In spite of available effective preventive, treatment and case management measures for a very long time, malaria remains a global public health problem and a leading cause of morbidity and mortality in many tropical countries. WHO [1] estimates for the year 2010 showed that there were 219 (154–289) million malaria cases and 660,000 (490,000–836,000) deaths due to malaria. Out of these about 80% of the cases and 90% of the deaths occurred in Africa while the remaining cases and deaths occurred mainly in South-East Asian and Eastern Mediterranean regions. In Ethiopia, Anopheles arabiensis Patton is the primary vector responsible for malaria transmission [2,3] whereas An. pharoensis Theobald, An. funestus Giles and An. nili (Theobald) are regarded as secondary vectors [4].

The initial malaria eradication campaigns against the mosquito vectors were very successful in the developed parts of the world, particularly with the coupled application of dichlorodiphenyltrichloroethane (DDT) and pyrethroid insecticides [5]. However, many countries in Africa, Asia and Latin America lacked the basic health infrastructure, and hence these campaigns could not yield the expected results [6]. In Ethiopia, as in several other countries, vector control through application of chemical insecticides was frustrated with the development of strong resistance of mosquitoes to the synthetic insecticides [7-9]. The discovery of DDT insecticidal properties in 1939 and the subsequent development of organochlorine and organophosphate insecticides limited plant product research since the answers to insect control were thought to have been found [10]. On the other hand, some inherent features and use patterns of the conventional synthetic insecticides lead to toxicity to humans, livestock, fish, birds, and other beneficial organisms as a result of adverse effects on the environment, resulting from contamination of soil, water, and air. Resurgence of insect vector populations because of the emergence and spread of physiological and behavioral resistance to a wide range of insecticides has been a serious impediment in malaria control. On top of this is the higher costs of spraying operations, and high degree of refusal by the public to permit indoor spraying have remained challenges to the malaria control programs in developing countries. Thus, the drawbacks characteristic of the synthetic insecticides, has necessitated the search for an environment friendly and less costly plant products as an alternative vector control tools for use in the integrated vector management programs. In additions, mosquito control using plant materials can be obtained from local sources and are likely to generate local employment, reduce dependence on expensive imported products, and stimulate efforts to enhance public health [11]. The value of local focus on medicinal pant research was also emphasized by Kuppusamy and Murugan [12] who showed that the selection of insecticidal plants could be optimized through initial ethnobotanical screening of plants that indigenous local communities use in their traditional medical systems. It is known that some plant derivatives have been in use as insecticides for quite some time. Among these, pyrethrum produced from Chrysanthemum cinerariifolium (Asteraceae), rotenone from Derris elliptica (Fabaceae), azadirachtin from Azadirachta indica (Meliaceae), and nicotine from Nicotiana tabacum (Solanaceae) were widely used [13]. While these botanical bioactive products have the advantage of a relatively low mammalian toxicity and a fairly broad spectrum of activity compared with many synthetic insecticides, their short half-life of activity is the disadvantage [14]. The low persistency of the existing botanical insecticides has been shown to be due to their susceptibility to the effects of light and air as demonstrated for pyrethrum [15]. The present study was designed with the objective of evaluating the adulticidal activities of purified fractions of an indigenous plant Oreosyce africana Hook.f. against adult An. arabiensis Patton, the primary vector of malaria in Ethiopia.

Plant materials

The leaves of Oreosyce africana Hook.f. (Cucurbitaceae) were collected from Oromia Region, Ethiopia, in October 2011. The leaves were stored dry at room temperature and were extracted by 80% aqueous methanol (SCP, England) within two months. Ethnobotanical information on the medicinal uses of the species were collected by interviewing local traditional herbalists and knowledgeable elders after explaining the purpose of the research and obtaining consent through blessings that constituted the traditional ethical clearance. The dried plant material was stored at room temperature until the extraction process started. Along with the collection of plant materials for extraction and analysis, voucher specimens were collected, pressed, dried and authenticity confirmed by taxonomic experts at the Department of Plant Biology and Biodiversity Management and deposited at the National Herbarium of Ethiopia, Addis Ababa University for further reference.

