Analysis by GC (Ir), GC/MS and mosquito repellent effect of essential oils against Anopheles gambiae: Case of stem bark of Sterculia tragacantha Lindl (Sterculiaceae) from Côte d'Ivoire

The objectives of this study were to determine the chemical composition and to evaluate the mosquito repellent property of essential oils extracted from the bark of the stem of Sterculia tragacantha Lindl on Anopheles gambiae, vector of malaria. The essential oil of the stem bark of S. tragacantha Lindl is extracted by hydrodistillation and their chemical compositions were identified by GC (Ir) and GC-MS. This oil has been tested on sensitive « kisumu » strains of A. gambiae adults, according to the World Health Organization (WHO) guidelines for laboratory testing of the repellency of chemicals. The extraction yield obtained from this oil is 0.12%. By means of GC (Ir) and GC-MS, 29 compounds representing 88.4% of the HE were identified. β-elemol (40.54%) is the major compound of this oil. It is rich in sesquiterpenes (79.47%) with a predominance of oxygenated sesquiterpenes (55.86%). As for the oxygenated monoterpenes, they represent 8.93%. The repellency test against adult female Anopheles, revealed that the EO of S. tragacantha (10%) has repellent properties against mosquitoes. Indeed, it induces an average reduction of 98.2% of the blood meal (TIRS) of mosquitoes on guinea pigs with a complete protection time (CPT) of 5 h compared to 3 h 30 min for the natural reference substance (Cymbopogon citratus). In view of its extractive value and prolonged repellent properties, S. tragacantha could be a good alternative in the vector borne diseases control if used as natural repellents for skin application.


INTRODUCTION
In recent years, global health authorities have alerted us to the worrying upsurge in vector-borne diseases.
These pathologies, transmitted from one individual to another via a vector, are at the origin of many historical pandemics. Today, malaria is the most important of these diseases in such a way that 3.2 billion people are at risk, representing half of the world's population (WHO, 2015). It thus constitutes a real public health problem for the underprivileged tropical areas of Africa, Asia and Latin America. Children under 5 and pregnant women are the most vulnerable because, for example, a child dies from malaria every 30 s in Africa (WHO, 2019). In Côte d'Ivoire, malaria accounts for approximately 32.53% of all causes of mortality (Kouadio et al, 2006) and in 2020 remains the leading cause of consultations in health centers (MSHP, 2020). Malaria is an infectious parasitosis caused by a protozoan Plasmodium transmitted to humans by an infected female mosquito of the genus Anopheles (Pasteur Institute, 2021). The current national strategy to eradicate this disease in Côte d'Ivoire is based on vector control which consists of the distribution and use of long-acting insecticidetreated mosquito nets and spraying intra-domiciliary pesticides in highly endemic areas to supplement the use of impregnated mosquito nets (MSHP). Unfortunately, these synthetic chemicals have harmful effects on humans and their environment because of their persistence in the environment (Corbel et al., 2009;Fradin and Day, 2002;Paluch et al., 2010). But overall, they may be faced with limited action following resistance developed by vectors (Chandre et al., 1999;Lundwall et al., 2005;Katz et al., 2008); hence, the interest in finding other arsenals of natural active ingredients. Our choice fell on Sterculia tragancatha, a plant traditionally used in the south-east of Côte d'Ivoire in the lagoon region as an anti-mosquito repellent in order to extract its essential oil, to study its chemical composition and to compare its repellent activity to that of Cymbopogon citratus taken as a natural reference substance because it is recognized for its insect repellent properties (Moore et al., 2007;Pushpanathan et al., 2006).

Plant and biological material
The plant material is composed of essential oils extracted from the bark of the stem of S. tragancatha collected on April 27, 2019 in the Department of Adzopé (latitude: 6.1069400 and longitude -3.8619400) and leaves of C. citratus (reference substance) collected in Bingerville (50 21'20.9 '' North,30 53'7.3 '') in the district of Abidjan on April 29, 2019. These plants were identified at the National Floristic Center of the University Félix Houphouët Boigny of Côte d'Ivoire.
The repellency was evaluated using a bio test on unfed females of Anopheles gambiae sensu lato aged 5 to 7 days after emergence (Kisumu strains sensible to insecticides) from the medical entomology laboratory of Swiss Center of Scientific Research in Ivory Coast (SCSR).

