Salvia Spp. Essential Oils against the Arboviruses Vector Aedes albopictus (Diptera: Culicidae): Bioactivity, Composition, and Sensorial Profile—Stage 1

Mosquito-borne arboviruses diseases cause a substantial public health burden within their expanding range. To date, their control relies on synthetic insecticides and repellents aimed to control the competent mosquito vectors. However, their use is hampered by their high economic, environmental, and human health impacts. Natural products may represent a valid eco-friendly alternative to chemical pesticides to control mosquitoes, and mosquito-borne parasitic diseases. The aim of this work was to combine the chemical and sensorial profiles with the bioactivity data of Salvia spp. essential oils (EOs) to select the most suitable EO to be used as a repellent and insecticide against the invasive mosquito Aedes albopictus (Diptera: Culicidae), vector of pathogens and parasites, and to describe the EOs smell profile. To do this, the EOs of four Salvia species, namely S. dolomitica, S. dorisiana, S. sclarea, and S. somalensis were extracted, chemically analyzed and tested for their bioactivity as larvicides and repellents against Ae. albopictus. Then, the smell profiles of the EOs were described by a panel of assessors. The LC50 of the EOs ranged from 71.08 to 559.77 μL L−1 for S. dorisiana and S. sclarea, respectively. S. sclarea EO showed the highest repellence among the tested EOs against Ae. albopictus females (RD95 = 12.65 nL cm−2), while the most long-lasting, at the dose of 20 nL cm−2, was S. dorisiana (Complete Protection Time = 43.28 ± 3.43 min). S. sclarea EO showed the best smell profile, while S. dolomitica EO the worst one with a high number of off-flavors. Overall, all the EOs, with the exception of the S. dolomitica one, were indicated as suitable for “environmental protection”, while S. dorisiana and S. sclarea were indicated as suitable also for “Body care”.


Introduction
Mosquitoes (Diptera: Culicidae) are among the most serious threats for humans because of their ability to transmit viruses and parasites. In particular, the arboviruses dengue, yellow fever, chikungunya, and Zika have recently expanded their geographical distributions and caused severe disease outbreaks in many urban populations [1][2][3]. Since the transmission of these viruses depends on the presence of the competent mosquito vectors Aedes aegypti and Aedes albopictus [3][4][5], measures such as repellent-or insecticide-treated nets, indoor spraying, or personal protection measures are needed to protect people from mosquito-borne infections [6]. To date, mosquitoes are mainly controlled by synthetic insecticides and repellents, but besides the quick development of resistance by insect pests, their use is often prohibitively expensive, unsustainable, and it poses relevant risks to humans and environmental health [7].
In this regard, essential oils (EOs) for their effectiveness, minimal toxicity to mammals, and low impact on the environment [8][9][10] have been recognized among the best alternative to synthetic chemicals. However, despite their insecticidal and repellent properties, EOs still do not have the expected broad use. In fact, besides the composition variability and the high volatility, their strong smell prevents their widespread application [11,12]. For these reasons, the acceptance of EOs to the human sensorial system is an important feature for their success as an ingredient in commercial products for topical use or for environmental protection.
The genus Salvia comprises many easily cultivable species well-known in traditional medicine, all around the world [13]. Besides the medicinal effects, Salvia spp. EOs have also been extensively reported for their antimicrobial and antifungal activity [14,15] , as well as for their repellent and insecticidal properties [16][17][18][19].
The aim of this work was, therefore, to evaluate the efficacy as repellents and insecticides of the EOs extracted from four cultivated Salvia species, namely S. dolomitica Codd, S. dorisiana Standl., S. sclarea L., and S. somalensis Vatke, against the filariasis vector Aedes albopictus Skuse (Diptera: Culicidae). Besides, we described the EOs smell profile for their possible use as active ingredients in the formulation of products for the environment or for personal use.