Extraction and solvent fractionation

The powdered leaves of O. africana was extracted with 80% aqueous methanol (SCP, England) at room temperature for 72 hr in a rotator shaker (VWR, USA) within two months of collecting. The residue was filtered and the extraction procedure repeated three times under the same condition. The filtrates were combined and evaporated to dryness under reduced pressure at 45°C by using a rotary evaporator (Stuart®RE300, UK). The extract was fractionated by solvent partitioning with dichloromethane (Carlo Erba, France) and deionized water (water deionizer, EASYpureII, USA) and partitioned three times. After drying over anhydrous sodium sulfate, the dichloromethane extract was evaporated by using a rotary evaporator to give the dichloromethane fraction.

Flash column chromatography

The dichloromethane fraction obtained by solvent partitioning was purified based on the procedure Alkofahi et al. [16] as shown in Figure 1. The residue of dichloromethane fraction was chromatographed in a column of silica gel (Merck 25-100 mesh size, Germany) and a glass column (40×5 cm) successively eluted with hexane, hexane-chloroform (95:5, v/v), hexane-chloroform (90:10), hexane-chloroform (80:20), hexane-chloroform (70:30), hexane-chloroform (60:40), and hexanechloroform (50:50). In each case, 100 ml of solvent mixture was added and 20 ml fractions were collected using 50 ml conical flasks placed under the stopcock.

Figure 1: Purification procedure for the extracts from Oreosyce africana leaf.

Thin-layer chromatography

Isolation of the metabolite of O. africana was monitored during the separation stages with thin-layer chromatography (TLC) on Kieselgel gel 60 F254 plates (Merck) which has a 0.5 mm thickness of silica sorbent (Merck no. 64271, Darmstadt, Germany) using chloroform-methanol (9:1, v/v) following the method of Okeke et al. [17]. The components of TLC plate were visualized by spraying with 1% sulfuric vanillin solution, then heating on a hot plate at 110°C for 5 min. The retention factor (R_f) value of the fraction was determined using the formula described by Ettre [18].

Preparative thin-layer chromatography

The concentrate obtained by solvent evaporation of dichloromethane fraction of O. africana leaf extract was dissolved with chloroform and then applied to 16 sheets of Kieselgel 60 F₂₅₄ TLC plates (20 × 20 cm, 0.25 mm thickness; Merck, Germany) following the method described by Jiao et al. [19] and Sasidharan et al. [20]. These were developed in chloroform-methanol (5:1) mixed solvent. 1.5 cm zone of the distinct bands of the purified fractions was scraped off from its plate and extracted eight times with 100 ml of chloroform-methanol (9:1) solvent. The extraction solvent was evaporated under rotary vacuum evaporator and extracted fractions were coded as B2’O and B2’’O. These two fractions were stored at -20°C until used in the mosquito bioassay tests.

Anopheles arabiensis laboratory rearing

Adult female An. arabiensis mosquitoes used in the bioassays were reared according to the WHO [21] protocol. The adult colonies were reared and maintained at 25-27°C and 70-80% relative humidity under a photoperiod cycle of 12 hrs light and 12 hr dark in the insectary of the Department of Zoological Sciences, Addis Ababa University. 10% sucrose solution was provided to the adult mosquitoes in soaked cotton pad placed on top of the mosquito nets. Adult female An. arabiensis were periodically blood-fed on restrained rabbits shaved back and belly. The larvae were fed on ground Tetramin® fish food pellets (Tetra holding Inc., Blacksburg, VA, USA); the feed applied on alternate days and the larvae and pupae were kept at 25-27°C and 70-80% relative humidity under a photoperiod cycle of 12 hr light and 12 hr dark.