Extraction of essential oils
The leaves or husks dried for 3 to 4 days under air conditioning were used for extraction. Thus a quantity (0.5 kg) of the plant organ (leaves or shell) was placed on a grid, kept about 5 cm from the bottom of the casserole dish, above the water level in a pressure cooker with a capacity of 10 L containing 1 L of water connected to the Clevenger type device. Using a hot plate, the pressure cooker was heated to a boil. The duration of the extraction for each sample is 3 h 30 min from the time of obtaining the first drop of EO for each of the two plants. During this time the water vapor destroys the structure of plant cells, releases the molecules contained and carries these volatile components to the cylindrical column and then to the refrigerant where the gas phase condenses. The wateressential oil mixture obtained is collected in a test tube through the gasoline. The essential oil separated from the water by decantation is removed using a Pasteur pipette in a tube and then dried over anhydrous sodium sulfate (NaSO 4 ). Finally, the essential oils of each plant organ are collected separately in glass bottles and then stored in the refrigerator at 4°C. Essential oils were coded S and C, respectively for S. tragancatha and lemongrass.
The essential oil yield is the ratio between the mass of oil extracted and the mass of the plant material used for hydrodistillation (AFNOR, 2000). The yield (R) expressed as a percentage of the two organs was calculated by the following formula: R (%) = (Mass of essential oil extracted (g) / Mass of plant material (g)) × 100

Analysis by GC (Ir)
The analyses were carried out using a gas chromatograph of the Perkin-Elmer Clarus 500 type, equipped with a divider injector, two columns of nonpolar and polar silica 50 × 0.22 mm in internal diameter, film thickness 0.25 µm, filled with polydimethylsiloxane (C2H6OSi)n and polyethylene glycol, respectively, and two flame ionization detectors.
The operating conditions are as follows: the carrier gas is hydrogen; the pressure at the top of the column is 20 psi; the temperature of the injector and the detectors is 250°C. The temperature program is 60°C up to 220°C at a rate of 2°C/min, with a step of 20 min at 220°C; the injection was done by divider mode with a 1/60 ratio. The amount of sample injected is 0.5 µL from a solution containing 50 µL of EO in 350 µL of CDCl 3 .
The volatile substances were identified by comparing their retention indices (Ir) on polar and non-polar columns. These Ir were calculated from the retention times of the separated compounds and the linear n-alkanes (Browning, 1971).

Analysis by GC/MS
GC/MS analyses were performed using a Perkin Elmer autosystem XL chromatograph coupled to a Perkin Elmer Turbo Mass detector. The chromatograph is equipped with an automatic injector and an apolar column (Rtx-1) with an internal diameter of 60 m × 0.22 mm with a film thickness of 0.25 µm.
The carrier gas is hydrogen (1 mL) which exerts a column head pressure of 25 psi. The temperature of the injector is 250°C and that of the detector 280°C. The temperature is programmed at 60°C followed by an increase to 230°C at a rate of 2°C/min, with a plateau of 45 min at 230°C. The injection is done by split mode with a division ratio of 1/50. The amount of sample injected is 0.2 µL. Mass spectra are obtained using a quadrupole filter detector and ionization is done by electron impact with a 70 eV electron beam.

Identification of constituents of EO
The identification of EO compounds was made by: (1) The comparison of their GPC retention indices, obtained on polar and non-polar columns (Ira and Irp), determined as a function of the retention times of a series of n-alkanes by linear interpolation with those of the authentic compounds or of the data literature; (2) The correspondence of their spectral data with those of commercial mass spectra libraries and the laboratory (König et al., 2004;Adams, 2007;McLafferty, 1994).
For the analysis of the spectroscopic data, the AMDIX32 processing software was used. The Automatic Mass Spectral Deconvolution and Identification System (AMDIS) allows you to automatically find any of a set of target compounds in a gas chromatography/mass spectrometry (GC/MS) data file. The program first deconvolutes the GC/MS data file to find all of the separate components. Each of these components is then compared against a library of target compounds. The match factor between the target spectrum and the deconvoluted component spectrum is then reported, if it is above a user set value, and the matched target spectrum is called a library hit.

Biological test
The main experiments were carried out according to the guidelines of the World Health Organisation (WHO) related to laboratory testing of the repellency of chemicals (WHO, 2009(WHO, , 2013. The general objective of these laboratory studies is to estimate the effective dose of a repellent and the complete protection time provided by a repellent after application to the skin by determining the specific objectives which are: (1) The dose-response and effective dose of the repellent; (2) Protection (p); (

3) Full protection time (CPT)
This test is preferably performed on human subjects (final users of the repellant) in order to obtain results that are relevant to the actual conditions of use (WHO, 2013). Laboratory animals were chosen in this study to perform the first tests for safety and then apply to human skin if the product is found to be effective.

