The study was conducted at the International Center for Research and Development on Livestock in Subhumid Areas (CIRDES) located in the city of Bobo-Dioulasso (11˚11'00'' North and 4˚17'00'' South) in Burkina Faso. All flies were supplied by the CIRDES insectary. The CIRDES colony is maintained under controlled environmental conditions of 25˚C ± 1˚C and 75% ± 5% relative humidity.

Glossina palpalis gambiensis (Vanderplanck, 1949) was the tsetse species used in this study. All flies were supplied by the CIRDES insectary where they were maintained under controlled environmental conditions of 25˚C ± 1˚C and 75% ± 5% relative humidity. For each experimental exposure, newly emerged teneral flies were selected. Flies are routinely fed using trays containing bovine blood covered with silicone membranes. A heating system maintains the plate temperature between 40˚C and 42˚C, ensuring a blood temperature of 36˚C - 37˚C beneath the membrane. The bovine blood used for tsetse feeding is collected at the Bobo-Dioulasso refrigerated slaughterhouse. After collection, the blood is defibrinated and subjected to both bacteriological and biological quality-control tests to confirm its suitability for fly consumption. It is subsequently irradiated using a Cesium-137 irradiator and stored under cold conditions until use.

During this study, we used three essential oils derived from two locally available plant species: Cymbopogon citratus (Cc) essential oil, Mentha piperita (Mp) essential oil, and a combined formulation (Cc_Mp), consisting of 80% Cc and 20% Mp, as defined in previous investigations [15]. This combination aimed to improve the effectiveness of the oils’ antioxidant effects while ensuring the pleasant aroma of these aromatic plants [15].Cymbopogon citratus DC. (lemongrass) belongs to the Poaceae family, which comprises approximately 660 genera and 9000 species [16]. It is a perennial, unbranched, lemon-scented grass that grows in dense clumps, characterized by light green, pubescent, strongly aromatic, tapering leaves united in a sheath along part of their length, with hyaline margins bearing numerous small, apically oriented teeth. Its subterranean system consists of a bulb or rhizome. The essential oil of C. citratus is primarily composed of Geranial and Neral [15].

Mentha piperita L. (peppermint) is a perennial, highly aromatic herbaceous plant belonging to the Lamiaceae family [17]. Mints have been used since antiquity and play a significant role in traditional therapy. As diffusible stimulants and sedatives, they are reputed to alleviate nervousness and related disorders. Their biological activity and characteristic scent derive largely from their essential oils [18]. The essential oil of M. piperita is mainly composed of Menthol, Menthone and Menthofuran [15].

All essential oils used in this study were produced and supplied by the Laboratory of Chemistry and Natural Substances at Nazi Boni University. Oils were extracted from dried leaves by hydrodistillation (steam distillation) using a Clevenger-type apparatus.

The oils were diluted according to the principle of conservation of mass in the solvent, using the equation Ci × Vi = Cf × Vf, where Ci represents the initial concentration of the pure essential oil or mixture, Vi the volume required to obtain the desired concentration, Cf the target concentration, and Vf the final volume of solution. Serial dilutions of each essential oil and the combined formulation were prepared in dimethyl sulfoxide (DMSO) to assess their insecticidal activity against tsetse flies. For each oil, three concentrations (100%, 50%, and 25%) were prepared through stepwise dilution. A volume of 200 µl per oil and per concentration was prepared in glass vials for the insecticidal assays. After dilution under fume hoods, all solutions were stored at 4˚C - 8˚C until use.