Evaluation of adulticidal activity

Plastic tube bioassay: The mosquitocidal activities of the purified fractions were determined based on WHO [22] plastic tube bioassay test method. These consisted of holding and exposure cylindrical plastic tubes having a length of 125 mm and a diameter of 44 mm. Concentrations ranging from 4 to 16 ppm of each fraction, which included fractions II,III and IV collected from flash chromatography and B2’O and B2’’O collected from preparative TLC were applied to WHO plain papers. Each concentration was prepared in 0.05% DMSO (Carlo Erba, France) in deionized water and shaken vigorously to achieve homogeneity of the contents. The WHO papers treated with the O. africana fractions were left to dry at room temperature overnight and then inserted into each exposure tube and the WHO papers treated with 0.05% DMSO in deionized water served as a negative control. The bioassays were performed at 25-27°C and 70-80% relative humidity with non-blood-fed An. arabiensis adult mosquitoes of known age (2 to 5 days post-emergence) in batches of 20 in each concentration. The mosquitoes were allowed to acclimatize in the holding tube for 1 hr and then exposed to the fractions on the impregnated paper and the unimpregnated control for 1 hr. At the end of the exposure period, the mosquitoes were transferred back to the holding tubes and kept for 24 hrs recovery period. A pad of cotton soaked with 10% sucrose solution was placed on the mosquito net during the holding period. Each treatment was replicated three times for each concentration.

Cage bioassay: Aerosol bioassays were performed by modifying the method of WHO [21] on adult An. arabiesnsis held in laboratory cages covered with mosquito netting (19 × 19 × 19 cm). Twenty five nonblood- fed adult 2 to 5 days old, female An. arabiensis were transferred into the test cages with a mouth aspirator. The adult mosquitoes were allowed to acclimatize in the cages for 1 hr before treatment. Each treatment was applied based on the method [23] using a small handheld compression sprayer that discharged the purified fractions (II, III, and IV) collected from flash chromatography and B2’O and B2’’O purified fractions collected from prep. TLC were sprayed into mosquito cages in separate experiments, at a concentration of 8 ppm. The test sample concentrations were prepared in 0.05% DMSO in deionized water and shaken vigorously to achieve homogeneity of the contents. The control cages were sprayed with the same amount of 0.05% DMSO in deionized water. 10% sucrose solution soaked on cotton pad, as a source of nutrition, was placed on top of the mosquito netting, in the test as well as in the control group. The Bioassay tests were carried out at 25-27°C and 70-80% relative humidity in the laboratory and mosquito mortality was recorded 24 hrs post-treatment.

Cone bioassay: Polyester fabric nets were purchased from retail shops in Addis Ababa and were cut into 25 × 25 cm size for use in the experiments. The nets were folded and soaked in disposable plastic petri dishes containing solutions of B2’O or B2’’O in 0.05% DMSO at concentrations of 4, 8, and 16 ppm. Nets were treated one day before bioassay and stored at ambient room temperature [24]. The impregnated and the control nets were dried in shade by laying them flat on the table. The impregnated nets were fixed to a cardboard and three standard WHO plastic cones were placed and fixed with tape over the area of the treated net. Ten non-blood-fed, 2 to 5 days old, female An. arabiensis each were introduced into the three test cones through a hole on the mosquito nets. The same numbers of non-bloodfed adult female mosquitoes were also introduced into the cone that was fixed over a net impregnated with 0.05% DMSO in deionized water as a negative control. Both test and negative control mosquitoes were exposed for 15 minutes under ambient room temperature. At the end of the exposure period, the mosquitoes were transferred into recovery paper cups covered with untreated nets and were held for 24 hrs with access to 10% sucrose solution. Mortality was determined after 24 hrs recovery period.

Chemical characterization of the mosquitocidal constituents

High performance liquid chromatography analysis: The active fractions of O. africana were further purified by high performance liquid chromatography (HPLC) using Waters LC-2000 model equipped with Waters 600 pump controller (Milford, MA, USA). 10 μl of each sample fraction-B2’O and B2’’O injected manually into a column of Tracer Extrasil ODS2, TR416059 (particle diameter, 5 μm; pore size, 30 nm; column size, 25 × 0.46 cm) (Teknokroma, Barcelona, Spain) and eluted by use of solvent system: methanol-water (80:20, v/v) with a running time of 0-30 min. Column temperature of 20°C and flow rate of 1.0 ml/min were maintained and the eluate detected at 254 nm using Waters 2487 dual absorbance UV detector. All data were recorded and processed by Millennium 32 software from Waters (Milford, MA, USA).

