The Antiplasmodial Potential of Medicinal Plants Used in the Cameroonian Pharmacopoeia: An Updated Systematic Review and Meta-Analysis

Malaria is a real public health problem. It is the leading cause of morbidity and mortality in the world. Research in herbal medicine has so far shown that the use of plants against malaria is not to be neglected. This review aims to highlight the antiplasmodial potential of Cameroonian plants. In order to achieve this objective, we conducted a bibliographic search in April 2022 using the PubMed search engine. This research included both the published and unpublished studies. A narrative approach was used to describe the antiplasmodial potential of the various species of plants investigated. Quantitative data were analyzed using R studio 4.1.1 software and random effects model was used to estimate the effect size. The research of the antiplasmodial activity of Cameroonian plants dates back to 2000. This area of research has since provided extensive data to indicate the antiplasmodial potential of several plants, most of which originate from the central region. Despite the heterogeneity observed between the different plant families studied in Cameroon for their in vitro antiplasmodial effect, there is strong evidence that 17 active compounds from these plants would be ideal candidates for the synthesis of new antimalarial drugs. The Dacryodes edulis species could be considered as the best natural alternative in the treatment of uncomplicated malaria according to its properties. It is clear that the traditional Cameroonian pharmacopoeia has many species that contain compounds with antiplasmodial activity. More studies need to be conducted to explore the multitude of unexplored plants that are used in traditional medicine. These studies should take into account the nature of the cell model used for cytotoxicity assessment.


Introduction
Malaria remains a global public health problem with about 228 million cases worldwide and 213 million cases (93%) recorded in Africa [1]. Multiple control strategies against this endemic, including vector control through the widespread use of long-lastinginsecticide-treated nets or indoor residual spraying on the one hand and chemoprevention on the other hand, have so far largely contributed to reduce the incidence of malaria in the world [2,3]. Unfortunately, these advances are constantly threatened by the emergence of resistance not only of the vectors to the insecticides used [4][5][6], but also of the parasite to the drugs. In the 1990s, the emergence of chloroquine resistance was associated with a dramatic increase in malaria mortality [7]. At the end of the last century, introduction of the artemisinin combination therapies (ACTs) provided a much needed, highly efficacious antimalarial treatment, which became the first-line treatment for uncomplicated falciparum malaria in all endemic countries [8]. e extremely rapid development of resistance to many antimalarials, and even the most recent, such as ACTs in five countries of the Greater Mekong subregion [9,10] and in Africa [11,12], justifies continued research on the factors causing this resistance. Like antibiotic resistance, antimalarial drug resistance is caused by the massive and uncontrolled use of certain molecules that could lead to a selection of resistant strains of Plasmodium over time. Diversification of effective antimalarial drugs would therefore be a solution to significantly reduce the rapid progression of resistance and thus the malaria-related mortality.
It has been highlighted that the richness of plant biodiversity and the knowledge of traditional therapies are likely to open new avenues for antimalarial therapy [13]. is was for example the case of quinine and artemisinin, which are the two currently prescribed antimalarials from medicinal plants, traditionally used in their country of origin against fevers and malaria. Quinine is from the bark of a tree from the flanks of the Andean cordillera (Cinchona calisaya and other species of Cinchona) [14] and artemisinin is from a herb native to China, Artemisia annua [15]. e search for new antimalarial drugs could therefore be undertaken within plant biodiversity using ethnopharmacology.
rough this approach, the potential antimalarial activity of plants could guide the scientific community towards more in-depth research. is review aims to highlight the antiplasmodial potential of the plants of the Cameroonian pharmacopeia while evaluating their ability to inhibit in vitro chloroquine-sensitive and chloroquine-resistant strains with the least cytotoxicity possible.

Methods
e proposed systematic review was conducted in accordance to the Cochrane Handbook [16] and PRISMA statement (i.e., Preferred Reporting Items for Systematic Reviews and Meta-analyses [17]). e following research question was formulated to address the literature and outline the search strategy: are Cameroonian plants species or family able to be more effective with low toxicity in vitro against Plasmodium resistant-chloroquine strains compare to sensitivechloroquine one? No publication year or language limit was considered.

Selection, Inclusion, and Exclusion Criteria.
Following the search, all identified citations were collected and uploaded into the Zotero software and duplicates were removed. Titles and abstracts were then screened by one reviewer for assessment against the inclusion criteria for the review. Potentially relevant studies were retrieved in full and their citation details were imported into the Rayyan software [18]. e full text of selected citations was assessed in detail against the inclusion criteria by one reviewer. Reasons for exclusion of full text studies that do not meet the inclusion criteria were recorded and reported in the systematic review. Review considered studies that included Cameroonian plants assessed for their in vitro antiplasmodial activity. Only primary studies assessing the in vitro 50% inhibitor concentration (IC 50 ) were included in the review. All review articles were excluded. e results of the search were reported in full in the final systematic review and presented in a Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flow diagram [17].