Mosquito breeding
Mosquito rearing was carried out according to the standard procedure described in the book for participant guide entitled Malaria Entomology and Vector Control (WHO, 2014). Indeed, instead to use wild strains which were likely in contact with repellents or synthetic insecticides and might carried some resistances genes, sensible strains from the medical entomology laboratory of SCSR were used. Mosquito larvae were reared with ground kibble for cats and then kept at 27°C plus or minus 3°C and at 80 plus or minus 10% relative humidity under a photoperiod at 12:12 (night: day). The adults were kept in cages, and fed with a 10% sucrose solution ( Figure 1A), but 24 h before the experiments, the sucrose solution was removed from the female mosquito storage cages ( Figure 1B).

Repellency test
Initially, the readiness of mosquitoes to land and/or probe was assessed by holding an untreated Guinea pig abdominal skin (that is, treated with simple petroleum jelly) kept still on top of a first cage, whose shaved face was exposed to mosquitoes for 30 s or until that 2 to 5 landings/soundings are counted. The procedure is repeated with another untreated animal in a second cage. If this landing and/or sounding level is not reached in either cage, the experiment must be abandoned (WHO, 2013).

Repulsion test
Each essential oil was used to make two ointments of 5 and 10% concentration which are tested to identify the most effective dose. Ointments formulated with the EO of S. tragancatha have been coded PS 5 and PS 10 . Those designed with the essential oil of the reference substance C. citratus at concentrations of 5 and 10% noted PC 5 and PC 10 . So for each test, 10 unfed female mosquitoes were placed in different cages. A 5×8 cm 2 area of abdomen hair was removed from laboratory-reared male Guinea pigs weighing an average of 400 to 500 g then washed, ethanol cleaned and dried. PC 5 , PC 10 , PS 5 and PS 10 ointments were applied to different animals. Each treated animal was kept on top of the cage in which the treated face is exposed to mosquitoes for 7 h. The number of bites was counted by examining the abdomen of all mosquitoes for blood fed. Thus, the number of bitches in the trials was recorded every minute for the first 15 min and then every 15 min for 7 h. Each test was repeated 2 times while using new mosquitoes and a new animal (WHO, 2009(WHO, , 2013.

Data analysis
The parameters for evaluating the effectiveness of the products are determined as follows (WHO, 2009(WHO, . 2013): (1) The dose-response lines and the effective doses (ED) of a repellent corresponding to 50% (ED50) and 99.9% (ED99.9) of protection against landing and/or detection of mosquitoes. They are estimated by calculating the protection P.
(2) The protection P offered by the ointments for each dose was calculated according to the following formula: where T: number of mosquitoes on the treated animal and C: number of mosquitoes on the control animal.
(3) Full Protection Time (CPT) is calculated as the number of minutes from the time of repellent application to the first landing and/or polling of the mosquito.
For the calculation of these parameters, data processing software was used. This is Kaplan Meier's online survival function calculator.
Entomological standards of efficacy are defined by the WHO and state that: (1) If a product has protection P ˂ 90% then it is ineffective.
(2) If a product has P ≥ 90% protection then it is effective.
(3) If several products have P ≥ 90% protection, the most effective at the longest CPT.

B
Each cage contains 10 females Anopheles gambiae subjected to a 24 h

Extraction yield
The essential oil extracted from the stem bark of S-coded S. tragancatha is light yellow with an aromatic odor. This oil is liquid at room temperature. The oil is extracted with a yield of 0.12%. According to AFNOR standards (2008), the stem barks of S. tragacantha are rich in essential oil. The extraction yield obtained in our study is lower than that reported by Aboaba et al. (0.73%) working on the species from Nigeria.
These dominant compounds identified in the study are different from those reported by Aboaba et al. (2016). They are hexahydrofaresylacetone (32.5%), viridiflorol (18.9%) and 1,8-cineole (14.79%). This difference can be linked to several factors such as climate, harvest period and area, etc.

Repellent activity of the essential oil of S. tragancatha (S)
The results of the mosquito repellent activity tests carried out using the product P S5 and P S10 are shown in Table 3. In view of Table 3 and Figure 4, the protection offered by P S5 is low, very rapidly falling over time. Indeed PS 5 offers a protection of 60 to 50% which extends over 1 h 30 min and drops to 30% in 3 h before stabilizing at 20% from 3 h 30 min until the end of the experience. PS 5 has a protection (P) of 32.5% and zero CPT (Table 4).
For P S10 , the 100% protection period lasts for 300 min or 5 h which corresponds to his CPT. During this time the unfed mosquitoes remained calm and away from the animal. The second period corresponds to 95% protection. PS 10 has a medium protection of 98.12% while 7 h (Table 4).
In view of the efficacy parameters P S10 is more effective than PS 5 . Indeed according to WHO standards, the most effective product has a medium protection ≥ 90% and the longest CPT.