The insecticidal properties of essential oils and their combinations were evaluated by tarsal contact of tsetse flies on impregnated glass microfiber paper. Before each exposure, the flies were lightly anesthetized in a freezer for five (05) minutes. Each fly was exposed for five seconds on the impregnated paper using soft tweezers; corresponding to the average contact time of tsetse flies on impregnated screens [19]. Pieces of glass microfiber paper with a surface area of 4.3 cm² were impregnated with 49 microliters of solution per concentration and per oil. Impregnation was carried out just a few minutes before exposure. At each repetition, the tsetse flies were exposed to the different essential oils one after the other. However, for each oil, exposure to the different concentrations was carried out simultaneously. The experiment was repeated three (03) times. Ten (10) males of tsetse were used per repetition and per concentration. One repetition involved exposing the tsetse flies to the different essential oils and the different controls (positive and negative). A screen impregnated with Permethrin was used as the positive control and a piece of glass microfiber paper impregnated only with DMSO was the negative control. After exposure, the tsetse flies were placed in Roubaud cages by concentration and by essential oil and transported from the handling room to the storage room. The parameters measured were the number of flies knocked down at different times after exposure: 1 h, 2 h, 3 h, and 4 h; and mortality 24 h after exposure.

The experiment consisted of releasing tsetse flies into the release cage and observing their behavior in response to the treatments. Three (03) test treatments consisting of: Cc vs Untreated Hand (Cc/UH); Mp/UH; and Cc_Mp/UH and two (02) control treatments (Empty/Empty and UH/UH) were the different combinations highlighted. The purpose of the controls was to demonstrate the reliability of the device by measuring the activation of flies in response to odors with Empty/Empty and untreated human hands in order to compare the effects of the treatments on fly activation. The treatments with essential oils consisted of soaking the palm of one hand with 50µl of essential oil and placing it on one of the collection boxes, while the other untreated hand (UH) was placed on the other box of the device. A repetition with each essential oil involved treating both hands one after the other in each test, which minimized the “arm” effect in attracting flies. The Vacuum/Vacuum control treatment was performed at the beginning and end of each repetition with the EOs.

For each test, twenty-two-day-old male tsetse flies that had never fed on blood were released for twenty-five minutes (latency time + observation time). Knowing that tsetse flies are naturally attracted to light, a light was shone on the “arms” of the device at the collection boxes five (05) minutes after the flies were released in order to stimulate their natural activation, and the observation time for the behavior of the tsetse flies according to the treatments was twenty (20) minutes. The number of activated flies (i.e., those that moved from the release cage to the glass tunnels) was recorded per arm in order to calculate the activation and attraction of Glossina palpalis gambiensis according to the three essential oil treatments. The number of flies knocked down according to the different treatments in the release cage was also noted in order to consider the number of potentially active flies in the evaluation of tsetse flies’ activation. Activation corresponds to the number of flies counted in the two collection arms relative to the total number of potentially active flies released into the cage. Attraction (or preference) for an essential oil is the ratio between the number of tsetse flies counted in the “test arm” (containing the treatment with that oil) and the total number of flies activated in both “arms”. To do this, one of the handler’s hands was impregnated with the oil to be tested and the other hand, which was not treated (UH), was considered the control.

As flies are diurnal insects, all tests were conducted during the day between 8 a.m. and 1 p.m., and the release cage was covered with a black cloth to create darkness inside the release cage, which increased the flies’ natural attraction to the light emitted from the collection boxes. For each essential oil, the tests were repeated three (03) times.

The arm-test repellency assays were conducted using a single adult human volunteer who participated voluntarily after providing informed consent. According to institutional guidelines, formal ethical clearance was not required for this type of bioassay.

All statistical analyses were performed using R.4.4.2 software [20]. The Shapiro test was used to test the normality of the data, and Pearson’s chi-square test was applied at a 5% significance level. Statistical analyses of the effects of treatments on knock-downs (KD) were performed using the non-parametric Kaplan-Meier estimator. Multiple pairwise comparisons were made using the (glht) function of the (Multcomp) package when there was a significant effect of treatment or concentration on fallen tsetse flies or mortality.

Statistical analyses of the effects of treatments on fly activation and attraction were performed using the generalized linear binomial model, followed by likelihood ratio tests (LRT) to evaluate the effects of different treatment combinations on tsetse fly activation. The analysis of tsetse fly attraction was performed by “combination” according to the treatment, and odds ratios were calculated to compare the rate of tsetse flies that were attracted to the “treated arm” compared to the “untreated arm” according to the different combinations tested.