Data Extraction.
Data was extracted from papers and included in the review by one reviewer using a data extraction tool developed by the reviewer. e data extracted included specific details about the plants species, family, place of harvest, part use, extract, metabolite, used parasite, cell model, IC 50 , 50% cytotoxic concentration (CC 50 ), standard deviation of each quantitative variable, sample size. e selectivity index (SI) of each extract was calculates as follows: CC 50 /IC 50 . e extracted data was presented in tabular form to align with the objective of this review. A narrative summary accompanied the tabulated results and described how the results relate to the reviews objective and question.

Risk of Bias Assessment.
Eligible studies were critically appraised by one reviewer considering a score described in previous systematic reviews of in vitro studies [19]. e description of the following parameters was checked in each study: clear extraction method, appropriate in vitro method for antimalarial activity, appropriate number of replicate, resistant vs. sensitive Plasmodium strain comparison, appropriate in vitro method for cytotoxicity, culture of Plasmodium and control cells in the same condition, availability of all required outcome (IC 50 , CC 50 and SDs) and quality control valid. If the parameter was described on the text, the study received a "yes" on that specific parameter, otherwise it had a "no." e risk of bias was classified according to the sum of "yes" received as follows: 1-3 = high, 4-5 = medium, 6-8 = low risk of bias. e results of critical appraisal is reported in narrative form and in a table. Only the low risk of bias studies was included in the meta-analysis.

Data Synthesis.
Selected studies were pooled in statistical meta-analysis using R Studio software V4.1.1. Effect sizes were expressed as standard mean difference (SMD) for selectivity index and 95% confidence level was considered for analysis. Heterogeneity was assessed statistically using the standard Chi-squared and I squared tests. Statistical analyses were performed using random effects model [20]. Plants or metabolites with SMD <0.8 were considered as good antiplasmodial drug candidate against both chloroquine/multi-drug-resistant (experimental) and sensitive (control) Plasmodium strain. Forest plots were created to illustrate the meta-analysis. Where statistical pooling was not possible, the findings were presented in narrative form including tables to aid in data presentation where appropriate. A funnel plot was generated to assess publication bias. Egger's test for funnel plot asymmetry was performed where appropriate.

Search and Selection.
A total of 220 articles were retrieved by automatic search on PubMed. Manual search based on reference screening completed our search with another 12 articles (Figure 1). From the 232 articles downloaded, 14 were duplicates and were removed, 46 were reviewed and therefore automatically excluded. e titles and abstracts of 218 articles were screened, 90 were excluded as not being on the topic of interest. Of the 128 eligible articles, 86 were excluded for various reasons such as: study population (28), type of publication (52), study design (7), and unavailability of full text (1). e remaining 42 articles were included in the qualitative analysis and assessed for risk of bias. Only 10 articles with low risk of bias were included in meta-analysis (Table S1). Tables S2 and 1 show descriptive extracted data from the included studies in systematic review and meta-analysis, respectively. All studies were published between 2000 and 2021. More studies were conducted in the center region of Cameroon (n = 18) followed by the western region (n = 6). e rest of the plants were collected from South-west, Littoral, East, North-west, and Far north region. No plants were yet investigated in the North and Adamaoua regions ( Figure 2). Almost 90 plants species ( [32]. Secondary metabolites isolated, such as 2,3,6-trihydroxy benzoic acid and 2,3,6-trihydroxy methyl benzoate, demonstrated low inhibitory effects against P. falciparum strains, with IC 50 values of 16.47 and 13.04 μM against P. falciparum W2, and 35.41 and 6.09 μM against falcipain-2, respectively [32]. Otherwise, aqueous and ethanol extract of bark or leaves of Mangifera indica exhibited a high selectivity index for their antiplasmodial activity (SI > 50) [21].

