Comparative Study on the Phytometabolites, in vitro Antiplasmodial Activity and Cytotoxicity of Stem Bark Extracts of Annickia

The similar-looking Annickia affinis and Annickia chlorantha are two closely related species of the genus Annickia that are difficult to tell apart. Literature to date has jointly referred to both as Enantia chlorantha. Amongst the many pharmacologic benefits ascribed to E. chlorantha, is its usefulness in the management of malaria. Given the closeness of both species, there is the likelihood of swapping one for another in the market of herbal drugs. This study thus set out to compare the phytometabolites, antiplasmodial activity and cytotoxicity of both species. E. chlorantha is known to be dominated by protoberberine alkaloids which are thought to confer it with its antiplasmodial property. Indeed, TLC metabolite fingerprinting of methanol and water stem bark extracts of both species revealed the presence of protoberberine alkaloids. However, these alkaloids were more abundant in A. affinis as compared to A. chlorantha. This was corroborated by the stark >2.82 fold difference in antiplasmodial IC 50 of the respective methanol extracts being <0.78 µg/ml for A. affinis versus 2.2 µg/ml for A.chlorantha. Interestingly the selectivity indices for A. affinis (22.7) and A. chlorantha (24.3) determined using human hepatoma cell line were only marginally different. In conclusion, our study suggests that although both plants may be useful for managing malaria, consumers may get better and faster antimalarial relief with A. affinis.