Results show that all essential oils tested exert knock-down (KD) effects on Glossina palpalis gambiensis(Figure 1). The highest knock-down rates were observed at the highest concentration (100% or 10,000 ppm). These rates were 16.1%, 48.4%, and 45.2% for Cymbopogon citratus (Cc), Mentha piperita (Mp), and their combination (Cc_Mp), respectively. The KD rates recorded in the controls were 9.6% and 100% for the negative control (DMSO-impregnated glass microfiber paper) and the positive control (impregnated screen), respectively. A low KD effect on tsetse flies exposed to essential oils compared to the positive control (impregnated screen) was observed (P < 0.0001). The KD probability curves show that the sensitivity rate of the essential oils tested on tsetse flies is very low, as the tsetse flies that fell in the oil treatments were spread out over time (1 h, 2 h, 3 hours, and 4 hours) of KD observation, compared to 100% of tsetse flies knocked down in the positive control group from the first hour of observation. The analyses showed an interaction between the knock-down effect and the treatment (P < 0.001). However, no significant interaction was observed between the knock-down effect and the concentration of essential oils, nor between the interaction of concentration and treatment (P= 0.6). A multiple comparison between the different essential oils tested showed a significant difference between the essential oil of C. citratus and that of M. piperita (P< 0.01). A significant difference was also observed between C. citratus essential oil and the combination essential oil (Cc_Mp) (P = 0.01). However, no significant difference was observed between M. piperita essential oil and the combination essential oil (P= 0.99).

DMSO: Dimethyl Sulfoxyde, Ecran: Impregnated screen, Cc: Cymbopogon citratus, Mp: Mentha piperita, Cc_Mp: Combination of Cymbopogon citratusand Mentha piperita.

Figure 1. Probability of knock-down in glossina according to essential oils and concentrations.

The insecticide test results for essential oils (Figure 2) show that all the essential oils tested cause mortality in tsetse flies. With the essential oils tested (for all concentrations combined), mortality rates for G.palpalis gambiensis ranged from 29.0% to 61.3%. The highest mortality rate recorded on G.palpalisgambiensis was obtained with M.piperita essential oil at a concentration of 100%. Mortality rates of 16.1% and 54.8% were recorded for the negative control and positive control, respectively. Significance tests at the 5% threshold revealed that there was a highly significant effect of all treatments on fly mortality (P < 0.001) while there was no significant effect between concentrations (P = 0.65) or interaction between concentrations and treatments (P = 0.16) on the different mortality rates. A multiple comparison between essential oils showed that there was no significant difference between the different essential oils tested on the mortality rates of Glossinapalpalisgambiensis, with P (Cc_Mp/Cc) = 0.82, P (Mp/Cc) = 0.99, and P (Mp/Cc_Mp) = 0.55, respectively.

Ctrl_N: negative control, Ctrl_p: positive control, Cc: Cymbopogon citratus essential oil, Mp: Mentha piperita essential oil, Cc_Mp: Combination of Cymbopogon citratus and Mentha piperita essential oil.

Figure 2. Effect of different concentration of essential oils on the mortality of tsetse.

The activation of tsetse flies in the experimental device was evaluated according to treatments with essential oils (Cc; Cc_Mp; and Mp) and controls (Empty = Ctrl1; and Untreated hand = UH = Ctrl2). The results on activation (Figure 3(a) and Figure 3(b); and Table 1 and Table 2) show that tsetse flies were activated in the device regardless of the type of combinations tested. With all combinations, activation rates ranged from 46.4% to 63.9% (Table 2). The presence of essential oils in the combinations resulted in lower activation rates compared to the controls (Figure 3(b)). The presence of C.citratus EO resulted in a decrease in tsetse fly activation to below 50% ( X ¯ = 4 6 . 4 6 % ). Fly activation was higher in the vacuum ( X ¯ = 6 3 . 1 5 % ) and untreated hands ( X ¯ = 6 3 . 9 9 % ) compared to the essential oils tested (Table 2).

