Mosquitoes (Diptera: Culicidae) are widely known for their role as vectors of disease-causing pathogens in the world [1]. Culex quinquefasciatus are involved in transmitting viral, bacterial and protozoan diseases around the world [2]. The adults of C. quinquefasciatus prefer to inhabit areas where there is human dense settlement, larvae dwells in polluted standing water [2]. A study by Berticat et al [3] reported that mosquitoes of C. quinquefasciatus have developed multiple resistance towards carbosulfan and permethrin insecticides, and that has being caused by resistant gene mutation, wrongful application methods and longer use period of same insecticide. The conventional insecticides have also been reported to be of stable chemical structure with associated deleterious effects on humans and other non-targeted animals [4,5]. Specifically, the resistance shown by Culex mosquito larva and adults to insecticides have prompted researches on botanicals to discover new human and eco-friendly insecticides [5]. In addition to being easily biodegradable, non-toxic to other organisms, safer to use, easily available and cost effective [4], the botanical insecticides have been used as alternative methods to control vectors of biological importance [5].

Asteraceae family has a global distribution with 246 genera that are represented by 2300 species in South Africa, and 17 tribes are used for medicinal purposes [6]. The Asteraceae are commercial and traditionally known to produce pesticides, essential oils, ailment medicine, edible food, and some species are used as ornamentals. Cosmos bipinnatus Cav. is a half hardy annual herb, with finely cut and thread-like leaves, the solitary inflorescence with pink, purple, red or white flowers [7]. The traditional medicinal uses of C. bipinnatus Cav. flower include treatment of various diseases like jaundice, intermittent fever and splenomegaly, and while its triterpene alcohols such as helianol were reported to display anti-inflammatory activity [7]. The Basotho tribe in South Africa traditional make use of C. bipinnatus to manage headaches and stomach disorders, while the Afrikaans ethnic group use C. bipinnatus to control bed bugs and head lice [8]. On the other hand, the larvicidal and insecticidal properties of C. bipinnatus have not being scientifically reported.

The annual hard-woody Tagetes minuta L. shrub is used traditionally as a biological pesticide to control agricultural, medicinal and veterinary important insect pests [9]. Larvicidal and insecticidal effects of essential oils from various parts of T. minuta against Aedes aegypti, Anopheles stephensi, Hyalomma rufipes (tick) and Pediculus humanus capitis (head louse) have been documented [9,10]. Extracts from its floral and foliar parts have also been demonstrated to potentiate good insecticidal activities toward Phlebotomus duboscqi (sandfly) [11].

Apiaceae is represented by 78 genera with 368 species in the sub-Saharan region, and genus Foeniculum is found worldwide and adds to the 19 endemic Apiaceae genera in southern Africa [12]. The essential oils from F. vulgare seeds showed larvicidal activities against mosquitoes of Anopheles stephensi and Culex pipiens [13].

The present study evaluated leaf formulations of C. bipinnatus, F. vulgare and T. minuta for their secondary metabolite constituents, larvicidal, pupicidal and insecticidal activities against Culex quinquefasciatus mosquitoes.

Plant collection

Plants were collected from their plantation in March 2013 around Qwaqwa Township, eastern Free State Province, South Africa. They were identified and authenticated by Dr. AOT Ashafa of the Department of Plant Sciences, University of the Free State, Qwaqwa campus, and voucher specimens ModMed.2013/3, ModMed.2013/4, and ModMed.2013/5 for Cosmos bipinnatus, Foeniculum vulgare, and Tagetes minuta, respectively, were thereafter deposited in the University’s Herbarium for future reference.

The fresh leaves of each plant were dried to constant weight in Ecotherm oven (Laboratory Consumables Pty, RSA) at 40 °C prior to pulverizing into fine powder using electric blender (Nanning Mainline Food Machinery Company Ltd, China).

Extract preparation

The powdered samples of each material were divided into three portions of 5 g each and extracted with 150 mL each of distilled water, ethanol (95 %) and hexane, and kept on orbital Labcon Platform shaker (Laboratory consumables, PTY, Durban, South Africa) in lidded 500 mL flasks at 110 rpm for 48 h. The resulting infusion in each case was filtered and evaporated to dryness in a rotary evaporator (Cole-Parmer, Laboratory Consumables and Chemical Supplies Co. Ltd, China) in respect of the organic solvent extracts. The aqueous extracts on the other hand were concentrated over water bath (45 °C). The crude extracts obtained in each case were kept refrigerated (4 °C) prior to use.

Secondary metabolites screening

Adopting standard procedures [14-16], eight secondary metabolites were screened for.