Plant Material, EOs Extraction and GC-MS Analysis
The EOs were obtained from air-dried aerial parts of Salvia dolomitica Codd, Salvia dorisiana Standl., Salvia sclarea L., and Salvia somalensis Vatke (Lamiaceae) plants cultivated at the Centro di Ricerca Orticoltura e Florovivaismo, (CREA) (Sanremo, Italy) and collected during the Summer 2017 (Table S1). The EOs were obtained by extraction for 2 h in a Clevenger apparatus. After the hydro-distillation, the EOs were dehydrated using anhydrous sodium sulphate and stored at −4 • C until analysis. The EOs were chemically analyzed by gas chromatography-mass spectroscopy (GC-MS) by an Agilent 7890B gas chromatograph equipped with an Agilent HP-5MS capillary column (30 m × 0.25 mm; coating thickness 0.25 µm) and an Agilent 5977B single quadrupole mass detector (Agilent Technologies Inc., Santa Clara, CA, USA). Analytical conditions were as follows: carrier gas helium at 1 mL/min; injection of 1 µL (0.5% HPLC grade n-hexane solution); oven temperature programmed from 60 to 240 • C at 3 • C min −1 ; split ratio 1:25, injector and transfer line temperatures 220 and 240 • C, respectively. The parameters that were acquired were as follows: full scan, scan range of 30-300 m/z; scan time of 1 sec. The identification of the constituents was based on a comparison of the retention times with those of the authentic samples, comparing their linear retention indices relative to the series of n-hydrocarbons. Computer matching was also used against commercial (NIST 14 and ADAMS) and laboratory-developed mass spectra library built up from pure substances and components of known oils and Mass Spectra (MS) literature data [20][21][22][23][24].

Aedes albopictus Rearing
Adults of Aedes albopictus Skuse (Diptera: Culicidae) were obtained from eggs collected in the open field on Masonite strips put in black pots filled with 1 L of tap water. The strips were daily collected, transferred in 500 mL beakers, and submerged in tap water under room conditions (26 ± 2 • C; 60% relative humidity), photoperiod of 14:10 h (L:D), for eggs hatching. The emerged larvae were fed with cat food until pupation. The pupae (about 300 per cage) were put in cylindrical cages (Plexiglas, 35 d7 60 cm) with a front cotton access sleeve. The emerged adults were kept under room conditions and fed with sucrose solution (20%) [17,25].

EOs Larvicidal Activity
Ten newly fourth-instar larvae (0-24 h) were put in a 250 mL beaker with 0.1% Tween 80 water solutions of the EOs. The EOs were tested at 50, 100, 150, 200, 300, 400, and 600 µL L −1 . As a control, 10 larvae were put in 0.1% Tween 80 tap water solution. Four replicates for each treatment were performed. The mortality of the larvae was recorded after 24 h. During the tests, no food was given to the larvae [26]. Abbott's formula [27] was used to adjust the mortality percentage rates of the treatments on the basis of the controls' mortality.

Essential Oils Repellent Activity
The repellence of the EOs was evaluated by the human-bait technique [28] with some modifications. The experiments were performed during the summer in the above-described cages. The cages contained about of 300 8-12 day-old adults (sex ratio 1:1). The mosquitoes were starved for 12 h and were not blood-fed or exposed to any form of repellent. The tests were performed by ten volunteers not allergic to mosquito bites, and that had no contact with perfumed products on the day of the bioassay. All volunteers were informed about the experiment and provided their written consent. After rinsing the hands in distilled water, the volunteer's forearms were protected with thick fabric sleeves and the hands with latex gloves in which a dorsal square area 5 × 5 cm was cut open. The mosquito-exposed skin of one hand was treated with 100 µL of ethanol as a negative control. The other hand was treated with 100 µL of EO ethanolic solution at concentrations ranging from 0.02 to 200 nL cm −2 . After ethanol evaporation, the control hand was inserted inside the cage and exposed to mosquitoes for 3 min. Immediately after, the other hand was treated with the EO solution and, after ethanol evaporation, exposed to mosquitoes in the same cage. The number of probing mosquitoes was recorded by two observers. All the tests were performed between 8:00 and 10:00 am. The complete protection time (CPT) for the concentration of the EO of 0.2 µL cm −2 of skin was calculated by tests performed every 15 min until either two bites occurred in a single exposure period or one bite occurred in each of two consecutive exposure periods. The complete protection times were calculated on an average of six replicates. To verify the mosquitoes' readiness to bite, the control and the EOs treated hand were regularly interchanged during each test. We considered the test valid if at least 30 mosquitoes landed on the control hand and attempted to bite. If the number of probing was < 30, a new mosquito's cage was used [11]. The study was approved by the ethical committee of the University of Pisa (Comitato Bioetico dell'Università di Pisa).