Asteraceae.
e carrot-like tubers of Vernonia guinensis are commonly used in ethnomedicine. Toyang et al.
investigated the antiplasmodial activity of crude extracts and pure compounds of V. guinensis. ese pure compounds and crude extracts from V. guinensis inhibited the growth of HB3 and Dd2 [41]. e IC 50 values of extracts were similar for HB3 and Dd2, and ranged from 1.64-27.2 μg/ml for Hb3 and 1.82-30.0 μg/ml for Dd2. e IC 50 values of vernopicrin, vernomelitensin, and pentaisovalerylsucrose isolated from V. guinensis were similar to HB3 and Dd2 and ranged from 0.47-1.62 μg/ml for HB3 and 0.57-1.49 μg/ml for Dd2 [41]. Similar result was found with Vernonia amygdalina and Vismia guinensis which exhibited antiplasmodial activity without cytotoxicity [26].

Caricaceae.
Leaves of Carica papaya are constituents of Nefang, a traditional drug used to treat malaria in Cameroon. ese leaves do not show any antiplasmodial activity against Dd2 strains of P. falciparum without toxicity [21].
3.3.9. Celastraceae. Crude dichloromethane/methanol and secondary metabolites extracts from Salacia longipes exhibited a very high antiplasmodial activity against W2 P. falciparum strain [44]. And cytotoxicity analysis was performed.

Clusiaceae.
Ndjakou Lenta et al. tested fruits extracts and isolates compounds from Pentadesma butyracea for their antiplasmodial activity in vitro against the W2 strain chloroquine-resistantP. falciparum and other antimalarial drugs. Pericarp extract showed good antiplasmodial activity, with an IC 50 of 1.83 μg/mL, while the seed extract was inactive. Among all isolated compounds, only the xanthones exhibited antiplasmodial activity against the W2 strain, with garcinone E showing the best potency and followed by α-mangostin, cratoxylone, and pentadexanthone [45].

Guttiferaceae.
ree species of Guttiferaceae (Allanblackia floribunda, Allanblackia monticola, and Allanblackia gabonensis) were tested for their antiplasmodial activity by Azebaze et al. [23,47,51]. ey found that, A. gabonensis did not show any antiplasmodial activity. However, A. floribunda an A. monticola exhibited strong antiplasmodial effects. Macluraxanthone isolated from A. floribunda was the most active compound on two strains of Plasmodium followed by volkensiflavone with a IC 50 of 0.46 and 0.99 μg/mL for the F32 and 0.33 and 0.93 μg/mL for the FcM29 strains respectively [51]. Allaxanthone B isolated from A. monticola was responsible of its antimalarial property with IC 50 of 3.70 and 3.93 μg/ mL for the F32 and FcM29 strains respectively [51]. Five of other prenylated xanthones (α-mangosine, tovophiline A, allaxanthone C, rubraxanthone, norcowanine) isolated from A. monticola previously tested for antiplasmodial properties had displayed after 24 h of contact with the parasite a significant antiplasmodial activity (IC 50

Lamiaceae. Neither Ocimum basilicum and
Ocimum canum, previously found as a repellent, and nor Ocimum gratissimum which is part of Nefang (a traditional remedy usually used in Cameroon to treat malaria) showed antiplasmodial activity in vitro [21,22,55].

Leguminoceae.
Only Kotschya speciosa was investigated in this family and was not found to be active against P. falciparum [28].

Loganiaceae. Tchinda et al. tested the stem bark of Strychnos malacoclados.
ey found that an ethyl acetate extract of this specie exhibited an antiplasmodial activity against the chloroquine-sensitive 3D7 strain of P. falciparum with IC 50 of 2.85 µg/ml [24]. All secondary metabolites extracted from S. malacociados displayed an antiplasmodial activity against the chloroquine-sensitive 3D7 strain of P. falciparum [24]. From the stem bark of S. malacoclados, one new bisindole alkaloid, 3-hydroxylongicaudatine Y, was isolated along with the known alkaloids vomicine, bisnordihydrotoxiferine, divarine, longicaudatine, longicaudatine Y, and longicaudatine F [24]. Strychnobaillonine from Strychnos icaja was found as a very high antiplasmodial compound with SI � 14 [56].

Monimiaceae.
A phytochemical study of the methylene chloride/methanol extract of leaves of Glossocalyx