Gas chromatography-mass spectrometry (GC-MS): The GC and GC-MS analyses were performed using the method described by Mohan and Maruthupandian [25]. Gas chromatography measurements were carried out with Hewlett-Packard HP5890 Series II (Agilent Technologies, USA) and GC-MS analysis was conducted with gas chromatograph model 7890B (Agilent Technologies, USA) coupled to Hewlett-Packard MS model 6977A (Agilent Technologies, USA) with mass selective detector. The analysis was carried out on a GC Clarus 500 Perkin Elmer system and gas chromatograph interfaced to a mass spectrometer (GC-MS) employing the following conditions: Column Elite-1 fused silica capillary column (30 mm × 0.25 mm 1D × 1 μm df, composed of 100% dimethyl poly siloxane). Helium (99.999%) was used as a carrier gas operating in electron impact mode at a constant flow of 1 ml/min and each sample injected contained a volume of 2 μl was applied manually (split ratio of 10:1) under injector temperature of 250°C and ion-source temperature of 280°C. The oven temperature was programmed from 110°C (isothermal for 2 minutes), with an increase of 10°C/min until 200°C and then rose at 5°C/min up to 280°C and was held for the last 9 minutes (isothermal at 280°C). The GC-MS instrument was used in the electron impact mode under an ionization voltage of 70 eV. The detection of mass spectra was completed in 38 min. Interpretation of the mass spectra from GC-MS data was conducted using the database of National Institute of Standards and Technology (NIST) library. The mass spectra of the unknown components were compared to the spectra of the known compounds and determined by searching in the NIST library.

Data analysis

The LC₅₀ and LC₉₀ values of O. africana fractions II, III, IV, B2’O, and B2’’O were determined by probit regression analysis using the statistical package PoloPlus (version 2.0, LeOra Software, Petaluma, California, USA; 2007). The 95% confidence limits of upper confidence limit (UCL) and lower confidence limit (LCL) for the lethal concentration in ppm (LC₅₀ and LC₉₀) were determined using probit regression analysis of the statistical package PoloPlus. Hypothesis testing to compare concentration-mortality probit lines was evaluated by likelihood ratio tests for equality and parallelism (chi-square tests and P values for comparing the slope and intercept). When the slopes of the concentration-mortality lines were not significantly different, lines were considered parallel, indicating that the two treatments had the same relative potency (i.e., the same variability in response) [26]. PoloPlus analysis was used to determine adulticidal efficacies among purified O. africana fractions II, III, B2’O, and B2’’O. Statistically significant differences between the compared LC₅₀ concentrations, the 95% confidence intervals for lethal concentration ratios were calculated. In this pair wise comparison, lethal concentrations were considered significantly different if the value ‘1’ did not fall within the confidence interval for the ratio [26].

Thin-layer chromatography analysis of Oreosyce africana fractions

Extracts of dichloromethane fraction of O. africana were applied onto the columns and the fractions with similar Rf values on the TLC plates were combined and designated as fractions II to IV as follows: O_6&7 as fraction II eluted with hexane-chloroform (95:5, v/v); O_8-10 as fraction III eluted with hexane-chloroform (90:10); and O_11-14 as fraction IV eluted with hexane- chloroform (80:20). The dichloromethane fraction of O. africana leaf extract was also applied to preparation. TLC for purification Sixteen plates of prep. TLC were used for the purification of the fractions and yielded two major yellow bands on prepared TLC. From the combined fractions two active purified fractions B2’O and B2’’O were obtained after separation by prep. TLC on silica gel plates with chloroform-methanol (5:1). In TLC analysis, two yellow spots corresponding to B2’O (R_f=0.5) and B2’’O (R_f = 0.46) were observed by spraying with 1% sulfuric vanillin reagent.

Bioassay of chromatographic fractions of Oreosyce africana on adult Anopheles arabiensis

Comparison of the adulticidal activities of purified potent fractions of O. africana collected following flash column chromatography and prep. TLC against the malaria vector, An. arabiensis adults are presented in Table 1. The hypothesis test for parallelism was rejected (χ²=22.90; P=0.001) showing that slopes differed significantly. Among the fractions tested, the highest adulticidal activity was observed with fractions of IV or B2’O with an LC₅₀ and LC₉₀ values of 2.206 and 7.811 ppm followed by fraction of B2’’O with an LC₅₀ and LC₉₀ values of 2.62 and 11.779 ppm (Table 1).

Table 1: Adulticidal activity of different fractions of O. africana (fraction II, III and IV using flash column chromatography; B2'O and B2''O using prep. TLC) against Anopheles arbiensis adults after 24 hr exposure period on impregnated papers in WHO test tubes (n=45 in each test).
* The codes used for the fractions.