. However, their great phenotypic similarity makes it difficult for herbal practitioners, their patrons as well as researchers to tell which is which (Olivier et al., 2015). Hence, there is likelihood of both plants being swapped for one another in the market of herbal drugs. Alternatively, both plants can be distinguished with confidence by DNA barcoding (Versteegh and Sosef, 2007). Besides, there are some salient features in their leaf and stembark which could be used to differentiate them (Versteegh and Sosef, 2007;Olivier et al., 2015). Thus, while A. chlorantha has a rough, grey bark, A. affinis possesses a smooth and greyish-brown bark (Figure 1) (Versteegh and Sosef, 2007). Further, while the simple hair under the leaf on the midrib of A. affinis point towards the apex, in A. chlorantha, hair on lower leaf surface is either simple, bifid, trifid or stellate and point in all directions (Versteegh and Sosef, 2007;Olivier et al., 2015). They can also be distinguished by the colour of their dry leaves since while the upper surface is brown to grey-green in A. chlorantha, it is grey-brown to almost black in A. affinis (Olivier et al., 2015). These were the criteria that were used by the botanist to authenticate A. affinis and A. chlorantha. Despite these differences between the two species, they are jointly referred to by a rather illegitimate name as E. chlorantha; a name substituted to Annickia in honour of Annick Le Thomas who previously revised the genus (Olivier et al., 2015). E. chlorantha is commonly known as African yellow wood. In Nigerian languages, it is known as Awopa, Osu pupa or Dokitaigbo (Yoruba), Osomolu (Ikale), Kakerim (Boki) and Erenba-vbogo (Edo). Given the differing reports of its antimalarial activity in publications (Agbaje and Onabanjo, 1991;Kimbi and Fagbenro-Beyioku, 1996;Adesokan and Akanji, 2010), this study set out to compare the phytometabolites, antiplasmodial activity and mammalian cell cytotoxicity of both species in vitro.  Water extracts of both plants were prepared by boiling 7 g/L of each plant in water for 1 hr at 100 o C as instructed by the herbal practitioner. Extracts were thereafter cooled and filtered through a 0.45-micron syringe filter after which they were frozen and lyophilized to obtain powdered extracts.
TLC Metabolite fingerprint of extracts This was done using pre-coated aluminium Merck kieselgel 60 F254 silica plates. Briefly, 0.1 to 2 μl of samples (25 mg/ml in methanol for methanol extracts or water for water extracts) was loaded on the origin of the TLC plates and allowed to dry. The sample loaded plates were placed in a pre-saturated chromatographic tank containing developing solvent of choice. The chamber was tightly closed, and the solvent allowed to run till about 90% of the plate length. Plate was thereafter taken out, solvent front marked with a pencil, allowed to dry and visualized using different methods including visible light, short UV and long UV (Erhunse et al., 2023). For 2D TLC, after the first run with the first solvent system, the plates were air-dried, turned 90 degrees and placed in a chromatographic tank containing the second solvent system and solvent allowed to run in a direction perpendicular to the first dimension run.
In vitro antiplasmodial study The SYBR Green I fluorescence-based assay was used for the antiplasmodial testing of samples (Smilkstein et al., 2004). Stocks (25 mg/ml) were prepared in either autoclaved water (water extracts) or DMSO (methanol extracts). The malaria parasite was cultured according to the method described by Trager and Jensen, (1976). The culture was synchronised at ring stage with 5% sorbitol and then exposed at 1% Parasitemia and 2% hematocrit to the extracts at various concentrations (0, 0.78, 1.56 and 3.13 µg/ml) in complete medium comprising of 16.2 g/L RPMI 1640 powder, 0.2% sodium bicarbonate, 0.5% Albumax II, 50 mg/L hypoxanthine, and 10 mg/L gentamicin, in fresh O + erythrocytes. Incubation was done using a mixed gas system (5% O2, 5% CO2, and 95% N2) for a total duration of 48 hr. at 37 o C. Thereafter, 100 μl of SYBR Green I solution {0.2 μl of 10,000 × of SYBR Green I (Invitrogen) per ml of lysis buffer} was dispensed into each well to stain the parasite's DNA in the lysed red blood cells. The 96well plate was then wrapped in foil and kept at 37 o C in an incubator for 1 hr. Lysis buffer was prepared by adding 20 mM Tris (pH 7.5), 5 mM EDTA (pH 8.0), 0.008% (w/v) saponin and 0.08% (v/v) Triton X-100 into Milli-Q grade water. Fluorescence was measured at 485 nm excitation and at 530 nm emission and the 50% inhibitory concentrations (IC50) were determined using the Antimalarial IC estimator version 1.2 software (Le Nagard et al., 2010).
In vitro cytotoxicity testing HUH cells were cultured using Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 5% foetal bovine serum until full confluence was achieved. Thereafter, they were trypsinized, seeded in a 96 well plate (10 4 cells/100 μl/well) and incubated at 37 o C and 5% CO2 for 12 hr. Next, spent media (90 μl) was replaced with fresh cDMEM (86 µl) followed by the addition of 4 µl extracts and plates were taken for a 24-hr incubation at 37 o C in a 5% CO2 incubator. DMSO (0.4% and 10%) were used as negative (full growth) and positive (zero growth) controls respectively. Post incubation, 20 μl of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) prepared in PBS solution (5 mg/ml) was added to each well. After a 3 hr, 37 o C incubation, media (120 μl corresponding to the amount of MTT (20 µl) plus 100 μl of extracts in DMSO/cDMEM) was aspirated out and 200 μl DMSO/well was added and solution mixed thoroughly to solubilize the formazan product. Absorbance was measured at 570 nm on a multi-well plate reader (Mosmann, 1983) and CC50 values were determined using the Antimalarial ICestimator version 1.2 software (Le Nagard et al., 2010).

TLC fingerprinting of test samples
The antiplasmodial activity of Annickia chlorantha has been linked to the presence of protoberberine alkaloids (Vennerstrom and Klayman, 1988;Imieje et al., 2017) which are berberinelike alkaloids that possess a number of pharmacological properties including antiplasmodial activity. Members of this subgroup include, berberine (the most common protoberberine alkaloid), palmatine, jatrorrhizine and columbamine. 1D and 2D TLC metabolite fingerprinting of extracts of both plants suggested that A. affinis has higher amounts of the protoberberine alkaloids (the green fluorescing spot under long UV) than A. chlorantha (Figures 2 and 3).