Essential Oils Sensory Analysis
The sensory analysis of the EOs was performed by a panel of 10 assessors (four males and six females aged from 23 to 60 years) selected and trained for sensory analysis of foods (mainly wine, vegetal oils, and bakery products) and non-foods (mainly essential oils), according to the internal protocol of the Department of Agriculture, Food, and Environment (DAFE) of University of Pisa [29,30]. For this general training protocol, five training sessions specific for the assessment of Sage spp. were arranged until the assessors familiarized with the main descriptors useful for the characterization of aromatic plants. With this aim, during these training sessions, assessors were asked to identify and describe the smell profile of different solutions prepared by infusion (12 h, 25 • C of temperature, inert atmosphere) in hydroalcoholic solution (13% v/v) of the main aromatic plants, spices, different flowers, fruits, and fresh vegetables.
The smell assessment of the EOs tested was performed as a blinded test, in a quiet, well-ventilated room in the morning. Each assessor was provided with a filter paper (2 × 2 cm) soaked with 20 µL of EO (1%). To avoid cross-contamination, the five samples were assessed separately in the same panel session (15 min waiting between two tests). The assessors were provided with a specific non-structured parametric descriptive scoring chart, and described the main odors of each sample on the basis of descriptors ranked on a scale of 0-10 in terms of "Smell intensity", "Smell persistency", and, "Overall pleasantness" as hedonic parameters [31]. The assessors also evaluated the possible use (Body care or Environmental protection) of the EOs, as well as the main emotions (Familiar, Relax, Exotic, Repulsion) elicited by them.

Statistical Analysis
Data of the smell profiles assessments of EOs were analyzed by one-way ANOVA with the EO as factor. Equality of variances was checked before the analyses by the Levene's test. The averages were separated by the Tukey's b post-hoc test. Larvae median and lethal concentration to 95% of tested organisms (LC 50 , LC 95 ), and adult median and total repellent dose (RD 50 , RD 95 ) were calculated by probit regression. Differences between LC 50 , LC 95 , RD 50 , and RD 95 values were evaluated by the relative median potency (rmp). The complete protection time (CPT) data were processed by the Kruskal-Wallis test with the time of protection as a factor. The means were separated by Dunn-Bonferroni pairwise comparisons. All the analyses were performed by the SPSS 22.0 software (SPSS Inc., Chicago, IL, USA).

Essential Oils Chemical Composition
In total, 108 compounds were identified in the four Salvia spp. EOs, with an identification percentage ranging from 96.9% to 99.8% (Table 1).

Essential Oils Larvicidal Activity
The Salvia EOs showed a wide range of toxicity against the Asian tiger mosquito larvae, depending on the species. Overall, the toxicity of S. dorisiana EO (LC 50 = 71.08 µL L -1 ) was in line with the toxicity previously reported for other aromatic plants' EOs (LC 50 ranging from 35 to 194 µL L -1 ) (reviewed by Pavela, 2015), while a much lower toxicity against Ae. albopictus larvae was recorded for S. dolomitica, S. dorisiana, and S. somalensis. (LC 50 ranging from 315.52 to 559.77 µL L -1 ) ( Table 2). a , Concentration of the EO that kills 50% of the exposed larvae; b , concentration of the EO that kills 95% of the exposed larvae. Data are expressed as µL L -1 ; in bracket, confidence interval; df, degrees of freedom. P, significance level of Pearson goodness of fit test.
The comparison of the relative toxicity of Salvia spp. EOs against Ae. albopictus by rmp analysis of probits showed that the S. dorisiana EO was significantly more toxic to the mosquitoes' larvae than the other Salvia EOs. In addition, S. dolomitica was significantly more toxic than S. sclarea EO (Table 3).

Essential Oils Repellent Activity
All the Salvia spp. EOs showed a repellent activity against the Ae. albopictus adults. RD 50 values ranged from 0.56 to 5.03 nL cm −2 for S. dorisiana and S. somalensis, respectively while RD 95 values ranged from 12.65 to 8308.54 nL cm −2 for S. sclarea and S. somalensis, respectively (Table 4). The comparison of the relative toxicity of Salvia spp. EOs by rmp analyses did not show significant differences among the EOs, with the exception of S. somalensis EO that was significantly less repellent than the others (Table 5). The results of the complete protection time (CPT) assay indicated that the Salvia EOs applied to human skin at the dose of 20 nL cm −2 protect from the mosquito's bites for a time ranging from 4.60 ± 2.7 to 43.28 ± 3.43 min (for S. somalensis and S. dorisiana, respectively) ( Table 6). The Kruskas-Wallis test indicated significant differences in the duration of the protection by Salvia EOs (χ2 = 14.432; df = 4; P = 0.006). The Dunn-Bonferroni pairwise comparisons of the CPT values indicated that the protective effect of S. dorisiana was significantly longer-lasting than the one of the S. somalensis EO (Table 6).