Repellent activity of essential oil of C. citratus (reference substance)
The results of the mosquito repellent activity tests carried out using the P C5 and PC 10 products are shown in Table  5. Table 5 and Figure 5 show the rapid drop in protection offered by P C5 over time. Indeed, it offers 50% protection for the first 30 min and drops to 45% (from 30 min to 1 h) then to 20% (1 to 5 h 30 min) to stabilize at 10% (6 to 7 h). P C5 has an average protection percentage of 26.87%, a zero full protection time and an effective time of 30 min ( Table 6).
As for P C10 , its 100% protection extends over 210 min or 3 h 30 min which corresponds to its Full Protection Time (CPT = 210 min). This protection drops to 95% and is maintained until the end of the experience. The medium protection of P C10 is 97.5% (Table 6).
In short, P C10 is more effective than PC 5 because it has IrapExp / Bd: Retention index of experimental Kovats on a non-polar column calculated using C7-C28 nalkanes / literature databases (Kondjoyan and Berdague, 1996). IrpExp / Bd: retention index of experimental Kovats on a polar column calculated using C7-C28 n-alkanes / literature databases (Kondjoyan and Berdague, 1996). a high protection (97.5%) compared to that of P C5 (26.87%) and better protection with CPT. 210 min versus 0 min for P C5 on adults of A. gambiae s.l.
Comparative study of the persistence of the repellent effect of PC 10 and PS 10 ointments Insect repellent tests have shown that PS 10 has a remarkable repellent effect like that of the reference substance (PC 10 ) on adults of A. gambiae s.l because their protection (P) is greater than 90%. However, PS 10 is more persistent than PC 10 because it has 100% complete protection (CPT) which lasts 5 h 30 min while PC 10 provides the same protection for 3 h 30 min ( Figure 6). However, according to WHO standards, the persistence of the effectiveness of a repellent is linked to its CPT; the higher it is, the more effective the repellent. This Percentage Figure 4. Evolution of the repellency of P S5 and P S10 as a function of time.
persistence of the repellent effect of PS 10 (essential oil of S. tragancatha) against PC 10 (essential oil of C. citratus) is linked to the chemical composition of its essential oil in particular to molecules such as myrcene (0.37%), limonene (2.32%), α-copaene (7.94%), (E) -caryophyllene (8.14%) and β-elemol (40.54%) recognized for their repellent property (Miora, 2017). The synergistic action of these compounds could justify the activity of PS 10 . This is supported by Odalo et al. (2005) by showing that the individual repellent activity of compounds of some oils from Kenya on A. gambiae is below the repellent activity of total essential oil. This PS 10 activity is also linked to oxygenated sesquiterpenes, in particular the β-elemol, the major compound, the repellent effect of which has  Figure 6. Comparison of the protection time of the HE of PC 10 and PS 10 already been proven by Nerio et al. (2010). Indeed, PS 10 is rich in less volatile oxygenated sesquiterpenes, which explains its prolonged action compared to the more volatile oxygenated monoterpenes contained in PC 10 (Kaloustian et al., 2012) (Figure 7). The repellent property observed previously is explained by the fact that the compounds are for the most part endowed with an aromatic property that can neutralize body odors (lactic acid, carbon dioxide, ammonia, carboxylic acids, oct-1-en-3-ol and other sweat compounds) that female mosquitoes use to locate their prey (Guillaumot, 2009;Keswani and Bellare, 2006;McBride, 2016). These volatile substances confuse the female Anopheles by inactivating her receptors just enough to disturb her, or even render them insensitive to odors that she would find attractive, according to Vosshall's team (Pellegrino, 2011).

Conclusion
The essential oil of the stem bark of S. tragacantha (S) was extracted by continuous hydrodistillation for 3 h with a yield of 0.12%. S is light yellow in color with an aromatic odor and liquid appearance at room temperature. Compounds (29) representing 88.4% of the EO have been identified, including a major compound βelemol (40.54%). HE is rich in sesquiterpenes (79.47%) with a predominance of oxygenated sesquiterpenes (55.86%). As for the oxygenated monoterpenes, they represent 8.93%. The essential oil of S. tragacantha (S) is endowed with repellent property at 10%. At this dose, it offers a protection of 98.12% on adult females of A. gambiae s.l and TPC at 5 h versus 3 h 30 min for the natural reference substance. This activity is due to the presence of certain molecules such as myrcene (0.37%), limonene, α-copaene, (E)-caryophyllene and β-elemol. This search is a scientific evaluation of the repellent activity against adults of A. gambiaes s.l of the essential oil of the stem bark of S. tragacantha. Based on the good extraction performance and proven repellent properties against A. gambiae s.l, S.tragacantha should be strongly recommended for the development of natural skin repellants after in deep further analysis.