Plant material and compounds isolation

P. hookeri was collected in Dschang, Western Region, Cameroon, in December 2012. A voucher specimen (No. 43234 HNC) was deposited at the National Herbarium of Yaoundé, Cameroon. At room temperature, the dried and whole powdered plant (8.30 kg) was extracted three times (20 L each) with a CH₂Cl₂/MeOH (1:1) mixture. After filtering the solution, the solvent was removed under vacuum, yielding 310.6 g of crude extract (a greenish solid). To obtain the evaluated compounds, 300 g of the crude extract was subjected to successive column chromatography (Tchamgoue et al., 2016). A part of this greenish solid (10.0 g) was extracted with n-hexane (100 mL), the insoluble portion was taken up in ethyl acetate and methanol (100 mL/each), respectively. The organic solvent-soluble fractions were lyophilized, weighed, and kept at 4°C until use (Melaku et al., 2017).

In vivo acute toxicity assay

The acute toxicity investigations followed OECD standards 425 (OECD, 2001). Female Swiss albino mice aged 8–12 weeks and weighing 23–29 g were procured from the Animal House of the College of Natural Sciences, Addis Ababa University. The mice were kept in typical lab conditions with free access to food and water. Twenty-four mice were separated into four groups of six and given a week to acclimate. Treatment and control mice were fasted three (03) hours before and two (02) hours after oral administration of crude extract and vehicle (only food was withheld, while water was given ad libitum). A dose of 500, 1500 or 2500 mg/kg body weight of crude extract was given orally. The control group received 0.28 mL of vehicle (3 percent tween 80). After treatment, animals were examined for mortality and toxicity for 30 minutes, every hour for 24 hours, and once daily for 14 days. On day 14, the surviving mice were weighed. The major indicators utilised to assess acute oral toxicity in mice were body weights and behavioural alterations (conditions to adopt humane endpoints). Every effort was made to avoid or minimise mouse death or severe pain and misery.

Malaria parasite and inoculation

The in vivo antiplasmodial assay used CQ sensitive P. berghei strain ANKA. To preserve strain variety, parasites were passed weekly from infected to non-infected mice. Each experiment used blood from donor mice with a parasitaemia of 30–40%, diluted in 0.9 percent physiological saline to contain 1 x 10⁷ P. berghei parasitized erythrocytes per 0.2 mL. We used male Swiss albino mice from Addis Ababa University’s Animal House. Each mouse received 0.2 mL of 1 x10⁷ P. berghei infected RBCs intraperitoneally (Fidock, Rosenthal, Croft, Brun, & Nwaka, 2004).

Figure 1

Structures of flavonoids isolated from P. hookeri and evaluated for antiplasmodial activity

In vivo antiplasmodial assay

Antiplasmodial efficacy of fractions and pure substances from the whole plant of P. hookeri (Figure 1) was assessed in vivo on early P. berghei infection using the method described by Peters et al., 1975. Sixty-three male Swiss albino mice (6–11 weeks, 20–27 grammes) were separated into treatment and control groups and given a week to acclimate. The groups used for the antiplasmodial assessment of compounds and fractions constituted three and five mice, respectively. The animals were then inoculated with 0.2 mL of 1 x10⁷ P. berghei infected red blood cells intraperitoneally. Therefore, treatment groups received daily dosages of 500 mg/kg of fractions and 15–35 mg/kg of compounds, based on the amount available and reported cut off or threshold values for in vivo compound testing. Chloroquine at a dose of 25 mg/kg and a vehicle at a dose of 1 mL/100 g (3 percent Tween 80) were used as positive and negative controls, respectively. Extracts and compounds were given intragastrically 3 hours after infection on day zero (D₀) and for 4 consecutive days. Five days later, thin smears of each mouse's tail blood were fixed in methanol and stained with 10% Giemsa. Five identical fields from the tailed region of each stained slide were viewed using a 100X oil immersion objective to assess each sample's percent suppression compared to the negative control group.Muluye, Melese, and Adinew (2015) computed % parasitaemia and percent suppression as follows:

Each mouse was watched to see how much weight they lost and how long they lived. On D₀ and D₄, their weights were recorded. The MST of the groups were computed as follows:

On D₀ and D₄, blood was drawn from each mouse's tail using heparinized capillary tubes. The tubes were filled to three-quarters of their volume with blood and sealed with sealing clay. The tubes were then centrifuged for 5 minutes with the sealed end outwards. The following formula was used to determine the PCV (Amelo, Nagpal, & Makonnen, 2014).

The mice showed no evidence of toxicity at the end of the experiment, so they were left to die naturally.

LC-MS analysis

The EtOAc fraction was diluted in HPLC grade methanol (5 mg/ml) and filtered. Five microlitre aliquots were injected into the LC–DAD/MS Dionex Ultimate 3000 HPLC (Germany) for analysis. HRESI-QTOF Spectrometer (Bruker, Germany) acquired high-resolution mass spectra with a UV–vis absorbance detector. Using Na Formate as a calibrant, the spectrometer was set to the positive mode of operation (mass range: 100-1500, scan rate: 1.00 Hz). Positive ionisation electrospray mass spectra were collected simultaneously. The spray voltage was 4.5 kV, the capillary temperature was 200 °C, and the sheath gas was nitrogen (10 l/min). UV spectra between 190 and 600 nm were recorded using a UV spectrometer. The spectrometer was connected to a Thermo Fisher Ultimate 3000 HPLC system equipped with an LC-pump, an autosampler (injection volume 5 μl), and a column oven (35.0 °C). Separation of solutes was accomplished using a Synergi MAX-RP 100A (50 x 2 mm, 2.5 m particle size) gradient of H₂O (A)/acetonitrile (B) (flow rate 500 L/min). The following program was used to analyse samples: for 1.5 min, 95% of A was used, and the polarity was gradually increased to 100% of B. The system recovered to its initial condition (90 percent A) and was equilibrated for 1 minute. All compounds were identified using HPLC–MS and comparing UV and HRMS data from sample peaks to pure compound data.