Mosquito collection

Mosquito larvae were collected from Kroonstad in Free State of South Africa from standing water pools, at sites (27°38’27.30” S, 27°11’20.40” E and 1372 m latitude) and (27°39’36.30” S, 27°10’18.90” E and 1338 m latitude). Larvae and female mosquitoes were identified using identification keys as previously described [17]. Museum voucher specimens were prepared for fourth instar larvae (UFSCulex01/Mod2013) and adult female (UFSCulex02/Mod2013) type specimens and deposited at the Department of Zoology and Entomology Museum, University of the Free State, Qwaqwa campus. Mosquito fourth instar larvae were used for larvicidal bioassay evaluation according to WHO [18]. The other earlier larval developmental stages were allowed to develop into adults, while held at temperature of 27 ± 2 °C, humidity of 70 ±1% and a photoperiod regime of 14:10 h (light/dark). To maintain the larval colonies, the aquatic larvae were reared in dechlorinated water and yeast solution.

Larvicidal bioassays

Different extract concentrations (0.1, 0.5, 1, 2, 5, 10 mg/mL) were prepared into separate opened McCartney (20 mL) glass vial for each plant extract. While ten fourth instar larvae were separately pipetted into each graded concentration and served as experimental treatments, the control experiments were conducted using 5 mL of double distilled water without plant extracts. Experimental treatments and control were kept under same conditions that were used to maintain Culex mosquito larval colonies as mentioned above. The larval mortalities were counted and recorded after 24 h for analysis. Mosquito larvae were declared dead when it lay flat at the bottom of the vial without motion, or did not respond to any stimuli.

Pupicidal bioassay

Ten two-day old pupa were placed separately into each graded concentration (0.1, 0.5, 1, 2, 5, 10 mg/mL) of each plant extract and represented test treatments. Control experiments were conducted in similar way using 5 mL of double-distilled water. Experimental treatments and control were also kept under same conditions as mentioned above. Pupa mortalities were counted and recorded after 24 h for analysis. Mosquito pupas were declared dead when just floated above water surface without motion, or does not respond to any stimuli.

Insecticidal bioassay

Varying amounts (0.25, 0.5, 1, 1.5, 2 g) of dried powdered leaf material were used. A Whatman No-1 filter paper was cut into a circular (25 mm diameter), and placed under plastic vials cap (25 mm diameter) to fit. The graded concentrations of the powdered leaf were poured onto the filter paper and covered with perforated vial (65.66 mm diameter, area of 13546.55 mm²) in an inverted position. Ten two day old adult Culex mosquitoes were incubated inside the perforated plastic vial. The steps were repeated three times for every tested plant containing the dried plant material and the best two results were recorded at 6 and 24 h and used for analysis.

Data analysis

Probit analysis was used to determine the median lethal concentration (LC₅₀), median lethal dose (LD50), and mean knock-down (KD₅₀). Repetitive-measure of analysis of variance (ANOVA) test was used to determine statistical difference between concentration mortality counts.

Crude extract yield and secondary metabolite constituents

The percentage yield for crude extract of Cosmos bipinnatus, Foeniculum vulgare and Tagetes minuta were higher in aqueous extracts with 27.00, 23.10 and 13.60 %, followed by ethanolic extracts with 7.00, 11.00 and 7.40 % while hexane extracts had 1.20, 6.0 and 2.40 %, respectively ().

Saponins, flavonoids, steroids and terpenoids were common for C. bipinnatus and T. minuta extracts. Tannins were present in aqueous and ethanolic extracts of F. vulgare, and alkaloids were detected mostly in T. minuta and F. vulgare ().

Larvicidal activity

A similar pattern of complete larvicidal activity was observed at both 5 and 10 mg/µL concentrations of all the investigated extracts. Generally, the ethanol and hexane extracts exhibited larvicidal activity for F. vulgare with LC₅₀ values of 0.10 and 1.03 mg/mL, T. minuta with LC₅₀ values of 1.17 and 1.01 mg/mL, while C. bipinnatus had the least larvicidal activity with LC₅₀ values of 1.18 and 1.27 mg/mL respectively, against Culex quinquefasciatus mosquito larvae after 24 h of exposure (). The aqueous extracts of all plants had no fatal effects on the larvae at all test concentrations. For the hexane extracts, statistically significant differences existed between larval percentage mortalities at F_5,2 = 6.46 (p < 0.05) and also between concentration dependent mortalities at F_5,2 = 5.77 (p < 0.05) (). There was significant difference for ethanol extracts between plant concentration dependent mortalities at F_5,2 = 3.57 (p < 0.05), and no significant difference between larval percentage mortalities. A positive correlation was displayed between extract concentrations and larval mortality of larvae for the tested ethanol and hexane extracts (A and B).

Pupicidal activity

Except for hexane extract of C. bipinnatus and T. minuta, all other extracts of the two plants (at 5 and 10 mg/µL) elicited uniform and complete 100 % larvicidal activity against the mosquito pupa. Overall, the ethanol and hexane extracts displayed pupicidal activity for T. minuta with LC₅₀ value of 1.11 and 1.12 mg/mL, C. bipinnatus extracts had LC₅₀ value of 1.14 and 1.16 mg/mL, and F. vulgare with LC₅₀ value of 1.31 and 1.07 mg/mL, against Culex quinquefasciatus mosquito pupa after 24 h of exposure (). The aqueous extracts had no fatal effects towards the pupa at all tested concentrations. While the ethanol extracts had significant difference only between plant mortality at F_5,2 = 4.29 (p < 0.05), the hexane extracts had significant difference only between pupa percentage mortalities at F_5,2 = 3.73 (p < 0.05) (). Ethanol and hexane extract concentrations displayed positive correlation with percentage pupa mortalities (A and B).