Effects of purified potent fractions of O. africana against adult An. arabiensis

The effect of the purified potent fractions, B2’O and B2’’O isolated from O. africana were compared with WHO papers treated with 0.05% lambdacyhalothrin (positive control) against An. arabiensis adults (Figure 2). The adulticidal activities of purified fractions B2’O and B2’’O at a concentration of 8 ppm each against An. arabiensis showed 89% and 80% adult mortality, respectively while with 0.05% lambdacyhalothrin, the WHO standard, the mortality of adult An. arabiensis was 74% (Figure 2). There was no mortality in the untreated controls.

Figure 2: Comparison of activities of potent fractions of Oreosyce africana (B2’O and B2’’O) at 8 ppm with 0.05% lambdacyhalothrin (positive control) on adult Anopheles arabiensi after 24 hrs exposure period on impregnated papers in WHO test tubes (n=45 in each test).
Note: Positive control (0.05% lambdacyhalothrin); negative control (0.05% DMSO in deionized water).

Efficacy of active components of O. africana as aerosol sprays on adult An. Arabiensis

Five purified fractions of O. africana exhibited different levels of adulticidal activity. With regard to treatment by aerosol sprayed inside the cage against adult An. arabiensis with O. africana purified fractions II, III, IV, B2’O and B2’’O at 8 ppm concentration each resulted in 49, 41, 89, 89 and 80% mortality, respectively (Figure 3).

Figure 3: Measure of mortality of Anopheles arabiensis adults following aerosol spray treatment with fractions of Oreosyce africana after 24 h exposure period (n=75 in each test).

Cone bioassay of B2’O and B2’’O fractions impregnated nets on Anopheles arabiensis

Based on high adulticidal potency of fractions B2’O and B2’’O of O. africana, dose-response bioassay for the adulticidal activities of the two fractions against adult An. arabiensis was conducted and mortality was observed after 24 hrs exposure period. The percentage adult mortality against the logarithmic concentration values was plotted based on the mean of three replicates (Figure 4). The results of adulticidal activity tests of B2’O and B2’’O clearly showed the percentage of adult An. arabiensis mortality to be directly proportional to the concentration of the fractions.

Figure 4: Parallel case regression plot showing percent mortality of adult Anopheles arabiensis after 24 hr exposure to different concentrations of purified fractions, B2’O and B2’’O, of Oreosyce africana using impregnated net in cone bioassay (n=45 in each test).

High performance liquid chromatography analysis of O. africana fractions

Purification of B2’O and B2’’O of O. africana fractions collected from the prep. TLC were determined by using HPLC. This B2’O fraction gave main and minor peaks by HPLC analysis (Figure 5). The retention times (Rt) of the two peaks were 6.398 and 10.471 min with 91% and 5.4% area under the peak, respectively. The HPLC chromatogram showed two peaks of which the first one was the major peak and the second was lesser peak (Figure 6). The first peak appeared at retention time (Rt) 3.565 min with area % of 72.30 and the second peak appeared at Rt 5.175 min with a % of 24.76.

Figure 5: High performance liquid chromatograms for B2’O fraction of O. africana, detection at 254 nm.
Note: Chromatographic conditions: column, Tracer Extrasil ODS2, TR416059 (particle diameter, 5 μm; pore size, 30 nm; column size, 25 × 0.46 cm); eluent, methanol: water (80:20, v/v); flow rate, 1 ml/min; temperature, ambient.

Figure 6: High performance liquid chromatogram for B2’’O fraction of O. africana extract detected at 254nm.
Note: Chromatographic conditions are the same as in Figure 5.