Essential Oils Sensory Analysis
The panel tests indicated S. dorisiana as the EO with the highest "Smell intensity" (Figure 1), closely followed by S. dolomitica and S. somalensis as the EO with the lowest intensity. Such differences, however, were not statistically significant (F (4.25) = 2.12. p = 0.109). On the contrary, the assessors attributed to the different EOs a significantly different "Smell persistency" as well as "Overall pleasantness" (F (4.25) = 5.80, p = 0.002; F (4.25) = 10.83, p < 0.001, respectively).
As for smell persistency, the highest value was attributed to S. dorisiana, followed by S. dolomitica and S. sclarea EO, with the lowest values attributed to the S. somalensis EO (Figure 1). The best smell profile (highest value of overall pleasantness) was attributed to the S. sclarea EO, followed by the S. somalensis and the S. dorisiana EOs. The worst smell profile was attributed to the S. dolomitica EO that was characterized by a high number of off-flavors (Table 7).
Furthermore, S. sclarea and S. somalensis EOs showed the highest complexity in flavor descriptors (Table 7) referring to both family of vegetative and Spicy odors.
All the assessors associated the S. dolomitica EO to negative emotion "Repulsion". On the contrary, the other Salvia spp. EOs were associated with positive sensations "Familiar" and "Relax" together with an "Exotic" feeling.
As for the possible use of the EOs, the assessors indicated that all the Salvia spp. EOs, with the exception of S. dolomitica, were suitable for "Environmental protection", while only the S. dorisiana and S. sclarea were suitable also for "Body care" (Figure 2). The best smell profile (highest value of overall pleasantness) was attributed to the S. sclarea EO, followed by the S. somalensis and the S. dorisiana EOs. The worst smell profile was attributed to the S. dolomitica EO that was characterized by a high number of off-flavors (Table 7).
Furthermore, S. sclarea and S. somalensis EOs showed the highest complexity in flavor descriptors (Table 7) referring to both family of vegetative and Spicy odors.
All the assessors associated the S. dolomitica EO to negative emotion "Repulsion". On the contrary, the other Salvia spp. EOs were associated with positive sensations "Familiar" and "Relax" together with an "Exotic" feeling.
As for the possible use of the EOs, the assessors indicated that all the Salvia spp. EOs, with the exception of S. dolomitica, were suitable for "Environmental protection", while only the S. dorisiana and S. sclarea were suitable also for "Body care" (Figure 2).

Discussion
The use of EOs as insecticides and insect repellents is raising increasing interest for their low impact on environmental and human health, and for their perception by consumers as safe, natural products [32]. As for their use as repellents, however, in order for them to be successfully utilized as an ingredient in commercial products for topical or for environmental protection use, the EOs need to have a good level of acceptance to the human sensorial system. In this work, we observed that the Salvia EOs tested showed a wide range of toxicity against the Asian tiger mosquito larvae, depending on the Salvia species and were able to show a repellent activity against the Ae. albopictus adults, depending on the concentration, while not all of them were indicated by the panelists as suitable to be utilized in the formulation of products for environmental protection or personal protection.
In order to get insights about the molecular components of the tested EOs that may act as chemical cues for mosquitoes and the human sensorial system, their chemical composition was characterized by GC-EIMS. The EO profiles of the studied species were completely in agreement with what was previously reported on S. somalensis [33] and S. dorisiana [17]. The S. sclarea EO evidenced results in partial agreement with those observed by Aćimović and co-workers [34] on clary sage EO from Tajikistan. The S. dolomitica EO was a subject of discord, in fact, Kamatou et al. [35] and Bassolino et al. [36] found a composition similar to what we found in the current study, while in a more recent study by Ebani [33], eucalyptol was the main compound. Caser underlined a completely different profile where limonene (19.8%), δ-3-Carene (9.1%), and germacrene D (8.6%) dominated together with (E)-β-ocimene (7.39%) and β-caryophyllene (7.9%) [37].
Overall, the toxicity of S. dorisiana EO against the Ae. albopictus larvae recorded in this work (LC50 = 71.08 μL L −1 ) agrees with previous studies performed with other aromatic plants EOs (LC50 ranging from 35 to 194 μL L −1 ) (reviewed by Pavela [38]) and was shown to be significantly higher than the

