-2 were active, with IC 50 values of varying from 2.3 to 11.6 μg mL −1 for W2, and 1.1-21.9 μg mL −1 for falcipain-2. Purified compounds (40 and 41) also showed inhibitory effects against P. falciparum W2 (IC 50  42.2 mg kg −1 , respectively. Active fractions were found to be safe in mice after oral administration of 7 g kg −1 body weight. These results suggest that further investigation of the anti-malarial activities of natural products from S. juglandifolia will be appropriate. The Ethiopian medicinal plant Combretum molle (Combretaceae), reported to possess genuine anti-malarial activity, was investigated by Asres et al. [72]. The fractionation of the stem bark extract yielded punicalagin (42) as the active compound. This compound exhibited in vitro activity against the 3D7 strain of P. falciparum with IC 50 values of 2.19 μg mL −1 . Ellagic acid (43), derived from the leaves of Alchornea cordifolia (Euphorbiaceae), showed good activity against P. berghei in mice with an ED 50 in the range of 0.2-0.151 μg mL −1 [73]. Cheplogoi et al. investigated the roots of Vepris uguenensis (Rutaceae), harvested from the Baringo District, Kenya [74]. Syringaldehyde (44) was identified as an active compound, exhibiting moderate antiplasmodial activity against two strains of P. falciparum, with IC 50 values of 13.0 μg mL −1 (chloroquinesusceptible 3D7 strain) and 21.4 μg mL −1 (chloroquineresistant FCM29 strain), respectively.
From the methanol extract of roots of Garcinia polyantha, Lannang et al. isolated Isoxanthochymol (45), which exhibited in vitro anti-malarial activity against P. falciparum and showed strong chemosuppression of parasitic growth [75]. The compound exhibited antimalarial activity with an IC 50 of 2.21 μM. This was lower than the IC 50 of the other five co-occurring compounds (garcinane, smeathxanthones A and B, chefouxanthone, isoxanthochymol, magnificol, and β-sitosterol and garciniaxanthone I), which ranged from 2.5 to 4.1 μM. The compounds were administered over a period of four days to the culture and the number of parasites was determined daily. Control experiments were performed either without treatment or with administration of 0.032 μM chloroquine in the same solvent. Gossypol (46), derived from the seeds of cotton plant (Gossypium sp., Malvaceae), exhibits a variety of biological activities, including antispermatogenic, anti-cancer, antiparasitic and antiviral activity. Deck et al. demonstrated that this compound also showed antimalarial activity against both chloroquine-sensitive and chloroquine-resistant strains of P. falciparum, with IC 50 values in the order of 10 μM [76]. The presence of aldehyde functional groups renders gossypol toxic and in the light of this fact, authors further investigated synthetic analogues of compound 46 for biological activity. It was found that the synthetic analogues lost toxicity while retaining antiplasmodial activity.

Quinones
Quinones also exhibit diverse pharmacological properties, including anti-malarial activity. Four quinones have been isolated from the root bark of Hoslundia opposita by Achenbach et al. treatment of malaria [78]. The isolation of these compounds was carried out as a result of an ethnomedical use of H. opposita in the treatment of malaria. The n-hexane extract root bark gave an IC 50 of 5.6 μg mL −1 and also exhibited a 26% inhibition of growth of P. berghei in mice, at a daily dose of 190 mg kg −1 body weight, for four days [78]. Only compound 51 was tested and showed significant in vitro activity against the multidrug-resistant K-1 strain and the chloroquine-sensitive NF54 strain of P. falciparum, with IC 50 values of 0.4 and 0.22 μg mL −1 , respectively. The other metabolites were not screened due to the limited amount available [78]. Cassia siamea (Fabaceae) was identified from southwest Nigerian ethnobotany as a remedy for febrile illness. This led to the bioassay-guided fractionation of stem bark of the plant extract, for assessing the in vitro anti-malarial activity. Emodin (55) and lupeol were isolated from the ethyl acetate fraction. Both compounds were found to be the active principles responsible for the antiplasmodial property with IC 50 values of 5 μg mL −1 respectively [79].
Makinde et al. investigated the action of extracts of the stem bark of Spathodea campanulata (Bignoniaceae) from Nigeria on Plasmodium berghei berghei in mice [82]. The blood schizontocidal activity of the extracts was studied in early and established infections using chloroquine as the reference drug. The prophylactic action of the extracts was also investigated with pyrimethamine as the standard drug. The hexane and chloroform extracts of the stem bark showed blood schizontocidal action in both the four-day test and Rane test. The chloroform extract demonstrated some prophylactic properties while the aqueous extract did not show any significant anti-malarial property. In addition, these authors were able to identify the active anti-malarial ingrediant to be lapachol (67). The other anti-malarial quinones identified were knipholone (68) and anthrone (69) from Kniphophia foliosa (Asphodelaceae) [83,84], as well as 2-(1-hydroxyethyl)naphtho[2,3-b]furan-4,9-dione (70) and isopinnatal (71)   (Bignoniaceae) [85]. These compounds were tested against chloroquine-sensitive (T9 − 96) and -resistant (K-1) P. falciparum strains and for cytotoxicity using KB cells. Compound 70 possessed good activity against both strains [IC 50 values 627 nM (K1), 718 nM (T9 − 96)]. Isopinnatal (71) and the co-occurring kigelinol and isokigelinol exhibited lower activity against both strains. Bringmann et al. also reported that knipholone (68) and three of its natural derivatives from the same plant, as well as seven structurally related but simplified compounds, have been examined for their antiplasmodial activity against asexual erythrocytic stages of two strains of P. falciparum in vitro (K1/chloroquine-resistant and NF 54/chloroquine-sensitive) [84]. All the phenylanthraquinones showed considerable activity, with only little cytotoxicity, while their anthraquinone and phenyl moieties were completely inactive.