Gas chromatography-mass spectrometry (GC-MS) analysis of B2’O and B2’’O

The mass spectrum presented in Figure 7 revealed the structure of 9,12-Octadecadienoic acid (Z,Z), more commonly known as linoleic acid, as the major chemical constituent of the purified isolate, B2’O. The mass spectrometer analysis showed a parent peak at m/z 280, which corresponded to a molecular formula of C₁₈H₃₂O₂. There also were other fragmented minor peaks at m/z 264, 110, 95, 81, 67, 55, 41 and 29 (Figure 7). In comparison with the library values, the fragmentation peaks were assigned chemical name, molecular formula, molecular weight, and approximate relative percentages for the compounds. Thus, the GC-MS analysis confirmed the identification of twelve compounds with different retention times from the purified B2’O fraction of O. africana leaf extract. The major compounds identified in the purified B2’O fraction were 9,12-Octadecadienoic acid (Z,Z)-, which constituted 98.34847% belonging to the most intense peak, followed by E,E,Z-1,3,12-Nonodecatriene-5,14-diol (0.15081%), Ethanol,2- (9,12-Octadecadienyloxy)-, (Z,Z)- (0.57465%), 10,13-Octadecadienoic acid, methyl ester (0.1329%), 10-Octadecenoic acid, methyl ester (0.11585%), and 9,12-Hexadecadienoic acid, methyl ester (0.11205% (Table 2). The gas chromatographic analysis of purified B2’’O fraction revealed seven peaks. The mass spectrometer analysis showed a parent peak at m/z 278 and fragments at m/z 223, 205, 149, 104, 57, and 29 (Figure 8). The major compound with a molecular formula of C₁₆H₂₂O₄, appeared as a peak at scan time 25.376 min and this corresponded to the chemical compound, dibutyl phthalate. The characteristics of the seven compounds identified from B2’’O purified fraction of the leaf extracts of O. africana by GC-MS analysis, including the retention time (Rt), molecular formula, molecular weight and the approximate relative percentages of this B2’’O fraction are presented in Table 3. Thus, In GCMS analysis of the B2’’O fraction seven compounds were identified, among which, dibutyl phthalate constituted the largest proportion (97.7535%) followed by Ethaneperoxoic acid, 1-cyano-1-[2-(2-phenyl- 1,3-dioxolan-2-yl)ethyl]pentyl ester (0.38466%), Phthalic acid, butyl tetradecyl ester (0.32408%), and 1,2-Benzenedicarboxylic acid, butyl octyl ester had the least proportion of 0.053% (Table 3).

Figure 7: Mass spectrum from GC-MS scan revealing the 9,12-Octadecadienoic acid (Z,Z)- and fragmentation pattern of the peak that eluted at 27.94 min for purified fraction, B2’O from Oreosyce africana leaf extract.

Figure 8: Mass spectra of the dibutyl phthalate and fragmentation patterns of the peak at scan time of 25.376 min for B2’’O fraction isolated from Oreosyce africana leaf.

Table 2: Compounds identified in the isolated fraction of B2'O of Oreosyce africana by GC-MS.

Table 3: Compounds identified in the isolated fraction of B2''O from O. africana by GC-MS.

The biological activity of the secondary metabolites from O. africana extracts might be due to a variety of compounds in the plant. These compounds may jointly or independently contribute for the adulticidal effects against An. arabiensis adults. The enhanced adulticidal activity of purified fraction of O. africana compared to their respective crude extracts and semi-purified fractions is evidence that purification of plant extracts would significantly improve the bioefficacy of the active constituents in the extract. Accordingly, the HPLC method enabled assessment of the levels of the active components of O. africana extract purified fractions, B2’O and B2’’O, obtained from prep. TLC. The linoleic acid, which was determined by the GC-MS analysis in the B2’O fraction has previously been shown to possess antibacterial potency. Evidence for this comes from investigations on the extracts of the plant Schotia brachypetal, in which linoleic acid was the major antibacterial constituent [27]. In addition, the bioactivities of long chain unsaturated fatty acids, such as linoleic acid, have been reported from bioassays against bacteria [28]. Furthermore, the broad spectrum bioactivity of linoleic acid is also evident from its inhibition of parasitemia in mice infected with Plasmodium vinckei and Plasmodium yoelii in a 4-day suppressive test [29].