Meta-Analysis.
We evaluated the impact of various potential interfering factors on the results of this metaanalysis. ese included the part of the plant used, the nature of the crude extract, the strain of resistant or susceptible P. falciparum used, and the cell model used for cytotoxicity evaluation. Our results show that only the cell model used could lead to a significant heterogeneity (p < 0.01) between the different groups ( Figure S1 and S3).
e Asteraceae V. amygdalina and V. guinensis on the one hand and the Burseraceae D. edulis on the other hand, respectively, presented almost similar selectivity index in resistant and susceptible strains (I 2 : 0%; SMD: −1.60 and −0.06 respectively) (Figure 3). Meta-analysis of the antiplasmodial activity of various metabolites extracted from Cameroonian plants, highlights the strong antiplasmodial potential of metabolites 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16, 17, 19, 20, and 21 (I 2 < 30 and SMD <0.8) (Table 2; Figure 4).    Evidence-Based Complementary and Alternative Medicine    Eggers' test showed an asymmetry between the crude extract plant data (P � 0.289), suggesting a high risk of publication bias for these data (Table 3). In contrast, the secondary metabolite data did not show asymmetry (P � 0.048) suggesting a low risk of publication bias for these data (Table 3). Figure 4 shows the funnel plot of the selectivity index of the different secondary metabolites evaluated in these studies. e detailed analysis of this funnel plot shows a reservation on two of the compounds (7 and 10) previously considered as good candidates for their antiplasmodial activity (P < 0.1). Figure S2 shows the plants selectivity index to chloroquine resistant and susceptible strain, using random effect model.

Discussion
e fight against malaria is a great challenge characterized on the one hand by the resistance of the vector to the insecticides used and on the other hand by the resistance of the parasite to conventional drugs. It has to be noted that the discovery of new drugs against malaria is most often based on the results of research in natural pharmacopoeia as was the case for artemisinin and quinine [14,15]. Some synthetic drugs such as dihydroartemisinin and chloroquine are based on active ingredients from natural plants. Given the rise of resistance,  especially to artemisinin and its derivatives used as the first line of defense against malaria, it is important to go back to the source of natural plants to look for potential candidates that could supplement this first line of defense while reducing the rapid emergence of resistance due to the massive use of a single type of drug. is review aimed at screening the antiplasmodial potential of Cameroonian plants and through a meta-analysis to bring out all the potential candidate active ingredients. Taking into account the main goal of this systematic review, 42 in vitro studies were selected and 10 were submitted for meta-analysis.
ere is strong evidence that the dichloromethane extract of Vernonia amygdalina leaves [25], the dichloromethane and dichloromethane/methanol extracts of Vismia guinensis stem bark [26] and also the dichloromethane/ methanol extracts of the leaves and stem bark of Dacryodes edulis [28] could be used as an antiplasmodial drug on chloroquine-sensitive and chloroquine-resistant strains (SI > 10). Despite the high risk of publication bias as revealed by Egger's test for crude extract results, the funnel plot showed us that studies on these above plants had a low risk of publication bias. Despite the fact that these plants had in common the type of extract used, our meta-analysis data did not support a significant effect of the type of extract used on the selectivity index of the plants for their antiplasmodial activities. e antiplasmodial potential of these plants would thus be particularly due to the nature of the active principles they contain. Indeed, the interactions between the compounds contained in the crude extracts of plants are often at the origin of a more or less high bioactivity of these crude extracts [65]. Antagonistic and synergistic interactions are the main causes. is review does not highlight the effect of these interactions on plant extracts, but we were able to demonstrate that 17 compounds extracted from plants, among which 5 (3 polyphenols and 2 terpenoids) extracts were from Dacryodes edulis [28], could be selected as ideal candidates for their antiplasmodial effect not only in chloroquine-sensitive strains, but even more so in chloroquineresistant strains. Dacryodes edulis (Safou) is known for its dietary properties via its edible fruit; its curative and suppressive properties on a mouse model infected with Plasmodium berghei were demonstrated by maximum inhibition of Plasmodium at 57% and chemosuppression of the parasite at 71% [66]. In addition, this plant has antioxidant [67], anticancer [68], antidiabetic [69] properties that make it a good research model for its multiple effects. e limitation of this review is that it does not allow to conclude on the effect of these molecules on artemisinin-resistant strains which for several years has been considered as a first-line drug instead of chloroquine.

Conclusion
Despite the heterogeneity observed between the different plant families studied in Cameroon for their in vitro antiplasmodial effect, there is strong evidence that 17 active compounds from these plants would be ideal candidates for the synthesis of new antimalarial drugs. e Dacryodes edulis species, containing 5 of these active compounds, could be considered as a natural alternative in the treatment of uncomplicated malaria because of its inhibitory and suppressive capacities on the one hand and its relatively low cytotoxicity on the other hand.

Data Availability
All data generated or analyzed during this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest. Figure S1. Analysis of potential confounding factors; Figure  S2. Plants' selectivity index to chloroquine resistant and susceptible strain, using random effect model; Figure S3. Funnel plot for plants species ; Table S1. Risk of bias assessment; Table S2. Characteristic of studies included in the systematic review. (Supplementary Materials)