In vitro antiplasmodial activity and cytotoxicity of extracts
The antiplasmodial activity of the extracts were tested against different strains of the parasite. Both aqueous and alcoholic extracts of A.affinis showed better antiplasmodial activity (IC50 <0.78 -1.7 µg/ml) than the corresponding extracts from A.chlorantha (IC50 2.08 -> 3.33 µg/ml) against all strains of the parasite studied (Table 1). Further, both extracts were quite selective (SI = >12) for the parasite as CC50 HUH ranged from 10 to >100 µg/ml. Aa meOH= methanol extract of Annickia affinis; Ac meOH = methanol extract of Annickia chlorantha, Ac aq. = aqueous extract of Annickia chlorantha; Aa aq. = aqueous extract of Annickia affinis; Nd = not done

Discussion and Conclusion
Annickia affinis and Annickia chlorantha are closely related look alike species making it difficult to identify one from the other. This is the first study which compares the metabolite fingerprints and antiplasmodial activity of these two plants. Plant samples are said to be highly active against the parasite if IC50 < 5 µg/ml, promising if IC50 is between 5 -15 µg/ml, moderately active if IC50 is between 15 -50 µg/ml and inactive if IC50 > 50 µg/ml (Kraft et al., 2003). Further, a selectivity index (SI) < 10 indicates nonspecificity of action (Peeters et al., 2022). Thus, with IC50 <0.78 to >3.13 µg/ml and SI >12, extracts of both plants were highly active against various strains of the malaria parasite as well as selective. This is consistent with reports of good antiplasmodial activity of Enantia chlorantha by other researchers (Vennerstrom and Klayman, 1988;Boyom et al., 2009;Imieje et al., 2017). However, in the present study we have found that A. affinis has much better antiplasmodial activity than A. chlorantha. Further the results of our study suggest that the antiplasmodial activity reported for E. chlorantha in the literature (IC50 < 1 µg/ml) may likely be that of A. affinis. Indeed, A. affinis is the more frequently occurring species (Olivier et al., 2015). However, since both plants are frequently confused, literature reports of E. chlorantha may refer to A.chlorantha or even other species of the Annickia genus (Olivier et al., 2015).
Although E. chlorantha is reported to contain protoberberine alkaloids that confer promising antiplasmodial activity in vitro (Vennerstrom and Klayman, 1988;Phillipson and Wright, 1991;Imieje et al., 2017), these alkaloids have not been proven to demonstrate antimalarial activity in vivo (Phillipson and Wright, 1991). Their limited oral bioavailability as substrates of Pglycoprotein (P-gp) is however established (Maeng et al., 2002;Tarabasz and Kukula-Koch, 2020). P-gp is an efflux pump which is expressed in a wide variety of tissues where this pump plays a role in the absorption, distribution and excretion of xenobiotics. Given the role of P-gp in influencing cellular concentrations of drugs, several methods including the use of inhibitors have been exploited to overcome P-gp mediated efflux (Dewanjee et al., 2017). P-gp inhibitors are classified into four generations with plant-derived P-gp inhibitors regarded as 4 th generation Pgp inhibitors (Dewanjee et al., 2017).
Against P. berghei (NK 65 strain), 400 mg/kg aqueous stembark extract administered via oral route showed an activity similar to chloroquine and caused 100% parasite clearance with 60% mortality after 17 days (Adesokan and Akanji, 2010). Apart from solvent of extraction, route of administration may have played a role in these differing reports since the bioactive agents (protoberberine alkaloids) are substrates of P-gp. For instance, it was recently reported that a hepta-herbal antimalarial Agbo-iba formula comprising of A. affinis and six other plants displayed a better antimalarial activity than A. affinis alone even though the latter had an over 50-fold better in vitro antiplasmodial activity than the herbal combo (Erhunse et al., 2023). It is also possible that one of either A. chlorantha or A. affinis contains P-gp inhibitor(s) or that one has a higher amount of P-gp inhibitor(s) than the other because some berberine containing plants have been reported to possess Pgp inhibitory properties thereby improving bioavailability (Stermitz et al., 2000;Ma et al., 2016). Hence, a study to compare the in vivo antimalarial activity of both plant extracts administered via various routes is advocated as this may guide their proper use.