Others
Lippialactone (105), derived from the ethyl acetate extract of aerial parts of Lippia javanicais, harvested from South Africa, was shown to be active against the chloroquinesensitive D10 strain of P. falciparum with an IC 50 value of 9.1 μg mL −1 , and is also mildly cytotoxic [98]. Helihumulone (106) was derived from extracts of the whole plant of Helichrysum cymosum (Asteraceae) from South Africa by Jakupovic et al. [99] and Vuuren et al. [100].
The dichloromethane extract of the leaves of Vernonia staehelinoides (Asteraceae) showed in vitro activity (IC 50 3 μg mL -1 ) against the chloroquine-sensitive D10 and the chloroquine-resistant (K-1) strains of P. falciparum [101]. Pillay et al. further investigated the extract by bioassay-guided fractionation and two structurally related hirsutinolides displaying in vitro antiplasmodial activity (IC 50~0 .2 μg mL −1 against D10) were isolated. These were 8α-(2-methylacryloyloxy)-3-oxo-1-desoxy-1,2-dehydrohirsutinolide-13-O-acetate (107), and 8α-(5′-acetoxysenecioyloxy)-3-oxo-1-desoxy-1,2-dehydrohirsutinolide-13-O-acetate (108). These were found to be cytotoxic to mammalian Chinese hamster ovarian (CHO) cells at similar concentrations, but proved to be attractive scaffolds for structure-activity relationship studies. Two main privileged substructures, a 2(5H)furanone unit and a dihydrofuran-4-one unit, were identified as potential pharmacophores, which may be responsible for the observed biological activity. Mucochloric and mucobromic acids were selected as appropriate 2(5H)-furanone substructures and these were shown to have comparable activity against the D10 and superior activity against the K1 strains relative to the hirsutinolide natural product. Mucochloric and mucobromic acids (109 and 110) also showed selective cytotoxicity to the malaria parasites compared to mammalian (CHO) cells in vitro. The antiplasmodial data obtained with respect to these two acids suggest that the 2 (5H)-furanone substructure is a key pharmacophore in the observed antiplasmodial activity. The identification of antiplasmodial hirsutinolides from V. staehelinoides suggests that they may play a role in the medicinal properties of the plant, but their potential for the development of anti-malarial drugs is limited due to inherent cytotoxicity and lack of selectivity. The results did however lead to the identification of potential pharmacophores, a 2(5H)-furanone unit and a dihydrofuran-4-one unit.

Conclusions
In this review, an attempt has been made to summarise the main finding of several research groups engaged in the search for naturally occurring active principles from African medicinal plants against P. falciparum. With multiple resistance developed by the malaria parasite, the cry has been towards obtaining new effective drugs. Attempts to develop 'green pharmacies' for improved phytomedicines against malaria are being encouraged by some NGOs and governments as part of their efforts to control malaria [105]. Additionally, modern hit/lead discovery efforts for specific anti-malarial drug targets are being encouraged. The trend has been towards accelerating this process by employing computer-based methods such as docking, virtual screening, pharmacophore modelling and binding-free energy calculations for hit/lead identification and combinatorial design of novel inhibitors against known anti-malarial drug targets. The practice of virtual screening is beginning to occupy the centre of drug discovery efforts [106] and it has been verified that developing NP libraries containing readily available compounds for screening virtual hits could be highly useful [107]. The authors of this paper have been developing NP databases containing three-dimensional structures of compounds derived from plants used in ATM [108][109][110] and using computed molecular descriptors to attempt to predict the pharmacokinetic profiles of NPs [110][111][112]. Since the role of NPs in drug discovery cannot be overemphasized [111][112][113][114][115][116], efforts are aimed at providing tools for research groups engaged in anti-malarial drug discovery, beginning with NPs derived from African medicinal plants. This is aimed at cutting down the cost of drug discovery when computational and 'wet lab' approaches are combined [117,118]. The intention is to make the current collection of threedimensional structures of naturally occurring antimalarials from African medicinal plants available for virtual screening. This shall be the scope of part III of this series.