The broad bioactivity of the leaf extracts of O. africana reported by several workers [25,30,31], characteristics such as their antiinflammatory, hypocholesterolemic, anticancer, hepatoprotective, nematicidal, insectifuge, antihistaminic, antieczemic, antiacne, 5-α reductase inhibitor, antiandrogenic, antiarthritic and immune enhancing may be explained by the presence of various low quantity compounds, which were identified using the GC-MS in the present work. Additional evidence for the insecticidal effects of linoleic acid also comes from Citrullus colocynthis (L.) Schrad (Cucurbitaceae) extract that had potent larvicidal activity against fourth instar larvae of An. stephensi Liston with LC₅₀ value of 9.79 ppm and LC₉₀ value of 37.42 ppm [32]. Sayed et al. [33] and Basalah et al. [34] had isolated and identified linoleic acids from Citrullus colocynthis spectroscopically. Another study by Edriss et al. [35] had also shown the presence of linoleic acid in the seed extracts of Jatropha curcas, a plant known to have larvicidal effects against An. arabiensis [36]. Likewise, evidence for dibutyl phthalate, the major B2’’O chemical component, detected by GC-MS as a mosquitocidal compound, exists whereby the substance isolated from Ipomoea carnea stem extracts that possessed larvicidal activities [37] and from Urtica dioicaleaf extract that showed antimicrobial activity [38]. The identification of six additional minor compounds from O. africana leaf extract would most likely be responsible for the broad spectrum of bioactivity reported by several workers such as antimicrobial and antifouling activities [39]. The fact that percentage mortality of An. arabiensis adults increased significantly with exposure time in the cone bioassay test of purified fractions of (B2’O and B2’’O), indicates the exposure-time dependence of their lethal effect. Maharaj et al. [40] in a cone bioassay test with dichloromethane extracts of Ptaeroxylon obliquum (Ptaeroxylaceae) and Pittosporum viridiflorum (Pittosporaceae) against An. arabiensis had also found that the more than 50% adult mortality could be achieved in an exposure time dependent manner. In the use of mosquito nets for protection against malaria, physical barrier provided by the nets, usually needs to be supplemented by a chemical barrier consisting of a long-lasting deposit of insecticides on the netting. However, as the most commonly used insecticides are inorganic chemicals that cause concern as environmental hazards, the need for their replacement with organic insecticides, such as B2’O and B2’’O, isolated in the present study from O. africana, may have to be considered. Similar findings from botanical sources, whereby the phenylpropenoid compounds from Uvariodendron pycnophyllum with long term mortality effects to adult An. gambiae on impregnated mosquito nets had been suggested as an alternative material for net impregnation [41]. Ramos-López et al. [42] reported that the linoleic acid isolated from Ricinus communis exhibited insecticidal activity against Spodoptera frugiperda. Furthermore, linoleic acid obtained from extract of Dirca palustris had insecticidal activity against Aedes aegypti [43]. Treating the nets with active ingredients of O. africana is relatively cheap and they are appropriate for an integrated, sustainable community based primary health care program. Therefore, the use of the products of O. africana an indigenous plant to Ethiopia for the impregnation of mosquito nets must be seriously considered in malaria control and elimination effort currently in place. This is particularly relevant because of the ever increasing failure of the conventional residual sprays with DDT or other insecticides for malaria vector control in Ethiopia [4,7].

The results of this study indicated that the plant-based compounds such as octadecadienoic acid and dibutyl phthalate have a potential to be developed as an effective alternative to conventional synthetic insecticides for the control of mosquitoes. The use of active compounds isolated from O. africana can be used for mosquito net impregnation for adulticidal activity against mosquitoes and can reduce man-vector contact leading to transmission of malaria. The fact that the purified fractions of O. africana (B2’O and B2’’O) were more potent than the standard WHO recommended 0.05% lambdacyhalothrin, in their toxicity to adult An. arabiensis mosquitoes, is a good indication of their potential for use as mosquito control agents.

The authors would like to thank the Department of Microbial, Cellular and Molecular Biology and Thematic Research of Malaria and other Parasitic Diseases for their financial assistance. We are grateful to Dr. Birkinesh Ameneshewa, Integrated Vector Management Protection of Human Environment at WHO Regional Office for Africa WHO-IST-ESA Harare, Zimbabwe, for providing the WHO test kit for the bioassay test work. We thank to Dr. Asfaw Debella, the Centre for Traditional Medicine and Drug Research, Ethiopian Public Health Institute, for availing their research laboratories for the plant extract isolation and purification. We also thank to Mr. Yehualashet Belete for assist in plant extract isolation and Mr. Fitsum Tesfaye for assist about details of rearing An. arabiensis during the experiment.