Discussion
The use of EOs as insecticides and insect repellents is raising increasing interest for their low impact on environmental and human health, and for their perception by consumers as safe, natural products [32]. As for their use as repellents, however, in order for them to be successfully utilized as an ingredient in commercial products for topical or for environmental protection use, the EOs need to have a good level of acceptance to the human sensorial system. In this work, we observed that the Salvia EOs tested showed a wide range of toxicity against the Asian tiger mosquito larvae, depending on the Salvia species and were able to show a repellent activity against the Ae. albopictus adults, depending on the concentration, while not all of them were indicated by the panelists as suitable to be utilized in the formulation of products for environmental protection or personal protection.
In order to get insights about the molecular components of the tested EOs that may act as chemical cues for mosquitoes and the human sensorial system, their chemical composition was characterized by GC-EIMS. The EO profiles of the studied species were completely in agreement with what was previously reported on S. somalensis [33] and S. dorisiana [17]. The S. sclarea EO evidenced results in partial agreement with those observed by Aćimović and co-workers [34] on clary sage EO from Tajikistan. The S. dolomitica EO was a subject of discord, in fact, Kamatou et al. [35] and Bassolino et al. [36] found a composition similar to what we found in the current study, while in a more recent study by Ebani [33], eucalyptol was the main compound. Caser underlined a completely different profile where limonene (19.8%), δ-3-Carene (9.1%), and germacrene D (8.6%) dominated together with (E)-β-ocimene (7.39%) and β-caryophyllene (7.9%) [37].
Overall, the toxicity of S. dorisiana EO against the Ae. albopictus larvae recorded in this work (LC 50 = 71.08 µL L −1 ) agrees with previous studies performed with other aromatic plants EOs (LC 50 ranging from 35 to 194 µL L −1 ) (reviewed by Pavela [38]) and was shown to be significantly higher than the other Salvia spp. EOs. In line with our findings, a strong variability in the larval toxicity of Salvia spp. EOs was observed also by Ali [16] against the mosquitoes Anopheles quadrimaculatus and Aedes aegypti (Diptera: Culicidae). These results indicate a strong variability in the toxicity of the EOs among species of the genus Salvia that is consistent with the high variability in the chemical composition of the EOs.
Besides the high variability in the chemical composition of the EOs, a limitation in the use of the EOs as mosquito repellent is represented by their high volatility. This is why the determination of the protection time is an important parameter for the screening of EOs for possible practical use. In this experiment, S. dorisiana EO showed a very good persistency (CPT = 43 min at the dose of 20 nL cm −2 of skin) of the repellent effect that was longer than the one reported for the synthetic repellent N,N-Diethyl-meta-toluamide (DEET) (CPT = 10 min at the dose of 40 nL cm −2 of skin) [11]. Overall, S. dorisiana showed the highest toxicity against Ae. albopictus larvae coupled with the strongest protection effect against the Ae. albopictus bites that were about 60% longer than that of DEET. On the contrary, S. somalensis, was the least toxic and persistent EO among the Salvia species tested.
These results are in line with previous experiments by Conti et al. [17] who found that S. dorisiana and S. sclarea EOs were able to protect the human skin from Ae. albopictus bites up to 31 and 21 min (at 40 nL cm −2 ), respectively. The repellent effect showed by Salvia spp. EOs in this work is consistent with one of EOs extracted from aromatic plants belonging to other families. For example, the EOs extracted from Curcuma longa L. (Zingiberaceae), Pogostemon heyneanus Benth. (Lamiaceae), and Zanthoxylum limonella Alston (Rutaceae) were found to repel the Asian tiger mosquito up to 23 min at a dose of 5% (about 10 times more concentrated than the solutions tested in the present study) [39]. Moreover, Nasir [40] found that the Zingiber officinale Rosc. (Zingiberaceae) Mentha piperita L., and Ocimum basilicum L. (Lamiaceae) showed 34-98 min of protection (EOs, 10%).
The sensorial analyses of the Salvia spp. EOs indicated that, overall, the best smell profile (highest value of overall pleasantness) was the one of the S. sclarea EO, followed by the S. somalensis and S. dorisiana EOs. The preference showed by the assessors to the S. sclarea EO should be probably due to the green spicy, fresh mint, menthol-like and citrus-like fragrances detected in all the samples. Those fragrances may be associated with the presence of the alcoholic volatile compounds' linalool and p-cymen-8-ol (Table 1).
On the contrary, the worst smell profile attributed to S. dolomitica EO is probably due to the off-flavors detected by panelists in EO that can be well explained by the presence in volatile fraction of borneol (smell character: camphoraceous odor [41]) in association with β-caryophyllene (smell character: dry-woody-spicy, clove-like odor [41]) and β-pinene (smell character: dry-woody, resinous odor [41]).

Conclusions
The control of mosquitoes is paramount in the efforts of stopping the spread of mosquito-borne diseases. According to our results, Salvia species may represent a valid source of repellents and insecticides alternative to the synthetic ones for the control of mosquitoes. In fact, unlike synthetic chemicals, the low-cost and the increasing demand for effective and safe natural products may make EO-based products well accepted by consumers, provided that they are compatible with their sensory system. The multidisciplinary approach of this study showed that the selection of EOs as ingredients for effective and good-smelling anti-mosquito formulations is feasible by combining the data of the biological activity with their chemical and smell profiles.