Insecticidal activity

Although, T. minuta exhibited a dose-dependent knock-down rate with total and most prominent effects at 1.5 and 2 g/air concentration, complete knock-down effects were evidently displayed by F. vulgare at concentration range of 1-2 g/air (). After 6 h of exposure, F. vulgare leaves performed the most with an average rate of knock-down effect with KD₅₀ = 7.52 min at a knock-down range between 65 - 77 %, while T. minuta displayed at KD₅₀ = 8.64 min with a knock-down range of 43-55 %, towards adults of C. quinquefasciatus mosquitoes. Knock-down rate had significant difference between plant leaves with F_4,2 = 10.09 (p < 0.01), while there was no significant difference between plant dose concentration.  Also, it is noteworthy that leaves of F. vulgare and T. minuta killed all evaluated mosquito adults within 12 h of exposure with LD₉₉ = 0.25 g/air against C. quinquefasciatus mosquitoes adults. However, C. bipinnatus leaves had no knock-down or fatal effects on mosquito adults.

The aquatic immature larvae stage is recognised as the most vulnerable and best control strategy to effectively reduce mosquito population densities during infestations [2]. Ethanol and hexane leaf extracts of F. vulgare, T. minuta and C. bipinnatus displayed good larvicidal and pupicidal activities, and may be considered effective for larvae control since their LC₅₀ values were lower than 100 mg/mL [19]. C. bipinnatus and T. minuta ethanol leaf extracts contained terpenoids, flavonoids, saponins, steroids and tannins that play a role in plant defence against insect pests, and might have been responsible for larval and pupa deaths. Ethanol extracts of Citrus sinensis have displayed both larvicidal and pupicidal activities against Aedes aegypti, Anopheles stephensi and Culex quinquefasciatus mosquitoes [20], and similar bioactivities were observed for F. vulgare, C. bipinnatus and Tagetes minuta ethanol extracts in this study against larva and pupa of C. quinquefasciatus mosquitoes. F. vulgare seed essential oil extract was reported for larvicidal activity on both 2nd and 4th larval instar of Culex pipiens [21], and the non-polar compounds present in the hexane extract can similarly act on larvae of C. quinquefasciatus mosquitoes.

F. vulgare, C. bipinnatus and T. minuta aqueous leaf extracts had no fatal effects against C. quiquefasciatus larvae and pupa. Similar trends were observed in previous larvicidal activity studies, where ethanol and hexane leaf extracts of Cassia occidentalis and Lantana camara displayed better larvicidal activities toward Anopheles stephensi and Aedes aegypti larvae, respectively than aqueous extracts [21,22]. The aquatic immature life stages (larvae and pupa) prefer to inhabit water pools that are polluted [23], and such behaviour may act to assist larvae and pupa to develop metabolic defence strategies against plant aqueous extracts that they might have been exposed to previously. This observation may be attributed to the differences in polarity of the extractants used in this study [7,22].

Generally, the attributes elicited by all the investigated plants in this study may be adduced to their secondary metabolites as revealed by the results of the phytochemical analyses. The biologically active terpenoids of T. minuta extracts such as dihydrotagetone, tagetones, ocimenones and piperitone have been reported for possible synergistic larvicidal effects [9].  The phytonutrients of F. vulgare foliage extracts were also found to be toxic and effective against Anopheles dirus and Aedes aegypti mosquitoes [24]. These phytonutrients have been opined to penetrate mosquito integument to disrupt important metabolic reactions. Such disruption may deprive adult mosquitoes of oxygen and death resulting from suffocation could occur [25]. This was observed at higher investigated concentrations of T. minuta and F. vulgare in this study. Extracts or chemicals with respective KD₅₀ and LD₅₀ values of lower than 20 min and 80 g/air have been pharmacologically adjudged to be potent insecticides [5]. Therefore, T. minuta and F. vulgare leaves powder could be said to have exhibited rapid and effective insecticidal activity towards C. quinquefasciatus adult mosquitoes.

F. vulgare, C. bipinnatus and T. minuta leaf extracts exhibit larvicidal and pupicidal activities against aquatic C. quinquefasciatus larvae and pupa, but the aqueous extracts do not. Only F. vulgare and T. minuta dried extracts has insecticidal activity against adult C. quinquefasciatus mosquitoes. Further studies should be conducted to monitor and evaluate the mode of action at the molecular level for each plant extract to clarify the observed effects in this study. Since F. vulgare, C. bipinnatus and T. minuta plants as well as C. quinquefasciatus mosquitoes are widely distributed around the world, these plants have a potential to be used in local communities for mosquito control during breeding and infestation seasons, thus promoting the use of natural pesticides.