Statistical analysis

SPSS version 20 was used to analyse data, expressed as mean followed by the standard error of the mean. The Tukey post hoc test and paired sample t-test (two-tailed) were used to compare the means of measured parameters. We considered P values equal to or less than 0.05 as statistically significant. The experimental groups were not excluded.

The yield of extraction and structural characterization of compounds

The highest yield of extraction was obtained with methanol (3.51 g), followed by n-hexane (2.47 g) and ethyl acetate (1.65 g). The assessed flavonoids (Figure 1) were previously isolated from the CH₂Cl₂/MeOH (1:1) crude extract of the whole plant of P. hookeri employing repeated silica gel and Sephadex LH 20 column chromatography. Compounds 6 and 7 resulted from chemical transformations on 6-prenylpinocembrin (8) (Tchamgoue et al., 2016). The structures of the compounds were determined using their spectroscopic data.

Metabolite profiling of the ethyl acetate fraction

Chromatographic profiling of the EtOAc fraction using liquid chromatography-mass spectrometry (LC-MS) led to the identification of nine flavonoids, of which compounds 8 (Rt 4.932 min) and 11 (Rt 4.898 min) were the major components (Figure 2). Compounds 1, 2, 4, 5 and 10 stood out clearly on the chromatogram at retention times of 3.566, 3.771, 3.600, 5.155 and 4.386 min, respectively, whereas compounds 9 (Rt 3.634 min) and 3 (Rt 4.215 min) had low peak intensities.

Figure 2

LC-MS chromatogram of the ethylacetate fraction from P. hookeri

In vivo acute toxicity test

The crude extract's acute oral toxicity was determined, and no toxic symptoms or death were found in any experimental animals. The rats survived for more than 14 days following oral administration of a single dosage of the CH₂Cl₂/MeOH crude extract of the entire plant (up to 2500 mg/kg body weight). We investigated physical and behavioural changes in the animals and found no significant differences between the vehicle (Tween 80) and extract-treated groups.

Effect of fractions and compounds on parasitaemia and mean survival time

The 4-day suppressive test findings showed that Hex and EtOAc fractions had significant in vivo antiplasmodial activity Table 1, but the MeOH fraction had no activity at the tested dose. The fractions, as mentioned above, reduced parasite burden significantly (p ≤ 0.05) at the tested dose compared to the negative control. Moreover, the EtOAc fraction considerably extended the mean survival time (p ≤ 0.05) of infected mice compared to non-treated mice, supporting its antiplasmodial effect Table 1. Chemosuppression of 97.76 percent was observed with chloroquine (the standard medication), higher than the tested fractions.

The assessed flavonoids exhibited mild to significant in vivo antiplasmodial activity against the chloroquine-sensitive strain of P. berghei with % parasitaemia suppression ranging from 13 to 85% Table 1. As per the literature, any compound is considered active when percent suppression in parasitaemia is 30% or more (Muluye et al., 2015). Thus, eight (1–3, 5–8 and 11) out of the eleven assessed compounds showed chemosuppressive activity against P. berghei with percent suppression in parasitaemia greater than 30% by the end of the four days dosage period. Parasitaemia reduction of these active compounds was statistically significant (p ≤ 0.05) compared to their respective negative controls. Moreover, compounds 1 (15 mg/kg/day) and 11 (35 mg/kg/day) exhibited the highest percentage suppressions, these are, 85.01% and 70.39%, respectively. Compound 11 dose-dependently reduced the parasite load of the infected mice. Comparison of survival rates Table 1 between groups revealed a significant effect (p ≤ 0.05) of compounds 1, 6 and 11 on suival time of P. berghei infected mice. Almost all flavonoid-treated mice lived longer than the respective negative controls. However, the chloroquine treated mice displayed the highest mean survival time.

Effect of fractions and compounds on PCV and body weights

A significant (p ≤ 0.05) body weight loss was observed between D₀ and D₄ in mice treated with MeOH fraction and the vehicle (3% tween 80) (Table 2). At the same time, no bodyweight reduction was perceived in mice treated with Hex and EtOAc fractions. The effect of the EtOAc fraction in preventing body weight loss of P. berghei infected mice was comparable to that of chloroquine and might be due to the observed activity of this fraction. Analysis of PCV change between D₀ and D₄ showed that the EtOAc fraction of P. hookeri significantly reduced PCV compared to the negative control. Neither body weight loss nor PCV reduction was observed in the chloroquine treated group.

Unlike compounds 4, 9, 10 and 11 (at lower doses), all the tested flavonoids prevented body weight loss associated with the infection (Table 2). The observed effects of these compounds in preventing body weight loss are in accord with their respective in vivo antiplasmodial activities. Compounds 1 and 11 at a single dose of 15 and 35 mg/kg, respectively, significantly prevented parasite-induced body weight reduction in infected mice compared to the vehicle-treated mice. The latter caused a significant attenuation of body weight loss of infected mice in a dose-dependent manner. Furthermore, all the active compounds (% suppression of parasitaemia ≥ 30%) reverted the reduction of PCV to a variable extent compared to the vehicle treated mice, which showed significant PCV reduction between D₀ and D₄. Further analysis of the percent PCV change revealed no PCV reduction in infected mice treated with compounds 1, 6 and 11 at the doses of 15, 20, and 35 mg/kg, respectively.