A multiple comparison showed that there was a significant difference between all the test treatments combined on the activation of Glossinapalpalisgambiensis compared to the different control treatments (Treated/Ctrl1: P < 0.01; Treated/ Ctrl2: P = 0.02). However, there was no significant difference between the two controls (P = 0.98) or between the essential oils tested on the activation of glossina in the experimental device, with P > 0.05 respectively (Table 1). Furthermore, Cymbopogoncitratus essential oil showed a highly significant effect on activation compared to the different controls combined (P = 0.01), but there was no significant difference in the repellent effects of M.piperita essential oil and the combination essential oil on fly activation compared to all controls (Table 1).

Panela. Comparison of the activation rate of all control against different treatments with essential oils; Panelb. Comparison of the activation rate of all treatment with essential oils against different control arm. Ctrl_N: negative control, Ctrl_p: positive control, Cc: Cymbopogoncitratus essential oil, Mp: Menthapiperita essential oil, Cc_Mp: Combination of Cymbopogoncitratus and Menthapiperita essential oil.

Figure 3. Comparison of activation rate of G.palpalisgambiensis between the treatments.

Preference or attraction is the ratio of the number of tsetse flies attracted to the treated arm to the total number of tsetse flies attracted to both arms of the experimental device. The preference of tsetse flies in choosing between the arms (“test arm” and “control arm”) was assessed based on combinations of each essential oil with an untreated control (Cc/UH; Cc_Mp/UH; Mp/UH). Fly releases were carried out using only control treatments (untreated or empty hand) to compare the effectiveness of the two arms of the olfactometry device. The preference results (Figure 4 and Table 2) showed that with all the essential oils tested, preference rates ranged from 21.7% to 52.6% (Table 2). In fact, all essential oils were less preferred except for M.piperita EO (Table 2). The lowest attraction rate (Rate =21.7%) was obtained with Cymbopogoncitratus EO.

Table 1. Comparison between the treatments on the activation of Glossinapalpalisgambiensis.

SE = standard error; OR = Odds ratios; Cc: Essential oil from Cymbopogoncitratus; Mp: Essential oil from Menthapiperita; Cc_Mp: Combination of Cymbopogoncitratus and Menthapiperita essential oil; Ctrl1: Control with empty box; Ctrl2: Control with untreated hand.

Table 2. Average activation and attractivity rate of glossina according to the treatments.

Cc: Essential oil from Cymbopogoncitratus; Mp: Essential oil from Menthapiperita; Cc_Mp: Combination of Cymbopogoncitratus and Menthapiperita essential oil; Ctrl1: Control with empty box; Ctrl2: Control with untreated hand.

Multiple comparisons showed differences between the oils themselves and between the oils and controls on preference (Table 3). There was a significant difference between Cymbopogoncitratus EO and the controls in terms of tsetse fly preference, and C.citratus essential oil was slightly less preferred than the two controls (with Ctrl1/Cc: P < 0.01; OR = 3.6; Ctrl2/Cc: P < 0.01; OR = 4.0). Furthermore, we noted a significant difference in the attractiveness of tsetse flies between C.citratus and M.piperita essential oils (P = 0.04; OR = 3.6). However, there is no significant difference between Cc and Cc_Mp (P = 0.59; OR = 1.8) or between Mp and Cc_Mp (P = 0.44; OR = 1.9). The analysis showed a significant difference between Cc_Mp and Ctrl1 (P < 0.01; OR = 1.9) but not with Ctrl2 (P = 0.1; OR = 2.1). Furthermore, there was no significant difference between Menthapiperita essential oil and the two controls (P ˃ 0.05).

Table 3. Comparison between different treatments in the preference of de G.palpalisgambiensis.

SE = Standard Error; OR = Odds ratios; Cc: Essential oil from Cymbopogoncitratus; Mp: Essential oil from Menthapiperita; Cc_Mp: Combination of CymbopogoncitratusandMenthapiperita essential oil; Ctrl1: Control with empty box; Ctrl2: Control with untreated hand.

UH: Untreated hand; Mp: hand treated with Mentha piperitaessential oil; Cc_Mp: hand treated with combined essential oil from Mentha piperita and Cymbopogon citratus.

Figure 4. Preference of glossina according to the arm of test.